Interview Analysis

Tension Analysis aims to identify the tensions in transcribed and translated interview transcripts by leveraging a list of potential indicators of tension including the use of hedging and boosting in the language, the sign of reticence, and emotion of the interviewees from their interview responses. We also consider prosodic cues such as laugh and silence as signs of reticence. This tool also identifies traces of tension in the situations when an interviewee gives unusually long or short answers for a specific question type. In summary, this tool identifies possible tension places which the researchers can review to identify the actual tension occurrences.

Oral history interviews involve complex social interactions and different factors highly influence the interview situation such as complexity of human lives, age, intelligence, personal quality, etc. Both the interviewer and the interviewee contribute to these components during this dialectical process and often it generates a situation which can be difficult for both parties, causing unpredictable emotional responses by the interviewee which changes the interview dynamics. Tension can thus occur, as a result of conflict of interest or uneasiness. Understanding these factors and the turning points in an interview is important for the interviewers to understand the process of interviewing and also for self-awareness.

This tool was developed by Jumayel Islam as his master’s thesis at the University of Western Ontario, with the collaboration of Robert Mercer, Lu Xiao, and Steven High. Download his thesis here.



  • Upload a transcript in .docx format
  • The transcript must contain a timestamp before the conversation begins in the following format: xx:yy:zz
  • The tool will only process conversations following the timestamp
  • Questions and responses must be preceded by speakers’ names or aliases (see template)
  • Interviewer's questions must be in bold text (see template)
  • Narrator's responses must be in regular text (see template)
  • This tool requires that the provided transcript follows the template below
  • The tool will ignore anything outside the template