Qualitative methods can be used for interpretive, theory-building research projects. The qualitative data used in such projects may consist of pictures, artwork, words, audio files, and similar non-numerical information. This chapter begins with an overview of two ways to analyze qualitative data: the grounded theory approach and thematic analysis. The focus then shifts to practical techniques for preparing various kinds of data, such as fieldnotes, interview transcripts, and audio files. Next, the stages of coding qualitative data and identifying themes are outlined. The chapter concludes with a discussion on credibility and trustworthiness.

While this chapter will provide you with clear steps you can use to analyze your qualitative data, it is important to keep in mind that analyzing qualitative data is not like using a recipe book; however, you can think of it as a recipe you can play around with and change as you see fit. As Renata Tesch says, “qualitative analysis can and should be done artfully, even ‘playfully,’ but also requires a great amount of methodological and intellectual competence” (quoted in Hesse-Biber & Leavy, 2011, p. 302). Ideally, qualitative analysis should also be done in close relationship with the community one is working with. This is especially true when working with Indigenous communities, where research has long taken the form of unreciprocated extraction by settler researchers (Kovach, 2021).

A Human and Iterative Process

Qualitative data analysis has an inherent human component: it is always performed by a person, and all people occupy different social positions that shape their perspective of the world (Braun & Clarke, 2006, 2021, 2023). Before beginning data analysis, consider reflecting on your own intersecting privileges and exclusions. How may these identities (gender, class, ethnicity, etc.) impact how you analyze data? Journaling can help you keep track of how your social positionality shapes your research throughout your project. Journaling can also help relieve anxiety as it provides a place for you to vent your anxiety and frustrations – feelings we all experience while conducting research.

Note that your analysis will not emerge all at once so be sure to write yourself memos as you go. These memos can be jotted down at any time, and sometimes our best insights come to us while walking alone in the woods. As discussed in chapter 10, the theoretical/analytical notes you took during your research can provide a good starting point for thinking about possible codes and serve as early instances of memos. Continue to write yourself memos throughout your project so that you don’t forget any of your good ideas. It’s also helpful to date these memos so you can remember when and why you thought the way you did.

Qualitative data analysis is an iterative process that occurs throughout the research process, which is why memo writing is so valuable. By “iterative,” we mean that the researcher is constantly looping back on previous stages. Data collection does not simply lead to a period of analysis where we forget about collecting data anymore; data analysis also shapes (and reshapes) the data collection, research design, and research questions. Unlike surveys, which are difficult to change once they are distributed, qualitative researchers revisit early stages again and again in a spiral or looping process.

Ultimately, the goal of data analysis is to reach some inferences, lessons, or conclusions by condensing large amounts of data into relatively smaller, more manageable bits of understandable information. Each type of qualitative data requires somewhat different analytic techniques, and it is impossible to cover every type of data in this textbook. Instead, we’ll focus on the most common types of data: fieldnotes, audio recordings of interviews, talking circles or focus groups, and texts for content analysis (see chapter 10 for a more detailed discussion of content analysis). Following the discussion on preparing the data, we will review the ways to code the data and, lastly, ways to analyze common themes that result from the coding.

Setting the Stage

Qualitative methods such as focus groups, talking circles, field research, and qualitative interviews generate enormous amounts of data. In focus groups, the researcher might end up with audio or video recordings of the group discussions, fieldnotes about participants’ interactions, pieces of artwork created during the focus group, and demographic information about participants. In field research, the data might be in the form of fieldnotes, audio recordings of interviews, and texts gathered for content analysis. Regardless of which forms of qualitative data the researcher ends up with, they all need to be prepared for analysis. While many different analytic strategies can be used by qualitative researchers, we will focus on two in particular: the grounded theory approach and thematic analysis.

The grounded theory approach (Charmaz, 2006; Glaser & Strauss, 1967) occurs, as you might imagine, from the “ground up,” with the researcher beginning with an open-ended and open-minded desire to understand a social situation or setting. The process of using a grounded theory approach to analyze data involves a systematic process whereby the researcher lets the data guide the inquiry rather than guiding the data using preset hypotheses. Grounded theory is an inductive approach because the researcher is looking to develop theory directly out of the data rather than testing out an already-existing theory (see chapter 3 for more on the inductive approach). The goal of a grounded theory approach is to generate theory that is grounded in the data. Its name implies not only that discoveries are made from the ground up but also that theoretical developments are grounded or rooted in a researcher’s empirical observations and a group’s tangible experiences.

