Do You Really Need a Framework? The Case for Starting with the Data
In research, there's a quiet pressure to appear certain. Before the first interview is conducted or the first transcript analysed, scholars are often expected to articulate the theoretical framework that will guide their interpretation. It sounds responsible. Rigorous. Necessary.
But here’s a contrarian thought… what if it’s sometimes premature?
What if the tidy framework we commit to at the outset ends up shaping what we’re allowed to see?
The Comfort of the Lens
Theoretical frameworks are helpful. They give us language, focus, and depth. They connect our work to broader scholarly conversations. They make sense of mess.
But they also carry assumptions. Any lens - whether it’s Foucault, grounded theory, or intersectionality - is as much filter as it is guide. It organises the chaos, yes, but it can also domesticate the wildness of the data. It encourages us to look for what fits, and to explain away what doesn’t.
The risk here is we stop listening to the data and start hunting for examples that support the framework we’re trying to prove we understand. We analyse in reverse.
Starting with Curiosity, Not Theory
What happens when we put the framework aside?
We allow the data to speak before we tell it what it should be saying. We shift from applying knowledge to generating it.
This doesn’t necessarily mean abandoning theory but it does mean delaying commitment. It means exploring patterns, surfacing tensions, and allowing contradictions to emerge without rushing to explain them away.
Data-First Doesn’t Mean Directionless
There’s a common misconception that avoiding an initial framework is somehow sloppy or unsystematic. But exploratory analysis can be deeply structured. It can involve careful attention to word patterns, recurring ideas, and silences. It just doesn’t rush to label them.
You can still ask rigorous questions:
What surprises you?
What doesn’t sit neatly?
What’s repeated but never explained?
What’s avoided?
These kinds of questions build toward theory. They don’t start with it.
When Frameworks Come In
Eventually, a framework can be a powerful tool. It can help you theorise what your data is doing. It can connect your work to wider ideas… power, identity, epistemology.
But now you’re not using it to shape the data - you’re using it to make sense of the shape the data has already shown you.
That distinction matters.
A Caution Against Framework-Driven Analysis
In some fields, especially social sciences and humanities, researchers feel immense pressure to signal their theoretical allegiances early, lest their work be seen as naive or unanchored. But there’s value in intellectual humility. In saying, “I’m still figuring out what this is.”
Frameworks should support discovery but often times researchers act more like they’re eager to support the framework.
So, Do You Really Need a Framework?
You certainly can. But maybe not yet.
A theoretical framework can illuminate your findings, anchor your work, and connect your insights to larger conversations. But if you lead with it too early, you risk narrowing your vision to what the framework is prepared to see.
Let the data breathe first. Let it surprise you. Let it tell a story you weren’t expecting.
Then, when you reach for a framework, you’ll do it not out of obligation but out of genuine analytical need. That’s when theory comes alive.
