Log in / Register
Home arrow Environment arrow Research Methods in Anthropology: Qualitative and Quantitative Approaches


In experience sampling (ES), people respond at random times during a day or a week to questions about what they’re doing, or who they’re with, or what they’re feeling at the moment. Some researchers ask informants to carry around a beeper and, when the beeper goes off, to jot an entry into a diary or talk about their actions, feelings, and surroundings into a small digital recorder. Some researchers ask informants to respond to a telephone interviewer (Kubey et al. 1996), and some researchers ask informants to fill out a form on an Internet-enabled cell phone or PDA (Foo et al. 2009; Wenze et al. 2007).

ES offers two big advantages. First, it combines the power of random spot checks with the relative ease of having people report on their own behavior. Csikszentmihalyi and Larson (1987) demonstrated the reliability of ES in a number of studies. Validity is another matter, but when people record or talk about what they’re doing and how they’re feeling on the spot, this should lessen the inherent inaccuracy of recall data. The second advantage is that, working on your own, you can only be in one place at a time, but with those beepers or cell phones, you can collect spot-observation data from lots of people at once.

In anthropology, Garry Chick used ES in his study (1994) of a small machine-tool company in Pennsylvania. Chick wanted to test Marx’s theory that automation would eliminate the need for skilled labor by turning complex tasks into a series of routine steps that could be performed by drones. Drones should find their work boring, unsatisfying, and unimportant.

There were 11 machinists, all men, in the small shop. Four of them worked on traditional, manually controlled lathes in turning out machine tools. Three worked only on a newly installed, computer-controlled lathe (you program it to do a job and then you more-or-less stand by while it executes your instructions). And four worked on both types of lathes, depending on the job.

Each man was beeped 30 times—six times a day over a 5-day week—and at each beep, he stopped what he was doing within 10 minutes and filled out a 2-minute questionnaire. (Each man kept a spiral-bound booklet of questionnaires handy in the shop.) There should have been 330 questionnaires (11 men filling out 30 each), but one man was beeped while he was in the bathroom and one was beeped while he was filling out a questionnaire (having been beeped 2 minutes before). That’s what happens when you collect data at random points, but random means random and no tinkering with it.

For each of the experience samples, workers indicated what they were doing. If they were operating a machine (or more than one), they indicated which one(s). Then they answered 14 attitude questions. Here they are:

On a scale from 1-10, where 1 means ‘‘definitely not’’ and 10 means ‘‘definitely yes,’’

what I was doing:

  • 1. was enjoyable.
  • 2. was interesting.
  • 3. was complex/technical.
  • 4. was fun.
  • 5. was under my control.
  • 6. was monotonous.
  • 7. was machine paced.
  • 8. was tricky.
  • 9. held my attention.

On a scale from 1-10, where 1 means ‘‘definitely not’’ and 10 means ‘‘definitely yes,’’ at

the time I was signaled, I was:

  • 10. pressed for time.
  • 11. working on my own.
  • 12. thinking about things other than work.
  • 13. doing something that I felt was important.
  • 14. doing something that required a lot of skill.

The results were mixed. The men found working on the computer-controlled machines more interesting, more satisfying, and more important than working on the manual machines, and they found programming the computer-controlled machines even more interesting than running them. But these machinists also felt more in control when they worked on manual lathes.

If you use self-administered questionnaires, you need literate informants to do experience sampling. But, as Chick says, ‘‘with more and more anthropological research in modern and modernizing societies, experience sampling can be a valuable addition to the anthropologist’s tool kit’’ (1994:6). And if you give people little voice recorders, you may be able to use the ES method with nonliterate populations (Further Reading: experience sampling).

Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
Business & Finance
Computer Science
Language & Literature
Political science