Grounded theory is the older of the two approaches covered here, and there has been some debate over how much someone can really approach their data without any preconceived theoretical ideas (Corbin & Strauss, 1990). For our purposes, you can think of grounded theory as a more inductive approach that is attempting to develop concepts and theories that are firmly grounded in the data themselves.

As exciting as it might sound to generate theory from the ground up, this experience can also be intimidating and anxiety-producing. Without hypotheses to guide their analysis, as is typically the case in quantitative research, researchers engaged in grounded theory work may experience feelings of frustration or angst. At the same time, the process of developing a coherent theory grounded in empirical observations can be quite rewarding, not only to researchers but also to their peers, who can contribute to the further development of new theories through additional research, as well as to research participants, who may appreciate getting a bird’s-eye view of their everyday experiences.

Thematic analysis is similar to grounded theory in terms of the analytical process, but it is not restricted to generating new theory; rather, thematic analysis can be used to refine existing theories and/or draw upon theories to help identify and illuminate themes within your data. Thematic analysis is therefore more flexible than grounded theory in that it can be more deductive (i.e., guided by theory) or inductive (i.e., themes are developed from the data themselves), while grounded theory is more clearly inductive. Thematic analysis is also newer than grounded theory, being a 21st-century approach, and it draws on practices developed in that tradition. While grounded theory developed in sociology, thematic analysis emerged from psychology; both are used by criminologists and criminal justice researchers today.

When doing thematic analysis, a researcher first familiarizes themselves with the data, generates a set of codes, and then identifies themes that emerge from the data (Moustakas, 1990). Unlike grounded theory, the codes themselves can be shaped more by the theoretical perspective used by the researcher rather than strictly from the data themselves. When conducting Indigenous research, thematic analysis is particularly valuable as the work can remain rooted in an Indigenous theoretical perspective (Kovach, 2021). It can also be used by other researchers who are interested in developing existing theory, for example, when used by feminist researchers (Yercich, 2021).

As mentioned at the outset of this chapter, there are many ways to analyze qualitative data. Each approach/researcher has their own take on the process and how it could be undertaken. A researcher who is doing discourse analysis, for example, might proceed with the same steps described here, starting with open coding and moving to more focused coding, all the while focusing on how language and power create and shape social reality (see Ashley, 2014). A researcher interested in narrative analysis might likewise use open and focused coding to identify how people construct stories to make sense of their own experience (see Sandberg, Tutenges, & Copes, 2015). Whatever your approach, and there are many others, the steps for coding described below can help you organize and make sense of your data.

Preparing the Data

Writing Fieldnotes

Analyzing field note data is an iterative process that begins the moment a researcher enters the field. Proper fieldnotes (i.e., the ones you write up at your desk) are often developed from jottings taken in the field, though some field settings allow researchers to take more detailed notes on the spot. Fieldnotes are the fully developed, rich, thick descriptions you produce when you exit a given observation period and sit down at your computer to type your observations into a more readable and formal format.

We’ve already noted that carefully paying attention while in the field is important; so too is noticing what goes on immediately upon exiting the field. Field researchers typically spend several hours typing up fieldnotes after each observation. During this process of creating and preparing the data for analysis, the researcher also begins analyzing their data. Once they are out of the field, researchers take time to reflect on their experiences in the field and what their observations might mean.

Transcribing Audio Recordings

The analysis of audio data typically begins with transcribing the audio into written form. When you transcribe an audio file, you create a complete written copy of the recording by playing the recording back, typing in each word spoken on the recording, and noting who spoke which words. In general, researchers aim for a verbatim transcription that reports everything said in the recording exactly as the speakers said it. In addition to the words spoken, a verbatim transcription should also include verbal cues, such as laughing, and filler words (e.g., uh’s, um’s) as well as notes on the tone of voice and when and how respondents emphasized specific spoken words. It’s very important that the punctuation the transcriber uses maintains the meaning of what is said – for example, “I love hunting, my kids, and my dogs”  is quite different from “I love hunting my kids and my dogs.”

Transcribing audio files can be very time-consuming. Some researchers pay for transcription services, while others transcribe audio themselves. On average, it takes four to five times longer to transcribe interview data than to listen to it. That means that a one-hour interview could take five hours to transcribe! Despite the time it takes to transcribe audio files, it is worthwhile to do it yourself. When researchers transcribe their own files, they become immersed in the data and their patterns. Listening to conversations you participated in or observed may spark recall of nonverbal cues or other interactions that you may have forgotten to include in your fieldnotes. These can contribute to richer data and aid in more robust analyses.

