The first step in developing a training plan is to determine what needs exist. For example, if employees do not know how to operate the machinery necessary to do their job, a training program on how to operate the machinery is clearly needed. On the other hand, when a group of office workers is performing poorly, training may not be the answer. The problem could be motivation, aging equipment, poor supervision, inefficient work design, or a deficiency of skills and knowledge. Only the last could be remedied by training. As training programs are being developed, the manager should set specific and measurable goals specifying what participants are to learn. Managers should also plan to evaluate the training program after employees complete it.

Many different training and development methods are available. Selection of methods depends on many considerations, but perhaps the most important is training content. When the training content is factual material (such as company rules or explanations of how to fill out forms), assigned reading, programmed learning, and lecture methods work well. When the content is interpersonal relations or group decision making, however, firms need to use a method that allows interpersonal contact, such as role-playing or case discussion groups. When employees must learn a physical skill, methods allowing practice and the actual use of tools and materials are needed, as in on-the-job training or vestibule training. (Vestibule training enables participants to focus on safety, learning, and feedback rather than on productivity.)

Web-based and other digital media-based training are especially popular today. Such methods allow a mix of training content, are relatively easy to update and revise, let participants use a variable schedule, and lower travel costs.⁹ On the other hand, they are limited in their capacity to simulate real activities and facilitate face-to-face interaction. Xerox, Massachusetts Mutual Life Insurance, and Ford have all reported tremendous success with these methods. In addition, most training programs rely on a mix of methods. Boeing, for example, sends managers to an intensive two-week training seminar involving tests, simulations, role-playing exercises, and flight-simulation exercises.¹⁰

Finally, some larger businesses have their own self-contained training facilities, often called corporate universities. McDonald’s was among the first to start this practice with its so-called Hamburger University in Illinois. All management trainees for the firm attend training programs there to learn exactly how long to grill a burger, how to maintain good customer service, and so on. The cult hamburger chain In-N-Out Burger also has a similar training venue it calls In-N-Out University. Other firms that use this approach include Shell Oil and General Electric.¹¹

Finally, the effectiveness of training and development programs should always be evaluated. Typical evaluation approaches include measuring one or more relevant criteria (such as attitudes or performance) before and after the training, and determining whether the criteria changed as a result of the training and development. Evaluation measures collected at the end of training are easy to get, but actual performance measures collected when the trainee is on the job are more important. Trainees may say that they enjoyed the training and learned a lot, but the true test is whether their job performance improves after their training.

Two basic categories of appraisal methods commonly used in organizations are objective methods and judgmental methods. Objective measures of performance include actual output (number of units produced), scrap rate, dollar volume of sales, and number of claims processed. Objective performance measures may be contaminated by “opportunity bias” if some persons have a better chance to perform than others. For example, a sales representative selling snowblowers in Michigan has a greater opportunity than does a colleague selling the same product in Alabama. Fortunately, adjusting raw performance figures for the effect of opportunity bias and thereby arriving at figures that accurately represent each individual’s performance is often possible.

Judgmental methods, including ranking and rating techniques, are the most common way to measure performance. Ranking compares employees directly with one another and orders them from best to worst. Ranking has a number of drawbacks. Ranking is difficult for large groups because the individuals in the middle of the distribution may be hard to distinguish from one another accurately. Comparisons of people in different work groups are also difficult. For example, an employee ranked third in a high-performing group may be more valuable than an employee ranked first in a lower-performing group. Another criticism of ranking is that the manager must rank people on the basis of overall performance, even though each person likely has both strengths and weaknesses. Furthermore, rankings do not provide useful information for feedback. To be told that one is ranked third is not nearly as helpful as to be told that the quality of one’s work is outstanding, its quantity is satisfactory, one’s punctuality could use improvement, or one’s interpersonal skills are excellent.

Rating differs from ranking in that it compares each employee with a fixed standard rather than with other employees. A rating scale provides the standard. Figure 10.2 gives examples of graphic rating scales for a bank teller. Each consists of a performance dimension to be rated (punctuality, congeniality, and accuracy), followed by a scale on which to make the rating. In constructing graphic rating scales, performance dimensions that are relevant to job performance must be selected. In particular, they should focus on job behaviors and results rather than on personality traits or attitudes.

Figure 10.2

Sample Performance Evaluation Form

Errors or biases can occur in any kind of rating or ranking system.¹⁴ One common problem is recency error, the tendency to base judgments on the subordinate’s most recent performance because it is most easily recalled. Often a rating or ranking is intended to evaluate performance over an entire time period, such as six months or a year, so the recency error does introduce error into the judgment. Other errors include overuse of one part of the scale—being too lenient, being too severe, or giving everyone a rating of “average.”

Halo error is allowing the assessment of an employee on one dimension to “spread” to ratings of that employee on other dimensions. For instance, if an employee is outstanding on quality of output, a rater might tend to give her or him higher marks than deserved on other dimensions. Errors can also occur because of race, sex, or age discrimination, intentionally or unintentionally. The best way to offset these errors is to ensure that a valid rating system is developed at the outset and then to train managers in how to use it.

One interesting approach to performance appraisal used in some organizations today is called 360-degree feedback, in which managers are evaluated by everyone around them—their boss, their peers, and their subordinates. Such a complete and thorough approach provides people with a far richer array of information about their performance than does a conventional appraisal given by just the boss. Of course, such a system also takes considerable time and must be handled so as not to breed fear and mistrust in the workplace.¹⁷

The last step in most performance appraisal systems is giving feedback to subordinates about their performances. This is usually done in a private meeting between the person being evaluated and his or her boss. The discussion should generally be focused on the facts: the assessed level of performance, how and why that assessment was made, and how it can be improved in the future. Feedback interviews are not easy to conduct, however. Many managers are uncomfortable with providing candid feedback, especially if the feedback is negative and subordinates are disappointed by what they hear. Properly training managers, however, can help them conduct more effective feedback interviews.¹⁸