Personnel selection is the systematic process of making decisions about which individuals to employ to fill open positions within an organization. The main goal of selection is to identify and employ those individuals who have the necessary knowledge, skills, and abilities (KSAs) to successfully complete the work of the organization. The volume and quality of this work directly affects organizational success. The individuals selected may be new entrants to the organization identified through an external recruitment process. Alternatively, those selected may be internal applicants—current employees who move to new positions within the organization and who are identified through the internal recruitment process. Internal applicants may be interested in advancing their career through a promotion or may want to change career paths.
The usefulness of selection depends on the (1) number and quality of individuals in the applicant pool, (2) identification of specific KSAs necessary to perform the job, (3) appropriate job-related content of selection instruments, and (4) psychometric properties of those instruments.
Number and Quality of Applicant Pool
Selection begins when the company requires additional employees or a different type of employee to meet production or service demands or when a current employee leaves the organization, resulting in a job opening. Some organizations have a formal human resource planning (HRP) process in which forecasts are made about the number and types of employees needed to achieve future organizational goals. Based on these forecasts, the company engages in recruitment activities designed to attract applicants for employment. New employees are selected from the resulting applicant pool.
The organization is in a good position if the pool contains more applicants than there are open positions and if the large majority of individuals in this pool have the minimum KSAs for success on the job. In such situations, selection identifies those individuals who have the highest level of the KSAs required and who, as a result, would be expected to perform at a high level on the job. If the applicant pool is relatively small or has relatively few qualified applicants, selection may not yield an adequate number of high-performing individuals.
Identification of Knowledge, Skills, and Abilities
The development of a selection program begins with a job analysis to detail the nature of the work the job incumbent will be doing. Job analysis information is commonly obtained from interviews, observation, archival records, reference books, or surveys completed by job incumbents and supervisors. A job description is developed from this information and describes the tasks, duties, responsibilities, and accompanying KSAs a job incumbent will need. Additional descriptive information may be included about the working conditions of the job, such as physical requirements, noise levels, and hazards.
The job analysis information is used to determine the appropriate KSAs required to perform essential job duties: Knowledge is the body of information necessary for performance (e.g., knowledge of career development theories), skill is competency in performing a specific job task (e.g., use of Excel), and ability is a general, job-related proficiency (e.g., ability to read at Grade Level 8).
Content of Selection Instruments
The KSAs determine the format and content of selection instruments. Using the previous examples, applicants could be given a written theory test, asked to prepare an Excel spreadsheet, and then given a verbal test of reading comprehension. Research studies have found that selection has the best results if (1) instruments have a defined scoring system and (2) scores are used as the primary data on which applicants are judged for their potential for future work performance. The value of the selection instrument is directly related to the extent to which the content of the questions or activities making up the instrument reflects job activities.
Almost all selection programs start with the use of an application form. This brief document requests descriptive information about topics such as education, previous work experience, awards, and contact information. The form is limited, providing space for only general, brief statements. Studies have determined that approximately one-third of applicants provide inaccurate information to at least some of the questions on these forms.
A training and experience form is a type of application form that does provide useful information. This form contains questions that directly reflect job tasks and includes a scoring system measuring the extent of the knowledge of the task operations. For a college instructor position, a training and experience form could provide descriptions of five tasks. One task might be the following: “Develops course lesson plans covering class topics, methods of presentation of each topic, and learning objectives.” The applicant is asked to indicate if he or she has ever performed or been trained in the performance of this task. If the answer is yes, the applicant is asked to provide a brief description of what was done during this work or training experience. Trained judges may be used to grade responses.
Applicants generally react more favorably to job interviews than to other types of selection measures. Unstructured interviews are those for which there is no set of specified questions or defined scoring system. The interviewer asks whatever questions are thought to be appropriate and evaluates the applicant on his or her judgment of the applicant’s general performance in the interview. Structured interviews developed from job analysis information have proven to be more valuable in selection. These interviews are composed of job-related questions used with all interviewees, along with a defined scoring system for evaluating responses.
The behavioral description interview is one of the most widely used types of structured interviews. Applicants are asked to describe specific examples of how they acted in past situations similar to those they will encounter on the job if hired. The underlying philosophy is that past behavior is the best predictor of future behavior.
