Design, Evaluation, and Analysis of Questionnaires for Survey Research, 2nd Edition

7

SURVEY ITEMS IN BATTERIES

In the last chapter, we discussed the forms that single survey items can take. However, in Chapter 4, we mentioned that researchers in the social sciences often bring items together in batteries. In that case, the different survey items do not stand alone anymore but often are connected by one introduction, instruction, and one request for an answer and a set of answer categories. Since we treat each text unit that requires one response as a survey item, we have to give special attention to the definition of survey items of batteries. The problem is that the different survey items in a battery contain very different text components even though they are often assumed to be equal and treated the same.

What distinguishes batteries is the mode of data collection in which they have been placed. Therefore, we start this chapter with batteries that are used in oral interviews, followed by a discussion about batteries in mail surveys, and finally batteries employed in computer-assisted self-interviews (CASI) are discussed. In each case, we will discuss which components should be seen as belonging to each survey item. In the summary and discussion, we also will give some recommendations.

We will discuss the different battery types, not because we think that batteries are a good tool for survey research, but because they are so popular. As we have indicated in Chapter 4, we think that the standard batteries of agree/disagree responses have done more harm than good ...