Behavioral Performance Management

By Luthans, F.

Edited by Paul Ducham


The most traditional and researched theory of learning comes out of the behaviorist school of thought in psychology. Most of the principles of learning and organizational Reward Systems and the behavioral performance management approach discussed in this chapter are based on behavioristic theories, or behaviorism.

          The classical behaviorists, such as the Russian pioneer Ivan Pavlov and the American John B. Watson, attributed learning to the association or connection between stimulus and response (S-R). The operant behaviorists, in particular the well-known American psychologist B. F. Skinner, give more attention to the role that consequences play in learning, or the response-stimulus (R-S) connection. The emphasis on the connection (S-R or R-S) has led some to label these the connectionist theories of learning. The S-R deals with classical, or respondent, conditioning, and the R-S deals with instrumental, or operant, conditioning. An understanding of these conditioning processes is vital to the study of learning and serves as a point of departure for understanding and modifying organizational behavior.

Classical Conditioning
Pavlov’s classical conditioning experiment using dogs as subjects is arguably the single most famous study ever conducted in the behavioral sciences. A simple surgical procedure permitted Pavlov to measure accurately the amount of saliva secreted by a dog. When he presented meat powder (unconditioned stimulus) to the dog in the experiment, Pavlov noticed a great deal of salivation (unconditioned response). On the other hand, when he merely rang a bell (neutral stimulus), the dog did not salivate. The next step taken by Pavlov was to accompany the meat with the ringing of the bell. After doing this a number of times, Pavlov rang the bell without presenting the meat. This time, the dog salivated to the bell alone. The dog had become classically conditioned to salivate (conditioned response) to the sound of the bell (conditioned stimulus). Thus, classical conditioning can be defined as a process in which a formerly neutral stimulus, when paired with an unconditioned stimulus, becomes a conditioned stimulus that elicits a conditioned response; in other words, the S-R (i.e., bell-saliva) connection is learned. The Pavlov experiment was a major breakthrough and has had a lasting impact on the understanding of learning.
      Despite the theoretical possibility of the widespread applicability of classical conditioning and its continued refinement and application to areas such as modern marketing, most contemporary learning theorists agree that it represents only a very small part of total human learning and behavior. Skinner in particular felt that classical conditioning explains only respondent (reflexive) behaviors. These are the involuntary responses that are elicited by a stimulus. Skinner felt that the more complex, but common, human behaviors cannot be explained by classical conditioning alone. When explaining why he was abandoning a stimulus-response psychology, Skinner noted, “The greater part of the behavior of an organism was under the control of stimuli which were effective only because they were correlated with reinforcing consequences.” Thus, Skinner, through his extensive research, posited that behavior was a function of consequences, not the classical conditioning eliciting stimuli. He felt that most human behavior affects, or operates on, the environment to receive a desirable consequence. This type of behavior is learned through operant conditioning.

Operant Conditioning
Operant conditioning is concerned primarily with learning that occurs as a consequence of behavior, or R-S. It is not concerned with the eliciting causes of behavior, as classical, or respondent, conditioning is. The specific differences between classical and operant conditioning may be summarized as follows:

1. In classical conditioning, a change in the stimulus (unconditioned stimulus to conditioned stimulus) will elicit a particular response. In operant conditioning, one particular response out of many possible ones occurs in a given stimulus situation. The stimulus situation serves as a cue in operant conditioning. It does not elicit the response but serves as a cue for a person to emit the response. The critical aspect of operant conditioning is what happens as a consequence of the response. The strength and frequency of classically conditioned behaviors are determined mainly by the frequency of the eliciting stimulus (the environmental event that precedes the behavior). The strength and frequency of operantly conditioned behaviors are determined mainly by the consequences (the environmental event that follows the behavior).

2. During the classical conditioning process, the unconditioned stimulus, serving as a reward, is presented every time. In operant conditioning, the reward is presented only if the organism gives the correct response. The organism must operate on the environment (thus the term operant conditioning) in order to receive a reward. The response is instrumental in obtaining the reward. Table 12.1 gives some simple examples of classical (S-R) and operant (R-S) conditioning.

