activity

Chapter 6 Objective Every organization should have some strategy to acquire new employees. This can be a formal or informal strategy. In this chapter, we talk about selecting employees and what goes into finding the right person for the right job.

HR Planning Organizations need a steady supply of people to fill vacancies and new positions. To make sure they’ll be available, HR professionals must estimate how many and what sort of person will be necessary and how many are likely to be available.

Forecasts list the number of persons needed per job category--secretaries, machine operators, etc. Likely that some categories will decrease and some will increase. Overall, unskilled jobs are being replaced with skilled ones. Two possible organizational responses:

Selection approach: Employees that are no longer needed are let go and new employees are selected to fit the new jobs. This approach is often less expensive to the organization.

Training approach: Old employees are trained to carry out the new jobs. This approach is more expensive; but less harmful to current employees. Necessary when few individuals with the necessary skills are available in the labor market.

Recruiting Applicants To hire good people, organizations must have a large number of possible employees to choose from. Further, scientific selection procedures tend to work best when there are several applicants.

Six common sources of applicants, differing in ease and cost:

1. Advertising

2. Employee referral

3. Employment agencies

4. School recruiters

5. Walk-ins

6. Web

The choice of source of applicants depends on how scarce they are. Some organizations have enough walk-ins to supply needs. Applicants from different sources may differ in the quality of the applicant.

INTRODUCTION 1/O PSYCHOLOGY CHAPTER 6

PAGE 1 OF 7 CHAPTER 6

Recruiting Source Effectiveness Which sources are the best? Inside sources (e.g., employee referrals, transfers within the organization) resulted in higher performing employees who stayed longer than outside sources (e.g., advertising, employment agencies) (Zotolli & Wanous, 2000).

Newspaper advertisements provided the worst source of applicants.

Efficient recruitment starts with the KSAOs for the job. Otherwise, you’re not recruiting the right people.Technology has impacted recruitment, not only listing jobs, but providing prescreening, on-line assessments, and systems to help manage recruitment.

Selecting Employees How do we select employees?

The purpose of employee selection is to hire people likely to be successful on the job. The more selective an organization can be, the better the chances that the person hired will be a good employee. To be selective an organization must know what KSAOs are critical for success. Once again, job analysis is critical for success of other HR systems.

Scientific selection requires two elements: the criterion and a predictor. The criterion is a definition of good job performance, as discussed in the chapter on performance appraisal. May be a question of what aspect of the criterion is most important for hiring- attendance, quality of work, etc.

The predictor is anything that relates to the criterion, such as measures of KSAOs. These measures can be used as predictors of the criterion of job performance.

Determining if a predictor relates to a criterion requires a validation study, a research study that collects data from a group of current employees on a criterion and predictors) and tests if they’re related.

If the correlation coefficient between a predictor and the criterion is statistically significant, the predictor is a valid way to forecast an applicant’s job performance. Example: collecting data from a group of machine operators, including a performance appraisal (the criterion) and a test of manual dexterity (predictor), then checking the correlation between the two.

Constructing a validation study: five steps

1. Job analysis: Provides information about the tasks in a job and about the KSAOs needed for success at those tasks. JA is I]important because for successful and legal selection, the KSAOs used in selection must be job-relevant--characteristics needed for successful performance. JA is effectively provides both the criteria of successful job performance and predictors of performance.

2. Specify Job Performance Criteria: The criteria for good job performance can be developed once an employer has a good idea about what a job entails.

3. Choose predictors: Potential predictors are chosen as the organization develops criteria for a job. Measures can be direct (a psychological test) or indirect (education level as a measure of verbal and math skills).

INTRODUCTION 1/O PSYCHOLOGY CHAPTER 6

PAGE 2 OF 7 CHAPTER 6

4. Validate the Predictors: Measures of predictor(s) and criteria are collected for a sample of people, usually in the organizational setting.

1. In a concurrent validation study, data on predictors and criteria are collected at about the same time, usually on current employees. If the variables are related for employees, we assume that scores on the predictor at the time of application predict later performance on the job.

2. In a predictive validation study the predictor is measured before the criterion. The predictor might be measured in applicants, and months later their performance measured. If the variables are related, the predictor is considered a valid selection device.

3. The two designs seem to be equally effective in validating predictors; however, the concurrent design is faster and more convenient.

5. Cross-validate 1. To cross-validate is to repeat the validation study on a new sample of employees. This is

done to make sure the results of the first sample are not due to chance. Statistically significant results can occur by chance--called Alpha or Type I errors--but are very unlikely to happen twice in a row. Thus, cross-validation adds to our confidence that the predictor can forecast the criterion of interest.

