chapter two of research methodology

Have a language expert improve your writing

Run a free plagiarism check in 10 minutes, generate accurate citations for free.

• Knowledge Base
• Dissertation
• What Is a Research Methodology? | Steps & Tips

What Is a Research Methodology? | Steps & Tips

Published on August 25, 2022 by Shona McCombes and Tegan George. Revised on June 22, 2023.

Your research methodology discusses and explains the data collection and analysis methods you used in your research. A key part of your thesis, dissertation , or research paper , the methodology chapter explains what you did and how you did it, allowing readers to evaluate the reliability and validity of your research and your dissertation topic .

It should include:

• The type of research you conducted
• How you collected and analyzed your data
• Any tools or materials you used in the research
• How you mitigated or avoided research biases
• Why you chose these methods
• Your methodology section should generally be written in the past tense .
• Academic style guides in your field may provide detailed guidelines on what to include for different types of studies.
• Your citation style might provide guidelines for your methodology section (e.g., an APA Style methods section ).

Table of contents

How to write a research methodology, why is a methods section important, step 1: explain your methodological approach, step 2: describe your data collection methods, step 3: describe your analysis method, step 4: evaluate and justify the methodological choices you made, tips for writing a strong methodology chapter, other interesting articles, frequently asked questions about methodology.

Here's why students love Scribbr's proofreading services

Discover proofreading & editing

Your methods section is your opportunity to share how you conducted your research and why you chose the methods you chose. It’s also the place to show that your research was rigorously conducted and can be replicated .

It gives your research legitimacy and situates it within your field, and also gives your readers a place to refer to if they have any questions or critiques in other sections.

You can start by introducing your overall approach to your research. You have two options here.

Option 1: Start with your “what”

What research problem or question did you investigate?

• Aim to describe the characteristics of something?
• Explore an under-researched topic?
• Establish a causal relationship?

And what type of data did you need to achieve this aim?

• Quantitative data , qualitative data , or a mix of both?
• Primary data collected yourself, or secondary data collected by someone else?
• Experimental data gathered by controlling and manipulating variables, or descriptive data gathered via observations?

Option 2: Start with your “why”

Depending on your discipline, you can also start with a discussion of the rationale and assumptions underpinning your methodology. In other words, why did you choose these methods for your study?

• Why is this the best way to answer your research question?
• Is this a standard methodology in your field, or does it require justification?
• Were there any ethical considerations involved in your choices?
• What are the criteria for validity and reliability in this type of research ? How did you prevent bias from affecting your data?

Once you have introduced your reader to your methodological approach, you should share full details about your data collection methods .

Quantitative methods

In order to be considered generalizable, you should describe quantitative research methods in enough detail for another researcher to replicate your study.

Here, explain how you operationalized your concepts and measured your variables. Discuss your sampling method or inclusion and exclusion criteria , as well as any tools, procedures, and materials you used to gather your data.

Surveys Describe where, when, and how the survey was conducted.

• How did you design the questionnaire?
• What form did your questions take (e.g., multiple choice, Likert scale )?
• Were your surveys conducted in-person or virtually?
• What sampling method did you use to select participants?
• What was your sample size and response rate?

Experiments Share full details of the tools, techniques, and procedures you used to conduct your experiment.

• How did you design the experiment ?
• How did you recruit participants?
• How did you manipulate and measure the variables ?
• What tools did you use?

Existing data Explain how you gathered and selected the material (such as datasets or archival data) that you used in your analysis.

• Where did you source the material?
• How was the data originally produced?
• What criteria did you use to select material (e.g., date range)?

The survey consisted of 5 multiple-choice questions and 10 questions measured on a 7-point Likert scale.

The goal was to collect survey responses from 350 customers visiting the fitness apparel company’s brick-and-mortar location in Boston on July 4–8, 2022, between 11:00 and 15:00.

Here, a customer was defined as a person who had purchased a product from the company on the day they took the survey. Participants were given 5 minutes to fill in the survey anonymously. In total, 408 customers responded, but not all surveys were fully completed. Due to this, 371 survey results were included in the analysis.

• Information bias
• Omitted variable bias
• Regression to the mean
• Survivorship bias
• Undercoverage bias
• Sampling bias

Qualitative methods

In qualitative research , methods are often more flexible and subjective. For this reason, it’s crucial to robustly explain the methodology choices you made.

Be sure to discuss the criteria you used to select your data, the context in which your research was conducted, and the role you played in collecting your data (e.g., were you an active participant, or a passive observer?)

Interviews or focus groups Describe where, when, and how the interviews were conducted.

• How did you find and select participants?
• How many participants took part?
• What form did the interviews take ( structured , semi-structured , or unstructured )?
• How long were the interviews?
• How were they recorded?

Participant observation Describe where, when, and how you conducted the observation or ethnography .

• What group or community did you observe? How long did you spend there?
• How did you gain access to this group? What role did you play in the community?
• How long did you spend conducting the research? Where was it located?
• How did you record your data (e.g., audiovisual recordings, note-taking)?

Existing data Explain how you selected case study materials for your analysis.

• What type of materials did you analyze?
• How did you select them?

In order to gain better insight into possibilities for future improvement of the fitness store’s product range, semi-structured interviews were conducted with 8 returning customers.

Here, a returning customer was defined as someone who usually bought products at least twice a week from the store.

Surveys were used to select participants. Interviews were conducted in a small office next to the cash register and lasted approximately 20 minutes each. Answers were recorded by note-taking, and seven interviews were also filmed with consent. One interviewee preferred not to be filmed.

• The Hawthorne effect
• Observer bias
• The placebo effect
• Response bias and Nonresponse bias
• The Pygmalion effect
• Recall bias
• Social desirability bias
• Self-selection bias

Mixed methods

Mixed methods research combines quantitative and qualitative approaches. If a standalone quantitative or qualitative study is insufficient to answer your research question, mixed methods may be a good fit for you.

Mixed methods are less common than standalone analyses, largely because they require a great deal of effort to pull off successfully. If you choose to pursue mixed methods, it’s especially important to robustly justify your methods.

Next, you should indicate how you processed and analyzed your data. Avoid going into too much detail: you should not start introducing or discussing any of your results at this stage.

In quantitative research , your analysis will be based on numbers. In your methods section, you can include:

• How you prepared the data before analyzing it (e.g., checking for missing data , removing outliers , transforming variables)
• Which software you used (e.g., SPSS, Stata or R)
• Which statistical tests you used (e.g., two-tailed t test , simple linear regression )

In qualitative research, your analysis will be based on language, images, and observations (often involving some form of textual analysis ).

Specific methods might include:

• Content analysis : Categorizing and discussing the meaning of words, phrases and sentences
• Thematic analysis : Coding and closely examining the data to identify broad themes and patterns
• Discourse analysis : Studying communication and meaning in relation to their social context

Mixed methods combine the above two research methods, integrating both qualitative and quantitative approaches into one coherent analytical process.

Above all, your methodology section should clearly make the case for why you chose the methods you did. This is especially true if you did not take the most standard approach to your topic. In this case, discuss why other methods were not suitable for your objectives, and show how this approach contributes new knowledge or understanding.

In any case, it should be overwhelmingly clear to your reader that you set yourself up for success in terms of your methodology’s design. Show how your methods should lead to results that are valid and reliable, while leaving the analysis of the meaning, importance, and relevance of your results for your discussion section .

• Quantitative: Lab-based experiments cannot always accurately simulate real-life situations and behaviors, but they are effective for testing causal relationships between variables .
• Qualitative: Unstructured interviews usually produce results that cannot be generalized beyond the sample group , but they provide a more in-depth understanding of participants’ perceptions, motivations, and emotions.
• Mixed methods: Despite issues systematically comparing differing types of data, a solely quantitative study would not sufficiently incorporate the lived experience of each participant, while a solely qualitative study would be insufficiently generalizable.

Remember that your aim is not just to describe your methods, but to show how and why you applied them. Again, it’s critical to demonstrate that your research was rigorously conducted and can be replicated.

1. Focus on your objectives and research questions

The methodology section should clearly show why your methods suit your objectives and convince the reader that you chose the best possible approach to answering your problem statement and research questions .

2. Cite relevant sources

Your methodology can be strengthened by referencing existing research in your field. This can help you to:

• Show that you followed established practice for your type of research
• Discuss how you decided on your approach by evaluating existing research
• Present a novel methodological approach to address a gap in the literature

3. Write for your audience

Consider how much information you need to give, and avoid getting too lengthy. If you are using methods that are standard for your discipline, you probably don’t need to give a lot of background or justification.

Regardless, your methodology should be a clear, well-structured text that makes an argument for your approach, not just a list of technical details and procedures.

If you want to know more about statistics , methodology , or research bias , make sure to check out some of our other articles with explanations and examples.

• Normal distribution
• Measures of central tendency
• Chi square tests
• Confidence interval
• Quartiles & Quantiles

Methodology

• Cluster sampling
• Stratified sampling
• Thematic analysis
• Cohort study
• Peer review
• Ethnography

Research bias

• Implicit bias
• Cognitive bias
• Conformity bias
• Hawthorne effect
• Availability heuristic
• Attrition bias

Methodology refers to the overarching strategy and rationale of your research project . It involves studying the methods used in your field and the theories or principles behind them, in order to develop an approach that matches your objectives.

Methods are the specific tools and procedures you use to collect and analyze data (for example, experiments, surveys , and statistical tests ).

In shorter scientific papers, where the aim is to report the findings of a specific study, you might simply describe what you did in a methods section .

In a longer or more complex research project, such as a thesis or dissertation , you will probably include a methodology section , where you explain your approach to answering the research questions and cite relevant sources to support your choice of methods.

In a scientific paper, the methodology always comes after the introduction and before the results , discussion and conclusion . The same basic structure also applies to a thesis, dissertation , or research proposal .

Depending on the length and type of document, you might also include a literature review or theoretical framework before the methodology.

Quantitative research deals with numbers and statistics, while qualitative research deals with words and meanings.

Quantitative methods allow you to systematically measure variables and test hypotheses . Qualitative methods allow you to explore concepts and experiences in more detail.

Reliability and validity are both about how well a method measures something:

• Reliability refers to the  consistency of a measure (whether the results can be reproduced under the same conditions).
• Validity   refers to the  accuracy of a measure (whether the results really do represent what they are supposed to measure).

If you are doing experimental research, you also have to consider the internal and external validity of your experiment.

A sample is a subset of individuals from a larger population . Sampling means selecting the group that you will actually collect data from in your research. For example, if you are researching the opinions of students in your university, you could survey a sample of 100 students.

In statistics, sampling allows you to test a hypothesis about the characteristics of a population.

Cite this Scribbr article

If you want to cite this source, you can copy and paste the citation or click the “Cite this Scribbr article” button to automatically add the citation to our free Citation Generator.

McCombes, S. & George, T. (2023, June 22). What Is a Research Methodology? | Steps & Tips. Scribbr. Retrieved November 1, 2023, from https://www.scribbr.com/dissertation/methodology/

Is this article helpful?

Shona McCombes

Other students also liked, what is a theoretical framework | guide to organizing, what is a research design | types, guide & examples, qualitative vs. quantitative research | differences, examples & methods, what is your plagiarism score.

Writing The Methodology Chapter

5 Time-Saving Tips & Tools

By: David Phair (PhD) and Amy Murdock (PhD) | July 2022

The methodology chapter is a crucial part of your dissertation or thesis – it’s where you provide context and justification for your study’s design. This in turn demonstrates your understanding of research theory, which is what earns you marks .

Over the years, we’ve helped thousands of students navigate this tricky section of the research process. In this post, we’ll share 5 time-saving tips to help you effectively write up your research methodology chapter .

Overview: Writing The Methodology Chapter

• Develop a (rough) outline before you start writing
• Draw inspiration from similar studies in your topic area
• Justify every research design choice that you make
• Err on the side of too much detail , rather than too little
• Back up every design choice by referencing literature

1. Develop an outline before you start writing 

The first thing to keep in mind when writing your methodology chapter (and the rest of your dissertation) is that it’s always a good idea to sketch out a rough outline of what you are going to write about before you start writing . This will ensure that you stay focused and have a clear structural logic – thereby making the writing process simpler and faster.

An easy method of finding a structure for this chapter is to use frameworks that already exist, such as Saunder’s “ research onion ” as an example. Alternatively, there are many free methodology chapter templates for you to use as a starting point, so don’t feel like you have to create a new one from scratch.

Next, you’ll want to consider what your research approach is , and how you can break it down from a top-down angle, i.e., from the philosophical down to the concrete/tactical level. For example, you’ll need to articulate the following:  

• Are you using a positivist , interpretivist , or pragmatist approach ?
• Are you using inductive or deductive reasoning?
• Are you using a qualitative , quantitative, or mixed methods study?

Keep these questions front of mind to ensure that you have a clear, well-aligned line of argument that will maintain your chapter’s internal and external consistency.

Remember, it’s okay if you feel overwhelmed when you first start the methodology chapter. Nobody is born with an innate knowledge of how to do this, so be prepared for the learning curve associated with new research projects. It’s no small task to write up a dissertation or thesis, so be kind to yourself!

2. Take inspiration from other studies 

Generally, there are plenty of existing journal articles that will share similar methodological approaches to your study. With any luck, there will also be existing dissertations and theses that adopt a similar methodological approach and topic. So, consider taking inspiration from these studies to help curate the contents of your methodology chapter.

Students often find it difficult to choose what content to include in the methodology chapter and what to leave for the appendix. By reviewing other studies with similar approaches, you will get a clearer sense of your discipline’s norms and characteristics . This will help you, especially in terms of deciding on the structure and depth of discussion.  

While you can draw inspiration from other studies, remember that it’s vital to pay close attention to your university’s specific guidelines, so you can anticipate departmental expectations of this section’s layout and content (and make it easier to work with your supervisor). Doing this is also a great way to figure out how in-depth your discussion should be. For example, word-count guidelines can help you decide whether to include or omit certain information.

Need a helping hand?

3. Justify every design choice you make

The golden rule of the methodology chapter is that you need to justify each and every design choice that you make, no matter how small or inconsequential it may seem. We often see that students merely state what they did instead of why they did what they did – and this costs them marks.

Keep in mind that you need to illustrate the strength of your study’s methodological foundation. By discussing the “what”, “why” and “how” of your choices, you demonstrate your understanding of research design and simultaneously justify the relevancy and efficacy of your methodology – both of which will earn you marks.

It’s never an easy task to conduct research. So, it’s seldom the case that you’ll be able to use the very best possible methodology for your research (e.g. due to time or budgetary constraints ). That’s okay – but make sure that you explain and justify your use of an alternate methodology to help justify your approach.

Ultimately, if you don’t justify and explain the logic behind each of your choices, your marker will have to assume that you simply didn’t know any better . So, make sure that you justify every choice, especially when it is a subpar choice (due to a practical constraint, for example). You can see an example of how this is done here.

4. Err on the side of too much detail

We often see a tendency in students to mistakenly give more of an overview of their methodology instead of a step-by-step breakdown . Since the methodology chapter needs to be detailed enough for another researcher to replicate your study, your chapter should be particularly granular in terms of detail. 

Whether you’re doing a qualitative or quantitative study, it’s crucial to convey rigor in your research. You can do this by being especially detailed when you discuss your data, so be absolutely clear about your:  

• Sampling strategy
• Data collection method(s)
• Data preparation
• Analysis technique(s)

As you will likely face an extensive period of editing at your supervisor/reviewer’s direction, you’ll make it much easier for yourself if you have more information than you’d need. Some supervisors expect extensive detail around a certain aspect of your dissertation (like your research philosophy), while others may not expect it at all.

Remember, it’s quicker and easier to remove/ trim down information than it is to add information after the fact, so take the time to show your supervisor that you know what you’re talking about (methodologically) and you’re doing your best to be rigorous in your research.

5. Provide citations to support each design choice

Related to the issue of poor justification (tip #3), it’s important include high-quality academic citations to support the justification of your design choices. In other words, it’s not enough to simply explain why you chose a specific approach – you need to support each justification with reference to academic material.  

Simply put, you should avoid thinking of your methodology chapter as a citation-less section in your dissertation. As with your literature review, your methods section must include citations for every decision you make, since you are building on prior research.  You must show that you are making decisions based on methods that are proven to be effective, and not just because you “feel” that they are effective.

When considering the source of your citations, you should stick to peer-reviewed academic papers and journals and avoid using websites or blog posts (like us, hehe). Doing this will demonstrate that you are familiar with the literature and that you are factoring in what credible academics have to say about your methodology.

As a final tip, it’s always a good idea to cite as you go . If you leave this for the end, then you’ll end up spending a lot of precious time retracing your steps to find your citations and risk losing track of them entirely. So, be proactive and drop in those citations as you write up . You’ll thank yourself later!

Let’s Recap…

In this post, we covered 5 time-saving tips for writing up the methodology chapter:

• Look at similar studies in your topic area
• Justify every design choice that you make
• Back up every design choice by referencing methodology literature

If you’ve got any questions relating to the methodology chapter, feel free to drop a comment below. Alternatively, if you’re interested in getting 1-on-1 help with your thesis or dissertation, be sure to check out our private coaching service .

