Monday, May 28, 2007

Writing Survey Questions

My offsider, Pauline, is embarking on the new world of question writing as we develop surveys for our Museum Shop and Science in the Suburbs. This got me thinking about what I would recommend as resources for question-writing.

Found a good online resource from the Committee on Audience Research and Evaluation called Visitor Studies 201 presented at the 2006 AAM, with a downloadable pdf on developing questions. There is a 'how to' on writing questions posted by the Evaluation and Visitor Studies SIG with a link to an online guide to designing and conducting visitor surveys by Julie Leones, The University of Arizona.

As mentioned before, Judy Diamond's 1999 book Practical Evaluation Guide: tools for museums and other informal educational settings (Walnut Creek: AltaMira Press) is a good starting point. For general advice on sampling and suggested questions for a whole range of areas try:
* McManus, P. (1991). Towards Understanding the Needs of Museum Visitors. In G. Lord & B. Lord (Eds.), The Manual of Museum Planning (pp. 35-52). London: HMSO. (I think there's a newer version of this book and it's in our Research Library).

You could also look at general social research texts - one I have used is:
* de Vaus, D. (1991). Surveys in Social Research (3rd ed.). London: UCL Press, especially Chapter 6 (also in our Research Library).

Good luck!

1 comment:

LyndaK said...

Found some more info about questionnaires in my thesis, as follows.

Questionnaires are widely used in educational and social research. They have the advantage of providing a large amount of information for relatively little time investment, and, if well designed, can be a fairly rigorous way of collecting data for further analysis and comparability across a population. Questions used can be either open-ended, where an interviewee responds in their own words, or closed, where respondents are provided with a set of choices. Closed questions include yes/no responses; tick boxes; rating scales (also called Likert-scales); agree/disagree statements or forced-choice responses.

Rating scales are commonly used in questionnaires. de Vaus (1991) identified three advantages in using rating scales. First, they encourage respondents to make a choice based on how strongly they feel about a complex subject area. Second, they increase validity by adding to data obtained from open-ended questions. Finally, reliability is met through obtaining a number of different sets of measures for similar question areas.

One aspect of rating scales is to decide between using a five or a seven points in the scale. In a review of this literature Cox (1980) concluded that ‘… the magic number seven plus or minus two appears to be a reasonable range for the optimal number of response alternatives’ (p.420). He also added that the number chosen needed to be enough to obtain the required information, yet not too refined to cause errors.

Another issue in rating scales is the potential for bias in designing the questions. For example, participants may give a biased response to a scale because of the way statements were ordered, responding more positively to statements that are asked at the beginning and more negatively later. This can be overcome by rotating questions and careful design of wording.

Further references:
* Argyrous, G. (1996). Statistics for Social Research. South Yarra: Macmillan Education Australia Pty Ltd.

* Cohen, L., & Manion, L. (1994). Research Methods in Education. London: Routledge.

* Cox, E. (1980). The Optimal Number of Response Alternatives for a Scale: A Review. Journal of Marketing Research, 17, 407-422.

* Frazer, L., & Lawley, M. (2000). Questionnaire design & administration. Brisbane: John Wiley & Sons Australia, Ltd.

I know my colleague, Carolyn from Museum Victoria, wrote a Museum Methods sheet on designing questionnaires, and I'll see if she can provide it for us.