2. Two samples are needed to conduct a cross-validation. The first sample is used to determine whether the criterion and the predictor are significantly correlated. A second sample is used to see whether the significant relation found in the first sample can be repeated in the second. The predictor is validated on the first sample and then cross-validated on the second.

Validity generalization: Validity Generalization means that validities of selection devices are generalizable from job to job and organization to organization. So it is not always necessary to collect new data to validate a selection measure. A measure that predicts success at drill press operation at one company should also work at another company. Validity generalizations are widely accepted as long as the jobs are comparable. Could also generalize to a job with the same KSAO requirements.

Alternatives to Conducting Validation Studies Many organizations do not do validation studies. May not hire enough people, or wish to spend the money and time. Alternative is to measure KSAOs with selection devices that have established validity. This relies extensively on existing research (e.g., validity generalization).

How predictor information is used for selection

Once we’ve validated a predictor or predictors, how do we use it?

Multiple hurdle approach sets a criterion for each predictor. If the applicant surpasses the criterion, the hurdle is passed. It is efficient to use hurdles in a specified order, one by one, eliminating applicants at each step. Cheapest to use the least expensive hurdles first; for example, check education level before giving a more costly aptitude test.

Regression approach uses scores for predictors in an equation (the regression equation) to provide a numerical estimate of the criterion. Individuals with the highest predicted

INTRODUCTION 1/O PSYCHOLOGY CHAPTER 6

PAGE 3 OF 7 CHAPTER 6

performance are hired. For one predictor, the form of the equation is Y = b x X + a , where X is the predictor, Y is the criterion, b is the slope, and a is the intercept. After the validation study, the values of a and b are known; to predict Y (performance), put in the applicant’s value for X (the predictor) and carry out the arithmetic.

For regression equations with more than one predictor, a multiple correlation (R) can be calculated, the correlation between the criterion and the predictors in combination. The magnitude of the multiple correlation depends on how strongly each predictor correlates with the criterion, and how strongly the predictors correlate with each other. Using the predictors combined can provide a more accurate prediction than the predictors alone.

For two predictors, the form of the equation is Y = (b1 x X1) +(b2 x X2) + a , where Xs are predictors, Y is the criterion, a is the intercept, and each b is a regression coefficient (slope). Each additional predictor adds another X and another b.

Accuracy of prediction is greater if the predictors correlate strongly with the criterion. When the correlation is small, prediction won’t be very accurate but might still be helpful (better than nothing, and probably better than a nonscientific approach). Each regression equation must be cross-validated.

An implication of the regression approach is that a low score on one predictor can be compensated for by a high score on another. The limitation of the regression approach can be overcome by combining it with the hurdles. First, applicants should be screened using the hurdles. Then, a regression equation should be applied only to those who make it past the hurdles.

Getting Applicants to Accept and Keep Jobs

Once an applicant is selected, they must be convinced to take the job. Some strategies: Salary must be comparable to that of competing jobs. Salary survey can determine if salary is sufficient. Negotiate Salary and other benefits, even exact content of the job, with the potential employee.

One approach offers a cafeteria benefits program, where employees choose benefits from a list of possibilities--different types of insurance, retirement programs, etc. Care must be taken to be truthful in describing job conditions.

Equally important for an applicant to stay on the job once hired. A false, positive view of an organization can result in high turnover. New employees can find out things that make the job unacceptable or unpleasant. Important that work conditions are presented honestly.

The realistic job preview (RJP) is used to give job applicants accurate information about the job and the organization. Usually a brochure or videotape. Gives both favorable and unfavorable aspects of a job. Meta-analysis has shown the RJP reduces acceptance of job offers, and also reduces initial expectations about the job and organization. RJP can increase job performance and satisfaction, but results with turnover have been small and inconsistent. Recently shown that effects of RJP may vary with applicants previous experience.

INTRODUCTION 1/O PSYCHOLOGY CHAPTER 6

PAGE 4 OF 7 CHAPTER 6

The Utility of Scientific Selection

Utility analysis is the study of the monetary effects of a practice (such as scientific selection) on an organization. What is the payoff to an organization from the cost and time invested in scientific selection?

How valid selection devices work:

To understand utility analysis, must understand how three factors determine the extent to which scientific selection will result in hiring better performing employees: baserate, selection ratio, and validity.

The baserate is the percentage of applicants who would be successful if all of them were hired. For some simple jobs, it will be close to 100%; for other, difficult ones, it may be close to 0%. A baserate of 50% gives the most room for improvement, (from 50% accuracy up to 100% accuracy, a difference of 50%.) Any other value has a smaller difference from 100% accuracy. For any baserate over 50%, can increase accuracy by guessing everyone will be successful; for any baserate less than 50%, can increase accuracy by guessing everyone will be unsuccessful.