   Operant conditioning has a much greater impact on human learning than classical conditioning. Today, even though Skinner died in 1990, he remains somewhat controversial and his views are commonly misrepresented, the operant theory is still being refined and expanded, historical analyses recognize some limitations but also definite contributions, and applications are being made in areas such as marketing and performance management. Operant conditioning also explains, at least in a very simple sense, much of organizational behavior. For example, it might be said that employees work eight hours a day, five days a week, in order to feed, clothe, and shelter themselves and their families. Working (conditioned response) is instrumental in obtaining the food, clothing, and shelter.
      Some significant insights can be gained directly from operant analysis. The consequences of organizational behavior can change the Environmental Situation and greatly affect subsequent employee behaviors. Managers can analyze the consequences of organizational behavior to help accomplish the goals of prediction and control. Some organizational behavior researchers are indeed using the operant framework to analyze specific areas such as escalation of commitment (where a tendency of decision makers is to “throw good money after bad”) as well as more generally the effectiveness of managers at work. In addition, this theory serves as the framework for operationalizing much of behavioral performance management presented in this chapter.

Table 12.1


For understanding organizational behavior in general, the cognitive theories can also be used to understand learning and, especially as an input into social and social cognitive theories, to better understand behavioral performance management. Edward Tolman is widely recognized as a pioneering cognitive theorist. He felt that Cognitive Learning consists of a relationship between cognitive environmental cues and expectation. He developed and tested this theory through controlled experimentation. In fact, even though behaviorists are mostly associated with animal subjects in their research, Tolman was one of the first to extensively use the now-famous white rat in psychological experiments. He found that a rat could learn to run through an intricate maze, with purpose and direction, toward a goal (food). Tolman observed that at each choice point in the maze, expectations were established. In other words, the rat learned to expect that certain cognitive cues associated with the choice point might eventually lead to food. If the rat actually received the food, the association between the cue and the expectancy was strengthened, and learning occurred. In contrast to the S-R and R-S learning in the classical and Operant approach, Tolman’s approach could be depicted as S-S (stimulus-stimulus), or learning the association between the cue and the expectancy.
      In another early, classic study to demonstrate cognitive learning, Wolfgang Kohler used chimps presented with a problem of obtaining an out-of-reach suspended banana. At first the chimps attempted to jump for it, but soon gave up and seized a box that had been placed in another part of the room, dragged it under the object, mounted it, and took down the fruit. Kohler called this more complex learning “insight.” The solution to the problem appeared as a whole, not as a series, gradual shaping of new responses as the operant approach would suggest. At the time (1927), famous social philosopher/critic Bertrand Russell concluded, “there are two ways of learning, one by experience, and the other by what Kohler calls ‘insight.’”
           Besides being the forerunner of modern cognitive theory, Tolman’s S-S connection and Kohler’s insightful learning also had a great impact on the early human relations movement. Industrial training programs starting after World War II (and in many respects still today) drew heavily on their ideas. Programs were designed to strengthen the relationship between cognitive cues (supervisory, organizational, and job procedures) and worker expectations (incentive payments for good performance). The theory was that the worker would learn to be more productive by building an association between taking orders or following directions and expectancies of monetary reward for this effort. The same is true for the creativity, problem-solving groups that have been so popular over the years; they have drawn heavily from the notion of insightful learning.
         Today, the cognitive sciences focus more on the structures and processes of human competence (for example, the role of memory and information processing) rather than on the acquisition and transition processes that have dominated learning theory explanations. In organizational behavior, the cognitive approach has been applied mainly to motivation theories. Expectations, attributions and locus of control, and Expectations, attributions and locus of control, and Goal Setting (which are in the forefront of modern work motivation) are all cognitive concepts and represent the purposefulness (which are in the forefront of modern work motivation) are all cognitive concepts and represent the purposefulness of organizational behavior. Many researchers are currently concerned about the relationship or connection between cognitions and organizational behavior.


Social Learning Theory served as the conceptual framework for the past several editions of this text. However, similar to the theory building in social psychology, primarily from the extensive work of widely recognized psychologist Albert Bandura, this edition of the text and this overview of learning recognizes the evolution to the more comprehensive social cognition. After first recognizing social learning, the discussion turns to social cognition and its derivatives of modeling and Self-Efficacy.

