Using Tests Effectively in the Public Sector
There are far more similarities than differences when it comes to effective testing in the public versus private sector. Nevertheless, there are some unique differences in employment testing for hiring and promotion purposes in the public sector. In particular, public sector testing occurs in a public arena, and therefore often requires greater openness regarding the testing process and greater public availability of information regarding individual test performance, test score use, test development, and other aspects of the testing process.
U.S. government agencies often base their hiring and promotional practices on a merit system, which uses a competitive examination process to assess the level or degree of qualification of candidates. These examinations may consist of one or more tests, such as multiple-choice tests, essays, interviews, physical ability tests, and performance tests.
Some tests may be evaluated on a pass/fail basis. One or more tests, however, must be scored on a more discrete basis, often percentage scores, to permit ranking of candidates. If more than one test is used, the scores can be weighted, and then combined to result in an overall score.
Once candidates final numeric examination scores are computed, they are placed on an eligible list in descending rank order based on this score and are then considered in this order for hiring or promotion. Different agencies use different rules for designating those candidates who are eligible for final hiring/promotion. One of the more common rules is to allow consideration of only the three highest scoring candidates. Some agencies permit consideration of only the highest scoring candidate. And, at the other extreme, some agencies apply a rule of the list wherein any candidate on the rank ordered list may be considered.
Civil Service Examinations
The civil service examination is the primary mechanism for selection into and promotion up through the ranks of many U.S. government agencies. This results in agencies conducting a large number of highly structured multi-faceted examinations. Because many individuals are attracted to this type of employment setting, and because many agencies conduct extensive outreach recruitment, large candidate groups are often the norm. This situation routinely presents considerable challenges in constructing and administering rigorous, comprehensive examinations for jobs to which hundreds or even thousands of candidates apply.
Public agencies tend to rely on content-oriented test construction to demonstrate the job-relatedness of employment tests. There are several reasons for this tendency, including:
- They follow a rigid job classification system, with different jobs being specifically and narrowly defined. Job descriptions or class specifications serve as a good starting point for constructing a more content-oriented test.
- They often prefer to hire employees fully proficient in a given type of work when possible, making a content validation strategy particularly applicable. Relative merit is often defined in terms of to do the job. Defining relative merit in this manner reflects the preference for a high level of content knowledge/skillful performance among those hired.
- Public sector employers may be required to have a relatively open system that allows for public scrutiny of processes, including candidate protest of examinations; and content-oriented tests, which usually exhibit a high degree of face validity that makes them more acceptable to laypersons, are less likely to attract criticism.
- Resources are often limited among public sector employers, so their ability to conduct more elaborate test validation research (or even to purchase tests) is restricted. The narrowness of job classifications also serves to deter conducting criterion-validation studies because most jobs have too few incumbents to yield consistent results statistically.