Chapter 9: Qualitative Methods

Welcome to Last Minute Lecture.

This free chapter overview is designed to help students review and understand key concepts.

These summaries supplement, not replace, the original textbook and may not be redistributed or resold.

For complete coverage, always consult the official text.

So imagine trying to measure human grief with a thermometer.

Good luck with that.

Right.

Or like trying to graph the exact statistical percentage of hope in a local community.

It doesn't work.

It's impossible.

Yeah.

Some of the most important questions we can ask about the human experience simply cannot be answered with a calculator, a standardized survey, or a sterile laboratory environment.

And yet, there is this lingering expectation that rigorous academic research requires mathematical precision.

It has to be engineering.

Exactly.

You run a survey, crunch the numbers in a statistical software program, and the data just spits out a clean, hard percentage.

Which is deeply comforting.

I mean, we like to look at a spreadsheet.

We like our information to be visible, binary, and neatly categorized into charts and graphs.

But the moment you decide to study human behavior in the real world, that clean numerical machinery just vanishes.

It does.

You're stepping into a methodological landscape that is inherently messy.

It requires placing the researcher's own deeply human, entirely subjective perspective right at the front and center of the scientific process.

So welcome to this special last minute lecture deep dive.

If you're a college student furiously studying for your research methods exam or trying to put together a research proposal that actually makes sense, pull up a chair.

You are in the exact right place.

Definitely.

Today, we are looking at chapter 9 on qualitative methods from the sixth edition of research

Qualitative, quantitative, and mixed methods approaches.

And my goal today is to help you understand the overarching architecture of a qualitative study.

We aren't just going to memorize a list of vocabulary words here.

Because that's what everyone tries to do, right?

Just flash card.

Right.

And it doesn't work.

We are going to look at why each methodological step logically demands the next.

We'll break down the philosophical foundations, the research frameworks, the sampling strategies,

and the data analysis.

All following the exact order of your text.

Exactly.

Because once you see how a foundational assumption dictates your design choices and how those choices dictate how you collect data, this entire chapter transforms from a dense academic textbook into a highly practical roadmap.

OK, let's unpack this.

Because before you can even begin to build a study, you have to understand the philosophical assumptions of the methodology.

You have to know what game you're playing.

Right.

So the chapter opens by defining the nine core characteristics of qualitative research.

And the very first one completely flips the script on traditional quantitative lab work.

It really does.

Qualitative researchers go into the field, the natural setting.

You weren't bringing subjects into a controlled environment.

No, you're going to where they actually experience the issue to gather their meanings, not what the academic literature tells you they should be feeling.

And that requires a massive shift in how you view the tools of research, right?

In quantitative work, your instrument is like a standardized survey or a blood pressure cuff.

But in qualitative work, the text emphasizes that the researcher is the key instrument.

Wait, OK, I have to push back here immediately.

Go for it.

If I am the instrument gathering the data,

doesn't that make the entire process fundamentally subjective and, well, biased?

It sounds like it, doesn't it?

Yeah, it feels like being an investigative journalist moving into a neighborhood to write an expose.

Rather than just flying an unfeeling drone over it.

Exactly.

My own baggage, my preconceived notions, they have to taint the data I'm collecting.

Well, qualitative research does not run away from that bias or pretend it doesn't exist.

It leans into it through a methodological characteristic called reflexivity.

Reflexivity, OK.

Yeah.

The text mandates that you must explicitly state your biases, your values, your personal background, like your gender, culture, socioeconomic status, and outline exactly how they might shape the study.

So you don't just write, like, a one -sentence confession at the end of the paper?

No, no.

You actively document how your worldview might influence your interpretations by writing memos, which are reflexive notes you take during the entire data collection process.

You're taking a potential weakness human subjectivity and turning it into rigorous, transparent context for your reader.

Oh, wow.

That actually explains the text's severe warning against something called backyard research.

Yes.

The backyard research warning is huge.

Because the book explicitly warns against studying your own workplace, your own organization, or your close friends.