Psst… there’s more (for free)

This post is part of our dissertation mini-course, which covers everything you need to get started with your dissertation, thesis or research project. 

You Might Also Like:

What data analysis method can work best for my study. I am using a mixed method in the study. I am developing a framework to address the challenges faced by the taxi operators as entrepreneurs. I will need to analyse both using qualitative and quantitative.

How to find standard deviation for non numerical data

Submit a Comment Cancel reply

Your email address will not be published. Required fields are marked *

Save my name, email, and website in this browser for the next time I comment.

• Print Friendly

Performance Specifications for Rapid Highway Renewal (2014)

Chapter: chapter 2 - research methodology.

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

12 C h a p t e r 2 Performance specifications emphasize desired outcomes and results, challenging owners and their industry partners to think in terms of user needs and to recognize that more than one solution may achieve the project objectives. Incorporating such concepts into a specification represents a distinct depar- ture from today’s build-to-print culture and demands a new approach to specification writing, contract administration, and construction execution. To help advance this new approach, the R07 research team developed guide specifications and associated implementation guidelines to support the application of performance speci- fications across a wide range of work and projects. In prepar- ing these documents, the team focused its research efforts on addressing the following fundamental questions: • What are performance specifications? • How are effective performance specifications developed and drafted? • Why use performance specifications? • What are the risks associated with using performance specifications? • When should performance specifications be used instead of method specifications? • Who is affected by the implementation of performance specifications and how are they affected? What are performance Specifications? Context drives how performance specifications are defined within the construction industry. For example, the U.S. Depart- ment of Defense (DoD) describes a performance specification as one that states requirements in terms of the required results and the criteria for verifying compliance, without specifically stating how the results are to be achieved. A performance specification describes the functional requirements for an item, its capa- bilities, the environment in which it must operate, and any interface, interoperability, or compatibility requirements. It does not present a preconceived solution to a requirement. (DoD 2009) In addition to addressing end-product performance, as contemplated by the DoD definition, requirements for a high- way construction project could conceivably extend to project- related performance in terms of work zone management, safety, and timely completion. Postconstruction and operational per- formance, as found in warranties and maintenance agreements, also could be included. The first task for the research team was therefore to con- duct a comprehensive literature review to establish what the term performance specifications encompasses when applied to the highway construction industry. Literature Review To provide focus to the literature review, the team first deter- mined which elements of a rapid renewal project would benefit from the development and implementation of performance specifications. Bearing in mind the objectives of rapid renewal (i.e., accelerated construction, minimal disruption, and long- life facilities), the team identified both physical products of con- struction (bridges, earthwork and geotechnical systems, and asphalt and concrete pavements) and project-level require- ments (work zone management, public relations, quality index- ing, and time incentives) as areas for possible application of performance specifications. To provide additional structure to the literature review, the team also established baseline definitions (presented in Appen- dix B) of specification types and contracting methods that would fall under the umbrella term performance specifications. As described in Chapter 1, performance specifications may be viewed in terms of a continuum. Categorizing specifications (e.g., as end-result specifications or PRS) helps identify the Research Methodology

13 advancement of performance specifications in a particular topic area. The literature review effort itself entailed collecting and reviewing reports, specifications, contract documents, and similar information to determine the status of performance specifying in each of the topic areas considered. The primary resources consulted included relevant FHWA, AASHTO, and NCHRP reports, as well as additional reports, contracts, and specifications from departments of transportation, industry, and international sources. Particular emphasis was placed on obtaining documents that addressed product performance measures, incentives, measurement and verification strategies, risk allocation techniques, legal and administrative issues, and other information relevant to the development and imple- mentation of performance specifications. Content Analysis The collected literature was classified according to specification type (e.g., end-result, PRS, warranty, and so on), topic area (e.g., pavement, bridge, work zone management, and so on), and project delivery approach (e.g., design-bid-build, design-build, design-build-operate-maintain). Then it was screened for per- ceived applicability to subsequent specification development efforts on the basis of containing or suggesting the following: • Progressive or creative performance parameters, measure- ment strategies, test methods (NDT or otherwise), or acceptance criteria appropriate to the rapid renewal environment; • Techniques to transfer performance responsibility from the owner to the contractor; • Actual or potential value of performance specifications; and • Conditions appropriate for the use of performance specifications. An annotated bibliography of documents is included in Appendix D. In addition, an index of existing performance specifications, collected as part of the literature review, is avail- able at the R07 report web page (http://www.trb.org/main/ blurbs/169107.aspx). how are effective performance Specifications Developed and Drafted? Historically, efforts at performance specifying (particularly in the pavement area) focused on the development and use of complex predictive models to establish specification require- ments. The research study undertaken for the R07 project adopted a more pragmatic approach that is amenable to, but not reliant on, the use of such models to define perfor- mance needs. The step-by-step process balances user needs and project goals against available technology and industry’s appetite for assuming performance risk, recognizing that such factors are often closely tied to the selected project delivery method. As illustrated by the suite of guide performance specifications prepared under this research study, the inherent flexibility of this approach makes it readily adaptable to different project elements and delivery methods. The complete performance specification development pro- cess is presented in the specification writers guide, Chapter 2. Agencies are encouraged to use the implementation guide- lines in conjunction with the guide specifications to tailor per formance requirements to project-specific conditions. Alter natively, agencies may develop additional performance specifications for needs not addressed by the current set of guide specifications. Specification Development Framework The primary function of a specification, whether prescriptive or performance oriented, is to communicate a project’s require- ments and the criteria by which the owner will verify confor- mance with the requirements. In this respect, performance specifications are similar to conventional method specifica- tions. Where they differ is the level at which performance must be defined. Figure 2.1—which was adapted from a model devel- oped by the Netherlands Ministry of Transport, Public Works, and Water Management—illustrates the possible requirement levels for a hypothetical pavement project (van der Zwan 2003). Taken as a whole, the pyramid depicted in the figure is intended to represent the entirety of knowledge and expe rience related to pavement design and construction. Taking and evaluating each level individually, the specifier can create a specification that is entirely prescriptive (if based solely on the material and workmanship properties defined on the lowest levels) or one that is more performance oriented (if based on the user needs and functional requirements described on the higher levels). For a particular project, the appropriate mix of perfor- mance requirements is driven by the project’s scope and objectives as well as the chosen project delivery approach and risk allo cation strategy. In practice, specifications typically include elements from several of the levels shown in Figure 2.1. Determining the appropriate balance between prescriptive and performance-oriented requirements is one of the main objectives of the eight-step specification development pro- cess illustrated Figure 2.2. Chapter 2 of the specification writ- ers guide describes this specification development framework in detail, systematically leading a specifier through each step in the process. However, as suggested by a review of the guide

14 specifications themselves, some steps are more critical to certain topic areas than to others. For example, although project deliv- ery approach (Step 3) plays a large role in shaping the develop- ment of a performance specification for pavements and bridges, it has less influence on establishing performance requirements for work zone management and geotechnical features. Application of the Performance Specification Framework To apply this framework to the main research areas of pave- ment, bridges, geotechnical systems, and work zone man- agement, the team first reviewed a cross section of existing performance specifications obtained through the literature review effort. Coordination with other SHRP and FHWA research projects provided additional information on topic areas that complemented the R07 effort to develop perfor- mance specifications for rapid renewal. The relevant projects addressed the following topics: • Advances in nondestructive testing techniques {e.g., SHRP 2 R06; FHWA Transportation Pooled Fund study [Project No. TPF-5(128)] on intelligent compaction}; • Innovative materials (e.g., SHRP 2 R19A); and • Mechanistic-based performance prediction (e.g., FHWA research study DTFH61-08-H-00005). The team carefully reviewed the collected literature, filter- ing existing performance specifications through the criteria established in the specification development framework to identify viable performance parameters and measurement strategies. Existing performance measures that met the frame- work criteria formed the basis for initial brainstorming ses- sions conducted among the team’s internal experts. Those existing measures, coupled with the team’s own project expe- rience, led to the development of draft performance require- ments which were then discussed and reviewed with external representatives from agencies and industry in formal work- shop settings. The input from external experts was used to refine and finalize the guide specifications and asso ciated commentary. Chapter 3 provides a more detailed summary of the findings from the literature review and outreach efforts in the context of the development of the guide specifications. To develop specifications that would be suitable for adop- tion by AASHTO, to the extent possible, the team adhered to the principles set forth in the National Highway Institute (NHI) Course No. 134001, Principles of Writing Highway Construction Specifications, and the FHWA Technical Advi- sory, Development and Review of Specifications (FHWA 2010). Even so, the team recognized that the typical AASHTO five- part format (Description, Materials, Construction, Measure- ment, Payment) may not be appropriate for every project delivery approach. For example, the lump-sum nature of a design-build contract may make measurement and payment sections unnecessary, whereas a warranty provision would require additional requirements related to bonding, distress evaluations, and required remedial action during the war- ranty period. Source: van der Zwan 2003 Figure 2.1. Pyramid of performance.

15 Figure 2.2. Performance specification development process.

16 Why Use performance Specifications? Successful implementation of performance specifications will likely require a departure from traditional project development and delivery processes. To gain support for necessary changes, best practice suggests first establishing a compelling business case as to why performance specifica- tions represent a desired addition to an agency’s contracting toolbox. Literature Review To establish the rationale for using performance specifica- tions, the team first performed a literature review to docu- ment any prior efforts to identify the actual or potential value received through the use of performance-based, incentive- based, and performance warranty contracts and specifica- tions in the highway construction industry. Recognizing that performance specifications have not been widely applied to transportation projects in the United States, the team expanded its literature search to include research and practice from outside the highway industry. For example, the use of performance-based service contracts has become a standard business practice for some federal agencies, such as the Department of Defense (DoD), and the benefits of these contracts have been validated by research studies and best practice guides (OFPP 1998a; OFPP 1998b; DoD 2000). Although the benefits may not directly translate to the value added or lost by applying performance specifications to a highway construction project, they do provide general insight into the advantages of using performance contracting strategies. Value Assessment Research and practice, particularly from outside the highway industry, suggest that implementing performance specifica- tions has the potential to provide several advantages, includ- ing decreased life-cycle costs, reduced inspection, and improved quality and customer satisfaction. However, the literature contains little data quantifying the actual value added or lost by implementing performance specifications. Despite the lack of quantitative data, the literature does reflect the perception that using performance specifications or a performance contracting system will result in enhanced value (or performance) for highway agencies and road users. The literature also makes evident that these enhancements can be attributed, at least in part, to alternative project delivery systems that provide more flexibility and shift more responsibility to the private sector to achieve perfor- mance goals. Comparative Framework The team felt it was necessary to develop a comparative struc- ture to assess performance specifications against a benchmark. That comparison would allow consideration of how project delivery approaches could affect the actual or potential value received from implementing performance specifications. On the basis of the literature review and consultation with subject matter experts, the team generated a list of viable delivery schemes for performance specifications. The results of that effort led the team to use the following delivery meth- ods as the basis for assessing the perceived value of perfor- mance specifications: • Prescriptive (method) specifications (benchmark); • Design-bid-build, with some performance requirements, but no warranty (DBB+P); • Design-bid-build, with short-term warranty (DBB+STW); • Design-build, with no warranty (DB); • Design-build, with short-term warranty (DB+STW); and • Design-build-maintain (DBM). Recognizing that project conditions could also affect the value received from performance specifications, the compar- ative framework considered the impact of different project characteristics such as the following: • Road class (local, state highway, interstate, toll); • Type of construction (preservation, reconstruction, new construction); • Traffic [low, moderate, or high annual average daily traffic (AADT)]; • Location (urban, rural); • Complexity (depending on project phasing, right-of-way requirements, utilities, environmental issues, etc.); and • Climate (depending on moisture and temperature, by region). In the context of these delivery approaches and project char- acteristics, the team turned to expert participation in surveys and workshops to assess the perceived value of using perfor- mance specifications. Such nonexperimental research tech- niques were found to be applicable given the nature of the study. Factors such as delivery methods and project character- istics can be shown to affect the perceived value placed on the implementation of performance specifications on highway construction projects. However, the effect or extent of the rela- tionship cannot be determined with precision, as any one of the other factors can lead to the same or a similar effect. There- fore, the team relied on nonexperimental techniques, includ- ing surveys and documentation of experts’ comments elicited in a workshop setting, as means for data collection.

17 Delphi Analysis Although the survey method is a detailed and systematic method of data collection, response rates can be poor and the participating experts can leave out vital information. To bol- ster this technique, the team applied the Delphi method. The Delphi method is an adaptation of the survey method and is used to obtain the judgment of a panel of experts on a complex issue or topic. It is a systematic method of data col- lection and structured discussion that aims to minimize the effects of bias given the characteristic lack of anonymity in interviews and general surveys. The method is particularly useful in situations in which empirical means are not suitable and research results rely heavily on the subjective opinions of experts. In brief, a Delphi analysis entails an iterative process in which experts’ opinions are processed and used as feedback for further refinement of opinions generated in earlier sur- vey rounds. The iterative nature of the process is expected to yield more reliable results than a single survey round. The Delphi analysis required the team to (1) assemble the Delphi expert group; (2) develop and administer survey questions; (3) receive and process the survey responses; (4) conduct a structured workshop to present, discuss, and clarify the sur- vey results; (5) conduct a second survey round assessment; (6) summarize the outcomes of the Round 2 assessment, and (7) conduct a Round 3 assessment and summarize results. Appendix E provides a detailed summary of the design and results of this data collection effort. The Delphi survey results are provided in Appendix G. Demonstration Projects Perhaps the most powerful way to identify and communicate the potential benefits of performance specifications is through demonstration projects. The SHRP 2 R07 project therefore included an implementation phase designed to validate the guidelines and performance specifications developed during the research effort. The first step toward this end was to identify candidate agen- cies that would be willing to participate in a demonstration project. A survey questionnaire was developed to gauge the interest and experience of a representative sample of highway agencies in the United States, particularly those known to have experience or interest in performance specifications or alter- native project delivery methods. The survey included a brief description of the R07 project, including the project objectives and scope of the demonstration program. The survey document further explained that the team was seeking to work with two or more transportation agencies in implementing performance specifications on demonstra- tion projects to test and validate the use of performance specifications for rapid highway renewal projects. The R07 team offered to provide resources to work with agency per- sonnel to select an appropriate project or projects, develop the necessary performance specifications and contracting pro- visions, and assist with the administration of the project dur- ing design and construction, and, if applicable, during the maintenance and operation phase. Most important, the survey sought information as to (1) the likelihood that the agency would have projects suit- able for a demonstration of performance specifications in the 2010–2011 construction seasons and (2) the areas for which the agency would be most interested in performance specifying. Ten agencies returned questionnaires or sent e-mail responses indicating interest in participating in a SHRP 2 R07 demonstra- tion project. From those responses, the team identified the following projects as viable candidates for demonstrations: • Virginia DOT Route 208 Lake Anna Bridge Deck Rehabilita- tion Project—a shadow demonstration of the use of perfor- mance parameters that related more to long-term durability and performance; • Missouri DOT Route 141 Roadway Improvement Project— a demonstration of the use of nondestructive roller-integrated compaction monitoring (RICM), or intelligent compaction, to provide real-time and improved quality control of soil compaction operations; and • Louisiana DOTD U.S. Frontage Roads—a demonstration of the use of RICM and mechanistic-based in situ point measurements on a new pavement section. A more detailed discussion of these demonstrations is pro- vided in Chapter 3. What are the risks associated with Implementing performance Specifications? Risk in the context of performance specifications relates to the existence of any uncertain event or condition that, if it occurs, has a positive or negative effect on the objectives of the specification. The FHWA’s Guide to Risk Assessment and Allocation for Highway Construction Management presents a continuous, cyclical approach to risk assessment, involving the following steps: identify, assess and analyze, mitigate and plan, allocate, and monitor and update (Ashley et al. 2006). A similar approach was used to address the risks associated with performance specifications. As a key component of the specification development frame- work, the discussion of risk related to performance specifica- tions (i.e., identification and evaluation of risks) is addressed in the specification writers guide. The entire process of developing

