What “Other” Is Hiding in Your Survey Data

You’re staring at a dashboard and it’s full of information that feels really meaningful.  And as your eyes scan a bar chart, they land on the biggest number. Eureka. That’s it, you know why all those farmers didn’t turn up for the training.

And then the euphoria turns to frustration. The largest selected option is ‘Others’. An opaque option that hides so many reasons, opinions, preferences..

Most  surveys, whether they are for research or impact assessment, have this annoying but necessary option of ‘Others’.

In some rare cases, the person designing the survey had the good sense to record the actual response by adding a ‘Please specify’ alongside the category. But that still leaves you making sense of all the responses buried under this obscure option. 

The team I was presenting a demo of Dots to last week highlighted exactly this issue. They had a little laugh about how much data was hidden (read lost) because it was marked ‘Other’. And it reminded me of how sampling is often done for design research studies.

Back when I was working as a design researcher, with small sample sizes, we often preferred to have a range of personas represented in the sample.

While a couple of people might represent the middle of the bell curve, there was always a preference for edge cases. These cases represented fresh perspectives, lending deep insight into how products and services could be used, where they failed, what affordances existed and much more. 

And that brings me back to the ‘Others’. This unassuming option hides the edge cases.

Imagine that you’re trying to find out why your stakeholders didn’t attend a series of training sessions that are a core component of your  program. The enumerators are out there for the survey, and you’ve got a few options that you could think of, either from past experiences or educated guesses.

But when you ask an open-ended question, simply, “why weren’t you able to attend?, you’d be able to figure out the edge cases.

Or maybe you’d find that a lot of the farmers have the same reason, just not something you thought of. These are the kind of insights that help pivot and iterate programs to be more responsive to the lived experiences of stakeholders, in this case the farmers.