Cognitive Ability Tests
In a research study of 19 selection measures, cognitive ability tests (also referred to as general intelligence tests) had the highest predictive validity for job performance. These tests measure several mental abilities that are part of the tasks of almost every job. Abilities measured include verbal comprehension, numerical fluency, general reasoning, memory span, and logical evaluation. Research has consistently found that cognitive ability tests can be used for all jobs, and that they increase in value as the job of interest increases in complexity. These tests also are among the lowest cost selection measures. The strong relationship between job performance and scores on general mental ability tests may exist because these tests are also a strong predictor of job-related learning and problem solving. Those individuals who have greater mental ability are likely to learn more quickly than others and also may be able to better solve problems that occur when a job operation is in error or when a new demand is placed on the job incumbent. Both job learning and job problem solving translate into better job performance. Caution is recommended when using these tests; adverse impact against legally protected (e.g., racial, ethnic) groups has resulted from their use.
Work samples require the applicant to perform one or more of the job tasks. A secretary could be asked to take a word processing test. Other jobs that frequently use work samples are skilled craft positions, construction, computer programming, and food service. Designed correctly, work samples closely approximate the actual job. The organization has a much better idea of the quality of work to expect from an employee based on the applicant’s performance on this measure. Work samples are usually better at predicting future job performance than other selection instruments but are costly to develop and use and require that the applicant already be trained to perform the job.
Assessment centers are job simulations commonly used for selection of managers or professionals, often for advancement within the organization. In an assessment center, a group of job candidates participates in a series of activities. Representative activities include the Leaderless Group Discussion and the In-Basket exercise. In the former, groups of six applicants are given an organizational problem and asked to interact in order to provide a solution to the problem. Raters observe the job candidates throughout the process and rate each candidate on his or her behaviors during the exercise. The In-Basket presents the candidate with a series of office memos. The individual is required to provide detailed responses and recommendations for each memo. Responses are scored by a trained evaluator. Often assessment centers last for multiple days and are quite expensive to develop and operate. Organizations have used these devices for measuring the strengths and weaknesses of employees in order to identify training needs.
Personality and Integrity Tests
Personality and integrity tests are self-report questionnaires in which the respondent provides information about his or her feelings or behaviors. Psychologists have agreed that the optimal personality test is one that measures the Big Five personality dimensions: extroversion, agreeableness, conscientiousness, emotional stability, and openness to experience. These personality dimensions have been theoretically linked to many different jobs and found to be statistically related to performance in various jobs. This latter is especially true for conscientiousness and extroversion. Literally hundreds of personality tests have been used for selection; however, tests of the Big Five are among a very small group of such tests that have demonstrated value in selection.
Integrity tests are intended to identify those applicants who have a high probability of engaging in acts such as stealing, embezzlement, sabotage, or violence. Integrity tests have steadily evolved in recent years and currently have demonstrated the ability to predict those behaviors. Most of these tests ask the same types of questions as do personality tests. They are scored in relationship to behaviors that are indicative of a general set of counterproductive actions that may be taken by employees. Responses that reflect or are related to these counterproductive actions are scored as signifying low integrity.
Psychometric Properties of Selection Instruments
Selection measures must demonstrate the test construction properties of reliability and validity in order to demonstrate their usefulness. Reliability refers to the consistency, dependability, and stability of the scores generated by the selection instrument. Selection methods that are reliable will yield the same results if administered at two different times or if used by different raters. Reliability is a necessary but not sufficient condition to ensure that the selection measures are useful.
Validity refers to the extent to which the selection measures actually do assess the knowledge, skills, and abilities required for successful job performance. Validity provides evidence that the measure can predict future job performance. In the context of selection, when KSAs have been identified for a particular job, the HR specialist has formulated testable hypotheses about the worker characteristics required for job performance. Validity refers to the process of determining if those worker characteristics are indeed related to job performance.
There are two major forms of validity. Content validity involves systematically using subject matter experts (e.g., practitioners and academics in the field of work) and job analysis information to develop the items to include in the selection measure. Accountants would be most knowledgeable about the questions to ask on an accounting test or the questions to ask in an interview for accounting jobs.
Empirical validity is the statistical demonstration that there is a significant correlation between the scores on the selection instrument and the scores on measures of job performance. Predictive and concurrent validity are types of empirical validity. In both types, scores are collected for a group of individuals on both the selection instrument and the job performance measure. Statistical analysis is then performed to determine if the correlation between these two scores is high enough to be greater than chance. If the correlation is statistically significant, there is evidence that applicants with higher scores on the selection measure will perform well on the job.