Preparing Texts

Preparing texts for analysis really depends on the type of texts the researcher has collected (e.g., social media posts, archival records, movie scripts). Texts should be compiled and organized in a way that makes sense for the aims of the research (e.g., by source, theme, year, or type of text). This may include organizing text into tables, collecting screen captures into a series of folders, organizing notes on archival exhibits into physical piles, or uploading interview transcripts into qualitative data analysis software.

Coding Qualitative Data

Once the researcher has prepared their qualitative data for analysis, they begin looking for patterns across the data by reading through their data files and trying to identify codes. A code is a word or short phrase representing some complex set of issues or ideas (Saldaña, 2016). The process of identifying codes in one’s data is often referred to as coding. Coding involves looking for patterns across data by reading and rereading (and rereading again) the data until the researcher has a clear idea about what sorts of themes come up across the data points. One might say you are searching and re-searching for codes in your data. As Kovach (2021, p. 209) reminds us, we must also be sensitive to coyote codes – outliers that don’t fit with the rest of our themes. Coyote codes are important and can contribute to the nuance and complexity of our analysis.

As you might imagine, wading through all these data can be a challenging process. Luckily, some computer programs can help qualitative researchers sort through, code, and analyze their data. Programs such as NVivo and ATLAS.ti are specifically designed to assist qualitative researchers with organizing, managing, sorting, and analyzing large amounts of qualitative data. The programs work by allowing researchers to import electronic documents and then label passages with codes, cut and paste passages, search for various words or phrases, and organize complex interrelationships among passages and codes, as shown in Figure 13.2. You can also code manually using highlighters or coloured pencils to mark off different sections. Or, you can do the same with Microsoft Word, Pages, or LibreOffice, as all have highlighting and comment functions.

Start your coding process by reading through all the data to get a general sense of what you have collected. After you’ve read through your data at least once, you can begin to assign initial codes. At a basic level, coding means identifying words or phrases that we can then use to “categorize chunks of the data so that we can work with them” (van den Hoonard, 2019, p. 174). This involves reading through your data carefully and coding each line or sentence with a descriptor.

This first stage is known as open coding. It usually begins with no pre-determined codes and requires the analyst to generate new codes throughout the coding process, which is why it is known as “open” coding. These codes are generally more descriptive and can include words that appear within the text itself. During open coding, researchers try not to let their original research question or expectations about findings influence the categories or themes they see. Open coding usually requires multiple go-rounds so that researchers can be sure they’ve identified all the possible codes they can think of and find.

If you are doing thematic analysis, the process of open coding may be more deductive than if you are doing grounded theory. If you are using a more deductive approach, one that is guided more by theory, your open coding might be guided more by the theory you are using (feminism, Indigenous, critical, etc.), though at this stage, your coding should still remain as close to your data as possible, with theory shaping your analysis in later stages of the coding process (i.e., focused coding, as described in the next section).

You Have Some Codes. Now What?

As researchers pore over their data, they begin to see patterns or commonalities. Once you begin to see some patterns, you can begin more focused coding. Focused coding involves collapsing or narrowing codes identified in the open coding process by reading through the notes and memos again. This process can involve identifying codes that seem to be related, and you might start merging some codes that seem too similar to warrant their own unique code. Over time, these mergers may create categories that might contain multiple codes. During the coding process, the researcher might also create a codebook, or a document that includes brief definitions or descriptions of each code. The codebook can help the researcher go back through their data to ensure they have accurately marked passages with the relevant codes. See Codebooks in Qualitative Content Analysis for examples of codebooks:

Let’s look at an example from a research project conducted by Ashley (2014) on the British Columbia polygamy reference that took place in 2011. Ashley knew that this case would be significant, but at the outset it was not clear how the arguments would be made or what exactly the court would decide. An open-coding approach was therefore used to code all the court transcripts. With the assistance of qualitative analysis software (TAMS analyzer), he conducted a line-by-line coding of the transcripts to create a set of codes that described what was being said in court. As you can see from the table below, these codes included “consequences of criminalization,” “fear of social service,” “psychological stress,” and “risks of insularity.” Ashley then moved on to more focused coding, which ended up creating nested hierarchies where “consequences of criminalization” became its own larger category, while “health risks” and “fear of social services” became subcategories. The following examples from the court transcript reflect words spoken by the lead lawyer for the BC Attorney General’s office:

You can see from this example how the coding process can help organize thousands of pages of data while providing a potential avenue for analysis.

