cognitive science
and more
What operating systems do experimental psychologists use?

OpenSesame is a program for creating psychological experiments. Over the past weeks I ran a poll on the OpenSesame documentation site. The aim of this poll was to find out what operating systems experimental psychologists use (or at least those that visit my site). Because psychologists generally develop experiments on their own computer, but run experiments (i.e. test participants) on a laboratory system, I presented two sub-questions:

  • Which operating system(s) do you use for creating experiments (i.e. developing the experimental script)?
  • Which operating system(s) do you use for running experiments (i.e. testing participants in the lab)?

Based on 327 respondents, the results are quite clear. Windows 7 is the most widely used operating system, both for developing and running experiments. Mac OS comes in second for development, but only third for running—Presumably because Mac OS is largely a consumer OS, and not often found on University lab PCs. Strikingly, more than a decade after its release, Windows XP is still used quite widely. But it's definitely losing ground to Windows 7 and will probably be phased out at most universities in the near future. Windows Vista, the ugly duckling that never quite turned into a swan, is hardly used at all.

A bit surprising (to me) is the fact that so many respondents indicate that they use Ubuntu Linux. Although I'm personally an Ubuntu user, I find it hard to believe that the Ubuntu marketshare among experimental psychologists is 15 to 20%. One confounding factor here may be that users of Linux and Mac OS are more inclined to respond to these types of polls—They like to boast about their favourite operating system. This reminds me of a recent study (from a documentary, so...) that claimed to show that Mac OS users experience feelings of religiosity when viewing Apple products. Perhaps Linux users experience something similar, but I doubt that very many Windows users do. So this poll may have underestimated the actual Windows marketshare.