11/20/08 / David Kennedy
I heard the following conversation a few nights ago while on the bus ride home between the bus driver and a passenger regarding a survey that RTD was administering to a few riders on each bus.
“That survey was like a book.”
“It even had chapters.”
“I think they get paid by the question.”
Of course they were referring to the rather long survey that the rider had just taken. Granted, she was on a regional bus and was a relatively captive audience (I have no idea how they’re doing this on city buses), but she still appeared put off by the long survey.
Surveys do appear to be getting longer. While there are many reasons for this, such as a greater need by businesses and organizations to get more information to base decisions on, and conducting complex analyses which need more data points, one key reason, in my opinion, is the expense of gaining cooperation.
Getting an individual to participate is one of the most costly parts of survey research. As participation has fallen for most modes, researchers have had to make more phone calls for telephone surveys, mail more letters for mail surveys, and send more invites for online surveys. As the number of contacts have increased – and often the incentives, too – so have the overall cost. So, since it costs the most to get someone to participate, once you have them – I believe the thinking goes – you might as well ask more questions to get your money’s worth.
And therein lies the problem in my opinion. Longer surveys turn people off (though in some instances a participant whom is highly interested in the topic may not mind) so next time they’re invited to participate, they’re less likely to do so. Gaining participation is then harder, driving costs up further. The process repeats.
While it is difficult – for both clients and us researchers – to limit the amount of questions we want to ask, we must appreciate a respondent’s time, even when they’re a captive audience.