A within-subjects repeated measures comparative study of the effect of two data collection methods on disclosure rates of sensitive behaviours in a tertiary student sample.

Abstract

Researchers primarily rely on self-report data collection methods to question participants about their behaviours, emotions, attitudes, and beliefs. The challenge of eliciting truthful answers is often affected by the sensitivity of the research. Research which investigates sensitive topics such as crime, drugs, politics, race, religion, and sex can be particularly challenging as participants are hesitant to disclose their own information truthfully. The current research was focused on three primary objectives. The primary objective of this research was to add to the existing knowledge surrounding the Unmatched Count Technique (UCT). This study aimed to investigate the efficiency of data collection methods: Unmatched Count Technique Type I and Type II (UCT Type I and Type II) in obtaining self-disclosure data on sensitive behaviours. This was done by investigating which data collection method (DCM), the UCT Type I and the UCT Type II yields higher rates of disclosure on sensitive sexual items as an analogue of validity as well as which DCM yields the lowest group rates of social desirability bias. Finally, the study aimed to understand the participant's experiences of each data collection method in terms of ease of use, anonymity, and protection of confidentiality It is imperative to improve DCMs methods to an accurate picture of specific social issues, especially those that are considered to be private, sacred or sensitive. The results of this study demonstrate a significant difference in the participants’ disclosure of sensitive items between the UCT Type I and the UCT Type II. The Unmatched Count Technique Type II did produce higher base rates than the Unmatched Count Technique Type I on several the sensitive questions. However, both UCT DCMs produced negative numbers. Within the present study, the social desirability test indicated that participants would choose to portray themselves in a favourable, or in a socially acceptable manner, regardless of the assurance of confidentiality. Significantly, participants agreed that their responses could not be linked to them as individuals for both UCT DCMs further demonstrating some of the protective factors the UCT is known for. This process clearly demonstrates the need for more research in this area to explore ways in which the UCT can be adapted to collect accurate data on sensitive behaviour.