Grounded Theory: Making It Up As You Go Along (But With Integrity)

This week, in a break from stalking the usual parade of sociological ghosts—Marx, Weber, Bourdieu and the like—I thought we’d do something different. Let’s talk about something alive, terrifyingly current, and capable of causing existential dread in postgraduate students across the globe: methodology. More specifically, Grounded Theory.

Now, if you’ve never come across Grounded Theory, count yourself lucky—or suspiciously under-researched. It’s a methodology often used in qualitative research and best described as the sociological equivalent of jazz: structureless at first glance, yet with an insistence that there is, in fact, a deeply rigorous method behind the chaos. It’s like improvising your way through a conversation and then later claiming it was all part of a brilliant strategy. And the best part? You get to call it academic.

The Origins of the Madness

Grounded Theory was developed in the 1960s by sociologists Barney Glaser and Anselm Strauss, who bravely stood up and declared, “You don’t have to start with a theory. You can just go out, collect data, and let the theory emerge.” This was revolutionary at the time, an intellectual slap in the face to the traditional top-down, theory-first approaches that dominated research. It was also, as many PhD students quickly discover, a slippery slope into a methodological abyss.

In theory, it’s beautiful. Go into the world with no assumptions. Talk to people. Observe behaviour. Record data. Let patterns emerge organically, like sourdough starter. In practice, it’s the equivalent of being handed a blank page, a broken compass, and a half-charged Dictaphone and being told, “You’ll know where you’re going when you get there.”

Stage One: The Blank Notebook of Hope

Your first step is to enter the field free from theoretical baggage. You are a tabula rasa, an open vessel ready to absorb whatever rich, juicy data your participants throw your way. Except what usually happens is this: you sit opposite your participant, press record, and realise you have absolutely no idea what you’re doing.

They talk about their cat. You wonder if this is relevant. You write down “cat = comfort?” and hope it’ll make sense later. They shift to talking about their experience of community, and you furiously scribble “belonging, liminality, place-making” as if you didn’t just spend five minutes drawing a cartoon dog in the margins.

You leave the interview confused, emotionally fragile, and mildly lactose-intolerant from all the tea you drank while pretending to be relaxed. But you’ve got data. Raw, unrefined, confusing data. Let the games begin.

Stage Two: Initial Coding, or Why Do I Have 738 Themes?

Initial coding is where you break down the data into chunks, labelling each piece with a “code” to capture its meaning. Except if you’re new to this—and even if you’re not—you’ll soon realise that everything looks like a potential code.

Participant mentions being tired? That’s work-life balance. Mentions toast? That’s domestic ritual. Mentions loving trains? That’s autonomy, nostalgia, industrial pride, transport as metaphor. You write them all down, because you don’t know what’s important yet. You end up with a sprawling mess of codes, some so specific they only apply to one sentence, others so broad they could apply to anything ever said, anywhere.

And yet, you feel productive. Look at all this data! All these colourful little codes! Surely something good will come of this. You save your file, confident in the lie that you’ll tidy it up later.

Stage Three: Focused Coding, or The Collapse of the Will

Now that you’ve created an encyclopaedia of initial codes, it’s time to group them into meaningful categories. Except nothing wants to group neatly. You start trying to merge codes, then panic and create new ones instead. You stare at your thematic map wondering why “emotional resilience,” “banter,” and “sandwich anxiety” have all ended up under the umbrella of resistance.

At this point, the idea of a “theory” feels laughable. All you have is a list of words and some half-baked memos that read like stream-of-consciousness poetry written during a sugar crash.

You begin to wonder whether any of this is salvageable. But instead of quitting, you press on—because there is no grant funding for giving up.

Constant Comparison: Sociological Paranoia as a Method

One of the core tenets of Grounded Theory is constant comparison. This means you must constantly compare new data to existing data, and new codes to existing codes. Sounds reasonable, until you realise that it creates a form of academic hypervigilance.

