According to R.D. Gatewood and H.S. Field, employee selection is the "process of collecting and evaluating information about an individual in order to extend an offer of employment." Employee selection is part of the overall staffing process of the organization, which also includes human resource (HR) planning, recruitment, and retention activities. By doing human resource planning, the organization projects its likely demand for personnel with particular knowledge, skills, and abilities (KSAs), and compares that to the anticipated availability of such personnel in the internal or external labor markets. During the recruitment phase of staffing, the organization attempts to establish contact with potential job applicants by job postings within the organization, advertising to attract external applicants, employee referrals, and many other methods, depending on the type of organization and the nature of the job in question. Employee selection begins when a pool of applicants is generated by the organization's recruitment efforts. During the employee selection process, a firm decides which of the recruited candidates will be offered a position.
Effective employee selection is a critical component of a successful organization. How employees perform their jobs is a major factor in determining how successful an organization will be. Job performance is essentially determined by the ability of an individual to do a particular job and the effort the individual is willing to put forth in performing the job. Through effective selection, the organization can maximize the probability that its new employees will have the necessary KSAs to do the jobs they were hired to do. Thus, employee selection is one of the two major ways (along with orientation and training) to make sure that new employees have the abilities required to do their jobs. It also provides the base for other HR practices—such as effective job design, goal setting, and compensation—that motivate workers to exert the effort needed to do their jobs effectively, according to Gatewood and Field.
Job applicants differ along many dimensions, such as educational and work experience, personality characteristics, and innate ability and motivation levels. The logic of employee selection begins with the assumption that at least some of these individual differences are relevant to a person's suitability for a particular job. Thus, in employee selection the organization must (1) determine the relevant individual differences (KSAs) needed to do the job and (2) identify and utilize selection methods that will reliably and validly assess the extent to which job applicants possess the needed KSAs. The organization must achieve these tasks in a way that does not illegally discriminate against any job applicants on the basis of race, color, religion, sex, national origin, disability, or veteran's status.
Employee selection is itself a process consisting of several important stages, as shown in Exhibit 1. Since the organization must determine the individual KSAs needed to perform a job, the selection process begins with job analysis, which is the systematic study of the content of jobs in an organization. Effective job analysis tells the organization what people occupying particular jobs "do" in the course of performing their jobs. It also helps the organization determine the major duties and responsibilities of the job, as well as aspects of the job that are of minor or tangential importance to job performance. The job analysis often results in a document called the job description, which is a comprehensive document that details the duties, responsibilities, and tasks that make up a job. Because job analysis can be complex, time-consuming, and expensive, standardized job descriptions have been developed that can be adapted to thousands of jobs in organizations across the world. Two examples of such databases are the U.S. government's Standard Occupational Classification (SOC), which has information on at least 821 occupations, and the Occupational Information Network, which is also known as O*NET. O*NET provides job descriptions for thousands of jobs.
An understanding of the content of a job assists an organization in specifying the knowledge, skills, and abilities needed to do the job. These KSAs can be expressed in terms of a job specification, which is an
|1. Job Analysis||The systematic study of job content in order to determine the major duties and responsibilities of the job. Allows the organization to determine the important dimensions of job performance. The major duties and responsibilities of a job are often detailed in the job description.|
2. The Identification of KSAs or
|Drawing upon the information obtained through job analysis or from secondary sources such as O*NET, the organization identifies the knowledge, skills, and abilities necessary to perform the job. The job requirements are often detailed in a document called the job specification.|
|3. The Identification of Selection Methods to Assess KSAs||Once the organization knows the KSAs needed by job applicants, it must be able to determine the degree to which job applicants possess them. The organization must Once the organization knows the KSAs needed by job applicants, it must be able to determine the degree to which job applicants possess them. The organization must Selection methods include, but are not limited to, reference and background checks, interviews, cognitive testing, personality testing, aptitude testing, drug testing, and assessment centers.|
|4. The Assessment of the Reliability and Validity of Selection Methods||The organization should be sure that the selection methods they use are reliable and valid. In terms of validity, selection methods should actually assess the knowledge, skill, or ability they purport to measure and should distinguish between job applicants who will be successful on the job and those who will not.|
|5. The Use of Selection Methods to Process Job Applicants||The organization should use its selection methods to make selection decisions. Typically, the organization will first try to determine which applicants possess the minimum KSAs required. Once unqualified applicants are screened, other selection methods are used to make distinctions among the remaining job candidates and to decide which applicants will receive offers.|
organizational document that details what is required to successfully perform a given job. The necessary KSAs are called job requirements, which simply means they are thought to be necessary to perform the job. Job requirements are expressed in terms of desired education or training, work experience, specific aptitudes or abilities, and in many other ways. Care must be taken to ensure that the job requirements are based on the actual duties and responsibilities of the job and that they do not include irrelevant requirements that may discriminate against some applicants. For example, many organizations have revamped their job descriptions and specifications in the years since the passage of the Americans with Disabilities Act to ensure that these documents contain only job-relevant content.
