Comparing interview and work sample methods for hiring decisions

HR Case Study Analysis

In order to test which method would work best, it is necessary to do some calculations with the numbers Randy and Professor Higgins have generated as an estimate of how much each process would be. In order to compare each method, this current research will use the Dollar Value Gain formula in order to determine which method would prove the best in regards to keeping costs down and optimizing the total dollar gained from the process, considering that one is slightly more reliable than the other. This would allow Randy to see how much potential gain would result from hiring candidates using the two hiring process methods, the interview, and the work sample trial.

The formula for the expected dollar value increase is as follows: NsTrSDyZs-NC. In this equation, T = tenure of selected group in years (assumed to be one year) = 1, Ns = number of applicants selected = 50 and r = correlation between predictor (interview) and job performance, in this case it would be two different sets, one at = .30 correlation between predictor (work sample) and job performance, and the other at = .50. Additionally, SDy = standard deviation of job performance (assumed to be 40% of base pay), Zs = average standard predictor score of selected group = .80 and N = number of applicants = 100. Finally, the C. would also be different for each method, which the interview set at 100 and the work sample set at 150.

Dollar Value Gain from Using Interviews =

50(1)(.30)(4800)(.80) -100(100) = 47,600

Dollar Value Gain from Using Work Samples =

50(1)(.50)(4800)(.80)-100(150) = $81,000

Question 2

Clearly, Randy would benefit more financially from using the work sampling test as the primary chosen process for selecting candidates to hire. With its greater reliability coefficient, it results in a much higher potential dollar gain when calculated with the equation. In fact, the dollar gain would be almost twice that of using the interview process. Ultimately, this means that Randy should think about using the work sampling process as his primary hiring and selection strategy. The research does show that this is an appropriate tool to use when hiring, especially for positions that require customer interaction and structured service and performance tasks (University of California Riverside, 2014). This is especially true for entry positions, like the ice cream makers that Randy would be hiring to service his customers. In these tests, there are cognitive, psychological, and physical assessments that are possible for Randy to observe (Bryan, 2013). Essentially, Randy would be able to see the potential employees in action, working with the equipment, customers, and other team members. Creating work sampling tests need to emulate the actual job as best as they can. This is how Randy can get the most value of the this testing process.

However, before Randy makes a final decision on what process to use, he should create very detailed job descriptions. These job descriptions would help form the backbone of the work sampling test, but would also tell Randy if there were other performance tasks that would not be able to be emulated in the context of the work sampling process. For example, managers and assistant manager positions may require skills that cannot be put into a traditional work sampling test (HR Council, 2014). For these, higher level positions, Randy should choose to use the interview process instead of the work sampling process. It is crucial that Randy empowers his selection process for higher level positions with a better sense of each candidate. An interview process can contain a fictional job scenario to simulate how the candidate would respond (University of California Riverside, 2014). Still, Randy would have to create his questions very carefully in order to avoid potentially falling into a bias during the interviewing process (HR Council, 2014). So, Randy should use a mixed methods approach to the hiring process methodology as needed by the specific details of each job he is filling positions for.

Question 4

Still, for both tests there are clear limitations to the estimates that were produced in these calculations. First, the standard deviation that represents job performance is a very rough estimate. This means that it is only an educated guess, and may not represent the reality close enough to make assumptions from the statistical testing. This means that the real gain on value could be skewed compared to the estimates. Overall, it reduces the accuracy in the calculations.

Additionally, there was no data for the 100 applicants provided. Ultimately, the unique characteristics of each candidate would impact the calculations. With no data, it is just a pure estimate. The level of accuracy could be greatly improved with data that represents the proficiency of the workers, the reliability of the testing methods, and the environmental conditions of the ice cream job for each location Randy is hiring for. The averages for the validity coefficients may not be as accurate as they could be if there was other data provided. In evaluating methods, "tests must be validated using statistical methods and administered consistently across the hiring process" (University of California Riverside, 2014). Without assurance in the data, the estimates are just that -- estimates. They may not represent reality with the level of accuracy that would empower Randy to make the best decision. For Randy to make a better decision, he could possibly use data from some of his competitors in order to set benchmarks for what is appropriate proficiency to hire candidates.