Does Non-Response Bias Compromise the External Validity of a Sample of College Students of Divorce?

It is possible that voluntary studies on the effects of divorce fail to capture the perspectives of offspring who may be deterred from volunteering by their negative experiences of the divorce of their parents. This issue of non-response bias would cause researchers to gather unrepresentative samples that ultimately create an unrepresentative picture on the effects of divorce. The problem of non-response bias may also be a possible explanation for why research shows that small differences in psychological problems exist between children of divorce and children from intact families. This study sought to identify if non-response bias compromises the external validity of a sample of college students of divorce. To answer this question we conducted this study through the use of the introductory psychology pre screening study that is administered every semester to introductory psychology students at Arizona State University. We surveyed undergraduate introductory psychology students, all of whom completed a required prescreen survey for research credit. The students who indicated they were from divorced families, or whose parents were “never married and not still together”, were invited to participate in a follow up study to “to understand young adults’ perspectives on their parents’ divorce”. The students who responded to our invitation were compared to the students who did not volunteer in terms of their prescreen data. Volunteers did not differ from non-volunteers on seven out of the ten dependent measures. Volunteers differed from non-volunteers in terms of their closeness to their fathers, in terms of the parents conflict they experienced during the two years before and the two years after their parents permanently separated. Volunteers were more likely to be closer to their fathers and more likely to have experienced more parent conflict than non-volunteers. We are unaware of any studies on the subject of divorce that have had a similar opportunity to address the issue of non-response bias and its effects on the external validity of a college sample of divorce. This study should be replicated to determine the reliability of the results.