Identifying Common Themes

As you start to compare and cluster various categories, you will start to see themes emerging from your data. Depending on your purpose, these themes may be more or less inductive. In grounded theory, the themes ideally develop into new theoretical insights that are closely tied to the data themselves (i.e., a more inductive form of analysis). For researchers who are interested in developing and testing out pre-existing theories, themes may be more deductive, reflecting questions and concerns of the theoretical tradition being used.

Regardless of one’s approach, the researcher’s subjective interpretation is always intertwined with data analysis and the entire research project (Braun & Clarke, 2021). While engaged in reflexivity, the researcher makes meaning of codes by generating initial themes, developing and reviewing themes as well as refining, defining, and naming themes (Braun & Clarke, 2019). Although these are described here in sequence, researchers will likely move through these stages of data analysis in a non-linear fashion: generating initial themes, developing and reviewing themes, and refining themes.

Generating Initial Themes

Once the researcher is ready to move beyond coding, they enter the first stage of theme development. Generating initial themes has some of the same traits of focused coding as it involves analyzing codes and grouping them into clusters of candidate themes. Candidate themes aim to tentatively explore meaning in relation to the research question(s) and the entire dataset. To help with this stage and future stages, Braun and Clarke (2021) suggest using thematic mapping. Thematic mapping refers to creating visual maps of candidate themes to explore relationships, interconnections, and meanings between candidate themes.  Below is a thematic map of the human-animal bond describing the different themes that arose from research interviews. The square boxes represent the main themes, with arrows indicating the sub-themes contained within. For example, under the theme of “Unconditional Love” are the sub-themes of no judgement and always happy to see you. The thematic map also helps to illustrate the connections between themes, as indicated by dotted lines. In the map, the themes of “Responsibility” and “Burden” are connected, as the same sub-themes of time, finances, and energy appeared in both themes. We can also see that the sub-themes may arise in more than one theme, such as the sub-theme of they saved me being contained in both the themes “Rescuing” and “Good for People.” In short, a thematic map helps to visually identify and make sense of connections between themes and sub-themes, allowing you to try out different connections, and helping you to deeply analyze your data.

Developing and Reviewing Themes

Once candidate themes have been identified, researchers need to develop them and check their accuracy. At this stage, candidate themes may be kept, changed, renamed, split up, combined, or discarded. When reviewing themes, ask yourself these questions: Is this pattern a viable theme? Can I identify the boundaries of this theme? Am I clear about what it includes and excludes? Is there enough (meaningful) data to support this theme ? Does this theme convey something important (Braun & Clarke, 2021, p. 90)?

Refining, Defining and Naming Themes

This stage continues the work of the previous with more finality. Here, themes are named, parameters are identified (where does one theme start and end?), and detailed analyses of the themes are presented (what is the core analysis of the theme?).

When Do I Stop?

It’s impossible to give clear guidelines on when to stop refining and working over your themes, but generally you can wrap it up when you aren’t finding anything new. That is, you can stop when refinements aren’t producing any new interpretations. This is known as saturation, the point where no additional issues, themes, or insights are being generated (Hennink, Kaiser, & Marconi, 2017).

Credibility and Trustworthiness

Determining if your interpretation is valid is different in qualitative analysis than in quantitative analysis. In qualitative research, as discussed in chapter 6c, validity speaks to how credible and trustworthy your conclusions are. Credibility and trustworthiness speak to how rigorous and systematic you were in your methodology and whether readers of your research believe you have established your credibility (Leavy, 2017). When reflecting on the validity of your study, ask yourself the following questions: Are you telling a convincing story? Have you reached your findings with integrity? Have you looked for and addressed contradictory cases or coyote codes? Are your interpretations available for discussion among participants and other knowledge holders? How do your findings impact those who participated in the research, and how do your findings impact the wider social context in which research occurred (Hesse-Biber & Leavy, 2011; Kovach, 2021)?