And I was thinking, wouldn't that just be super convenient?

It is highly convenient to just study the people in your own office.

But it's incredibly risky methodologically.

Right, because if you're a manager studying your own employees, there's an inherent power imbalance.

Exactly.

The data is compromised from day one because your participants might not tell you the truth.

And your reflexivity is blinded by your proximity to the subjects.

You're too close to see your own biases.

Precisely.

So once you accept these foundational characteristics, that your research must be naturalistic, that your design will likely emerge and change midstream, and that you must be deeply reflexive, you have to choose a design structure that actually accommodates those realities.

And the text uses a visual diagram to break down these approaches.

Figure 9 .1.

Right, so if you picture a pyramid of qualitative methods

at the wide foundational base of that pyramid, you have descriptive methods.

So in a descriptive method, you're staying incredibly close to the everyday literal language of the participants.

Exactly.

You use maximum variation in your sampling to get a wide variety of voices.

And you catalog the raw data into overarching themes without trying to force it into some preexisting theoretical box.

Got it.

And then sitting on top of that descriptive base, pointing upward, are the analytic frameworks.

These go beyond just everyday language and use deeply established academic traditions to view the data through a specific lens.

OK, the text lists five major ones.

Can we use a hypothetical example to see how they differ?

Sure.

Let's say you want to study college student burnout.

Very relevant for our listeners.

Very.

So if you choose a narrative study,

you are going to focus intensely on one or two students gathering their stories and restoring their lives into a chronological narrative with a plot and setting.

So showing the exact sequence of events that led to their burnout.

Exactly.

But if you choose phenomenology, you're doing something completely different.

You're looking for the core essence of a shared experience.

So I'd interview maybe 10 different students from completely different backgrounds who all experience severe burnout.

And I distill those interviews down to figure out what the universal fundamental feeling of burnout actually is.

Bingo.

Exactly the same topic, completely different design.

Then you have grounded theory, where you actually try to generate a brand new psychological or sociological theory from your data because one doesn't exist yet.

So I'd build a theoretical model of how burnout develops step by step.

Right.

Now, if you used ethnography, you would immerse yourself in a culture sharing group, maybe living in a specific fraternity house for a semester, to see how their shared cultural norms contribute to or prevent burnout.

Oh, that's fascinating.

And the last one is case study.

And a case study involves placing a strict boundary around your research.

You aren't studying all burnout everywhere.

You're providing a detailed, localized description of one specific burnout intervention program on one specific campus during one specific semester.

OK, looking at how this pyramid is built with descriptive methods at the bottom and these complex analytic frameworks on top, it's really easy to assume the descriptive method is just like the light version of qualitative research for beginners and the analytic frameworks are the pro version.

That is a huge misconception.

The recent American Psychological Association standards explicitly recognize the descriptive method as a distinct, highly credible method in its own right.

So it's not just training wheels.

Not at all.

It is not a lesser approach.

It is incredibly valuable when a researcher wants to stay exceptionally close to the raw data, purely coding and generating themes directly from the participant's mouths without filtering it through the heavy, complex lens of a tradition like phenomenology or grounded theory.

OK, but to do any of that, you actually need the raw material.

How do we get the data?

That brings us to section three of the chapter.

It shifts from design into the reality of sampling, permissions, and collection.

And right out of the gate, the text emphasizes purposeful sampling.

You aren't randomly picking 1 ,000 people to get a statistically significant margin of error for a mathematical formula.

No, you are purposefully hand -picking specific sites and individuals who actually possess deep knowledge of the problem you're studying.

Which immediately raises the question every student asks, how many people do I need to interview?

Always the first question.

And the text offers historical estimates.

You know, maybe one or two for a narrative, three to 10 for phenomenology, 20 to 30 for grounded theory.

But the ultimate methodological rule comes from a researcher named Sharmaz who developed the concept of saturation.

Yes, saturation is key.

And saturation makes so much sense when you think of it like eating at a Vegas buffet.

Oh, I love this analogy.