Social Learning
This theoretical approach to learning was the first to combine and integrate both behaviorist and cognitive concepts and emphasized the interactive, reciprocal nature of cognitive, behavioral, and environmental determinants. It is important to recognize that social learning theory recognizes and draws from the principles of classical and operant conditioning. But equally important is the fact that social learning theory went beyond classical and operant theory by recognizing that there is more to learning than direct learning via antecedent stimuli and contingent consequences. Social learning theory posits that learning can also take place via vicarious, or modeling, and self-control processes. Thus, social learning theory agrees with classical and operant conditioning processes, but says they are too limiting and adds vicarious, modeling, and self-control processes.

Social Cognition
This theory has emerged in recent years to go beyond social learning theory. Social cognitive theory extends learning and/or modifying behavior by giving more attention to the self-regulatory mechanisms. Specifically, social cognitive theory identifies five capabilities that people use to initiate, regulate, and sustain their behavior: (1) symbolizing, (2) forethought, (3) vicarious/modeling learning, (4) selfregulation, and (5) self-reflection. These human capabilities recognize cognitive processes, social learning, and self-efficacy. A closer look at social learning through the social cognitive derivatives of modeling and self-efficacy can lead to the better understanding of learning and contribute to the theoretical underpinning of behavioral performance management.

Modeling Processes
The vicarious, or modeling, processes essentially involve observational learning. “Modeling in accordance with social learning theory can account for certain behavior acquisition phenomena that cannot be easily fitted into either operant or respondent conditioning.”
            Many years ago, Miller and Dollard suggested that learning need not result from discrete stimulus-response or response-consequence connections. Instead, learning can take place through imitating (i.e., modeling) others. Bandura states:

Although behavior can be shaped into new patterns to some extent by rewarding and punishing consequences, learning would be exceedingly laborious and hazardous if it proceeded solely on this basis. . . . [It] is difficult to imagine a socialization process in which the language, mores, vocational activities, familial customs and educational, religious and political practices of a culture are taught to each new member by selective reinforcement of fortuitous behavior, without benefit of models who exemplify the cultural patterns in their own behavior. Most of the behaviors that people display are learned either deliberately or inadvertently, through the influence of example.

         Bandura has done considerable research demonstrating that people can indeed learn from others. This learning takes place in two steps. First, the person observes how others act and then acquires a mental picture of the act and its consequences (rewards and punishers). Second, the person acts out the acquired image, and if the consequences are positive, he or she will tend to do it again. If the consequences are negative, the person will tend not to do it again. These positive and negative consequences, of course, are where there is a tie-in with operant theory. But because there is cognitive, symbolic representation of the modeled activities instead of discrete response-consequence connections in the acquisition of new behavior or modifying existing behavior, modeling goes beyond the operant explanation. In particular, Bandura concludes that modeling involves interrelated subprocesses, such as attention, retention, and motoric reproduction, as well as reinforcement.

Self-efficacy has also been recognized as a construct in behavioral performance management. Bandura has defined self-efficacy as the “beliefs in one’s capabilities to organize and execute the courses of action required to produce given attainments.” In particular, when faced with a specific task or job, an employee’s efficacy will determine whether the necessary behavior will be initiated, how much effort will be expended and sustained, and how much persistence and resilience there will be when there are obstacles or even failure. In other words, people who believe they can perform well on a task (high self-efficacy) do better than those who think they will fail (low self-efficacy). Importantly for the field of organizational behavior, a stream of research studies meta-analyzed has found a strong relationship between self-efficacy and work-related performance. Also, those with high selfefficacy have the tendency to remain calm in a stressful situation. In other words, there is considerable evidence that those employees with high self-efficacy tend to persevere and end up doing a good job without suffering stress or burnout. Unlike predisposed personality traits, efficacy is a state that can be trained and developed. The input into efficacy is recognized to be mastery experiences, vicarious/modeling learning, social persuasion, and physiological/psychological arousal. Both managers and their employees who experience success, are trained through modeling, and are encouraged and aroused on a given task or job, will have their efficacy enhanced and will perform well. There seems to be considerable practical implications for understanding and developing self-efficacy in managers and employees for performance improvement.