Guba and Lincoln (1989) explain this as follows:

The basic issue in relation to trustworthiness is simple: how can an inquirer persuade his or her audiences (including self) that the findings of an inquiry are worth paying attention to, worth taking account of? What arguments can be mounted, what criteria invoked, what questions asked, what would be persuasive on this issue? (p. 398)

Validation is therefore a process whereby we as researchers interrogate our own findings and test out our conclusions in dialogue with others. Considering questions of who owns the stories we are working with and how their interpretations shape a study is particularly important when working with Indigenous communities, though the question of how our research represents any community is important to explore whichever community we are working with.

Conclusion

The overall goal of data analysis is to reach some inferences, lessons, or conclusions by condensing large amounts of data into relatively smaller, more manageable bits of understandable information. The analysis process for qualitative data is not distinct from the preparation and coding stages; rather, analysis begins early and continues throughout the project. For example, while transcribing audio files, the researcher is also beginning to identify themes (coding) and make sense of those themes (analysis). Even creating a codebook is a way of making sense of data and developing a way to talk about the findings. Thus, the analysis of qualitative data occurs throughout the entire qualitative research process. Researchers conducting these types of studies begin analyzing the moment they start a focus group or interview, enter the field, or gather texts to analyze. When the researcher has prepared their data, identified codes, marked passages with those codes, developed definitions of each code, and the data have been condensed into manageable form, it is time to begin the important work of writing up these findings in ways that make sense to a larger audience – whether it be the community that participated in the study, a government agency, the academic community, or the public at large.

References

Ashley, S. M. (2014). Sincere but naive: Methodological queries concerning the British Columbia polygamy reference trial. Canadian Review of Sociology, 51(4), 325-342. http://doi.org/10.1111/cars.12050

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa

Braun, V., & Clarke, V. (2019) Reflecting on reflexive thematic analysis. Qualitative Research in Sport, Exercise and Health, 11(4), 589-597. https://doi.org/10.1080/2159676X.2019.1628806

Braun, V., & Clarke, V. (2021). Thematic analysis: A practical guide. Sage.

Braun, V., & Clarke, V. (2023). Toward good practice in thematic analysis: Avoiding common problems and be (com)ing a knowing researcher. International Journal of Transgender Health, 24(1), 1-6. https://doi.org/10.1080/26895269.2022.2129597

Charmaz, K. (2003). Grounded theory. In J. A. Smith (Ed.), Qualitative psychology: A practical guide to research methods (pp. 81–110). Sage.

Charmaz, K. (2006). Constructing grounded theory: A practical guide through qualitative analysis. Sage.

Corbin, J., & Strauss, A. (1990). Grounded theory research: Procedures, canons, and evaluative criteria. Qualitative Sociology, 13, 3-21. https://doi.org/10.1007/BF00988593

Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research. Aldine.

Grounded Theory Institute. https://www.groundedtheory.com/.

Guba, E. & Lincoln, Y. (1989). Fourth generation evaluation. Sage.

Hennink, M. M., Kaiser, B. N., & Marconi, V. C. (2017). Code saturation versus meaning saturation: How many interviews are enough? Qualitative Health Research, 27(4), 591-608. https://doi.org/10.1177/1049732316665344

Hesse-Biber, S. N., & Leavy, P. (2011). The practice of qualitative research. (2nd ed.). Sage.

Kimmerer, R. W. (2013). Braiding sweetgrass: Indigenous wisdom, scientific knowledge, and the teachings of plants. Milkweed Editions.

Kovach, M. (2021). Indigenous methodologies: Characteristics, conversations, and contexts. (2nd ed.). University of Toronto Press.

Leavy, P. (2017). Research design: Quantitative, qualitative, mixed methods, arts-based, and community-based participatory research approaches. The Guilford Press.

Moustakas, C. (1990). Heuristic research: Design, methodology, and applications. Sage. https://doi.org/10.4135/9781412995641

Saldaña, J. (2016). The coding manual for qualitative researchers (3rd ed.). Sage.

Sandberg, S., Tutenges, S., & Copes, H. (2015). Stories of violence: A narrative criminological study of ambiguity. The British Journal of Criminology, 55(6), 1168-1186. https://doi.org/10.1093/bjc/azv032

van den Hoonard, D. (2019). Qualitative research in action: A Canadian primer (3rd ed.). Oxford University Press.

Yercich, S. (2021). Fathers investing in fatherhood: A qualitative examination of contemporary fathering in fatherhood groups in Canada. [Doctoral dissertation, Simon Fraser University]. Summit Research Repository. https://summit.sfu.ca/item/34712