18 a performance specification is in a sense a risk management exercise designed to identify, allocate, and mitigate the risks associated with implementing a performance specification. The generally accepted approach to project-level risk management—as described in FHWA’s Guide to Risk Assess- ment (Ashley et al. 2006) or the SHRP 2 R09 Guide for the Process of Managing Risk on Rapid Renewal Projects (Golder Associates et al. 2013)—is useful in developing a general frame- work for identifying risks; it is less useful in terms of analysis (e.g., quantifying the frequency and impacts of specification- related risks). In some cases the specification risks, such as gaps in performance measurement, are difficult to quantify given the current state of the practice (or level of understanding). For example, given the interest in the use of NDT and mechanistic properties for performance measurement, further research is needed to quantify the effects of risk related to variability or reliability of NDT versus traditional tests, or opportunities related to the use of mechanistic versus traditional perfor- mance measures. Additional long-term data collection will be needed to make valid quantitative risk assessments. The risk process described in the specification writers guide is geared to identifying risks and gaps and qualitatively deciding whether performance specifications are appropri- ate. Further, the guide assists in determining how to develop a performance specification to allocate risk among the proj- ect participants considering the current state of the practice. Further assessment of performance specification risks are needed to quantify the impacts or opportunities related to their use. When Should performance Specifications Be Used? Performance specifications are not ideal for every construc- tion contract or project circumstance. However, they may hold significant advantages over traditional method specifi- cations when certain criteria or conditions are met. To inte- grate performance specifications into an agency’s contracting toolbox, a process is needed to evaluate when to use or not to use performance specifications. The decision to use method or performance specifications is often a matter of degree (how much and at what level). Both approaches may be appropriate for specific elements within a project. In choosing the appropriate level of perfor- mance specifications, an organization’s culture, statutory restrictions, project objectives and characteristics, project delivery approach, and risk appetite all may play important parts in defining specifications. The interaction among these key factors will likely determine the preference for one type of specification over the other. The decision to use performance specifications versus method specifications can involve a relatively straightforward screening test, followed by a more in-depth analysis of the level and type of performance specifying appropriate for the project characteristics and contracting type. Thus, the imple- mentation guidelines (see the executive guide, Chapter 5) present a two-part decision process for evaluating when to use or not to use performance specifications. Part 1 of this decision process considers a project’s scope and goals. Part 2 addresses the project delivery considerations that could also affect the decision. Who Is affected by performance Specifications and how are they affected? For agency personnel, developing and implementing a scope of work in terms of user needs and end-result performance is often much more challenging and resource intensive than simply adhering to the agency’s standard specifications. For contractors, an initial investment may be needed to acquire the necessary knowledge, skills, and equipment to assume more responsibility for performance. While critical to a project’s success, a well-drafted perfor- mance specification will not in and of itself ensure that an agency’s performance goals will be met. Cultural, organiza- tional, and legal issues can also affect the successful imple- mentation of performance specifications. For this reason, the team prepared a set of implementation guidelines to accom- pany the guide specifications. In doing so, the team reviewed the existing literature, had discussions with practitioners from agencies and industry, and identified lessons learned from the demonstration projects. The goal was to address the following considerations: • The effect the decision to use performance specifications could have on an agency’s traditional project delivery phases, from project planning and preliminary engineering through to construction completion and possibly beyond to mainte- nance and asset management; • Any natural progression or transition from more traditional contracts and specifications that should precede the deci- sion to use performance specifications (i.e., a learning curve to attune both the agency and industry to a new business model); and the • General mechanics of administering performance con- tracts (e.g., procurement process, document and database management, and so on). This information, along with the key takeaways drawn from the other research tasks, was incorporated into both the imple- mentation guidelines and the guide specifications, as applica- ble, to provide agencies with the tools needed to develop and successfully implement performance specifications.

TRB’s second Strategic Highway Research Program (SHRP 2) Report S2-R07-RR-1: Performance Specifications for Rapid Highway Renewal describes suggested performance specifications for different application areas and delivery methods that users may tailor to address rapid highway renewal project-specific goals and conditions.

SHRP 2 Renewal Project R07 also produced:

• Implementation Guidelines: Volume I: Strategies for Implementing Performance Specifications: A Guide for Executives and Project Managers , which is designed to provide a broad overview of the benefits and challenges associated with implementing performance specifications.
• Implementation Guidelines: Volume II: Developing and Drafting Effective Performance Specifications: A Guide for Specification Writers , which presents a flexible framework that specifiers may use to assess whether performance specifying represents a viable option for a particular project or project element. If it is indeed a viable option, the guide discusses how performance specifications may then be developed and used to achieve project-specific goals and satisfy user needs.
• A pilot study , in partnership with the Missouri Department of Transportation, to investigate the effectiveness of selected quality assurance/quality control testing technologies.

A separate document, Guide Performance Specifications , includes model specifications and commentary to address implementation and performance targets (for acceptance) for 13 routine highway items. Agencies may adapt guide specifications to specific standards or project conditions. The commentary addresses gaps, risks, and options.

Welcome to OpenBook!

You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

Do you want to take a quick tour of the OpenBook's features?

Show this book's table of contents , where you can jump to any chapter by name.

...or use these buttons to go back to the previous chapter or skip to the next one.

Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

To search the entire text of this book, type in your search term here and press Enter .

Share a link to this book page on your preferred social network or via email.

View our suggested citation for this chapter.

Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

Get Email Updates

Do you enjoy reading reports from the Academies online for free ? Sign up for email notifications and we'll let you know about new publications in your areas of interest when they're released.

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

• We're Hiring!
• Help Center

CHAPTER TWO RESEARCH METHODOLOGY

Related Papers

International Journal of Accountancy (IJA) ISSN 2738-2249

Sandaya Kumari Jayasena

Abstract This study aims to determine the Impact of Board Structure on the Performance of listed Commercial Banks on the Colombo Stock Exchange. This study examines the relationship between Board Structure and Corporate Financial Performance of listed commercial banks using a sample of 12 banks listed on the Colombo Stock Exchange, based on data from 2015 to 2019. Descriptive, correlation and regression analyses were performed to achieve the study objectives. This study uses the company's performance based on the following financial performance measures: return on assets (ROA) and return on equity (ROE) and board structure consisting of board size, female board members, non-executive directors, and frequency of board meetings. Bank capital is used as the control variable in this study. The study's findings suggest that the frequency of board meetings and non-executive directors have a negative and substantial influence on ROE, while the frequency of board meetings has a significant and negative impact on ROA. Additionally, bank capital has a positive and significant impact on ROA. However, the board size, female board members, and bank capital have no significant impact on ROE while board size, non-executive directors, and female board members have no significant impact on ROA of CSE-listed commercial banks. This study contributes to the literature on corporate governance and firm performance by providing the framework that affects the relationship between board structure and firm performance in the context of listed commercial banks. The findings of the study are important for policymakers, investors, regulators, and other bankers of the country. Commercial banks listed in Colombo Stock Exchange would put more attention on the structure and quality of the board to improve their performance. Concerning the study's scope, the present study only included CSE-listed commercial banks. Thus, additional research is needed to study the entire Sri Lankan banking system and 20 business sectors. Further, this study only looked at accounting-based measures such as ROA and ROE. Tobin's Q, for example, can be employed as a dependent variable. Future studies should be done by including cross-border analysis of developing and developed countries. Keywords: Bank Capital, Board Size, Female Board Members, Frequency of Board Meetings, Non-Executive Directors ROA, ROE

Abstract Globalization has resulted in complex business procedures and activities as a result of global organizations. In the case of banking sector procedures, complicated banking transactions and an increase in banking activity fraud have prompted a focus on the Banking Sector's Internal Control System. Every organization relies on an effective Internal Control System. Accordingly, the purpose of this research is to examine and compare the level of effective internal control systems of state-owned and private domestic commercial banks in the Anuradhapura district. The level of effectiveness of the Internal Control System was assessed using the five components of the Committee of Sponsoring Organizations (COSO) framework (Control Environment, Risk Assessment, Information and Communication System, Control Activities, and Monitoring Activities). Structured questionnaires were used to collect data from fifty respondents from state-owned and thirty respondents from private domestic commercial banks, respectively. Managers (76.25 percent response rate) and secondary officers were the most common respondents (response rate was 23.75 percent). Descriptive analysis with one-sample t-test and independent sample t-test were used to analyze the data. The results of the study indicated that the level of the effective control environment, information and communication systems, and control activities are different between StateOwned and Private Domestic Commercial Banks in Anuradhapura district, whereas the level of effective risk assessment and monitoring activities are not different between State-Owned and Private Domestic Commercial Banks in Anuradhapura district. The findings of this study will help practitioners as a guideline to get the maximum benefit of their Internal Control System. Keywords: COSO Framework, Internal Control System, Level of Effective Internal Control System, Private Domestic Commercial Banks, State-Owned Commercial Banks

International Journal of Science and Research (IJSR)

STEPHEN I K E C H U K W U AGU

This study set out to ascertain the application of forensic accounting as a means of bridging audit expectation gap in Nigerian Deposit Money banks in Enugu State. A survey research design approach was adopted for the study. The population of this study comprises of two thousand eight hundred and sixty-five (2,865) accountants in the selected thirteen (13) deposit money banks domicile and operational in Enugu State, Nigeria. The sample size of three hundred and fifty-one (351) accountants/respondents was drawn using Taro Yamane sampling method but by aid of simple stratified sampling technique, twenty-seven (27) respondents which represents stratum from each bank was obtained from the core-study departments-auditing, treasury and accounting departments of thirteen (13) selected banks domicile and operational in Enugu state, Nigeria. Data were analyzed using the Pearson Product Moment Correlation Coefficient. The results showed that there is significant relationship between litigation support services and expectation gap among Nigeria Deposit Money Banks. The findings further revealed that there is significant relationship between forensic investigation and expectation gap among Nigeria Deposit Money Banks. Finally, the result established also that there is significant relationship between forensic audit and expectation gap among Nigeria Deposit Money Bank. Consequent upon the findings, this study recommends among others; that all organizations must be made to subject their transactions to forensic audit and evidence before the statutory auditor. Again, that the management of the banking industry should rely and consult more on forensic engagement as a way of bridging expectation gap.

Joshua Dickson

Research Journal of Finance and Accounting

Omoneye Olasanmi

ASHRAF Alhabsi

Digital Transformation in Times of Crisis Proceedings

International Conference on ICT in Our Lives

Managerial Auditing Journal

ROGER DEBRECENY

BENEDICT NOR

This research seeks to evaluate the effectiveness of internal control system in the private sector. Internal control is one of the vital tools used to control finances, safe guide assets and to checkmate incident of fraud, irregularities in every organization. Internal control can be liken to a network of pillars that holds a skyscraper without which the skyscraper will collapse like a park of card. It is in the light of foregoing that the researcher put up a study of its effectiveness in private companies using Nugget Hotel Limited, Abuja as a case study. The study was carried out through personal interviews, questionnaire and tabular analysis of findings. It became obvious that internal control system is not so effective in most private companies. hence useful recommendation were proffered which if utilized by the private company in this category will make internal control effective and guarantee the survival and profitability of private companies.

st monica university ,bae cameroun

MONDAY R I C H M O N D EFUT

ABSTRACT The purpose of this study was to investigate the impact of forensic education in strengthening the internal control mechanisms of the Ministry of Education Headquarters; survey research design was employed in gathering data and two hypotheses were formulated and tested; results of the hypotheses tested indicate that forensic education has significant impact in the detection and prevention of fraud in the Federal Ministry of Education and therefore can strengthen the internal control mechanisms of the Federal Ministry of Education Headquarters. The study therefore recommended the use of forensic education techniques as a panacea for the detection and prevention of fraud in the Federal Ministry of Education Headquarters.

RELATED PAPERS

Adeyeye Debam

The International Journal of Management

BARNABAS O AGOCHUKWU, BSc, MSc, PhD, FCA, CNA

KIU Interdisciplinary Journal of Humanities and Social Sciences

Babatunde Lawal

IAEME PUBLICATION

IAEME Publication

Future Business Journal

Osariemen Asiriuwa

Ojo A D U R A G B E M I Sunday

Barnabas Agochukwu

JASH MATHEW , MAURICE MUKINGINYI WEKESA

mehdi bouchetara

IOSR Journals publish within 3 days

Georgios L Vousinas

International Journal of Shari'ah and Corporate Governance Research

Fuad Salleh

Roselyn Haruna

Udisifan M I C H A E L Tanko

IOSR Journals

TAREMWA ABRAHAM NICHOLAS

African Journal of Accounting and Financial Research

Edward I D E M U D I A Agboare ACA

Boris Kuete

Dr Mohamed Zakari

Bonface Maina

ERN: Other Emerging Markets Economics: Firm Behavior & Microeconomic Issues (Topic)

Morteza Ramazani

EUROPEAN RESEARCH STUDIES JOURNAL

tulus suryanto

Journal of International Accounting, Auditing and Taxation

Mohammad Hudaib

•   We're Hiring!
•   Help Center
• Find new research papers in:
• Health Sciences
• Earth Sciences
• Cognitive Science
• Mathematics
• Computer Science
• Academia ©2023

• Financial Aid
• Campus Life
• About UNI
• Support UNI
• UNI Bookstore

• Rod Library

Research Methods

Chapter 1: introduction.

Whether you are studying communication, sociology, literature, history, psychology, music, biology, or any other major, that academic field relies on standardized practices to produce scholarly knowledge.  Scholarship  can be in the form of highly controlled laboratory research, observation of human activities in daily life, surveys, interviews, critical analyses of public documents or visual images, and creative work like music, videography, performance, or playwriting. Each field of scholarship is based on thousands, if not millions, of research studies or creative projects conducted by students and faculty. Sociologists know what they know about societies because of research. Biologists know what they know about the biological function of organisms because of research. Artists know what they know about drawing human forms because of previous artists' work. Communication Studies scholars know what they know about how people construct meaning through interactions because of research and creative projects. The overall purpose of this book is to help you understand  how  knowledge is constructed in Communication Studies. We hope to provide an appreciation of, and critical lens for examining, research and enable you to begin constructing your own contributions to the body of scholarly knowledge.

In this chapter, we first describe how developing a command of research methods can assist you in your careers and personal lives. Second, we provide a brief definition of our topic of study in this book – communication research. Third we identify the predominant research and creative methods used in the field of Communication Studies. Fourth, we explain the academic roots of the diverse methods used in communication studies: the humanities and social sciences. Fifth, we explain the implicit and explicit relationships between theory and research methods. Sixth, we describe how the choice of research methods influences the results of a study. Sixth, we provide a preview of the remainder of the book, and finally, seventh, we describe our approach to writing this book.

How Will Research Methods Help in My Life?

If you want to learn practical skills relevant to your professional, personal and community life, learn research methods. Given that daily life is full of decision-making opportunities and challenges, knowing how to effectively do research is essential. Ideally, any decision you make is based on research, and rigorous methods enable you to conduct better research and make better decisions. People who know how to ethically use research methods quickly become leaders in their workplaces and communities. Research also can inform creative expression. If you understand why things work the way they do, you can make more thoughtful, creative choices.

Consider how you make choices in everyday life such as the following:

• Which route to take to get to class on time
• What to eat for lunch
• How to make a major purchasing decision

Or, how you address more complex questions such as whether dishonesty is ever warranted, or if there is a God?

Brainstorm all the ways in which you think you know something for one or more of the examples listed above.

If you are like previous students in this course, you may have responded: "read," "observe," "intuit," "faith," "advice," "physical senses," "test it out," "compare," "Google it" and more.

What does this activity reveal about how you come to know something?

We hope the activity above reveals you already are a researcher, and use some informal research methods every day of your life. You likely use more than one way to know something. Multiple methods construct knowledge. And being educated includes questioning the results of each method. For example, if you use  Google  or  Wikipedia  to find information, how do you know the source is reliable? What clues should you look for?

Research methods will help you be a better ....

• Critical Consumer  — You will find you look at the world of information through a more refined lens. You may ask questions about information you never thought to ask before, such as: "What evidence is this conclusion based on?" "Why did the researcher interview rather than survey a larger number of people?" and "Would the results have been different if the participants were more ethnically and racially diverse?
• Competent Contributor  — When an organization you belong to wants to attain a group's input on a program, product or service, you will know how to construct, administer and statistically analyze survey results. Or if the project warrants small focus group discussions for information gathering, you will know how to facilitate them as well as how to identify themes from the discussions.
• Problem-Solver  — Research methods skills are nearly synonymous with problem- solving skills. You will learn how to synthesize information, assess a current state of knowledge, think creatively, and make a plan of action for original research gathering and application.
• Strategic Planner – Knowing research methods can teach you how to gather the necessary information to forecast and plan tactically rather than only react to situations, whether it is in your work place or personal life.
• Decision-Maker  — As you cross through life transitions and major decisions stare you in the face, such as how to keep a job, give the best care for aging parents, or select the least invasive medical treatment, you will have coping skills to help you break down the decision into manageable parts and approach the decision making process from more than one perspective.
• Informed Citizen  — As a person educated in how knowledge is constructed, you will have the skills needed to be vigilant for your community and to identify and address potential problems, be they environmental, political, social, educational, and/or about quality of community life.

For more specific ideas about how a command of research methods can broaden your life options, see the examples of practical research at  Communication Currents: Knowledge for Communicating Well . It is a reader-friendly magazine where communication scholars discuss research about current social problems ( The National Communication Association - Communication Currents ). Also check out the National Communication Association website  http://www.natcom.org/  for careers in communication.

The Topic of Study: Communication Research

You may have noticed we, the authors, use the singular form of the term communication to refer to the academic field of study on a wide variety of message types, rather than the plural form: communications. The distinction is a quick way to tell who understands communication is one specific field of study and who does not, so you will want to use the proper, singular form when referring to the field of study. Communications – plural is used only when referring to multiple media sources, as in "the communications news media" (Korn, Morreale, & Boileau, 2000).

