Guide to Advanced Empirical
Download 1.5 Mb. View original pdf
|
3299771.3299772, BF01324126
| 5. Survey Design Two common types of survey design are: ● Cross sectional: In this type of study, participants are asked for information atone fixed point in time. For example, we may poll all the members of a software development organization at 10 am on a particular Monday, to find out what activities they are working on that morning. This information gives us a snapshot of what is going on in the organization. ● Longitudinal: This type of study is forward-looking, providing information about changes in a specific population overtime. There are two main variants of longitudinal designs, you can survey the same people at each time period or you can survey different people. Recall the three survey examples we introduced in Sect. 2. The Lethbridge survey asked respondents about their levels of training and education (see Lethbridge, 1998, 2000). The Ropponen and Lyytinen (2000) study requested information about risk management practices from Finnish software projects. The Pfleeger- Kitchenham study sought to determine what kinds of evidence were used to support technology adoption decisions. All three surveys were all cross-sectional studies, in which participants were asked about their past experiences at a particular fixed point in time. It is not simply coincidence that all our examples are of this type in our experience, most surveys in software engineering have this kind of design. There are other more complex forms of survey design, for example designs that compare different populations, or designs that aim to assess the impact of a change. For information on such designs see, for example, Shaddish et al. The other issue to decide is the way in which the survey will be administered. Options include: ● Self-administered questionnaires (usually postal but increasingly Internet). ● Telephone surveys. ● One-to-one interviews. 3 Personal Opinion Surveys The questions that can be addressed are influenced by this factor. In addition, strategies for obtaining reliable data such as question ordering and wording differ according to the administration method. Fowler provides a detailed examination of the pros and cons of different administration methods (Fowler, 2002). In this chapter we concentrate primarily on self-administered questionnaires. Download 1.5 Mb. Share with your friends:
|