Once the necessary KSAs are identified the organization must either develop a selection method to accurately assess whether applicants possess the needed KSAs, or adapt selection methods developed by others. There are many selection methods available to organizations. The most common is the job interview, but organizations also use reference and background checking, personality testing, cognitive ability testing, aptitude testing, assessment centers, drug tests, and many other methods to try and accurately assess the extent to which applicants possess the required KSAs and whether they have unfavorable characteristics that would prevent them from successfully performing the job. For both legal and practical reasons, it is important that the selection methods used are relevant to the job in question and that the methods are as accurate as possible in the information they provide. Selection methods cannot be accurate unless they possess reliability and validity.
Validity refers to the quality of a measure that exists when the measure assesses a construct. In the selection context, validity refers to the appropriateness, meaningfulness, and usefulness of the inferences made about applicants during the selection process. It is concerned with the issue of whether applicants will actually perform the job as well as expected based on the inferences made during the selection process. The closer the applicants' actual job performances match their expected performances, the greater the validity of the selection process.
The organization must have a clear notion of the job requirements and use selection methods that reliably and accurately measure these qualifications. A list of typical job requirements is shown in Exhibit 2. Some qualifications—such as technical KSAs and nontechnical skills—are job-specific, meaning that each job has a unique set. The other qualifications listed in the exhibit are universal in that nearly all employers consider these qualities important, regardless of the job. For instance, employers want all their employees to be motivated and have good work habits.
The job specification derived from job analysis should describe the KSAs needed to perform each important task of a job. By basing qualifications on job analysis information, a company ensures that the qualities being assessed are important for the job. Job analyses are also needed for legal reasons. In discrimination suits, courts often judge the job-relatedness of a selection practice on whether or not the selection criteria was based on job analysis information. For instance, if someone lodges a complaint that a particular test discriminates against a protected group, the court would (1) determine whether the qualities measured by the test were selected on the basis of job analysis findings and (2) scrutinize the job analysis study itself to determine whether it had been properly conducted.
The attainment of validity depends heavily on the appropriateness of the particular selection technique used. A firm should use selection methods that reliably and accurately measure the needed qualifications. The reliability of a measure refers to its consistency. It is defined as "the degree of self-consistency among the scores earned by an individual." Reliable evaluations are consistent across both people and time. Reliability is maximized when two people evaluating the same candidate provide the same ratings, and when the ratings of a candidate taken at two different times are the same. When selection scores are unreliable, their validity is diminished. Some of the factors affecting the reliability of selection measures are:
In addition to providing reliable assessments, the firm's assessments should accurately measure the required worker attributes. Many selection techniques are available for assessing candidates. How does a company decide which ones to use? A particularly effective approach to follow when making this decision is known as the behavior consistency model. This model specifies that the best predictor of future job behavior is past behavior performed under similar circumstances. The model implies that the most effective selection procedures are those that focus on the candidates' past or present behaviors in situations that closely match those they will encounter on the job. The closer the selection procedure simulates actual work behaviors, the greater its validity. To implement the behavioral consistency model, employers should follow this process:
Three strategies can be used to determine the validity of a selection method. The following section lists and discusses these strategies:
When using a content-oriented strategy to document validity, a firm gathers evidence that it followed appropriate procedures in developing its selection program. The evidence should show that the selection devices were properly designed and were accurate measures of the worker requirements. Most importantly, the employer must demonstrate that the selection devices were chosen on the basis of an acceptable job analysis and that they measured a representative sample of the KSAs identified. The sole use of a content-oriented strategy for demonstrating validity is most appropriate for selection devices that directly assess job behavior. For example, one could safely infer that a candidate who performs well on a properly-developed typing test would type well on the job because the test directly measures the actual behavior required on the job. However, when the connection between the selection device and job behavior is less direct, content-oriented evidence alone is insufficient. Consider, for example, an item found on a civil service exam for police officers: "In the Northern Hemisphere, what direction does water circulate when going down the drain?" The aim of the question is to measure mental alertness, which is an important trait for good police officers. However, can one really be sure that the ability to answer this question is a measure of mental alertness? Perhaps, but the inferential leap is a rather large one.
When employers must make such large inferential leaps, a content-oriented strategy, by itself, is insufficient to document validity; some other strategy is needed. This is where a criterion-related strategy comes into play. When a firm uses this strategy, it attempts to demonstrate statistically that someone who does well on a selection instrument is more likely to be a good job performer than someone who does poorly on the selection instrument. To gather criterion-related evidence, the HR professional needs to collect two pieces of information on each person: a predictor score and a criterion score.