The forms communication can take are nearly endless. They include, but are not limited to: language, nonverbal communication, one-on-one interpersonal communication, organizational communication, film, oral interpretations of prose or poetry, theatre, public speeches, public events, political campaigns, public relations campaigns, news media, Internet, social media, photography, television, social movements, performance in everyday life, journalistic writing, and more. Yet, the theme that runs through almost all Communication Studies research is that communication is more than a means to transmit information. Although it is used to transmit information and get things done, more importantly, communication is the means through which people make meaning and come to understand each other and the world.

Because of this, communication scholars tend to operate with the assumption that reality is a social construction, constructed through human beings' use of communication, both verbal and visual (Gergen, 1994). Thus, when communication scholars conduct research, they ask questions not only about how to make communication more precise and/or effective, but they also ask questions about how communication is being used in a particular context to shape individuals' and groups' world views.

Research, as a form of communication, contributes to the social construction of knowledge. Knowledge does not come out of a vacuum that is free of cultural values. Instead, research results, or what society calls knowledge, is influenced by the values, beliefs, methods choices, and interpretations of those in a given culture doing the research. Knowledge and one's reality are constructed through an interactive, interpretive process. Although scholars from a more traditional natural science view might argue there are absolute truths and set realities, in the study of human interaction, there are few universal truths about communication and what is seen as knowledge changes across cultures and over time. Unique cultural contexts, social roles, and inequities create a wide spectrum of behaviors ( Kim ,  http://plato.stanford.edu/entries/weber/ ). That is what makes miscommunication common and why research in our field is so much in demand. It is highly practical and relevant work.

 Research  refers to the systematic study of a topic and can include social science and creative work. Research, quite simply, refers to people's intellectual work of gathering, organizing, and analyzing data, which enables them to create meaning they can then present to others. Research is conducted to answer questions or solve problems in a systematic way. Being  systematic  means that the steps of the study are guided by principles and theory, rather than just chaotic wandering; the data used is representative and not just anecdotal or random. Being systematic in a way that can be replicated is usually emphasized more in natural and social science research, such as organizational and interpersonal communication research, than the humanities and fine arts, such as rhetorical studies and performance studies, but, rhetoric scholars and artists also rely on methods and theoretical training to guide their work.

Communication Studies research has several unique characteristics:

• Communication research is the study of how people make meaning.  If one thinks of communication as the process of making meaning, then the study of communication is the study of this meaning making process.
• Communication research is the study of patterns  (Keyton, 2011). Communication and meaning are made possible through the creation of patterns. For example, languages are rule-based and construct recognizable patterns (such as sentences). Conversations have social norms of politeness to enable participants to build on each party's turns at talk; social media have unique patterns of interaction (such as the abbreviations used in text messaging on cell phones or the emoticons used in e-mail and social networks); and persuasive messages are built on patterns of communication strategies (such as advertisements showing sequences of visual appeals for destitute children to solicit donations).
• Communication research is practical knowledge construction.  The field of communication is highly applied. Scholars and practitioners try to do work that matters. Work that improves the quality of people's lives, that solves problems, and that is needed. Research in the field is pragmatic. Film makers tell a story that they believe needs to be told, performance studies students create interactive scenarios to draw the audience into needed cultural discussion, and public relations practitioners conduct market analyses as a basis for planning a client's communication strategies.
• A ll research builds an argument. Whether it is a creative, rhetorical, qualitative or quantitative project, the author necessarily has a point to make. The introductory rationale for a project, the choices the researcher makes in methods selection, the interpretations offered, and the significance she or he claims for the results are all a part of building an argument. If all knowledge is socially constructed, then all research or scholarship is a persuasive process.

Whether one is doing Creative, qualitative, rhetorical or quantitative work, the methods share the above characteristics, as they are inherent in the very communication process being studied.

The term  method  refers to the processes that govern scholarly and creative work. Methods provide a framework for collecting, organizing, analyzing and presenting data. Scholars use a range of methods in Communication Studies: quantitative, qualitative, critical/rhetorical, and creative. This text focuses on the first three, but the authors note connections to creative work when relevant.

Quantitative Studies  reduce data into measurable numerical units (quantities). An example would be a survey administered to determine the number of times first-year college students use social networking sites and for what purposes. Such a survey could provide general statistics on frequency and purpose of use. But, such a study also could be set up to determine if first and fourth year college students use social networks differently, or if students with smart phones spend more time on Facebook than students who rely on computers to check Facebook.

Qualitative Studies  use more natural observations and interviews as data. An example would be a study about a workplace organization's leadership and communication patterns. A researcher could interview all the members of the business, and then also observe the members in action in their place of work. The researcher would then analyze the data to see if themes emerge, and if the interview and observational data results are similar. The researcher might then propose changes to the organization to enhance communication and performance for the organization.

Critical/Rhetorical Studies  focus on texts as sources for data. The term  texts  is used loosely here to refer to any communication artifact --films, speeches, historical monuments, news stories, letters, tattoos, photos, etc. Here, the data collected is the text, and it is used by the researcher to support an argument about how the text participates in the construction of people's understanding of the world. An example would be an analysis of a presidential inaugural address to understand how the speech writers and speaker are attempting to reunite the nation after a hotly contested election and invest the new president with the powers of the office.

Creative Scholarship  in the field of Communication Studies refers most often to work done in performance studies, film making, and computer digital imagery, such as Dreamweaver and Photoshop (e.g. Camp Multimedia Begins Two-Week Run ).  Performance Studies  is a wide umbrella term used to refer to several methods and products of scholarship. It is distinct from theater in that it is the study of performances in everyday life. It involves students in script writing, acting, and directing productions based upon oral history and ethnographic qualitative research, as well as personal experience and creative performance techniques used to tell a story more evocatively.  Film Making  can also include interviews, oral histories, and ethnographies, as well as learning aesthetic methods to effectively present verbal and visual images. Our colleague Karen Mitchell has used qualitative methods of interviewing and ethnography to script performances on topics from the lives of undocumented immigrants in the U.S. to romance novel readers (1996).

Communication Studies Bridges the Humanities and Social Sciences

How is it that Communication Studies as an academic field came to embrace so many different methods, given most other disciplines tend to use only one, or maybe two? The answer lies in the history of the field.

Communication Studies is different from other academic fields because it is rooted in one of the oldest areas of scholarship (rhetoric is one of the original four liberal arts) and in several of the newest areas of scholarship (such as electronic media and intercultural communication). The study of rhetoric dates back to 350 B.C.E, the time of Aristotle and the formation of democratic governance in Greece. The study of intercultural communication dates back to the 1940s and emerged out of the commerce and political needs in the U.S. after World War II (Leeds-Hurwitz, 1990). The study of the Internet took shape in the 1990s as it became a popular medium for communication (Campbell, Martin & Fabos, 2010). Because the discipline of Communication Studies includes research on all forms of communication, the method of study needs to fit as the form of communication. However, just because new forms emerge (like social media), old forms (like public speaking) do not disappear. Thus, as students, future practitioners and scholars, we need to employ a wide range of scholarly approaches. (for more about the history of the field see:  Communication Scholarship and the Humanities ) .

The diverse origins of Communication Studies mean its scholars use a range of methods from the humanities (e.g., rhetorical criticism and performance) and the social sciences (e.g., quantitative and qualitative research). Both focus on the study of society, but the humanities embrace a more holistic approach to knowledge and creativity. The  humanities  are those fields of study that focus on analytic and interpretive studies of human stories, ideas and words (rather than numbers), and include philosophy, English, religion, modern and classical languages, and Communication Studies. When Communication Studies scholars analyze how communicative acts (like speeches or photographs or letters) create social meaning, they do so from a perspective that emphasizes interpretation.

The social sciences use research methods borrowed from previously established and recognized fields of natural science study, such as biology and chemistry.  Scientific methods  of knowledge construction are accomplished through controlled observation and measurement or laboratory experiments, and generally use statistics to form conclusions (Kim, 2007). The  social sciences  apply scientific methods to study human behavior, for example scholars use surveys to find out about people's communication patterns or create laboratory experiments to observe interruption patterns in conversation. In addition to Communication Studies, examples of other social science fields include economics, geography, psychology, sociology, and political science.

Studying human communication from the perspective of the social sciences differs in important ways from studying human communication from the perspective of the humanities. Social scientists typically are interested in studying shared everyday life experiences, such as turn-taking norms in conversation, how people build relationships through self-disclosure, and what behaviors contribute to a successful group, family or organizational culture. Social science researchers attempt to find generalizations about human behaviors based on extensive research that may be used to make predictions about that behavior. Take, for instance, research on communication in heterosexual married couples. Based on over twenty years of research, psychologist John Gottman found in 1994 he could predict with 94% accuracy which marriages will fail based on patterns of only five negative conflict behaviors among couples who ended in divorce(for updates on his work visit his website ( Research FAQs ). (Of course exceptions exist to generalizations, but for a social scientist, the exception to the rule may be ignored as an insignificant outliers, a random error.

Instead of seeking out generalizations about communication, scholars in the humanities tend to focus on the outliers, or what are considered distinctive human creations, such as Abraham Lincoln's "Gettysburg Address," Elizabeth Cady Stanton's "Solitude of Self," Shakespeare's  Romeo and Juliet , Lorraine Hansberry's  Raisin in the Sun , Beethoven's Fifth Symphony, or Ani diFranco's "Dilate." Humanities scholars tend to focus on understanding  how  something happened or how someone attempted to evoke meanings and aesthetic reactions in the receivers of a message, rather than describing what occurred and predicting what will occur.

As an example of how diverse methods have been used to research a topic, consider how researchers who want to try to reduce intimate partner violence have approached the problem drawing on methods from across fields of study.

Quantitative researchers administered the National Survey on Violence Against Women and found 1.5 million women are physically or sexually assaulted by their domestic partners annually in the U.S. (Tjaden & Thoennes, 2000). The survey identified the difficult reality about the enormous extent of the problem. Qualitative researcher, Loren Olson (2010), wrote an autoethnography of her personal experience as a battered woman. By doing so she put a face on the problem and demonstrated a way in which she was able to reconstruct her identity after the abuse.

Performance studies scholar M. Heather Carver and ethnographic folklorist Elaine Lawless (2009) conducted a qualitative study with women who are surviving intimate violence and generated a creative performance script from their observations. The theatrical performance developed with creative methods literally help to give voice to the experiences of the women in the qualitative study, raises awareness about the problem, and may motivate audience members to address the problem in their personal or community lives.

Researchers also have critically analyzed the way domestic violence is communicated in various media. For example, Cathy Ferrand Bullock (2008) studied media framing in domestic violence news stories in Utah newspapers and rhetoric scholar Nathan Stormer (2003) studied the play,  A Jury of Her Peers , to explore how collective memory is formed about acts of domestic violence. These samples of research into the complex social problem of domestic violence demonstrate how both humanities and social science approaches to scholarship are needed and valued. Because the social sciences and humanities provide different contributions to the construction of knowledge, together they create a fuller picture of a social problem or issue of study.

Given its multi-methodological research, Communication Studies is uniquely positioned to contribute to both of the two most prominent approaches to knowledge construction: humanistic and social scientific approaches. This is why, as the authors of this book, we believe Communication Studies provides a well-rounded education to prepare students to respond to the challenges and opportunities of the culturally, technologically, and economically complex 21st century.

The Interdependence of Theory and Research Methods

Whether you approach a topic of study from a humanistic or social science perspective, you will necessarily work with two-components: theory (explanations that guide or evolve from a study) and methods (application of tools to analyze texts or data). Even though the two serve distinct research functions, there is great interdependence between theory and method. Theory informs methods, and methods enable theory construction and revision.

At its essence, a  theory  is simply a person's attempt to explain or understand something. Individuals use theories to help make sense of their world and everyday lives. An academic theory is different from everyday theories only in the degree of rigor and research used to develop it and the depth of explanation it provides. Academic theories are more formal, with detailed explanations of the parts that make up the theory, and are usually tested (West & Turner, 2010). But as with theories for everyday life, they are subject to change and refinement. DeFrancisco and Palczewski (2007) emphasize, "A theory is not an absolute truth, but an argument to see, order, and explain the world in a particular way" (p. 27). For any topic of study, multiple theories could explain it, and research can be used to determine which theory offers the best explanation. Communication theories tend to focus on helping explain how and why people interact as they do in interpersonal relationships, small groups, organizations, cultures, nations, publics, and mediated contexts. Theories can help people understand their own and others' communication.

When you make decisions in daily life, you probably use an informal theory. You might collect some data (or try and recall what information you have), you might discard data that comes from non-credible sources, and then you might assess your options. You will likely make your assessment based on hunches or underlying assumptions you have about what makes sense. Those hunches or assumptions are a lay person's theory. They help you make sense of things and inform your decisions.

Activity Consider the following questions to determine if you use theories in your daily life:

• What is your advice for how to live on a college student's budget?
• Do you think advertising influences your purchasing decisions? If so, how?
• What is your approach to making a good first impression on a person to whom you are attracted?
• How do you know someone is attracted to you?
• Why do you think people tend to avoid relationships with others they perceive as different from them?

If you have ideas on the above topics, you are a theorist.

Now ask yourself: what do your answers to the specific questions consist of?

• Are they attempts to explain a phenomenon?
• How did you form the explanations?
• Are they based on prior experience, advice from others, and/or informal research?

Likely your answers are a little of each.

A further question to ask yourself is:

• Are other explanations possible besides the ones you developed?

Students have developed more than one way to survive on a college student's budget. For one thing, not all college students are living on a tight budget, many will survive through student loans and jobs, others may get allowances from their parents, have spouses or partners who are supporting them, etc. Some will delay gratification of purchases such as cars, I-Pads, smartphones, spring break trips, and more. Others may argue, "You only live once," and use credit cards to charge for their pleasures or life necessities. The point is people develop multiple theories for any topic of interest, and many are useful.

People develop theories through testing, academic debates, and scholarly/creative work. Natural science and social science researchers, in particular, believe that the best research is directed or driven by academic theory. This means the research methods chosen are not random but are firmly based in a credible theoretical approach that has been tested over time.

Theories often guide research. When studying presidential campaigns for example, scholars often use Thomas Burke's theories on how speakers create identification to explore the ways in which candidates create connections with their audience (Burke, 2002).

Sometimes the research will extend or challenge the legitimacy of the theory. For example, intercultural communication scholar, William Gudykunst extended Berger and Calabrese's (1975) assumed universal Uncertainty Reduction Theory (URT) regarding what people do to reduce uncertainty anxiety when communicating with strangers. During 30 years of research, Gudykunst tested URT in cross cultural interactions and developed a new intercultural theory, Anxiety/Uncertainty Management (AUM) with 47 axioms or specific distinctions that help explain the universal and cultural variances he found (2005). Contrary to the original URT, Gudykunst now proclaims cultures vary in terms of comfort with uncertainty and the methods they use to manage it. These cultural differences contribute to unique cultural identities and help explain communication problems with other groups.

In the field of gender studies in communication there are countless examples of research that has disproven the commonly held theoretical assumption that universal gender differences exist between all women and men (e.g. Tannen, 1990; or in the popular press:  Men are from Mars and Women are from Venice  (Gray, 1992). In fact, communication scholars Kathryn Dindia and Dan Canary (2006) published a series of quantitative  meta-analyses  (a statistical way to control for differences across studies to directly compare the results) on just about every presumed communication difference previous researchers have studied. What did they find? While some differences were present, the variances  among  women's behaviors and among men's behaviors were greater than those  between  the sexes, and furthermore, women and men communicate in many more ways that are similar rather than different. Finally, they found that the assumption of two distinct sets of behavior is far too simplistic. It ignores the fact that people have the ability to adjust their behaviors according to situational needs and that gender identity does not affect one's behavior alone. It is also influenced by one's race, ethnic, age, nationality, sexual orientation and more.

A useful way to think of the relationship between theory and scholarly/creative work is that it is synergistic – each influences the other, almost simultaneously. As the illustration below shows, the theories selected direct the types of  research questions  posed to guide a study, the questions dictate the appropriate research methods needed, which then affect the results produced, which in turn contributes to theory building, thus the cycle repeats.

The General Research Process:  Circular and Interdependent

The diagram is circular rather than hierarchal because the starting point for different types of research will vary. For example, qualitative work begins with research questions with an end goal of producing theory, whereas quantitative work often begins with theory with an end goal of producing results. Rhetorical research and creative scholarship do not typically use research questions but the research process is still synergistic, and decisions made at each part of the cycle influences the others. The parts are interdependent. The circular model also reminds one that the process of theory construction, conducting research, and producing knowledge are never ending.

Knowledge  generally refers to a command of facts, theory and practical information. There is not one agreed upon approach for constructing knowledge as is illustrated in the above discussion of diverse research methods. Indeed, there is an entire field of philosophy,  epistemology , which focuses on debates about how knowledge is attained. Epistemologists ask "how does one know something?" Is knowledge found or created? These are questions we encourage you to ask as you learn about the various research methods. The methods researchers use to construct knowledge are generally called  methodology . The term simply means an approach being used to form knowledge is assumed to have both a theory and a method. Here again the interdependence between theory and method are evident.

Finally, throughout the research process the ability to think critically is essential. To be critical means to examine material in more depth, to peel back layers of meaning, to look beyond chunks of information to the context in which the information is presented, to look for multiple interpretations, to attempt to identify why a piece of information or perspective is important and/or not important. It requires doing a close reading or investigation of the topic of study in a more nuanced, systematic way. It does not mean to always be negative, but rather to question even common assumptions.

