Anonymisation – what’s in a name?


Many researchers find themselves inventing names because it’s standard ethical procedure to anonymise the people we’ve talked with and the places we’ve been. And naming is of course a simple and straightforward process. Well, maybe. Well, not all the time. Perhaps hardly ever. Names can be very troublesome and researchers can spend surprising amounts of time working out what to call things.

Sometimes it is possible, and desirable, to get people to choose their own research name, their pseudonym. Sometimes it is not possible or desirable – people can’t see the point and/or they’ve already spent enough time on your research thank-you very much and/or they don’t care what they are called as long as they won’t be recognised and/or there just isn’t the relationship between you and/or there is no time or, or, or…  If your research is one of the or, or ors, then you will likely find yourself faced with a decision about what to call people and places.

Now first off, a warning. It is possible to get too-clever-by-half with names. Once upon a time, long long ago, a colleague and I decided to give the schools we were studying the names of English trees. We’d done a study of one school that we called Hollytree so we decided to stick with an arboreal convention. So we had Elder and Hawthorn and Oaktree and Plumtree and so on. We allocated these tree names randomly to our thirty case studies. We didn’t need to do this. The tree names told us nothing about each school.

Hollytree had been a fine anonymisation. It was one school and we’d spent two years researching it and we wrote a lot about it. But this wasn’t the case here. We had thirty schools, our interest was in identifying patterns across them and we didn’t write much about about each of them individually. We would have been a lot better off if we’d just stuck with numbers or with something that combined region and school type – Primary Midlands, Secondary Northern for example. But no, we went with trees. The names had zip, zilch and zero benefit for analysis but did mean we always had to have our naming key to hand whenever we did any writing. Lesson learnt!

Don’t be us. In order to avoid the too-clever-by-half option, it’s helpful to have a bit of a handle on the choices you have in naming. There are some predictable options.

There are simple, descriptive choices:

The name is relevant to the kind of person or place selected for study e.g. University 1, University 2 etc.

There are more descriptive choices – where the name bears some relationship to the research question and analysis: 

  • The names uses chronology e.g. university ( new), university (old)
  • The name is relevant to demography e.g. academic (female), academic (non-binary) academic (neurotypical)
  • The name is relevant to location e.g. university ( south) university (London)
  • The names is relevant to size e.g. faculty (small), faculty ( medium)
  • The name is relevant to the organisation  e.g. academic ( temporary), academic (tenured) academic (senior) administrator, clerical staff
  • The name is relevant to magnitude e.g. academic (most cited) academic (least cited)

There are descriptive choices of names which aim to humanise the person so that the reader will see the words and get an idea of the person speaking e.g. people are given new names – Maria, Fred. Re-naming can be tricky. Researchers of course need to take care not to use a name or nickname of anyone in the study. But they also need to consider class and cultural questions. Are they stereotyping people by giving them particular names? Are they steering readers’ expectations – for instance, what would you assume about an undergraduate called Rainbow, or one called Tarquin? ( Apologies to any Rainbows or Tarquins reading this.) And it is always important to consider whether naming is going to be confusing – do readers really need this person to have a name, or would a number or some other less human name be just as good and potentially less risky?

And places and organisations can be given names that carry information. e.g. Royal Livery University (probably old and posh), University of the Southern Islands (possibly remote and involved in distance education with a community focus). But the researcher has the (often difficult) question of what information gets carried with the name. How would you respond to a Midlands Entrepreneurial University or Northern Trades College? If names are central to the analysis carried out, then they might be apt. Or they may skew the readers’ impressions. And of course if there is only one organisation which fits the name, then apparent anonymity might give away far too much. What would a google search of Midlands Entrepreneurial University produce for example?

Yes, this is not all there is to say about naming and anonymisation but I hope, dear anonymous, this is an answer of sorts to your question. Use the categories listed above as a starting point to think through what kind of naming you will do. And do remember, naming is often a lot more complex than it appears at first glance. It’s worth spending that extra time thinking about it.