Validity is calculated by statistically correlating predictor scores with criterion scores (statistical formulas for computing correlation can be found in most introductory statistical texts). This correlation coefficient (designated as r ) is called a validity coefficient. To be considered valid, r must be statistically significant and its magnitude must be sufficiently large to be of practical value. When a suitable correlation is obtained ( r > 0.3, as a rule of thumb), the firm can conclude that the inferences made during the selection process have been confirmed. That is, it can conclude that, in general, applicants who score well during selection turn out to be good performers, while those who do not score as well become poor performers.
A criterion-related validation study may be conducted in one of two ways: a predictive validation study or a concurrent validation study. The two approaches differ primarily in terms of the individuals assessed. In a predictive validation study, information is gathered on actual job applicants; in a concurrent study, current employees are used. The steps to each approach are shown in Exhibit 3.
Concurrent studies are more commonly used than predictive ones because they can be conducted more quickly; the assessed individuals are already on the job and performance measures can thus be more quickly obtained. (In a predictive study, the criterion scores cannot be gathered until the applicants have been hired and have been on the job for several months.) Although concurrent validity studies have certain disadvantages compared to predictive ones, available research indicates that the two types of studies seem to yield approximately the same results.
Up to this point, our discussion has assumed that an employer needs to validate each of its selection practices. But what if it is using a selection device that has been used and properly validated by other companies? Can it rely on that validity evidence and thus avoid having to conduct its own study? The answer is yes. It can do so by using a validity generalization strategy. Validity generalization is established by demonstrating that a selection device has been consistently found to be valid in many other similar settings. An impressive amount of evidence points to the validity generalization of many specific devices. For example, some mental aptitude tests have been found to be valid predictors for nearly all jobs and thus can be justified without performing a new validation study to demonstrate job relatedness. To use validity generalization evidence, an organization must present the following data:
The extensiveness and complexity of selection processes vary greatly depending on factors such as the nature of the job, the number of applicants for each opening, and the size of the organization. A typical way of applying selection methods to a large number of applicants for a job requiring relatively high levels of KSAs would be the following:
One viable strategy for arriving at a sound selection decision is to first evaluate the applicants on each individual attribute needed for the job. That is, at the conclusion of the selection process, each applicant could be rated on a scale (say, from one to five) for each important attribute based on all the information collected during the selection process. For example, one could arrive at an overall rating of a candidate's dependability by combining information derived from references, interviews, and tests that relate to this attribute.
Decision-making is often facilitated by statistically combining applicants' ratings on different attributes to form a ranking or rating of each applicant. The applicant with the highest score is then selected. This approach is appropriate when a compensatory model is operating, that is, when it is correct to assume that a high score on one attribute can compensate for a low score on another. For example, a baseball player may compensate for a lack of power in hitting by being a fast base runner.
In some selection situations, however, proficiency in one area cannot compensate for deficiencies in another. When such a non-compensatory model is operating, a deficiency in any one area would eliminate the candidate from further consideration. Lack of honesty or an inability to get along with people, for example, may serve to eliminate candidates for some jobs, regardless of their other abilities.
When a non-compensatory model is operating, the "successive hurdles" approach may be most appropriate. Under this approach, candidates are eliminated during various stages of the selection process as their non-compensable deficiencies are discovered. For example, some applicants may be eliminated during the first stage if they do not meet the minimum education and experience requirements. Additional candidates may be eliminated at later points after failing a drug test or honesty test or after demonstrating poor interpersonal skills during an interview. The use of successive hurdles lowers selection costs by requiring fewer assessments to be made as the list of viable candidates shrinks.
SEE ALSO: Human Resource Management
Lawrence S. Kleiman
Revised by Tim Barnett
Barrick, M.R., and R.D. Zimmerman. "Reducing Voluntary Turnover Through Selection." Journal of Applied Psychology 80, no. 1 (2005): 159–66.
Gatewood, R.D., and H.S. Field. Human Resource Selection. 5th ed. Fort Worth, TX: Dryden Press, 2001.
Hausknecht, J.P., D.V. Day, and S.C. Thomas. "Applicant Reactions to Selection Procedures: An Updated Model and Meta-Analysis." Personnel Psychology 57, no. 3: 639–83.
Kleiman, L.S. Human Resource Management: A Tool for Competitive Advantage. Cincinnati: South-Western College Publishing, 2000.
Occupational Information Network. Available at < http://online.onetcenter.org >.
Potosky, D., and P. Bobko. "Selection Testing Via the Internet: Practical Considerations and Exploratory Empirical Findings." Personnel Psychology 57, no. 4: 1003–1034.
Ryan, A.M., and N.T. Tippins. "Attracting and Selecting: What Psychological Research Tells Us." Human Resource Management 43, no. 4 (2004): 305–18.