Identifying Survey Interviewer Applicants with a High Probability of Early Termination

• Activity Number: 476
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 8:30 AM to 10:20 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Heather M. Schroeder*+ and Frost Hubbard and Kenneth Szmigiel
• Companies: Institute for Social Research and Institute for Social Research and Institute for Social Research
• Keywords: proportional hazards model ; interviewer attrition ; survey research

Abstract

Each year, survey organizations spend thousands of dollars recruiting, hiring and training interviewers who resign or are let go within the first few months of being hired. If an organization could identify individuals who are likely to attrit based on information that is learned during the hiring process, this could lead to significant savings in both time and money. Previous research shows that older phone interviewers are less likely to attrit than their younger counterparts (Schalk and van Rijckevorsel, 2007), while mixed results are found concerning education and attrition (Porter and Steers, 1973; Mobley, 1982). In addition to demographic characteristics, the University of Michigan's Survey Research Center also asks applicants to self-report their proficiency on different tasks and gives applicants an exam testing their computer skills. Analyses including proportional hazards models are conducted to investigate if these factors are associated with applicants who attrit within the first few months of being hired.