Research Methods Influence Results

The research methods one chooses for a study are critical. The methods will largely determine the results or what is called knowledge. The influence of methods choices is more visible when comparing social science and humanities approaches to the construction of knowledge, as will be discussed in chapter 2. The two are designed to answer different types of research questions. Together, they will offer you a wealth of methods choices.

For example, consider the relatively simple task of measuring the floor area of a room. We assigned small groups of students to measure the square footage in a room. Each group was provided different measurement tools. One group used a tape measure 12 feet long, another used a tape measure 40 feet long, another group used their own feet, and another used a metric tape measure. As you can imagine, the groups' results differed every time. Some used feet rather than inches to calculate square feet, some did not measure the same exact places in the room, metric measurements produced different results than the U.S. measuring system, and human feet produced varying results. The point here is not that one method was superior to another or that the groups made errors. The point is that even a slight change in methods can produce significant changes in results (Turman, personal communication, January 27, 2010). (If you would like to see more on metrics versus U.S. units conversion, see for example,  Metric to U.S. units conversion .

If diverse results can be produced when measuring the floor area of a room, imagine how different research methods may influence the study of processes as complex as human communication. Leslie Baxter has studied interpersonal relationship development and maintenance for nearly 20 years. Most of her early research was based on quantitative surveys of romantic partners in an attempt to identify the specific tensions or stresses in their relationship. By using standardized surveys she was able to identify three dominant tensions most couples struggled with: connection/independence, openness/closedness, and predictability/spontaneity. From this she developed what is now a well known theory in the field, Dialectic Tensions Theory. However, more recently she revised her theory based on qualitative studies of relational partners' conversations. Baxter now argues that by examining tensions in actual discourse rather than surveys, she is not only able to identify common tensions, but move beyond identification to see why some relationships successfully negotiate the tensions and why others do not (2011). We offer this example not to argue qualitative methods are superior to quantitative ones, but simply to make the point that the two serve different functions.

As you will learn in the coming chapters, each method used to collect data carries with it a different implied theory about how knowledge should be, or is formed. When researchers use surveys they value the ability to solicit a larger number of people and make generalizations from the responses. When researchers analyze conversations or use interviews, they value the ability to probe individual perspectives in more depth and are less concerned with generalizations. As teachers, scholars and practitioners, the authors of this book believe a command of research methods is central to developing one's unique expertise.

Preview of Chapters

In this book, three general research approaches are included: quantitative social science research methods, qualitative social science research methods, and critical rhetorical research methods. This does not mean these three are the only approaches to knowledge construction used in the field of Communication Studies or that they are necessarily independent or opposite of each other. Communication Studies is a wonderfully diverse field of study. In addition to rhetorical methods, other humanities scholarship include performance studies and film making. Because of the extreme interdisciplinary nature of film-making and performance studies, no one research method or chapter is dedicated to them. Instead we integrate examples throughout the collection, and readers should keep in mind how such work pushes the boundaries of traditional academic fields. Below are summaries/previews of the remaining chapters in this book.

Chapter One Summary : In the present chapter, we overviewed the interdisciplinary nature of the field of Communication Studies and demonstrated how this provides a broader choice of research methods for students and faculty members in the field. We introduced basic concepts necessary to have a foundation for the study of research methods. Even though scholars use diverse research methods in the field, they are built on common premises. One is that knowledge is constructed. The way it is constructed is influenced by the theoretical approaches used and the related research methods chosen. Understanding these fundamental relationships will help students be more informed critical consumers and contributors to the field of Communication Studies, their chosen professions, and society.

Chapter Two: General Comparisons . In chapter two, we offer basic points of comparison for the research methods taught in this book. This comparison should help provide a structure to understand how the diverse methods are distinct from each other before you are introduced to the specifics of conducting research in each method in subsequent chapters. The comparison is based on the two general orientations to knowledge construction introduced in chapter one: humanistic and social scientific.

Chapter Three: Ethical Research, Writing, and Creative Work . In this chapter, we discuss the importance of researcher ethics. This chapter is placed at the front of the book to stress this importance. Regardless of the method chosen, researchers have ethical choices to make in writing honestly, citing other sources, and treating human subjects fairly. Good research is, at its core, based on ethical principles.

Chapter Four: Quantitative Methods . The first research approach presented is quantitative research methods from the social sciences. The rules involved in doing quantitative methods are very clear, with a linear research process. Reading this chapter will teach you how to plan and conduct a quantitative study, and make sense of your findings once you have collected your data.

Chapter Five: Qualitative Methods . Qualitative research methods can be placed in the middle of a continuum of research methods from the scientific to the humanistic. Qualitative methods are usually considered to be a social science approach, but in more recent years researchers have been pushing these boundaries to embrace multiple ways of knowing.

Chapter Six: Critical/Rhetorical Methods . The core assumption of rhetorical criticism is that symbolic action (the use of words, images, stories, and argument) are more than a means to transmit information, but actually construct social reality, or people's understanding of the world. Learning methods of rhetorical criticism enable you to critique the use of symbolic action and understand how it constructs a particular understanding of the world by framing a concept in one way rather than another. The more adept you become at analyzing others' messages, the more skilled you become at constructing your own.

Chapter Seven: Presenting Your Results . This chapter teaches you how to present the results of your study, regardless of the choice made among the three methods. Writing in academics has a basic form and style that you will want to learn not only to report your own research, but also to enhance your skills at reading original research published in academic journals. Beyond the basic academic style of report writing, there are specific, often unwritten assumptions about how quantitative, qualitative, and rhetorical studies should be organized and the information they should contain. In this chapter students will learn about the functions of each part of a report (e.g. introduction, methods and data description, and critical conclusion) and find useful criteria to help guide the writing of each part in a research report.

Approach to Writing this Resource Book

When the faculty in the UNI Department of Communication Studies decided to make research methods a required course for all students majoring in the department (starting Fall, 2010), we searched for a textbook that equitably covered methods used in the humanities  and  social sciences. We could not find one, so we decided to write our own. This resource book is the product of a collaborative effort by faculty in the department. Although five of us wrote and organized the chapters, everyone in the department was invited to contribute ideas and examples.

The result is not a traditional textbook. For one thing, rather than one voice, the authors hope you will hear their distinct voices in each chapter influenced, in part, by the methods chosen and the values these methods reflect. The differing styles should help prepare you for the differing writing styles you will find when you read original research in journals that feature quantitative, qualitative, or critical/rhetorical studies. Consequently, the citation systems we use to document sources differ across chapters. In chapters on social science research (quantitative and qualitative research methods) we use the American Psychological Association (APA) (2010) style, because it is the format of choice for most journals publishing social science research. In the chapters on ethics and rhetorical methods we use the format prescribed by the Modern Language Association (MLA) (2009) because rhetoric is rooted in the Humanities, and rhetorical research often is published in journals that also include scholarship from performance studies, English literature and the fine arts. As one reads the coming chapters, it can be insightful to attempt to identify how the methods and values are reflected in the writing styles.

Another distinction is that because the text is digital, rather than paper, we are able to make the book more interactive, including additional websites and other resources to hopefully help make the methods come alive. Perhaps most importantly for you as a student, using a digital delivery system means far less expense. The digital delivery system also means we have the ability to update material continuously.

Through this book, we hope you will become excited by the possibilities of participating in the construction of knowledge in Communication Studies. We also hope to help demystify the research process and reveal underlying assumptions of each process. Contrary to what some public figures, educators and media sources would have the public believe, most knowledge is not absolute. We invite your critical voice to this learning process.

American Psychological Association. (2010).  Publication manual of the American Psychological Association  (6th ed.). Washington, DC: American Psychological Association.

Baxter, L. (2011).  Voicing relationships: A dialogic perspective . Thousand Oaks, CA: Sage Publications, Inc.

Berger, C. R., & Calabrese, R. (1975). Some explorations in initial interactions and beyond: Toward a development theory of interpersonal communication.  Human Communication Research , 1, 99-112.

Bullock, C. F. (2008). Official sources dominate domestic violence reporting.  Newspaper Research Journal , 29(2), 6-22.

Burke, K. (1966).  Language as symbolic action: Essays on life, literature, and method . Berkeley, CA: University of California Press.

Burke, T. (2002).  Lawyers, lawsuits and legal rights: The battle over litigation in American society . Berkeley, CA: University of California.

Campbell, R., Martin, C. R., & Fabos, B. (2010).  Media & culture: An introduction to mass communication  (8th ed.). Boston, MA: Bedford/St. Martin's.

Carver, M. H., & Lawless, E. J. (2009).  Troubling violence: A performance project . Jackson, MI: University of Mississippi Press.

DeFrancisco, V. P., & Palczewski, C. H. (2007).  Communicating gender diversity: A critical approach . Thousand Oaks, CA: Sage.

Dindia, K., & Canary, D. J. (Eds.). (2006).  Sex differences and similarities in communication  (2nd edition). Mahwah, NJ: Erlbaum.

Dues, M., & Brown, M. (2004).  Boxing Plato' s shadow: An introduction to the study of human communication . Boston, MA: McGraw Hill.

Gergen, K. J. (1994).  Realities and relationships: Soundings in social construction . Cambridge, MA: Harvard University Press.

Gray, J. (1992).  Men are from Mars, women are from Venus: A practical guide for improving communication and getting what you want in your relationship . New York: HarperCollins.

Gottman, J. M. (1994).  What predicts divorce? The relationship between marital processes and marital outcomes . Hillsdale, NJ: Erlbaum.

Gudykunst, W. B. (2005). An anxiety/uncertainty management (AUM) theory of effective communication: Making the mesh of the net finer. In W. B. Gudykunst (Ed.),  Theorizing about intercultural communication  (pp. 281-322). Thousand Oaks, CA: Sage.

Keyton, J. (2011).  Communicating research: Asking questions, finding answers  (3rd ed). New York: McGraw Hill.

Kim, B. (n.d.) Social constructivism. In M. Orey (Ed.),  Emerging perspectives on learning, teaching, and technology . Department of Educational Psychology and Instructional Technology, University of Georgia. Retrieved from  http://projects.coe.uga.edu/epltt/index.php?title=Social_Constructivism

Kim, S. H. (2007). Max Weber.  Stanford encyclopedia of philosophy . Retrieved from  http://plato.stanford.edu/entries/weber/

Korn, C.J., Morreale, S.P., and Boileau, D.M. (2000). Defining the field: Revisiting the ACA 1995 definition of communication studies.  Journal of the Association for Communication Administration , 29, 40-52.

Leeds-Hurwitz, W. (1990). Notes in the history of intercultural communication: The foreign service institute and the mandate for intercultural training.  Quarterly Journal of Speech , 76, 262-281.

Merriam-Webster's Word of the Year 2006. (2006). Merriam-Webster Online. Retrieved from  http://www.merriam-webster.com/info/06words.htm

Mitchell, K. S. (1996). Ever after: Reading the women who read (and re-write) romance.  Theatre Topics 6.1  (1996) 51-69

The Modern Language Association. (2009).  MLA handbook for writers of research papers  (7th ed.). New York: The Modern Language Association of America.

National Communication Association. (2007). Communication scholarship and the humanities: A white paper sponsored by the National Communication Association. Retrieved from  http://www.natcom.org/uploadedFiles/Resources_For/Policy_Makers/PDF-Communication_Scholarship_and_the_Humanities_A_White_Paper_by_NCA.pdf

Olson, L. N. (2010). The role of voice in the (re)construction of a battered woman's identity: An autoethnography of one woman's experiences of abuse. Women's Studies in Communication 27(1), 1-33. DOI: 10.1080/0749/409-2004.10162464

Stormer, N. (2003). To remember, to act, to forget: Tracing collective remembrance through "A Jury of Her Peers".  Communication Studies , 54(4), 510-529.

Sunstein, C. (2001).  Republic.com . Princeton, NJ: Princeton University Press.

Tannen, D. (1990).  You just don't understand: Women and men in conversation . New York: William Morrow.

Tjaden, P. & Thoennes, N. (2000, July). Extent, nature and consequences of intimate partner violence: Findings from the National Violence Against Women Survey. (NCJ 181867). Washington, DC: National Institute of Justice/Centers for Disease Control and Prevention. Retrieved from  http://www.ncjrs.gov/pdffiles1/nij/181867.pdf

Walker, L. (1979).  The battered woman . New York: Harper & Row Publishers.

West, R., & Turner, L. H. (2010).  Introducing communication theory: Analysis and application . Boston, MA: McGraw-Hill.

Research Methodologies

• What are research designs?
• What are research methodologies?

What are research methods?

Quantitative research methods, qualitative research methods, mixed method approach, selecting the best research method.

• Additional Sources

Research methods are different from research methodologies because they are the ways in which you will collect the data for your research project.  The best method for your project largely depends on your topic, the type of data you will need, and the people or items from which you will be collecting data.  The following boxes below contain a list of quantitative, qualitative, and mixed research methods.

• Closed-ended questionnaires/survey: These types of questionnaires or surveys are like "multiple choice" tests, where participants must select from a list of premade answers.  According to the content of the question, they must select the one that they agree with the most.  This approach is the simplest form of quantitative research because the data is easy to combine and quantify.
• Structured interviews: These are a common research method in market research because the data can be quantified.  They are strictly designed for little "wiggle room" in the interview process so that the data will not be skewed.  You can conduct structured interviews in-person, online, or over the phone (Dawson, 2019).

Constructing Questionnaires

When constructing your questions for a survey or questionnaire, there are things you can do to ensure that your questions are accurate and easy to understand (Dawson, 2019):

• Keep the questions brief and simple.
• Eliminate any potential bias from your questions.  Make sure that they do not word things in a way that favor one perspective over another.
• If your topic is very sensitive, you may want to ask indirect questions rather than direct ones.  This prevents participants from being intimidated and becoming unwilling to share their true responses.
• If you are using a closed-ended question, try to offer every possible answer that a participant could give to that question.
• Do not ask questions that assume something of the participant.  The question "How often do you exercise?" assumes that the participant exercises (when they may not), so you would want to include a question that asks if they exercise at all before asking them how often.
• Try and keep the questionnaire as short as possible.  The longer a questionnaire takes, the more likely the participant will not complete it or get too tired to put truthful answers.
• Promise confidentiality to your participants at the beginning of the questionnaire.

Quantitative Research Measures

When you are considering a quantitative approach to your research, you need to identify why types of measures you will use in your study.  This will determine what type of numbers you will be using to collect your data.  There are four levels of measurement:

• Nominal: These are numbers where the order of the numbers do not matter.  They aim to identify separate information.  One example is collecting zip codes from research participants.  The order of the numbers does not matter, but the series of numbers in each zip code indicate different information (Adamson and Prion, 2013).
• Ordinal: Also known as rankings because the order of these numbers matter.  This is when items are given a specific rank according to specific criteria.  A common example of ordinal measurements include ranking-based questionnaires, where participants are asked to rank items from least favorite to most favorite.  Another common example is a pain scale, where a patient is asked to rank their pain on a scale from 1 to 10 (Adamson and Prion, 2013).
• Interval: This is when the data are ordered and the distance between the numbers matters to the researcher (Adamson and Prion, 2013).  The distance between each number is the same.  An example of interval data is test grades.
• Ratio: This is when the data are ordered and have a consistent distance between numbers, but has a "zero point."  This means that there could be a measurement of zero of whatever you are measuring in your study (Adamson and Prion, 2013).  An example of ratio data is measuring the height of something because the "zero point" remains constant in all measurements.  The height of something could also be zero.

Focus Groups

This is when a select group of people gather to talk about a particular topic.  They can also be called discussion groups or group interviews (Dawson, 2019).  They are usually lead by a moderator  to help guide the discussion and ask certain questions.  It is critical that a moderator allows everyone in the group to get a chance to speak so that no one dominates the discussion.  The data that are gathered from focus groups tend to be thoughts, opinions, and perspectives about an issue.

Advantages of Focus Groups

• Only requires one meeting to get different types of responses.
• Less researcher bias due to participants being able to speak openly.
• Helps participants overcome insecurities or fears about a topic.
• The researcher can also consider the impact of participant interaction.

Disadvantages of Focus Groups

• Participants may feel uncomfortable to speak in front of an audience, especially if the topic is sensitive or controversial.
• Since participation is voluntary, not every participant may contribute equally to the discussion.
• Participants may impact what others say or think.
• A researcher may feel intimidated by running a focus group on their own.
• A researcher may need extra funds/resources to provide a safe space to host the focus group.
• Because the data is collective, it may be difficult to determine a participant's individual thoughts about the research topic.

Observation

There are two ways to conduct research observations:

• Direct Observation: The researcher observes a participant in an environment.  The researcher often takes notes or uses technology to gather data, such as a voice recorder or video camera.  The researcher does not interact or interfere with the participants.  This approach is often used in psychology and health studies (Dawson, 2019).
• Participant Observation:  The researcher interacts directly with the participants to get a better understanding of the research topic.  This is a common research method when trying to understand another culture or community.  It is important to decide if you will conduct a covert (participants do not know they are part of the research) or overt (participants know the researcher is observing them) observation because it can be unethical in some situations (Dawson, 2019).