Selection measures have been involved in a number of discrimination cases brought against organizations. A charge of discrimination may be filed when an individual(s) claims that he or she has been denied employment and can link that denial to one or more of the demographic characteristics identified in the equal rights federal legislation. The demographic characteristics often cited in selection discrimination cases are race, gender, religion, age, disability, and ethnic origin.
There are two forms of discrimination that may be charged by the plaintiff: disparate treatment and disparate impact. Disparate treatment occurs when membership in one of the previously noted demographic classes is used as the basis for the selection decision. If an employer decides that females should not be promoted into a particular job, disparate treatment has occurred. In this case, gender has been used illegally as the basis for the employment decision.
Disparate impact occurs when a protected group of people are treated differently from the majority. On a selection test, lower scores might be found for one demographic group than for others. A selection measure could be a weight-lifting requirement for a firefighter or a high school diploma requirement for a secretary. Both of these selection measures are likely to result in lower scores for specific demographic groups, thereby making individuals from that demographic group less likely to be hired than individuals from another group. Females, in general, are less likely to be able to meet a stringent weight-lifting requirement. Minority groups, such as African Americans or Hispanics, are less likely to meet the high school diploma requirement.
In either type of discrimination charge, the organization can only defend its selection program by clearly demonstrating that the scores produced by the selection instrument being questioned are related to job performance for individuals in the demographic group charging discrimination. The methods of determining validity discussed previously would be used to demonstrate the relationship between instrument scores and job performance scores.
Among the most important current selection issues is how to successfully employ technology. Many organizations have embraced the use of selection testing using the Internet and other electronic instruments. Uses include having applicants complete application and training and experience forms online and the administration of interviews, cognitive ability tests, and personality surveys. The major issues being raised include the following: (1) Are electronically administered tests equivalent to traditional tests? (2) How can the use of reference materials be prevented? and (3) How can the identity of the applicant be definitely established? The limited research to date comparing electronic and traditional testing has looked at testing times, applicant reactions, design of tests, and proctoring of tests. This research is inconclusive on most of these issues although some studies report more positive applicant reactions for electronic tests. Additional research is needed to provide more definitive information about the reliability and validity of electronically administered tests.
Value of Selection
Research over the past 85 years has provided a great deal of information regarding the monetary value to the organization of using valid selection measures. Such measures lead to increased learning of job-related skills and increased output on the part of the employees. Overall, the monetary gain from using valid selection methods is typically large, with the gain being directly proportional to the increase in validity of the better methods versus previously used methods. The more valid selection measures have greater return.
- Job interviews
- Knowledge, skills, and abilities (KSAs)
- Organizational staffing
- Unbiased hiring systems
- Barrick, M. R. and Zimmerman, R. D. 2005. “Reducing Voluntary, Avoidable Turnover through Selection.” Journal of Applied Psychology 90:159-166.
- Gatewood, R. D. and Field, H. S. 2001. Human Resource Selection. New York: Harcourt.
- Lievens, F., van Dam, K. and Anderson, N. 2002. “Recent Trends and Challenges in Personnel Selection.” Personnel Review 31:580-643.
- Posthuma, R. A., Morgeson, F. P. and Campion, M. A. 2002. “Beyond Employment Interview Validity: A Comprehensive Narrative Review of Recent Research and Trends over Time.” Personnel Psychology 55:1-81.
- Potosky, D. and Bobko, P. 2004. “Selection Testing via the Internet: Practical Considerations and Exploratory Empirical Findings.” Personnel Psychology 57:1003-1034.
- Robertson, I. T. and Smith, M. 2001. “Personnel Selection.” Journal of Occupational and Organizational Psychology 74:441-472.
- Ryan, A. M. and Tippins, N. T. 2004 “Attracting and Selecting: What Psychological Research Tells Us.” Human Resource Management 43(4):305-318.
- Schmidt, F. L. and Hunter, J. E. 1998. “The Validity and Utility of Selection Methods in Personnel Psychology: Practical and Theoretical Implications of 85 Years of Research Findings.” Psychological Bulletin 124:262-274.
- Wilk, S. L. and Cappelli, P. 2003. “Understanding the Determinants of Employer Use of Selection Methods.” Personnel Psychology 56:103-124.