2017 ASEE Annual Conference & Exposition

This research paper explores the premise that surveys are not neutral data collection instruments. Surveys are often used in educational research to gather information about respondents without considering the effect that answering the questions may have on the survey-takers themselves. However, during longitudinal studies, where the same subjects complete multiple surveys or interviews, a phenomenon called ‘panel conditioning’ has been observed in various fields including political science, public opinion and social science. ‘Panel conditioning’ refers to the participants’ experience taking surveys or interviews for a study influencing their responses to subsequent surveys or interviews for the study. The evidence on panel conditioning suggests that participating in surveys or interviews may have an effect not only on respondents’ subsequent responses, but also on their actual behavior.

In order to investigate the effect that surveys might have on participants, we included an optional reflection question at the end of the first wave of a national longitudinal engineering education survey (total respondent n=7197). As the survey was intended to gather information on both participants’ past academic experiences and their intentions for future actions, including their career plans in the near and distant future, it seemed ideally suited to fostering reflection. Here, reflection is defined as intentional thinking whereby an individual recalls and engages in meaning making around past experiences and considers the implications on future actions. Thus the optional final question asked respondents to what extent the survey inspired them to think about their education in new or different ways.

A clear majority out of the 2374 responses to the reflection question affirmed and described how participants were thinking differently after the survey, for example increased awareness of variety in employment opportunities, interest in campus resources, and musings on the engineering education system. Some respondents even reported explicit action goals for the future, ranging from a desire to spend more time thinking about topics the survey brought up, to immediate plans to take advantage of opportunities, to future-oriented goals such as developing skills and preparing for a career. Students with lower class standing and those identified as Under-Represented Minorities were significantly more likely to report new or different thoughts as well as make explicit goals.

Our results also suggest that there may be an association between reflection on the survey and participation in subsequent surveys. Participants who took the time to answer the final question substantively had higher odds of marking that they could be contacted for the second wave of the study, although their actual response rate was not significantly different from that of all other respondents. The current results suggest that including reflective questions on surveys may offer benefits to both survey participants and researchers, although more research is needed. The results also caution educators to be mindful of the impact that the questions in surveys can have on students, highlighting the importance of considering the alignment between educational goals and the operationalization of research goals.