Open-Ended Questionnaires

These types of questionnaires are the opposite of "multiple choice" questionnaires because the answer boxes are left open for the participant to complete.  This means that participants can write short or extended answers to the questions.  Upon gathering the responses, researchers will often "quantify" the data by organizing the responses into different categories.  This can be time consuming because the researcher needs to read all responses carefully.

Semi-structured Interviews

This is the most common type of interview where researchers aim to get specific information so they can compare it to other interview data.  This requires asking the same questions for each interview, but keeping their responses flexible.  This means including follow-up questions if a subject answers a certain way.  Interview schedules are commonly used to aid the interviewers, which list topics or questions that will be discussed at each interview (Dawson, 2019).

Theoretical Analysis

Often used for nonhuman research, theoretical analysis is a qualitative approach where the researcher applies a theoretical framework to analyze something about their topic.  A theoretical framework gives the researcher a specific "lens" to view the topic and think about it critically. it also serves as context to guide the entire study.  This is a popular research method for analyzing works of literature, films, and other forms of media.  You can implement more than one theoretical framework with this method, as many theories complement one another.

Common theoretical frameworks for qualitative research are (Grant and Osanloo, 2014):

• Behavioral theory
• Change theory
• Cognitive theory
• Content analysis
• Cross-sectional analysis
• Developmental theory
• Feminist theory
• Gender theory
• Marxist theory
• Queer theory
• Systems theory
• Transformational theory

Unstructured Interviews

These are in-depth interviews where the researcher tries to understand an interviewee's perspective on a situation or issue.  They are sometimes called life history interviews.  It is important not to bombard the interviewee with too many questions so they can freely disclose their thoughts (Dawson, 2019).

• Open-ended and closed-ended questionnaires: This approach means implementing elements of both questionnaire types into your data collection.  Participants may answer some questions with premade answers and write their own answers to other questions.  The advantage to this method is that you benefit from both types of data collection to get a broader understanding of you participants.  However, you must think carefully about how you will analyze this data to arrive at a conclusion.

Other mixed method approaches that incorporate quantitative and qualitative research methods depend heavily on the research topic.  It is strongly recommended that you collaborate with your academic advisor before finalizing a mixed method approach.

How do you determine which research method would be best for your proposal?  This heavily depends on your research objective.  According to Dawson (2019), there are several questions to ask yourself when determining the best research method for your project:

• Are you good with numbers and mathematics?
• Would you be interested in conducting interviews with human subjects?
• Would you enjoy creating a questionnaire for participants to complete?
• Do you prefer written communication or face-to-face interaction?
• What skills or experiences do you have that might help you with your research?  Do you have any experiences from past research projects that can help with this one?
• How much time do you have to complete the research?  Some methods take longer to collect data than others.
• What is your budget?  Do you have adequate funding to conduct the research in the method you  want?
• How much data do you need?  Some research topics need only a small amount of data while others may need significantly larger amounts.
• What is the purpose of your research? This can provide a good indicator as to what research method will be most appropriate.
• << Previous: What are research methodologies?
• Next: Additional Sources >>
• Last Updated: Aug 2, 2022 2:36 PM
• URL: https://library.tiffin.edu/researchmethodologies

• Key Differences

Know the Differences & Comparisons

Difference Between Research Method and Research Methodology

Last updated on January 25, 2018 by Surbhi S

The research methods are often confused with research methodology , which implies the scientific analysis of the research methods, so as to find a solution to the problem at hand. Hence, it seems apt to clarify the differences between research method and research methodology at this juncture, have a look.

Content: Research Method Vs Research Methodology

Comparison chart, definition of research method.

Research method pertains to all those methods, which a researcher employs to undertake research process, to solve the given problem. The techniques and procedure, that are applied during the course of studying research problem are known as the research method. It encompasses both qualitative and quantitative method of performing research operations, such as survey, case study, interview, questionnaire, observation, etc.

These are the approaches, which help in collecting data and conducting research, in order to achieve specific objectives such as theory testing or development. All the instruments and behaviour, used at various levels of the research activity such as making observations, data collection, data processing, drawing inferences, decision making, etc. are included in it. Research methods are put into three categories:

• First Category : The methods relating to data collection are covered. Such methods are used when the existing data is not sufficient, to reach the solution.
• Second Category: Incorporates the processes of analysing data, i.e. to identify patterns and establish a relationship between data and unknowns.
• Third Category : Comprise of the methods which are used to check the accuracy of the results obtained.

Definition of Research Methodology

Research Methodology, as its name suggest is the study of methods, so as to solve the research problem. It is the science of learning the way research should be performed systematically. It refers to the rigorous analysis of the methods applied in the stream of research, to ensure that the conclusions drawn are valid, reliable and credible too.

The researcher takes an overview of various steps that are chosen by him in understanding the problem at hand, along with the logic behind the methods employed by the researcher during study. It also clarifies the reason for using a particular method or technique, and not others, so that the results obtained can be assessed either by the researcher himself or any other party.

Key Differences Between Research Method and Research Methodology

The differences between research method and research methodology can be drawn clearly on the following grounds:

• The research method is defined as the procedure or technique applied by the researcher to undertake research. On the other hand, research methodology is a system of methods, used scientifically for solving the research problem.
• The research method is nothing but the behaviour or tool, employed in selecting and building research technique. Conversely, research methodology implies the science of analysing, the manner in which research is conducted appropriately.
• The research method is concerned with carrying out experiment, test, surveys, interviews, etc. As against this, research methodology is concerned with learning various techniques which can be employed in the performance of experiment, test or survey.
• Research method covers various investigation techniques. Unlike, research methodology, which consists of complete approach aligned towards the attainment of purpose.
• Research method intends to discover the solution to the problem at hand. In contrast, research methodology aspires to apply appropriate procedures, with a view to ascertaining solutions.

The scope of research methodology is wider than that of research method, as the latter is the part of the former. For understanding the research problem thoroughly, the researcher should know the research methodology along with the methods.

In a nutshell, research method refers to the technique which can be adopted to explore the nature of the world that surrounds us. On the contrary, research methodology is the foundation, which helps us to understand the determinants influencing the effectiveness of the methods applied.

Related Differences

• Difference Between Bank Rate and Repo Rate
• Difference Between CV and Cover Letter
• Difference Between Distributive Negotiation and Integrative Negotiation
• Difference Between Lay and Lie
• Difference Between Spin-off and Split-off

You Might Also Like:

probably says

May 26, 2017 at 6:05 pm

Thank you for the definitions.

arsadan says

June 7, 2017 at 11:00 am

good and helpful content…. thank you

Uwezo Chahe says

November 19, 2022 at 5:50 pm

Richard Kisembo says

September 26, 2017 at 1:37 pm

It’s a good piece. Thanx

Elijah Olawale says

November 23, 2017 at 6:45 pm

I really thank you at this time,thanks a lot!

Amelia Williams says

February 28, 2018 at 4:17 am

Good and helpful thanks Surbhi, I would like to cite your work, can you please post a reference. thankyou

Surbhi S says

February 28, 2018 at 9:47 am

“Difference Between Research Method and Research Methodology” Keydifferences.com By Surbhi S. 28 Feb 2018 https://keydifferences.com/difference-between-research-method-and-research-methodology.html >

MUSA Auwal Ahmad says

November 4, 2018 at 9:03 pm

tamirat says

November 22, 2018 at 7:08 pm

please put in pdf for more reading

PRADEEP says

November 27, 2018 at 1:32 pm

You expressed the content in an elegant way. I wanted to know, What are the references from where you collected this content?

November 28, 2018 at 6:32 pm

Thank you for clarification

CHARLES SSENGONZI says

March 12, 2019 at 6:31 am

This is really helpful. Thanks for the explanation.

Jackilina Rafael says

March 25, 2019 at 7:08 pm

thanks for your good explanation

philomena wanyama. says

June 22, 2019 at 1:10 am

Thanks. very well put!

Er SOHAIL JAVED says

December 15, 2019 at 11:19 am

The description of the subject is mentioned quite well, easy to understand and it reduces the mental stress for understanding the concept.

Jewel Rana says

April 16, 2020 at 9:56 am

This writing is very helpful to understand. Thank you for this .

Emmanuel Boateng Donald says

February 12, 2021 at 3:01 pm

Thank you for this work.

DAVID MASELI says

March 20, 2021 at 3:27 pm

So helpful have not enough words to say thank you

Bola Lawson says

September 25, 2021 at 9:30 pm

This is very insightful. You’re efforts is appreciated

Jkmarafa says

December 2, 2021 at 11:17 am

Thanks for the post

James A Batey says

January 29, 2022 at 8:15 am

I appreciate your contribution to the facilitation of learning

Jasson J Nkongo says

February 9, 2022 at 11:17 pm

Thank you so much fo the explanations

March 13, 2022 at 7:07 pm

THANK YOU IT WAS QUITE HELPFUL ,PLEASE CAN YOU ALSO EXPLAIN RESEARCH TECHNIQUES?

Babasola Osoteku says

July 30, 2022 at 6:02 pm

Thank you, that was insightful

September 25, 2022 at 2:34 am

Thank you for the brief but deep explanation

square says

December 2, 2022 at 12:27 am

Thanks for the explanation its very helpful

Fatima Ahmad says

January 25, 2023 at 8:12 pm

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Save my name, email, and website in this browser for the next time I comment.

Difference Between Research Method and Research Methodology

Research methods and research methodology are two terms that are often confused as the same. Strictly speaking, they are not so and show differences between them.

If we zone in on the etymology of the word methodology , it refers to method + ology , ‘Ology’ typically means a discipline of study or a branch of knowledge. Thus, technically speaking, the methodology is the study of methods.

The most important difference between research method and research methodology is that the research method is the techniques and tools for research , whereas research methodology explains the research methods.

Understand the Differences between Research Method and Research Methodology

Research method.

Research method means the techniques or tools used for conducting research irrespective of whether the research belongs to physical or social sciences or other disciplines.

The methods include three broad groups.

• The first group includes methods dealing with the collection and description of data;
• The second group consists of techniques used for establishing a statistical relationship between variables ;
• The third group deals with methods used to evaluate the reliability, validity, and accuracy of the results discerned by the data.

For example, a physical scientist may employ tools such as an electron microscope or a radio telescope to obtain his data.

In contrast, a social scientist or a manager may use, as a technique, an opinion poll or sample survey with a mail questionnaire or conduct a personal interview to obtain his data.

He might conduct a telephonic interview , group discussion , and case study approach to gather data. Still, they are employing the same technique, ‘ observation ‘ of some kind, that generates data for research.

Nevertheless, scientists in their disciplines employ tools and techniques that may differ widely in nature and complexity.

Research Methodology

The research methodology is a way to study the various steps generally adopted by a researcher in systematically studying his research problems, along with the logic, assumptions, justification, and rationale behind them.

Whenever we choose a research method, we must justify why we prefer this particular method over others. The methodology seeks to answer this question.

Thus, when we speak of research methodology, we not only talk of research methods but also keep the logic and justification behind the method we use in the context of our research undertaking.

A researcher’s methodology aims at answering such questions as:

• Why was this particular group of people interviewed and not the other groups?
• How has the research problem defined?
• How many individuals provided the answers on which the researcher’s conclusions were based?
• Why were these particular techniques used to analyze data?
• In what way and why has the research hypothesis been formulated ?
• What evidence was used to determine whether or not to reject the stated hypothesis?

Difference between Research Method and Research Methodology

Suppose the subject into which you conduct research is a scientific subject or topic. In that case, the research methods include experiments, tests, the study of many other results of different experiments performed earlier about the topic or the subject, and the like.

On the other hand, research methodology about the scientific topic involves the techniques regarding how to conduct the research, the justification of the use of particular research tools, advanced techniques that can be used in performing the experiments, and the like.

A method is what you did. It is a simple description. You selected, for example, 100 rats and measured their weights. You fed some rats and some not.

A week later, you measured their weights again.

The methodology is why that should give you a meaningful result and why you used some specified method and not some other one.

This would, in particular, include how you have controlled for errors, e.g., why you fed the rats for a week rather than a month and why 100 rats you thought were enough.

How are the terms “research methods” and “research methodology” often misconstrued?

“Research methods” and “research methodology” are two terms that are frequently confused as being identical. However, they are distinct and have different meanings and applications in the realm of research.

What is the primary difference between the objectives of research methods and research methodology?

The primary objective of research methods is to find solutions to research problems, while research methodology ensures the employment of the correct procedures to address these problems, providing a comprehensive understanding of the research process .

Why is the distinction between research method and research methodology crucial?

The distinction is vital because while the research method pertains to the techniques and tools used in research, the research methodology explains and justifies these techniques and tools, providing a broader context and rationale for the research process.

In the context of scientific research, how do research methods differ from research methodology?

In scientific research, research methods include conducting experiments, tests, and studying results from previous experiments related to the topic. In contrast, research methodology involves the techniques on how to conduct the research, the justification for using specific research tools, and advanced techniques that can be employed in the experiments.

Following our deep dive into difference between research method and research methodology; use our total guide on research and research methodology concepts .

• Probability Sampling
• Evaluating Proposed Research Topic
• Marketing Research: Types, Process, Models
• Operational Definition in Research
• Terms of Reference (TOR) [9 Main Sections]
• Concept, Indicator, Construct in Research
• Data Analysis in Research
• Program Performance Indicators
• Measures of Association: Definition, Method
• Research Design: Definition, Types, Types, Guide & Examples
• Unstructured Interviews
• Document Study: Definition, Advantages, Disadvantages
• Hypothesis Testing: Definition, Examples
• Computer-Assisted Personal Interviewing (CAPI)
• Stapel Scale: Definition, Example

• Open access
• Published: 28 October 2023

Functional principal component analysis and sparse-group LASSO to identify associations between biomarker trajectories and mortality among hospitalized SARS-CoV-2 infected individuals

• Tingyi Cao 1 ,
• Harrison T. Reeder 2 , 3 &
• Andrea S. Foulkes 1 , 2 , 3  

BMC Medical Research Methodology volume  23 , Article number:  254 ( 2023 ) Cite this article

172 Accesses

2 Altmetric

Metrics details

A substantial body of clinical research involving individuals infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has evaluated the association between in-hospital biomarkers and severe SARS-CoV-2 outcomes, including intubation and death. However, most existing studies considered each of multiple biomarkers independently and focused analysis on baseline or peak values.

We propose a two-stage analytic strategy combining functional principal component analysis (FPCA) and sparse-group LASSO (SGL) to characterize associations between biomarkers and 30-day mortality rates. Unlike prior reports, our proposed approach leverages: 1) time-varying biomarker trajectories, 2) multiple biomarkers simultaneously, and 3) the pathophysiological grouping of these biomarkers. We apply this method to a retrospective cohort of 12, 941 patients hospitalized at Massachusetts General Hospital or Brigham and Women’s Hospital and conduct simulation studies to assess performance.

Renal, inflammatory, and cardio-thrombotic biomarkers were associated with 30-day mortality rates among hospitalized SARS-CoV-2 patients. Sex-stratified analysis revealed that hematogolical biomarkers were associated with higher mortality in men while this association was not identified in women. In simulation studies, our proposed method maintained high true positive rates and outperformed alternative approaches using baseline or peak values only with respect to false positive rates.

Conclusions

The proposed two-stage approach is a robust strategy for identifying biomarkers that associate with disease severity among SARS-CoV-2-infected individuals. By leveraging information on multiple, grouped biomarkers’ longitudinal trajectories, our method offers an important first step in unraveling disease etiology and defining meaningful risk strata.

Peer Review reports

Since the outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in December 2019, more than 670 million confirmed cases and 6.8 million associated deaths have been reported worldwide, with a large proportion of these deaths preceded by hospitalization [ 1 ]. The vast amount of data collected and stored in electronic health records among hospitalized patients provides an opportunity to identify early predictors of severe disease. Ultimately understanding the relationship between patient level in-hospital data, including early biomarker trajectories, and severe outcomes may inform disease etiology, risk stratification, and resource allocation.

Among SARS-CoV-2 infected individuals, multiple biomarkers are typically measured repeatedly over the duration of hospitalization. An extensive literature has identified correlations between biomarker levels and severe outcomes, including intubation, admission to intensive care units, and death among hospitalized SARS-CoV-2 infected individuals [ 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 ]. However, most existing studies considered each biomarker independently [ 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 ] with a few exceptions that applied penalized regression and other machine learning techniques [ 10 , 11 ]. None of these manuscripts, to our knowledge, accounted for the pathophysiological relationships among biomarkers. Moreover, repeatedly measured biomarkers were typically reduced to baseline or peak values [ 2 , 3 , 4 , 5 , 6 , 7 , 8 , 10 , 11 ], with again a small number of exceptions, including one report using linear mixed-effects models to account for the entire biomarker trajectory [ 9 ]. To our knowledge, analyses that simultaneously consider multiple biomarkers as well as their longitudinal trajectories in evaluating associations with severe SARS-CoV-2 outcomes have not been reported.

