Increasing Self-Disclosure in Interviews

Increasing Self-Disclosure in Interviews - a ECR blog by Christina Winters

CREST doctoral researcher Christina Winters explains how interview context and interpersonal tactics can produce more information about job candidates in sensitive roles.

Pre-employment vetting is the process used to screen job candidates who will see sensitive information in their role. The vetting process seeks to investigate a candidate’s character, integrity, lifestyle and attitude towards risk to support a judgment of suitability for employment. Can they be trusted to keep sensitive information confidential? Are they at low risk of being blackmailed? Pre-employment vetting relies on full and accurate self-disclosure concerning potential risks. Increasing disclosure of verifiable information is ideal, as this information can be corroborated by other sources.

Despite the importance of vetting to national security, there is almost no research on the interview process and what can be done to encourage disclosure. Most investigative interviewing research focuses on acquiring information related to event-specific investigations or ascertaining the reliability of information. Little investigative interviewing research focuses on autobiographical narrative. Most research on self-disclosure examines it as a function of relationship building in other contexts, such as romantic relationships, education, and healthcare. My studies bring these two areas together to pioneer research on self-disclosure in terms of interpersonal or organisational risk-management.

The effect of interview location on self-disclosure

Through existing academic literature and conversations with individuals who have been vetted, I have created a Sensitive Topics Questionnaire (STQ).

The STQ is analogous to questions asked during a vetting interview, with topics that range from questionable affiliations (i.e., “Have any of your romantic partners, friends, or family members spent time in prison?”) to substance use (i.e., “Have you ever used marijuana?”).

If a question is endorsed – that is, if the interviewee replies in a manner that indicates “yes”, it is followed with a prompt for more information (“Please tell me about that.”).

My studies examine two specific outcome measures: the number of endorsements interviewees provide, and the number of unique details they disclose during elaborations. The interviews are audio recorded, transcribed, and coded for these unique details.

Skype, a coffee cup, an office building.

My first set of studies examined the role of context, in terms of both interview location and interview medium, on self-disclosure. As there is reason to believe that the personalisation of such a personal interview would lead to a beneficial outcome, I was mainly interested in exploring the effect of home-based interviewing as opposed to other locations.

I found that interviewees disclosed more when interviewed in their homes compared to when interviewed in an office or public setting. This appears to be true, but to a lesser extent, even when the interview is undertaken over Skype compared to face-to-face.

People disclose more when they are at home during an interview, even if it takes place over Skype.

Findings from these studies coupled with a fascination for Scharff interviewing techniques (see CSR 2016, Issue 1) led me to investigate my second set of studies which examines interpersonal tactics. As entering a person’s home automatically leads an interviewer to gain familiarity and knowledge about a person, is it possible to artificially construct a facade of familiarity and knowledge?

The know-it-all illusion

The know-it-all illusion is an interviewing tactic used by the interviewer in which they allude to having more information than they actually do. In my third study I explored whether this could be effectively applied to vetting interviews.

One group of individuals downloaded a phone app, which examined basic phone analytics such as the number of times the phone was unlocked, the names of applications, and when they were accessed. Another group was asked to complete a personality questionnaire. A control group was not asked to do anything.

At the beginning of the interview, the interviewer told people who downloaded the app and people who completed the personality questionnaire that a ‘personality profile’ was created for them.

They were told this profile was based on either their phone analytics or the results of the personality assessment they took. The control group was not given any information.

At the end of each interview, interviewees were asked questions about how familiar and knowledgeable they felt the interviewer was of them.

Those interviewees who were told that a ‘personality profile’ had been created rated that they felt the interviewer was significantly more familiar with them and knowledgeable about them. Importantly, interviewees who were told a ‘personality profile’ had been created disclosed significantly more details about themselves to the interviewer than the control group.

Further study

An important part of vetting interviews concerns eliciting information from an interviewee’s referees. My final study will examine whether or not the interviewer’s report of information acquired from secondary sources (referees) will increase the interviewee’s self-disclosure during an interview.

Interviewees will be asked to nominate a close friend or relative to complete a personality questionnaire about themselves, and this will be used to draft a ‘personality profile’ that the interviewer will allude to having created.

This research has several applications, namely for use in vetting interviews. Understanding how to mitigate insider risk by increasing knowledge about an applicant is crucial to maintaining organisational security.

This research also poses several important questions such as:

  • Why does the home yield greater self-disclosure than other places?
  • What can interviewers do to make an interview space feel more ‘at-home’?
  • What exactly is the role of video-mediated communication?
  • Might these findings extend to suspect interviews in event-specific investigations?

There is still much to learn about this new area of research. You can read more about my research on my CREST profile page: https://crestresearch.ac.uk/people/christina-winters/

Christina Winters is a CREST-funded PhD student based at Lancaster University. She is supervised by Professor Paul Taylor and Dr Kirk Luther.


This blog is one of a number written by CREST’s Early Career Researchers whose work contributes to CREST’s mission of understanding, countering and mitigating security threats.

You can see all the blogs in the ECR series: https://crestresearch.ac.uk/tag/ecr-series/