Right.

You walk in, and everything looks amazing, so you keep putting different things on your plate.

But eventually, you've sampled the carving station, the pasta bar, the desserts.

You reach a point where absolutely nothing on the buffet looks new or interesting anymore.

You're full.

Exactly.

You stop collecting data when gathering fresh data no longer sparks new insights or reveals any new themes.

You've heard the same stories enough times that you are officially saturated.

But before you can even begin filling that plate, you have to navigate permissions.

Institutional Review Board, or IRB, approval is mandatory to ensure human subjects are protected.

And the text talks about dealing with gatekeepers.

Yes.

These are the individuals, a hospital administrator, a school principal, a tribal leader, who literally control access to the physical site you want to study.

So you can't just walk in.

You typically have to present them with a brief proposal answering crucial logistical questions, like will this study be disruptive to our daily operations?

And frankly, what will our organization gain from letting you in here?

OK, so let's say the gatekeeper grants you access.

The text provides a massive compendium of data collection types in tables 9 .2 and 9 .3.

There are a lot of options.

You've got interviews, which can be face -to -face over the phone, or in focus groups.

You've got documents, both public records like meeting minutes and private documents like personal journals.

You also have audio, visual, and digital materials.

Yeah, the text even mentions collecting ambient sounds, recording something like a child's laughter or the chaotic honking of car horns to build the setting.

But I want to linger on the observations category for a second.

OK, let's do it.

The text says a researcher can be a complete participant where they actively conceal their role as a researcher for the people they are studying.

Yes, that is an option.

Methodologically speaking, isn't a complete participant observer basically an undercover spy?

How on earth do you get an academic board to approve that?

Well,

it's tough.

Qualitative research frequently deals with vulnerable populations in sensitive environments.

If you're proposing to conceal your role to capture truly authentic, naturalistic behavior,

the IRB is going to scrutinize your methodology intensely to ensure absolutely no harm comes to the participants.

So that's why the chapter places such a heavy emphasis on ethical procedures.

Exactly.

Ethics in qualitative research isn't just a hurdle to jump over.

It's the foundation of validity.

It means protecting privacy through pseudonyms, building authentic trust, using bias -free language, and ensuring the participants eventually get credit and ownership of the final results.

OK, so to keep all of this ethical and organized in the field, you need strict protocols.

Figure 9 .2 shows a sample interview protocol, and it's not just a casual list of questions jotted on a notepad.

No, it has a rigid architecture.

You start with an introduction where you secure informed consent and explain the structure of the interview.

Then you move into the content questions, which are careful sub -questions exploring the phenomenon.

But crucially, the text says you must sprinkle these with probes.

Probes?

Yeah, these are reminders written directly on your protocol page, like, tell me more about that, or I need more detail.

Here you use them to force yourself to dig deeper when a participant gives a shallow answer.

Oh, that's smart.

And you also need an observation protocol.

The text advises drawing a dividing line straight down the middle of your page.

The left side is for descriptive notes, the raw objective documentation of exactly what is happening in the room.

Just the facts.

Right.

And the right side is for your reflexive notes, your own hunches, emotional reactions, and emerging prejudices as the observation unfolds.

So this physical separation on the page ensures you don't confuse your subjective feelings with the objective events.

OK, let's fast forward.

You spent six months in the field.

You have 40 hours of interview audio,

200 pages of observation protocols, personal journals, and photographs.

You have a mountain of raw data.

It's daunting.

Where do you even begin?

It feels completely overwhelming.

The chapter outlines seven steps of data analysis in Figure 9 .3, but it prefaces them with a massive paradigm shift.

Simultaneous procedures.

Right, because in quantitative research, the timeline is linear.

You collect all your data, you lock it in, and then you run the analysis.

But in qualitative research, analysis happens continuously while you are still collecting data.

You are writing reflexive memos and analyzing your very first interview while you're simultaneously calling to schedule your third interview.

So your early analysis actually informs what you ask the next person.

Exactly.

