Interviewers can be an important source of error in face-to-face surveys. To reduce this error source, survey organizations have used different quality control procedures including calling back respondents, audio-recording the interview, and accompanying interviewer and observing the interview process. These procedures have been recently supplemented by the increased use of paradata to achieve greater efficiency and improve data quality. This presentation will focus on using paradata to monitor interviewer’s behavior in an international survey in the Kingdom of Saudi Arabia.

The Saudi National Health and Stress Survey (SNMHS) is the first mental health survey covering all geographical regions of the Kingdom of Saudi Arabia. It is part of the cross-national World Mental Health (WMH) initiative where fieldwork is centrally monitored by the Survey Research Center at the University of Michigan. Interviews are conducted face-to face using a computer assisted personal interviewing (CAPI) mode with an audio computer assisted self-Interview (A-CASI) administration for a number of sections judged to be of sensitive nature to the Saudi community. A randomly selected male and a randomly selected female are chosen from each household and interviewed by an interviewer of the same gender. Since interviews are not recorded for cultural reasons, paradata is heavily relied on to monitor interviewers’ behaviors (in addition to call-back and evaluation). In this presentation we demonstrate the use of OLAP Cube to display and compare quality indicators among interviewers. Indicators include: time spent on survey questions, pauses taken by interviewers during the interview, interview length, the maximum string of identical consecutive answers, ACASI administration rate, rate of saliva consent, verification results, response rate, eligibility rate, number of contact attempts, time between interviews, and the number of completed interviews per day.

We demonstrate how making consolidated tools available at the beginning of data collection is extremely important when conducting international work that is centrally coordinate by a remote organization. Such tools when coupled with the use of paradata allow for real-time monitoring and control even in widely dispersed and geographically distant field operations.