Everything is now a potential code. Every utterance, gesture, sigh, or tea-making routine becomes a puzzle piece in a never-ending jigsaw. You can no longer speak to friends without mentally coding their speech. You begin comparing yourself to your own earlier self and worry you’re not being rigorous enough with your own contradictions.

In short, you become methodologically paranoid. But at least you’ve embraced the process.

Theoretical Saturation: When Even the Data Is Bored

Eventually, you hit that sacred milestone: theoretical saturation. This is the point at which no new data is adding anything new to your categories.

Or, in less formal terms: “If one more participant talks about belonging, I’m going to eat my fieldnotes.”

Whether or not you’ve truly reached saturation or have just become emotionally saturated is a separate issue. Either way, you write the line: “Theoretical saturation was achieved at Interview 12.” What you mean is: “I can’t face another transcript.”

It counts.

Glaser vs. Strauss: Academic Beef, Qualitative Edition

Of course, it wouldn’t be sociology without some inter-theorist drama. After co-inventing Grounded Theory, Glaser and Strauss had what can only be described as a methodological divorce. Glaser championed pure emergence—start from zero, let the theory rise like steam from a cup of peppermint tea. Strauss, slightly more organised, wanted structured coding paradigms and diagrams that resemble London Underground maps.

Which one is right? Nobody knows. But if you’re smart, you’ll mention both, hedge your bets, and claim you’re using a “pragmatic hybrid model.” Then sit back and enjoy the academic equivalent of saying “I’m a bit of both” in a Hogwarts house quiz.

NVivo: The False Hope of Order

No discussion of Grounded Theory would be complete without a nod to NVivo, the qualitative coding software designed to help manage your data. And by “help,” I mean “reveal the depth of your existential disorganisation.”

NVivo lures you in with its colour-coded nodes, tree hierarchies, and the seductive promise of order. Within hours, you’re lost in a digital jungle of overlapping codes and contradictory categories. You accidentally code an entire interview under “Miscellaneous” and pretend it was intentional.

You name one of your themes “Stuff.” You have a subcategory called “More Stuff.” You try to export a chart and the programme crashes. You do not cry. You simply return to your codes with dead eyes and the stoicism of someone who has accepted their fate.

Reflexivity: Yes, This Is About Me Now

Grounded Theory requires you to be reflexive—to acknowledge how your background, identity, and biases shape your research. At its best, this is a thoughtful and important practice. At its worst, it’s a self-indulgent descent into an academic diary:

“As a neurodivergent researcher with mild coffee dependency and an irrational fear of group emails, I brought a unique lens to the data.”

“My interpretation of the participant’s comments about ‘the break room’ may have been coloured by my own traumatic experiences with fluorescent lighting and passive-aggressive Post-it notes.”

Is it research? Is it therapy? We may never know.

What Do You Actually End Up With?

At the end of all this, what you (hopefully) have is a theory grounded in actual data—something original, emergent, nuanced. Or, if you’re less lucky, you have a bunch of quotes loosely clustered around concepts like “navigating uncertainty” and “sandwich culture” and a diagram you definitely made up at 2 a.m. while eating cereal.

And yet, there’s something oddly beautiful about it. You didn’t force a theory. You coaxed it out of conversations, moments, and lived experience. You let your participants shape the path. You resisted the urge to control, and in doing so, made peace with the mess.

You did Grounded Theory. And you lived to tell the tale.

Final Thoughts: Chaos, But Make It Sociological

Grounded Theory is, at its core, an act of intellectual optimism. It says: “Start with nothing. Observe. Listen. The theory will come.” It’s terrifying and liberating and almost always chaotic.

It’s the perfect method for people who enjoy wallowing in complexity, who prefer to see research as an unfolding mystery rather than a step-by-step guide. It’s a mess, yes—but it’s your mess. And somewhere in the middle of those coffee-stained transcripts and panic-coded NVivo nodes, something real does emerge.

Is it a theory? Maybe. Is it sociological? Almost certainly. Is it making it up as you go along?

Of course. But with integrity.