Methods for joint modeling of multiple longitudinal biomarkers and time-to-event outcomes have also been described [ 15 , 16 , 17 , 18 ]. However, as these methods generally require computationally intensive procedures, such as multi-dimensional numerical integration or complex Bayesian sampling schemes, approaches incorporating variable selection among multiple biomarkers into the joint modeling framework remain limited [ 18 , 19 ]. A scalable alternative involves application of multivariate functional principal component analysis (FPCA) [ 20 ] to reduce each biomarker trajectory to a set of scores and then using these scores as covariates in a survival model [ 21 , 22 , 23 ]. Application of FPCA and survival modeling has been limited to prediction of the time-to-event outcome. To allow for variable selection in this context, we propose a two-stage analytic strategy that combines FPCA and sparse-group LASSO (SGL) [ 24 ], abbreviated as FPCA-SGL, to characterize associations between multiple biomarker trajectories and mortality, while also leveraging the pathophysiological grouping of these biomarkers.

Study population

Data derived from a retrospective cohort of 12,941 patients infected with SARS-CoV-2 based on hospital record ICD-10 codes (U07.1, B34,2, and B97.29) and positive PCR tests between March 1, 2020 and November 30, 2021 were used for analysis (Table  1 ). All patients were hospitalized at Massachusetts General Hospital or Brigham and Women’s Hospital (MGB) within 5 days prior to and 30 days after a positive SARS-CoV-2 test. Patients hospitalized for less than 24 hours or with an unknown duration were excluded.

Data pre-processing

The primary outcome is 30-day mortality since hospital admittance. Death records from both the MGB Enterprise Data Warehouse (EDW) and the Massachusetts Registry of Vital Records and Statistics were obtained, and in the case of an inconsistency between the two sources, death dates from the Registry were adopted. Exposures are repeated laboratory measurements of \(m=20\) routine biomarkers collected during hospitalization up to 30 days. Biomarker data were extracted from MGB EDW and if there were multiple measurements of a biomarker for one patient within a 24-hour period, the mean value of the measurements was used. Censored laboratory measurements were treated as known at the cut-off value. Demographic information including age, sex, race/ethnicity and body mass index (BMI) was obtained from MGB EDW. Biomarkers were divided into six categories based on their pathophysiological functions, as shown in Supplementary Table 1 .

Statistical analysis

A two-stage analytic approach was considered. First, FPCA was performed separately on each of the \(k=1,2,\cdots ,K\) biomarkers. Each biomarker’s repeated measurements were treated as functional data, i.e., independent realizations of a smooth random function \(X_k(t)\) [ 20 ]. Through spectral decomposition of the covariance operator, FPCA reduces the functional data into eigenfunctions \(\phi _{km}(t)\) for \(m=1,\cdots ,M\) , referred to as functional principal components (FPCs). Each individual i has a set of coefficients for these eigenfunctions called FPC scores, denoted as \(A_{kim}\) . Thus the trajectory of one biomarker for patient i , \(X_{ki}(t)\) , can be expanded as

where \(\mu _k(t)=\mathbb {E}[X_k(t)]\) is the mean function. Then each patient’s FPC scores, \(A_{kim}, m=1,\cdots ,M\) , characterize the variation of individual level biomarker trajectories from the sample mean function. We adopted the PACE method which computes the FPC scores as conditional expectations because it is suitable for sparse and irregularly spaced longitudinal data like our biomarker data [ 25 ]. To implement FPCA using the PACE approach, we used the fdapace package in R [ 26 ]. Based on the cumulative percentage of variance explained, we determined the number of FPCs ( M ) to adopt, resulting in \(K\times M\) exposure variables. Missing FPC scores were imputed using MICE based on the FPC scores of all other biomarkers [ 27 ].

Second, using the SGL package in R, we performed Cox SGL regression with the \(K\times M\) FPC scores as the exposure variables, while adjusting for J demographic characteristics \(Y_{ij}\) including race/ethnicity, age (indicator for \(>50\) years), and BMI (orthogonal polynomials of degree 2):

where there are \(K\times M+J\) regression parameters to be estimated, collectively denoted as \(\varvec{\beta }=\{\delta _{km},\zeta _j\}\) for \(k=1,\cdots ,K\) , \(m=1,\cdots ,M\) , and \(j=1,\cdots ,J\) .

SGL estimates the \(\varvec{\beta }\) ’s using a weighted combination (controlled by a hyperparameter \(\eta\) ) of group LASSO \(l_1\) -penalty term and the standard parameter-wise LASSO \(l_1\) -penalty term to induce both groupwise and within-group sparsity [ 24 ]:

where the grouping \(l=1,\cdots ,L\) represents biomarkers’ six pathophysiological categories and one group for all demographic characteristics, i.e. for \(l=1,2,\cdots ,6\) , \(\beta ^{(l)} = \{ \delta _{km}\}\) for k over all biomarkers in that pathophysiological group and \(m=1,\cdots ,M\) for each biomarker’s M FPC scores; and \(\beta ^{(7)} = \{ \zeta _j\}\) for \(j=1,\cdots ,J\) .

This two-stage strategy, FPCA decomposition of the biomarker trajectoris followed by SGL, allowed us to identify biomarkers associated with 30-day mortality while accounting for within-group correlations between biomarkers as well as the time varying biomarker trajectories. Both stages of analysis were stratified by sex.

We selected the overall regularization parameter \(\lambda\) through a 10-fold cross validation (CV) from a pre-specified sequence of 100 \(\lambda\) values. The sequence of 100 candidate \(\lambda\) values were chosen such that the maximum, \(\lambda _\text {max}\) , was the smallest possible \(\lambda\) that shrunk all coefficients to zero, the minimum, \(\lambda _\text {min}\) , was set equal to \(\lambda _\text {max}/100\) , and all other \(\lambda\) values were spaced equally between \(\lambda _{\text {min}}\) and \(\lambda _{\text {max}}\) . Eventually, we selected \(\lambda _\text {1se}\) which was the largest value of \(\lambda\) such that the CV error, defined as the CV negative log likelihood, was within 1 standard error of the minimum.

We set the weight for the group LASSO and LASSO penalty terms to \(1-\eta = 0.7\) and \(\eta = 0.3\) , respectively to get a group LASSO structure with limited within-group sparsity. This reflects that biomarkers are expected to exhibit group structure due to their underlying pathophysiological relationships, while still allowing individual biomarkers or FPC scores to be excluded from the model to enhance sparsity. Alternatively, this weight parameter \(\eta\) could be tuned via an additional layer of CV. As a sensitivity analysis, we fit the SGL model with different weights, \(\eta = 0.05, 0.50, 0.70, 0.95\) respectively, to examine whether the results were sensitive to the choice of this hyperparameter.

For comparison, we considered application of SGL using only the baseline or peak measurements of each of the K biomarkers in place of the FPC scores. Here we log transformed the baseline and peak measurements to ensure they were approximately normally distributed, and again imputed missing values with available biomarker measures using the MICE package in R [ 27 ]. Analyses were again stratified by sex and adjusted for race/ethnicity, age, and BMI while accounting for the pathophysiological groupings of biomarkers.

Simulation studies

To characterize the performance of our proposed two-stage approach, we conducted simulation studies including 200 repetitions with sample sizes of \(n=2000\) for each condition [ 21 , 28 ] ( Supplementary Methods ). We first simulated trajectories of four biomarkers belonging to two groups, denoted as \(Z_{ki}(t)\) where \(k=1,2,3,4\) and \(i=1,2,\cdots ,n\) . Biomarkers \(k=1,2\) were in the first group with relatively low within-group correlation, and biomarkers \(k=3,4\) were in the second group with relatively high within-group correlation. \(Z_{ki}(t)\) were simulated under three models: Model 1 was a linear mixed-effects model (LME) with a linear time trend; Model 2 was a LME with a quadratic term for time; and Model 3 was a LME with a 3-knot spline function for time.

Death times were simulated based on \(Z_{ki}(t)\) using inverse transform sampling on the survival function derived from the following hazard function: \(h_i(t) = h_0(t) \text {exp} \left[ \alpha _1\times Z_{1i}(t) + \alpha _2\times Z_{2i}(t) + \alpha _3\times Z_{3i}(t) + \alpha _4\times Z_{4i}(t) \right]\) . The association parameters \(\alpha _1,\alpha _2,\alpha _3,\alpha _4\) , were specified based on four scenarios: Scenario 1, only the low correlation biomarker group was associated with mortality; Scenario 2, only the high correlation biomarker group was associated with mortality; Scenario 3, both biomarker groups were associated with mortality; Scenario 4, neither of the two biomarker groups was associated with mortality (Table  2 ). Measurement error was added to the true trajectories to simulate observed trajectories. Lastly, we simulated censoring by truncating observed trajectories at the patient’s time of death or discharge ( Supplementary Methods ).

For the primary analysis we applied FPCA-SGL using the top three FPC scores ( \(M=3\) ) for each biomarker as the exposure variables, i.e. \(K\times M=4\times 3=12\) variables. The weights for the group LASSO and LASSO penalty terms were set to \(1-\eta =0.95\) and \(\eta =0.05\) , respectively because our simulated scenarios were a group LASSO case. To assess performance of our proposed approach, we reported the true positive rate (TPR), defined as the proportion of simulations where truly non-zero coefficients were selected, and the false positive rate (FPR), defined as the proportion of simulations where truly zero coefficients were selected. We also compared FPCA-SGL to two simpler comparator approaches using baseline or peak measurements alone.

Application using MGB cohort

The MGB cohort was composed of 12,941 patients, \(32\%\) were \(\le 50\) years of age, \(20.5\%\) Hispanic, \(11.7\%\) Black/non-Hispanic, and \(35.8\%\) obese (BMI \(\ge 30\) ) (Table  1 ). 1,198 patients ( \(9.3\%\) ) died within 30-days of hospitalization, with a higher proportion in males than in females ( \(10.7\%\) vs. \(7.7\%\) ) (Table  1 ). Supplementary Table  1 summarizes laboratory measurements for 20 biomarkers. More than \(60\%\) of patients had at least one measurement on each of the 20 biomarkers, with a median number of measurements for each biomarker ranging from 1 to 6, except that d-dimer had fewer measurements.

We performed FPCA on each of the 20 biomarkers, stratified by sex. Supplementary Fig.  1 displayed the mean function and corresponding FPCs of each biomarker. Across the 20 biomarkers, the first three FPCs cumulatively explained a median of \(97.39\%\) [IQR = ( \(95.32\%\) , \(99.22\%\) )] and \(97.49\%\) [IQR = ( \(96.06\%\) , \(98.48\%\) )] of the total variance among females and males, respectively. Therefore, we picked \(M=3\) FPCs for each of the M biomarkers. FPC scores were approximately normally distributed (Supplementary Fig.  2 ).

To better illustrate how each patient’s \(M=3\) FPC scores could represent the variation of their individual biomarker trajectories from the mean function, Supplementary Fig.  3 plotted the trajectories of blood urea nitrogen (bun) of three male patients with different FPC scores \(A_{k=1,i=\{1,2,3\},m=\{1,2,3\}}\) . It shows how each individual’s trajectory decomposes into a linear combination of the mean function and three eigenfunctions, resulting in different individual-specific FPC scores.

The pairwise baseline biomarker correlations were similar among females and males (Fig.  1 A, B). The renal, hematological, and the hepatic groups exhibited high within-group correlation while the cardio-thrombotic, inflammatory and metabolic groups presented low within-group correlation. The across-group correlations were generally low (Fig.  1 A, B). The pairwise peak biomarker correlations showed similar patterns (Fig.  1 C, D). Biomarker peak and baseline values were approximately normally distributed after log transformation, imputation and standardization, with the exception of estimated glomerular filtration rate and total bilirubin (Supplementary Fig.  4 ).

Pearson correlations between biomarkers’ baseline and peak measurements, stratified by sex and masked by p -value under an \(\alpha -\) level of 0.05

Using the FPCA-SGL approach with \(K\times M=20\times 3=60\) FPC scores as exposure variables, we found biomarkers in the renal and inflammatory groups to be strongly associated with mortality in both males and females. In the cardio-thrombotic group, only d-dimer appeared to be associated with mortality. Biomarkers in the hepatic groups showed slight associations while the metabolic group was not associated with mortality. The hematological group was associated with mortality among males but not females (Fig.  2 ).

Estimated regression coefficients \(\hat{\varvec{\beta }}\) from SGL models fitted with the scores of the first 3 FPCs of each biomarker as exposure variables, tiles with no border or annotated numbers indicate \(\hat{\varvec{\beta }}\) being regularized to zero (The full names of the abbreviated biomarkers are listed at the end of the manuscript)

For comparison, we used each of the 20 baseline measurements and the 20 peak measurements as exposure variables, and fitted SGL Cox regressions stratified by sex. Using baseline measurement, all biomarker groups except for the hematological group among females and the metabolic group among males were associated with mortality, with some degree of within-group sparsity observed (Fig.  3 ). Using peak measurements, most biomarkers across all groups except for the metabolic group were associated with mortality, with almost no within-group sparsity (Fig.  3 ). Results from our simulation studies, as presented below, demonstrated that using baseline or peak measurements can result in high false positive rates.

Estimated regression coefficients \(\hat{\varvec{\beta }}\) from SGL models fitted with the baseline or peak measurement of each biomarker as exposure variables, tiles with no border or annotated numbers indicate \(\hat{\varvec{\beta }}\) being regularized to zero (The full names of the abbreviated biomarkers are listed at the end of the manuscript)

As a sensitivity analysis, we applied the FPCA-SGL approach with the same 60 FPC scores while changing the weight hyperparameter \(\eta\) , with larger \(\eta\) implying more LASSO than group LASSO structure. Supplementary Fig.  5 displayed four similar heatmaps as Fig.  2 for four different weight values: \(\eta =0.05, 0.50, 0.70, 0.95\) . The associations were similar with the main analysis: the renal and inflammatory groups were still strongly associated with mortality in both sexes; in the cardio-thrombotic group, d-dimer showed associations and creatine phosphokinase only showed trivial associations when a group LASSO structure was enforced under \(\eta <0.3\) ; the hepatic group was slightly associated with mortality and under small \(\eta\) the entire group was not selected, while larger \(\eta\) revealed that the aspartate aminotransferase and albumin in the group were the biomarkers driving these associations; the metabolic group was not associated with mortality; the hematological group showed associations still only among males but not females. Therefore, this analysis demonstrated that the results were not sensitive to the choice of the weight hyperparameter.

Both our proposed two-stage FPCA-SGL method and the simpler comparator methods using baseline or peak measurements, offered high TPR, i.e., high sensitivity, under Scenario 1-3 for Model 1-3: the approach using baseline measurements always gave TPR as high as \(100\%\) and never smaller than \(98.5\%\) ; the approach using peak measurements always gave TPR as high as \(100\%\) and never smaller than \(98\%\) ; our proposed approach using FPC scores gave comparable TPR over \(98.5\%\) for all first FPCs and relatively high TPR greater than \(87\%\) for the second and third FPCs (Table  3 ).

The FPCA-SGL approach gave relatively low FPR, as low as \(0\%\) in Scenario 2 in the case that the biomarker group with high within-group correlation was associated with the survival outcome, and no higher than \(12\%\) in the null case of Scenario 4. Notably, in every scenario under every model, this approach consistently showed smaller FPR than the approaches using baseline or peak measurements, especially under Scenario 1-2 for Model 2 (FPR ranging from \(17\%\) to \(22\%\) using baseline or peak measurements but as low as \(0-0.5\%\) using FPC scores). This demonstrated that our proposed two-stage FPCA-SGL approach gave much higher specificity than the simpler methods using baseline or peak measurements (Table  3 ).

The two comparator methods, especially the one using peak measurements, suffered from high FPR, which was particularly high in Scenario 4: the approach using peak measurements yielded an FPR of \(17-18\%\) under Model 1, \(13.5-18\%\) under Model 2, and \(40.5-41.5\%\) under Model 3; the approach using baseline measurements yielded an FPR of \(14-16\%\) under Model 1, \(15.5-17.5\%\) under Model 2, and \(16.0-17.5\%\) under Model 3 (Table  3 ). This again illustrated that these two simpler approaches using baseline or peak measurements suffered from low specificity.

To investigate further the inflated FPR in Scenario 4, we considered an additional Scenario 5, a complete null case in which neither of the two biomarker groups was associated with mortality ( \(\alpha _1=\alpha _2=\alpha _3=\alpha _4=0\) ) and observed biomarker trajectories were not censored by death times (Table  2 ). Simulation results showed that the FPR decreased slightly in this scenario, most notably under Model 3 (from \(40.5-41.5\%\) in Scenario 4 to \(16.5-20\%\) in Scenario 5 with peak measurements, from \(7.5-9\%\) in Scenario 4 to \(5-6.5\%\) in Scenario 5 with FPC scores), and also under Model 1 with peak measurements (from \(17-18\%\) in Scenario 4 to \(13.5-16\%\) in Scenario 5), as well as under Model 2 with FPC scores (from \(10-12\%\) in Scenario 4 to \(4-5\%\) in Scenario 5) (Supplementary Table  2 ). For the approach using baseline measurements, this additional Scenario 5 did not alter the baseline values thus the FPR remained similar (Supplementary Table  2 ).