The selection ratio is the proportion of job applicants that an organization must hire. It is found by taking the number of positions to fill divided by the number of applicants. Selection ratio is low when there are many applicants for each position, high when few applicants. Low selection ratios produce the greatest utility because they allow an organization to be more selective.

The validity of a selection device is the magnitude of correlation between it and the criterion. The larger the correlation, the more accurate the prediction, and thus the greater the utility-- because utility depends on increasing the success rate over the baserate.

How valid predictors increase success rates

The cutoff values for the predictor and the criterion divide the graph into four quarters. Refer to Figure 6.4 on page 159 in your textbook.

In the graph, the upper right quadrant contains true positives—those who would have been hired and successful;

The lower left contains true negatives--would not have been hired nor successful.

Upper left contains false negatives-would not have been hired but would have been successful.

Lower right holds false positives--would have been hired but not successful.

A valid predictor will show a pattern of points in the shape of an ellipse. The more valid the selection device, the closer the points fall to a straight line, in this case from lower left to upper right. On a straight line our prediction of the criterion would be perfect. The greater the validity, the greater the potential utility of scientific selection.

Computing the utility of scientific selection

Even if a predictor results in hiring better employees, not necessarily cost effective. Jobs that result in a significant investment in individual employees warrant the use of expensive selection devices.

INTRODUCTION 1/O PSYCHOLOGY CHAPTER 6

PAGE 5 OF 7 CHAPTER 6

Utility analyses can be used to estimate the utility of performance improvements that might result from valid selection.

There is a debate regarding the best approach for conducting an utility analysis. People are not accurate in their judgements, making utility estimates inaccurate.

Some analyses have indicated that scientific selection can have considerable monetary benefits-- millions of dollars to large organizations. Utility analyses could be applied to other organizational practices, such as giving job performance feedback.

Utility analyses provide theoretical estimates of how much gain in performance might occur. Might not happen in practice, due to organizational conditions and constraints such as lack of equipment or support.

Legal Issues

Many industrialized countries now have laws that protect people from discriminatory actions by organizations. Groups that are the focus of such laws vary with country; most include women, some include blacks. Large minority groups are likely to be protected.

Legal selection in the U.S.

Civil Rights Act of 1964 first made discrimination illegal. Protection against discrimination has since been broadened. Changes in laws and results of Supreme Court cases have produced a complex and confusing set of legal requirements for selection, so that implementing a legal selection system is complicated and uncertain.

Currently illegal to discriminate on the basis of: age (over 40), color, disability, gender, national origin, race, religion.

Certain groups have been the target of protection, called protected classes. Examples include victims of past discrimination such as women, African Americans, Hispanics, Native Americans.

Uniform Guidelines on Employee Selection, published by the U.S. government in 1978, serve as guidelines for legal selection. Defines important concepts, and outlines the proper way to develop a valid employee selection system.

Adverse impact refers to the impact on a class of a given selection practice, defined in terms of selection ratios of the protected class and the majority group. Adverse impact occurs when the selection ratio for the protected class is less than 80% of the group with the largest selection ratio. That is, if 60% of male applicants were offered a job, there would be adverse impact against females if fewer than 48% of them (80% of 60%) were offered a job.

It is not illegal to use a selection device with adverse impact. But to be legal it must be job- relevant, assessing a KSAO necessary for job success. Organization should be prepared to defend itself legally.

Americans with Disabilities Act of 1990 extended protection to people with disabilities. Essential functions, actions that must be done on a job. KSAOs that do not relate to essential functions are inappropriate in selection. Reasonable accommodation means that an organization must make reasonable allowances to enable a disabled person to do the job. Providing help with

INTRODUCTION 1/O PSYCHOLOGY CHAPTER 6

PAGE 6 OF 7 CHAPTER 6

nonessential functions, and providing help to get a wheelchair up and down steps, etc., are clearly included. Not yet clear what would be considered too difficult or expensive.

Affirmative action is used to increase the numbers of protected class members in targeted jobs, addressing past discrimination. Actions can range from recruitment to preferential treatment. Does not require quotas or hiring anyone without the necessary KSAOS. AA may be required by courts if the organization has a history of discrimination. [Executive order requires AA of all organizations with 50+ employees and federal contracts exceeding $50,000.] May have detrimental effects on the targeted groups, who may be seen--and see themselves--negatively (e.g., as incompetent). Can be viewed by most employees as fair and non-threatening when undertaken properly.

Future Issues

KSAO requirements for jobs are changing in the U.S. and elsewhere. Selecting for teams is one issue. In addition, organizations will most likely have to be able to identify people who are trainable rather than those who have the necessary KSAOs.

A second challenge is to end discrimination by making selection decisions based on job-relevant factors for as many jobs as possible.

INTRODUCTION 1/O PSYCHOLOGY CHAPTER 6

PAGE 7 OF 7 CHAPTER 6