At one time or another, virtually every organization needs to select and hire new employees. Selecting the wrong employee for the job can be costly in terms of the time and funds required for the selection process and training new employees. Therefore, the tools used to select new employees need to be demonstrably related to the job and must help the organization select those employees with the appropriate knowledge, skills, abilities, and other characteristics necessary for job success. Selection tools (including application forms, interviews, psychometric tests, and work samples) need to be empirically validated to determine the degree to which they are related to the requirements of the job. This process also helps the organization meet the various legal requirements prohibiting discrimination in selection procedures.
No matter what product or service an organization offers, a good human resources function is essential to its success. The human resources function comprises the activities and tasks associated with recruiting and managing the organization's personnel. Strong human resources capabilities are necessary in order to attract, hire, and manage the human capital necessary for successfully performing the business of the organization. One of the basic -- and most important -- responsibilities of the human resources function is the selection of new employees.
From an applicant's perspective, employee selection can often seem like a subjective or even random process. However, done correctly, employee selection is based on empirical data and the application of scientific method not only so that the organization is able to hire those individuals best suited to the tasks of the position but also so that the organization can ensure that it is meeting both the letter and spirit of employment law.
The Job Analysis
All selection procedures should be based on a systematic, thorough, empirically-based job analysis that is used to determine the actual requirements of the job. To be useful, a job analysis needs to be behavior-centered and describe the type of behavior expected of the employee. For example, a clerk in a retail clothing store may need a "good" personality or provide "good" customer service in order to successfully interact with customers. However, that description does little to state what the clerk must actually be able to do. A good job analysis would help operationally define a good personality by stating that a sales clerk needs to possess several abilities, such as:
- Being observant and noting when customers enter the store,
- Being able to multitask or switch tasks in order to determine when a customer needs help,
- Patiently helping the customer find what he or she is seeking,
- Successfully resolving customer problems without losing her or his temper, and so forth.
The job analysis would further break down this information to the level of the knowledge, skills, abilities, or other characteristics (KSAOs) an employee needs in order to do the job well. This information can be used to show the job-relatedness of selection decisions and provide the organization with the logical, empirical information necessary to support personnel decisions by providing a structure for determining the job-relatedness of these decisions.
Performing a thorough job analysis and developing empirically-based criteria for use in differentiating between applicants is one part of the employee selection equation. Equally important is the development of selection instruments that adequately and accurately measure these criteria. There are a number of commonly used categories of job selection instruments, including application forms, written psychometric instruments, interviews, and work sample and assessment centers. For any of these instruments to be meaningful, however, they need to be tied to the results of the job analysis.
As shown in Figure 1, there are several steps to developing a meaningful selection battery. The first is to perform a thorough job analysis to determine and operationally define what the criteria of success on the job are (e.g., ability to produce a minimum of widgets in an hour) and what KSAOs the employee needs in order to meet or exceed these criteria (e.g., ability to lift 20 pounds, ability to read 10-point type). The next step in developing a battery of selection tools is to choose instruments that will help the organization determine if the applicant has the appropriate level of the necessary KSAOs to do the job. This may involve choosing an off-the-shelf psychometric instrument (e.g., reading comprehension test, personality test) or determining or developing other ways to measure the applicants' KSAOs (e.g., work product, simulation).
Assessing the Predictors
Once the predictors of job success have been selected or developed through the job analysis, the next step in developing an adequate and accurate employee selection battery is to measure how well people perform on the predictors and how well they perform on the actual tasks. This can be done in several ways, including collecting data on how well existing employees do on the predictors and measuring how well they do on the job or collecting data on how well applicants do on the predictors and then -- after job training and experience -- measure how well they do on the job. Whichever method is chosen, the next step is to statistically determine the degree to which scores on the predictor (the test) predict how well the person will do on the job. Typically, this is done through correlation, a statistical technique that allows the organization to determine the degree to which the score on the predictor is consistently related to performance on the job.
Correlation may be positive (i.e., as the score on the predictor increases, job performance also increases), negative (i.e., as the score on the predictor increases, job performance decreases), or zero (i.e., the values of the predictor score and job performance are unrelated). When there is a strong correlation (whether positive or negative) between the predictor and performance, the predictor is said to be valid. This is this degree to which a survey or other data collection instrument measures what it purports to measure. In addition to being valid, the predictor must also be reliable. It must consistently measure what it is measures (i.e., consistently yield the same scores). If a predictor is not reliable, it cannot be valid.
If the correlation between the predictor(s) (tests or selection instruments) and the criterion (i.e., performance on the job) is found to be statistically significant, it can be used to help predict job success as part of a selection battery. However, jobs change from time to time, and it is important to periodically review the selection instrument(s) and statistically determine that they are still valid and reliable. If so, it is appropriate to continue to use them to choose applicants for a job. If not, then the organization needs to develop and validate another instrument for use in the selection process.
One goal of reliable and valid selection procedures is to make an optimal match between the new employee and the organization. Having the right employee in place can help ensure that the organization has the necessary human capital to perform its tasks and activities and to help it become a high performing organization. A good employee-organization match also helps ensure that the employee will be satisfied in the position and have the motivation to perform the tasks of the organization at a level that will help it succeed.
Following the Law
It is important to do a thorough...
(The entire section is 3422 words.)