Now, because you're dealing with so much dense text, the book says you have to do something called winnowing the data.

Yes, winnowing.

You essentially have to discard or set aside huge portions of your raw data to narrow it down to a manageable five to seven major themes.

But let me stop here.

If I'm throwing out data that doesn't neatly fit into my five to seven themes, isn't that just cherry picking the evidence I like?

I mean, if I'm studying burnout and one student tells me they actually feel incredibly energized and love their classes, do I just delete their transcripts so my burnout themes look cleaner?

That is the exact trap novice researchers fall into, and it absolutely destroys the integrity of the study.

I knew it sounded sketchy.

Winnowing is about distilling density, not ignoring reality.

If you find data that completely contradicts your emerging theme, you do not throw it out.

Keeping contrary data, what the text officially calls negative or discrepant information, is actually a core strategy for ensuring your study is valid.

So presenting the contradictions proves you're capturing the messy reality of the natural setting, not just proving your own assumptions.

Precisely.

Now with that in mind, let's look at the engine of this whole process, coding.

The seven steps of data analysis.

Steps one and two are just organizing the data -like transcribing interviews and reading through it all to get a general sense of the tone.

But step three is where the magic happens, coding the data.

Coding is essentially a translation mechanism.

It turns messy human emotion into categorized data architecture.

And the text outlines a specific eight -step coding process developed by a researcher named Tesh.

Yes.

Practically speaking, you read through a transcript, pick one interesting document, and ask yourself, what is this fundamentally about?

You list all the topics you find, cluster similar topics together, and abbreviate them into short codes.

And then you go back and write these codes directly next to the corresponding text segments.

Exactly.

And the book breaks down the types of codes you might discover during this translation.

Right.

You have expected codes.

If you're studying school bullying, you fully expect to find a code about attitudes toward oneself.

You also have surprising codes that you never anticipated.

And you have unusual or conceptual codes.

I love this.

The text cites a brilliant example from a real study of a campus response to a gunman.

The researchers discovered the code re -triggering.

Yes.

Where the current lockdown event vividly reminded individuals of completely unrelated past traumas, it was an unusual, deeply conceptual code that evolved into a major theme.

That's powerful.

It is.

Now, some disciplines, particularly the health sciences, might use a qualitative codebook with predetermined codes derived from existing theory, checking to see if the new data matches the old model.

But regardless of how you code, steps four through seven involve grouping those codes into your five to seven major themes, interconnecting those themes into a storyline interpretation, applying your analytic framework if you're using one, and finally, representing and interpreting the data in your narrative.

Which brings us to a huge hurdle.

You've coded the data, you've built your themes, and you've written your interpretation.

But how do you prove to an incredibly skeptical academic review board that you didn't just invent these themes in your head?

You need a system to prove your findings are accurate.

Exactly.

The text outlines eight specific strategies to check the validity or accuracy of your findings.

The most powerful is triangulation.

Think of a GPS system on your phone.

One satellite might give a vague estimate of your location.

But if you pull data from three different satellites, they triangulate to pinpoint your exact coordinates.

Qualitative triangulation works the same way.

If a personal interview, a private journal entry, and a public document all point to the exact same theme, that theme is rock solid.

Another validity strategy is member checking,

which honestly sounds terrifying.

It can be.

You take your final themes and written descriptions back to the actual participants and ask them, did I get this right?

You're actively inviting them to correct your interpretation of their own lives.

It's humbling, but crucial.

You also use rich, thick description to vividly transport the reader to the setting, making the findings undeniably resonant.

We already discussed clarifying your bias in presenting discrepant information.

You should also spend prolonged time in the field to build trust.

Yes, and you can rely on peer debriefing, having a colleague heavily critique your work, or even hire an external auditor who has no connection to the project to objectively assess if your logical leaps from raw data to final theme are sound.

Okay, so that's validity.

What about reliability?

Making sure your approach is consistent.

For reliability, the text suggests checking your transcripts for obvious typing mistakes and writing memos to prevent code drift.

Code drift.