Our proposed FPCA-SGL approach revealed associations between several biomarker trajectories and 30-day mortality among hospitalized SARS-CoV-2 patients. In particular, renal and inflammatory biomarkers were strongly associated with mortality risks. Several studies have examined incidence of acute kidney injury (AKI) among SARS-CoV-2 patients and discovered that AKI was related to more severe outcomes including death, respiratory failure, and disseminated intravascular coagulation [ 7 , 10 , 29 ]. Elevated blood urea nitrogen and creatinine, as well as lower estimated glomerular filtration rate were all markers of AKI and were reported to be correlated with worse outcomes [ 2 , 5 , 7 , 10 , 13 , 29 ]. Studies have indicated excessive inflammatory response as a contributory factor to SARS-CoV-2 disease severity [ 30 ]. Lymphocytes are crucial in modulating inflammatory response and maintaining immune homeostasis during viral infection [ 31 ], and research reported elevated white blood cell count and lymphopenia (low absolute lymphocyte count) among severe SARS-CoV-2 patients [ 3 , 4 , 7 ]. Elevated c-reactive protein levels were also closely related to inflammation and shown to be highly associated with disease severity [ 32 ].

A limited number of studies have specifically investigated the effect of sex on the associations between biomarker levels and disease severity [ 12 , 13 , 14 ]. We had a relatively large cohort of 12,941 patients, thus we conducted our analyses under stratification by sex so as to better explore any potential sex modification. Interestingly, we observed associations of hematological biomarkers (hemoglobin, hematocrit, and platelets) with 30-day mortality risks only among males. As males usually experienced more severe symptoms and worse survival outcomes during SARS-CoV-2 infection [ 12 , 14 ], our results may lend insight into the sex difference behind the cellular and molecular pathways underlying SARS-CoV-2 disease progression.

Methodologically, our proposed FPCA-SGL approach is an easy-to-implement and computationally efficient analytic strategy that is able to simultaneously consider multiple biomarkers as well as their longitudinal trajectories in evaluating associations with severe SARS-CoV-2 outcomes. It is a versatile alternative to existing methods concerning multiple longitudinal measurements and a survival outcome and could be applied in other areas. Using simulation studies, we demonstrated that FPCA-SGL retained high TPR and outperformed alternative approaches using baseline or peak values with respect to FPR. In particular, we observed a substantial “survival bias” (inflated FPR) in our simulations when using peak measurements, because they are endogenous covariates, meaning their values and future paths are directly affected by the survival outcome of interest [ 33 ]. For example, if a certain biomarker has a monotonically increasing trajectory during hospitalization, the peak value observed will be higher for patients surviving longer, causing spurious associations between lower peak biomarker values and higher mortality risks. This “survival bias” resulted in high FPR in our simulation study using the peak measurement approach (Scenario 4 in Table  3 ) and was moderately alleviated when we did not censor the simulated biomarker trajectories based on simulated death times (Scenario 5 in Supplementary Table  2 ). Our FPCA-SGL approach mitigated this “survival bias” (lower FPR in Scenario 4 in Table  3 ) because FPCA naturally imputed biomarkers’ unobserved future trajectories even after patients’ deaths. Nonetheless, we did still observe some false positives using this proposed approach with FPC scores (largest FPR as \(12\%\) in Scenario 4 in Table  3 ).

We presented a two-stage analytic approach that combined FPCA and SGL to study the associations between hospitalized SARS-CoV-2 patients’ multiple biomarker trajectories with their 30-day mortality rates. We demonstrated that this method had high TPR and outperformed simpler comparator approaches using biomarkers’ baseline or peak measurements with respect to FPR. Using data from a retrospective cohort of 12,941 patients, we showed that renal biomarkers (blood urea nitrogen, creatinine, and estimated glomerular filtration rate), inflammatory biomarkers (c-reactive protein, white blood cell count, and absolute lymphocyte count), cardio-thrombotic biomarkers (d-dimer) were associated with 30-day mortality rates among hospitalized SARS-CoV-2 patients. Our sex-stratified analysis also revealed that hematological biomarkers (hemoglobin, hematocrit, and platelets) were associated with higher mortality only among males. This study recognized the prognostic value of biomarkers as well as the underlying potential sex difference. These results provide insights into assessment of SARS-CoV-2 disease severity and effective risk stratification.

Availability of data and materials

The data is not publicly available. The code for the simulation studies (both data simulation and the application of our proposed FPCA-SGL method) is available in the Github repository, https://github.com/Aimeessn/FPCA-SGL .

Abbreviations

Blood urea nitrogen

Estimated glomerular filtration rate

Creatine phosphokinase

C-reactive protein

White blood cell count

Absolute lymphocyte count

Alanine aminotransferase

Aspartate aminotransferase

Alkaline phosphatase

Total bilirubin

Lactate dehydrogenase

Dong E, Du H, Gardner L. An interactive web-based dashboard to track COVID-19 in real time. Lancet Infect Dis. 2020;20(5):533–4. https://doi.org/10.1016/s1473-3099(20)30120-1 .

Article   CAS   PubMed   PubMed Central   Google Scholar  

Bivona G, Agnello L, Ciaccio M. Biomarkers for Prognosis and Treatment Response in COVID-19 Patients. Ann Lab Med. 2021;41(6):540–8. https://doi.org/10.3343/alm.2021.41.6.540 .

Malik P, Patel U, Mehta D, Patel N, Kelkar R, Akrmah M, et al. Biomarkers and outcomes of COVID-19 hospitalisations: systematic review and meta-analysis. BMJ Evid Based Med. 2021;26(3):107–8. https://doi.org/10.1136/bmjebm-2020-111536 .

Article   PubMed   Google Scholar  

Izcovich A, Ragusa MA, Tortosa F, Lavena Marzio MA, Agnoletti C, Bengolea A, et al. Prognostic factors for severity and mortality in patients infected with COVID-19: A systematic review. PLoS ONE. 2020;15(11):e0241955. https://doi.org/10.1371/journal.pone.0241955 .

Wang D, Hu B, Hu C, Zhu F, Liu X, Zhang J, et al. Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in Wuhan. China Jama. 2020;323(11):1061–9. https://doi.org/10.1001/jama.2020.1585 .

Article   CAS   PubMed   Google Scholar  

Gupta D, Jain A, Chauhan M, Dewan S. Inflammatory Markers as Early Predictors of Disease Severity in COVID-19 Patients Admitted to Intensive Care Units: A Retrospective Observational Analysis. Indian J Crit Care Med. 2022;26(4):482–6. https://doi.org/10.5005/jp-journals-10071-24171 .

Xu Z, Zhang Y, Zhang C, Xiong F, Zhang J, Xiong J. Clinical Features and Outcomes of COVID-19 Patients with Acute Kidney Injury and Acute Kidney Injury on Chronic Kidney Disease. Aging Dis. 2022;13(3):884–98. https://doi.org/10.14336/ad.2021.1125 .

Article   PubMed   PubMed Central   Google Scholar  

Petrilli CM, Jones SA, Yang J, Rajagopalan H, O’Donnell L, Chernyak Y, et al. Factors associated with hospital admission and critical illness among 5279 people with coronavirus disease 2019 in New York City: prospective cohort study. Bmj. 2020;369:m1966. https://doi.org/10.1136/bmj.m1966 .

Bowring MG, Wang Z, Xu Y, Betz J, Muschelli J, Garibaldi BT, et al. Outcome-Stratified Analysis of Biomarker Trajectories for Patients Infected With Severe Acute Respiratory Syndrome Coronavirus 2. Am J Epidemiol. 2021;190(10):2094–106. https://doi.org/10.1093/aje/kwab138 .

Boss AN, Banerjee A, Mamalakis M, Ray S, Swift AJ, Wilkie C, et al. Development of a Mortality Prediction Model in Hospitalised SARS-CoV-2 Positive Patients Based on Routine Kidney Biomarkers. Int J Mol Sci. 2022;23(13). https://doi.org/10.3390/ijms23137260 .

Syed AH, Khan T, Alromema N. A Hybrid Feature Selection Approach to Screen a Novel Set of Blood Biomarkers for Early COVID-19 Mortality Prediction. Diagnostics (Basel). 2022;12(7). https://doi.org/10.3390/diagnostics12071604 .

Haitao T, Vermunt JV, Abeykoon J, Ghamrawi R, Gunaratne M, Jayachandran M, et al. COVID-19 and Sex Differences: Mechanisms and Biomarkers. Mayo Clin Proc. 2020;95(10):2189–203. https://doi.org/10.1016/j.mayocp.2020.07.024 .

Lumish HS, Kim E, Selvaggi C, Cao T, Gupta A, Foulkes AS, et al. Biomarkers of Cardiac Injury, Renal Injury, and Inflammation Are Strong Mediators of Sex-Associated Death in COVID-19. Front Cardiovasc Med. 2022;9:809997. https://doi.org/10.3389/fcvm.2022.809997 .

Megiorni F, Pontecorvi P, Gerini G, Anastasiadou E, Marchese C, Ceccarelli S. Sex-Related Factors in Cardiovascular Complications Associated to COVID-19. Biomolecules. 2021;12(1). https://doi.org/10.3390/biom12010021 .

Lin H, McCulloch CE, Mayne ST. Maximum likelihood estimation in the joint analysis of time-to-event and multiple longitudinal variables. Stat Med. 2002;21:2369–82.

Hickey G, Philipson P, Jorgensen A, Kolamunnage-Donà R. JoineRML: A joint model and software package for time-to-event and multivariate longitudinal outcomes. BMC Med Res Methodol. 2018;18. https://doi.org/10.1186/s12874-018-0502-1 .

Rizopoulos D, Ghosh P. A Bayesian semiparametric multivariate joint model for multiple longitudinal outcomes and a time-to-event. Stat Med. 2011;30:1366–80. https://doi.org/10.1002/sim.4205 .

Chen Y, Wang Y. Variable selection for joint models of multivariate longitudinal measurements and event time data. Stat Med. 2017;36(24):3820–9. https://doi.org/10.1002/sim.7391 .

He Z, Tu W, Wang S, Fu H, Yu Z. Simultaneous variable selection for joint models of longitudinal and survival outcomes. Biometrics. 2015;71(1):178–87. https://doi.org/10.1111/biom.12221 .

Wang JL, Chiou JM, Müller HG. Functional Data Analysis. Ann Rev Stat Appl. 2016;3(1):257–95. https://doi.org/10.1146/annurev-statistics-041715-033624 .

Article   Google Scholar  

Li K, Luo S. Dynamic prediction of Alzheimer’s disease progression using features of multiple longitudinal outcomes and time-to-event data. Stat Med. 2019;38(24):4804–18. https://doi.org/10.1002/sim.8334 .

Lin J, Li K, Luo S. Functional survival forests for multivariate longitudinal outcomes: Dynamic prediction of Alzheimer’s disease progression. Stat Methods Med Res. 2021;30(1):99–111. https://doi.org/10.1177/0962280220941532 .

Jiang BS, Xie Y, Colditz GA. Functional Ensemble Survival Tree: Dynamic Prediction of Alzheimer’s Disease Progression Accommodating Multiple Time-Varying Covariates. bioRxiv. 2020. https://doi.org/10.1101/2020.02.17.952994 .

Simon N, Friedman J, Hastie T, Tibshirani R. A Sparse-Group Lasso. J Comput Graph Stat. 2013;22(2):231–45. https://doi.org/10.1080/10618600.2012.681250 .

Yao F, Müller HG, Wang JL. Functional Data Analysis for Sparse Longitudinal Data. J Am Stat Assoc. 2005;100(470):577–90.

Article   CAS   Google Scholar  

Gajardo A, Bhattacharjee S, Carroll C, Chen Y, Dai X, Fan J, et al. fdapace: Functional Data Analysis and Empirical Dynamics. 2021. R package version 0.5.8. https://CRAN.R-project.org/package=fdapace . Accessed Sept 2023.

van Buuren S, Groothuis-Oudshoorn K. mice: Multivariate Imputation by Chained Equations in R. J Stat Softw. 2011;45(3):1–67. https://doi.org/10.18637/jss.v045.i03 .

Yan F, Lin X, Huang X. Dynamic prediction of disease progression for leukemia patients by functional principal component analysis of longitudinal expression levels of an oncogene. Ann Appl Stat. 2017;09(11):1649–70. https://doi.org/10.1214/17-AOAS1050 .

Cheng Y, Luo R, Wang K, Zhang M, Wang Z, Dong L, et al. Kidney disease is associated with in-hospital death of patients with COVID-19. Kidney Int. 2020;97(5):829–38. https://doi.org/10.1016/j.kint.2020.03.005 .

Wang Y, Perlman S. COVID-19: Inflammatory Profile. Annu Rev Med. 2022;73:65–80. https://doi.org/10.1146/annurev-med-042220-012417 .

Wang F, Hou H, Luo Y, Tang G, Wu S, Huang M, et al. The laboratory tests and host immunity of COVID-19 patients with different severity of illness. JCI Insight. 2020;5(10). https://doi.org/10.1172/jci.insight.137799 .

Chalmers S, Khawaja A, Wieruszewski PM, Gajic O, Odeyemi Y. Diagnosis and treatment of acute pulmonary inflammation in critically ill patients: The role of inflammatory biomarkers. World J Crit Care Med. 2019;8(5):59–71. https://doi.org/10.5492/wjccm.v8.i5.59 .

Rizopoulos D. The R Package JMbayes for Fitting Joint Models for Longitudinal and Time-to-Event Data using MCMC. arXiv. 2014 . https://doi.org/10.48550/ARXIV.1404.7625 .

Download references

Acknowledgements

NIH/NHLBI R01HL162373 and NIH/NIGMS R01GM127862.

Author information

Authors and affiliations.

Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA, USA

Tingyi Cao & Andrea S. Foulkes

Biostatistics, Massachusetts General Hospital, Boston, MA, USA

Harrison T. Reeder & Andrea S. Foulkes

Department of Medicine, Harvard Medical School, Boston, MA, USA

You can also search for this author in PubMed   Google Scholar

Contributions

TC conceived the idea of the study. TC and AF designed the study. All authors participated in the development of the methods. TC performed the data analysis, interpreted the findings and drafted the manuscript. HR and AF interpreted the findings and revised the manuscript. All authors read and approved the final manuscript.

Authors’ informations

Corresponding author.

Correspondence to Tingyi Cao .

Ethics declarations

Ethics approval and consent to participate.

The Partners HealthCare Institutional Review Board (IRB) ( \(\#\) 2020P000829) approved this study, including methods, the collection of curated data based on comprehensive manual chart reviews, and data extractions from electronic health records (EHR) on patients who receive care through the Mass General Brigham (MGB, formerly Partners) system. All methods were carried out in accordance with relevant guidelines and regulations. All participants provided written informed consent to participate in research.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1..

Supplementary Figures and Tables.

Additional file 2.

Supplementary Methods.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and Permissions

About this article

Cite this article.

Cao, T., Reeder, H.T. & Foulkes, A.S. Functional principal component analysis and sparse-group LASSO to identify associations between biomarker trajectories and mortality among hospitalized SARS-CoV-2 infected individuals. BMC Med Res Methodol 23 , 254 (2023). https://doi.org/10.1186/s12874-023-02076-3

Download citation

Received : 22 May 2023

Accepted : 18 October 2023

Published : 28 October 2023

DOI : https://doi.org/10.1186/s12874-023-02076-3

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Sparse group LASSO
• Functional data analysis
• Functional principal component analysis

BMC Medical Research Methodology

ISSN: 1471-2288

• Submission enquiries: [email protected]
• General enquiries: [email protected]

• Newsletters
• Commentary & Analysis
• International
• Chapter 11 Filings
• Distressed Company Screener
• Firm Retention Summary
• Bonds, Loans and Ratings
• Agriculture
• Entertainment
• Manufacturing
• Mining & Natural Resources
• Oil & Gas
• Regulation/Government

This copy is for your personal, non-commercial use only. Distribution and use of this material are governed by our Subscriber Agreement and by copyright law. For non-personal use or to order multiple copies, please contact Dow Jones Reprints at 1-800-843-0008 or visit www.djreprints.com.

https://www.wsj.com/articles/helicopter-ambulance-operator-air-methods-to-shed-debt-in-chapter-11-1ba2809a

Helicopter Ambulance Operator Air Methods To Shed Debt in Chapter 11

A majority of creditors have agreed to a proposed reorganization that would cut debt to $553 million from $2.24 billion.

Private-equity backed helicopter ambulance business Air Methods has filed for bankruptcy, planning to cut its debt load by $1.7 billion amid struggles with inflation, reimbursement delays and severe weather.

The Greenwood Village, Colo.-based company sought protection from creditors Tuesday in the U.S. Bankruptcy Court in Houston with a proposed restructuring agreement in hand with majorities of secured lenders and unsecured bondholders to cut debt load to $553 million from $2.24 billion.

Copyright © 2023 Dow Jones & Company, Inc. All Rights Reserved. 87990cbe856818d5eddac44c7b1cdeb8