Tim Beebe
Mode-effects Experiment


Beebe, Timothy J., Harrison, Patricia A., McRae, James A. Jr., Anderson, Ronald E., and Jayne
A. Fulkerson. 1998. "An Evaluation of Computer-Assisted Self-Interviews in a School
Setting." Public Opinion Quarterly 62:623-632.


Theory. This study is a mode-effects experiment that compares different methods of collecting sensitive information from adolescents. Specifically, it examines two types of self-administered survey instruments: (1) a paper and pencil version, and (2) a computerized version. Following the existing literature, the researcher tested a primary and an alternative hypothesis. The primary hypothesis is that children will be more forthcoming to the computer. The alternative hypothesis is that children will provide fewer reports of sensitive information in the computerized version.

Design. The researchers sought to maximize the mode-equivalence of the two types of surveys presented to the adolescents. Accordingly, the main design questions focused on the actual user-interface of the computerized survey. A within school randomization was selected to test the two hypotheses.

Measurement. The study focused on Minnesota students aged twelve to seventeen. The surveys were administered either in classrooms (paper and pencil version) or in the schools' computer labs (computerized version). Students were asked to respond to sensitive questions that fell into three main categories. The first category concerns illicit drug use. The second category focused on antisocial behavior excluding illicit drug use. And the third component of the survey was antisocial behavior of the family.

Data Analysis. The hypotheses were tested with multiple regression analysis. All tests were conducted with the Pearson's chi-square. Dummy variables were used for the different sites and for the distance between children taking the computerized version.

Interpretation. The research determined that, contrary to the existing literature, fewer reports of sensitive information were obtained by the computer vis-à-vis the traditional paper and pencil version. However, this difference was largely a function of proximity. When the students, who were administered the computerized version of the survey, were seated at least five feet apart, their responses were not statistically different from those who were administered the paper and pencil version.

Back to abstracts & scripts menu.