That's when your internal definition of a code subtly changes over six months of analysis.

Exactly.

You also utilize intercoder agreement.

This is highly practical.

You and another researcher code the exact same transcript independently, compare notes, and see if you arrived at the same codes.

And the text notes that researchers Miles and Huberman recommend aiming for at least 80 % agreement.

Yes, but here's where it gets really interesting.

Qualitative generalization.

Right, I might spend six months studying this one specific ER, but qualitative research doesn't even try to say my findings apply to the whole country.

Why am I even doing it then?

I know, this is a massive paradigm shift for students accustomed to quantitative rules.

The goal of qualitative research is deep localized understanding.

The hallmark of good qualitative research is particularity, not generalizability.

Particularity.

Yes, you are trying to understand the intricate dynamics of this specific hospital, this specific culture sharing group, this specific phenomenon.

So you aren't trying to solve the problem for the whole world.

No, the only time generalizability really applies is through replication logic and case studies, where you repeat the exact same study methodology over and over again in dozens of new settings to see if the initial findings hold up across different environments.

Which perfectly sets up the final step.

Writing the report, the chapter concludes by showing us how to synthesize all these distinct procedures into a final academic narrative.

And it provides a fantastic sample method section from a 1992 study by a researcher named Miller.

The Miller case study is the perfect capstone for you, the student listener, because it applies every single rule we just discussed in real time.

She conducted an ethnography studying the first year experiences of a college president.

And you can see the textbook rules come to life in her writing.

She declares her bias right up front in the section titled the researcher's role, explicitly admitting she used to be a college administrator herself.

So she brings specific assumptions to the table.

She strictly bounds the study by defining the exact setting, the specific actors, the events, and the processes she will observe.

But you know what stands out to me the most about the writing strategies the text recommends for the final report?

What's that?

The sanctioned use of the first person I.

It is so incredibly refreshing to read an academic paper that says I observed this dynamic instead of the robotic passive the researcher observed.

It's like breaking the fourth wall in a movie.

It is refreshing, but it is also a methodological necessity.

Using the word I reflects that core philosophical assumption we started with.

The researcher is the instrument.

The writing style has to match the epistemology.

You are openly acknowledging your physical and interpretive presence in the natural setting.

Exactly, and to balance that, the text emphasizes weaving in short to long embedded quotes directly from the participants, ensuring their literal voices and meanings remain the central focus of the narrative, not just your analysis of them.

So if you are looking at your study guide right now, here is the life cycle of the chapter from top to bottom.

You start with your philosophical assumptions, embracing natural settings and committing to reflexivity.

That understanding guides you to pick a structural framework, whether it's the foundational descriptive method or an analytic one like ethnography or grounded theory.

That choice dictates your purposeful sampling to reach saturation.

You gain access, then collect data using carefully structured protocols.

You simultaneously code and winnow that massive pile of data using translation steps like Tekka's process.

You scientifically prove the validity of those themes through triangulation, member checking, and discrepant information.

And finally, you write it all up using rich, thick description, embedded quotes, and the first person I have.

It is a beautifully logical sequence once you strip away the academic jargon.

Qualitative design is a rigorous system built specifically to honor the messy complexity of human experience without ever sacrificing scientific credibility.

It really is.

But before we let you go back to your studying, we wanna leave you with a final provocative thought,

something to mull over that builds on the foundation of this chapter, but looks directly at the future of the field.

As we move rapidly into an era of artificial intelligence, we are seeing new software tools that claim they can analyze, translate, and code qualitative interview data in a matter of seconds.

Right, so if an AI analyzes and codes our qualitative data for us, what happens to the core tenant of the researcher as the key instrument?

Does handing the coding process over to an artificial intelligence actually remove human bias, making the study cleaner?

Or does it simply hide that bias deep inside an unreadable black box algorithm effectively destroying the very human reflexivity that makes qualitative research valid in the first place?

Something to think about as you review your notes.

Thank you for studying with the last minute lecture team.

Good luck in your exam, you've got this.