Most of us know people who have a knack for expressing strong opinions on topics without having to rely on any actual data or factual information. This tendency to make unsupported claims about the “truth” is particularly prevalent in emerging fields such as online staffing. You don’t have to look far to find individuals who will make authoritative statements about how online staffing must be done on the basis of few facts other than personal conjecture and anecdotal evidence. This article addresses one particularly dangerous myth that has plagued online staffing virtually since the design of the first internet based screening system: the myth that online assessments should be limited to a very small number of questions and should take less than 5 to 10 minutes to complete. I suspect this myth was originally expressed as someone’s informal opinion about the ideal length of assessments. But over time I have noticed that this myth tends to be most frequently repeated by people selling screening tools that just happen to have very few questions and take less than 10 minutes to complete (what an amazing coincidence!). The “10-minute myth” appears to be based on an assumption that using longer assessments will cause candidates to drop out of the hiring process. But what little systematic research has been done in this area suggests that most candidates do not hold strong opinions about how long assessments should or should not be. In fact, most professional applicants will spend substantially more than ten minutes completing online assessments, provided that they are interested in the job and that the assessments are perceived as job relevant1 2 (iLogos, 2001; Mael et al., 1996). Although this data is admittedly anecdotal, in my experience working at a variety of major organizations that use staffing assessments taking well over 60 minutes, and I have not once heard of a professional applicant dropping out of the hiring process because the assessment was too long. Research conducted with hourly applicants reveals similar findings. An analysis of data from 29,000 hourly applicants found that it was not the length of assessments that led to drop-outs, but instead the nature of the content3 (Unicru, 2003). Almost all of these hourly applicants answered as many as 100 personality type questions (e.g., “do you like to take risks?”) without dropping out midway through the questionnaire. In contrast, a much larger percentage of applicants dropped out of the hiring process when asked a single question about their willingness to provide reference information, take a drug test, or submit to a background check. In sum, its not the number of questions that lead to hourly applicant drop-outs, its the nature of the questions. In sum, empirical research studying applicant behavior does not support the belief that using online assessments longer than 10 minutes will cause more applicants to drop out of the hiring process. However, if staffing professionals persist in promoting the myth that online assessments should take less than 10 minutes, it’s possible that candidates may eventually start expecting applications to take 10 minutes or less. At this point, the 10-minute myth will change from a simple misguided assumption to a major limitation that staffing professionals have needlessly imposed upon themselves. The possibility of the 10-minute myth becoming a reality poses a major risk to organizations seeking to effectively identify and match candidates to jobs. Many of the most scientifically valid and useful online staffing assessments require between 30 and 60 minutes to complete. These tools often cannot be shortened without severely damaging their accuracy. Staffing organizations that limit themselves to 10-minute online assessments will not be able to get the information they need to make accurate hiring decisions. The best way to ensure this does not happen is for staffing professionals to take a firm stance when it comes to assessment length. This stance can be rooted in two empirical findings:
- There is a wealth of personnel selection research indicating that time spent collecting scientifically valid online assessment content during the hiring process results in much more effective hiring decisions.
- Applicant reaction research does not support the belief that applications longer than 10 minutes will necessarily lead to greater levels of applicant drop-out.
The purpose of this article is not to advocate that all online staffing assessments should take longer than 10 minutes. Assessments should only take as long as is needed to get the information required to make accurate staffing decisions. It is possible that screening decisions aimed out removing unqualified applicants from the candidate pool may require less than 10 minutes of applicant data. However, ensuring the accuracy of hiring decisions later on in the staffing process will almost always necessitate collecting significantly more than 10 minutes worth of applicant data. Asking how long an online assessments should take before selecting a candidate is like asking how long a housing inspection should take before purchasing a home. The length depends primarily on the financial value, scope, and complexity of the decision. Larger, more expensive houses typically require longer inspections. The same is true for staffing assessments used to select candidates for more complex, financially critical jobs. In contrast, the length depends relatively little on seller or candidate feelings about “how long” an inspection or assessment should take. Most candidates are willing to invest a fair amount of time applying for jobs they feel are attractive as long as the assessment questions are reasonable and job relevant. It’s up to staffing professionals to determine how much time is truly needed to get the information needed to ensure that the candidates are being hired into a job where they will succeed.
1 iLogos Research (2001). “Perception vs. Reality: Jobseeker Behavior On-Line.” Recruitsoft, Inc. San Francisco, CA. 2 Mael, F.A., Connerley, M. & Morath, R.A. (1996). “None of Your Business: Parameters of Biodata Invasiveness.” Personnel Psychology, 49, 613-650. 3 Unicru (2003). “Applicant Drop Out Rates in Different Phases of the Hiring Process.” Unicru Inc. Beaverton, OR.