A Comparison of Selection Methods - Research and Reality

A Comparison Of Selection Methods

Exhibit 7.1 is a comparison of the most commonly used selection methods. The comparison was compiled by Robert Levin and Joe Rosse¹⁴ and is based on published research studies. In reviewing these research findings, it can be clearly seen that some methods are more effective than others¹⁵. Beyond a comparison of selection methods, Levin and Rosse looked at the appropriate matching of selection methods with “attributes.” The results of their research are shown in Exhibit 7.2. This second exhibit is of interest to us because of the existence of the National Occupational Standards. The Standards, as we have stated, describe all of the tasks performed by a bus operator and within each of those tasks further outlines the knowledge and abilities required to carry out that task. Referring to Exhibit 7.2 we see that the preferred selection method for determining knowledge and abilities is tests. Secondary preference is given to background reviews and interviews for knowledge and interviews and work samples for abilities.

Candidate Likes and Dislikes with Selection Procedures

An important consideration in the development of selection procedures is the reaction of candidates to the procedure. Researchers have found that candidates react favourably to selection methods that have a strong relationship to job content. Such methods are seen as being necessary and fair because they have a high level of face validity and job-relatedness. In contrast, selection methods such as psychological assessments and handwriting analysis have been found to produce negative reactions in candidates due to skepticism about the need for such information and the ability to evaluate test results correctly.

Two final pieces of research should be considered before we move on to look at how we can use these research findings to develop a “best practices” selection process.

Where Does the Union Fit In?

In situations where a bargaining agent represents bus operators it is important to involve the union in the selection process because the candidates who are eventually hired as employees will also become union members. Clearly, using procedures that are valid and job-related is important in building union support for the selection process. Equally, selection staff should have credibility with the union. For this reason, many companies use current or former operators as part of their selection team. Involvement of the union in the development of the selection process will build support for it; as will listening to and communicating with employees about the process. Finally, in many companies, it is the Human Resources Department that has the responsibility for managing the recruitment process. This department is the one that is also responsible for labour relations and therefore may be viewed as the “enemy” by the union. It needs to be recognized that such an adversarial relationship will not engender support without a commitment to allow the union input into the recruitment process.

The first is a study done by the American Management Association in 1996 that determined almost 60 percent of its members did no psychological testing of job applicants and employees and that when just applicants were considered the figure rose to almost 70 percent¹⁶. The second piece of research done with Fortune 100 companies in 1997 found that virtually 100 percent of these companies used interviews as a selection tool but that only 40 percent used any kind of test with job applicants ^{17, 18}. With this research in mind we will look at three selection methods - Background Review, Tests and Interviews - and outline how we can incorporate “best practices” into each of these methods to select candidates for hiring effectively.

¹⁴ For an excellent in-depth discussion of the recruiting and selection research and ways to improve hiring effectiveness read Levin and Rosse’s book High Impact Hiring. 1997. Jossey-Bass, San Francisco.

¹⁵ One of the criteria is “reliability.” This term refers to the ability of a selection method to produce valid results over time or from one testing sequence to another. In other words, if the method produces valid results this month/year can we rely upon it to be valid next month/year? High reliability means that we can.

¹⁶ American Management Association, 1997, AMA Survey of Workplace Monitoring and Testing, New York, NY.

¹⁷ Tippens, N., Wunder, S. 1997. Entry-level Management and Professional Selection: Best and Most Common Practices - Are They the Same? Workshop presentation for the Society for Industrial and Organizational Psychologists and reported in Employment Risk Management by James Sharf and David Jones, chapter 8 of Managing Selection in Changing Organizations Jerard Kehoe, editor. 2000. Jossey-Bass, San Francisco.

¹⁸ Data on test usage in the United Kingdom and reported in Competency-based Recruitment and Selection by Robert Wood and Tim Payne, 1998. John Wiley and Son, Chichester, United Kingdom show a much higher proportion of British firms rely upon tests to make selection decisions. Survey results show that over 60 percent of