In Thorndike’s own words, the law of effect is simply stated as follows: “Of several responses made to the same situation, those which are accompanied or closely followed by satisfaction [reinforcement] . . . will be more likely to recur; those which are accompanied or closely followed by discomfort [punishment] . . . will be less likely to occur.” From a strictly empirical standpoint, most behavioral scientists, even those with a strict cognitive orientation, generally accept the validity of this law. It has been demonstrated time after time in highly controlled learning experiments and is directly observable in everyday learning experiences. Sometimes called the laws of behavior, desirable, or reinforcing, consequences will increase the strength of the preceding behavior and increase its probability of being repeated in the future. Undesirable, or punishing, consequences will decrease the strength of the preceding behavior and decrease its probability of being repeated in the future. Sometimes a third law is added: If the behavior is followed by no consequence (neither a positive nor a negative contingent consequence) the behavior will extinguish over time (thus this is called the extinction principle or law).


Although there is wide acceptance of the laws of behavior, there may be some occasions when a person’s cognitive rationalizations might neutralize them. For example, people with inaccurate self-efficacy beliefs may not be affected by the consequences of their actions. Inthe workplace, this is a real problem for managers. Those with inaccurate or false selfefficacy beliefs who experience performance failures time after time will not learn from their mistakes or respond to the manager’s comments on how to correct the problem. They have high self-efficacy (they believe that their behaviors are appropriate to successfully accomplish the task), but they are wrong. In addition to this type of cognitive processing that may neutralize the law of effect, there is some disagreement when it is carried a step further and used as an overall theory or an absolute requirement for learning.
      Both Tolman’s and Kohler’s classic studies providing initial support for cognitive theories, presented earlier, discounted the need for incremental reinforcement as necessary for learning to occur. For example, Tolman conducted place learning, latent learning, and transposition experiments in an attempt to demonstrate that reinforcement was not a precondition for learning to occur. Specifically, in the place learning he trained a rat to turn right in a T maze in order to obtain the reward of food. Then he started the rat from the opposite part of the maze. According to operant theory, the rat should have turned right because of past conditioning. However, the rat turned toward where the food had been placed. Tolman concluded that the behavior was purposive; the rat had formed a cognitive map to figure out how to get to the food. Over time, the behaviorists were able to counteract Tolman’s studies with more controlled (e.g., sterile mazes, etc.) experiments, and Kohler’s conclusions about insight were also explained away by a reinforcement history alternative explanation.
      More recently, Deci and Deci and Ryan, through their cognitive evaluation theory and laboratory research studies, have posited that external consequences (i.e., rewards) have a negative impact on intrinsically motivated behavior dealing with task persistence and creativity. These findings generated considerable follow-up research with mixed findings. One review of about 100 studies found some rewards may have a detrimental effect, but an equal number found no effect or a positive effect. The conclusion from this extensive review was that (1) the detrimental effects of rewards occur under highly restricted, easily avoidable conditions; (2) mechanisms of classical and operant conditioning are basic for understanding incremental and detrimental effects of reward on task motivation; and (3) positive effects of rewards on performance are easily attainable using procedures derived from behavioral theory.
       Finally, a meta-analysis of 96 studies found that the only detrimental effect of rewards was the time spent carrying out laboratory activity following a performance-independent (i.e., a noncontingent) reward. There is also systematic analysis that discounts cognitive evaluation theory when compared to operant theory explanations.Yet, despite this considerable empirical and theoretical counterevidence, an unconvinced few such as popular author Alfie Kohn continue to write (not do research) with titles such as Punished by Rewards and “Why Incentive Plans Cannot Work.” Based on his own assumptions and the now-countered Deci and Ryan theory and research, and in stark contrast to the large body of Reinforcement Theory and research, he makes unequivocal statements such as: “The bottom line is that any approach that offers a reward for better performance is destined to be ineffective.”
     Unfortunately, Kohn’s largely unsupported statements do not fall on deaf ears in the real world. This is because practicing managers have indeed experienced some implementation problems with pay-for-performance programs. For example, after an extensive review of the relevant literature, Lawler concluded that process/design problems, not the underlying theory of reinforcement or the supporting basic research, limit the effectiveness of pay for performance. There is also a research study that found that highly dispersed reward systems (i.e., very large differences between highest and lowest payouts) may have a negative effect on both individual and organizational performace, especially when collaborative efforts (such as in teams) are important. Yet, once again, it is not that the theory/research on reinforcement is wrong, but rather it is the implementation that can cause problems. As Bandura points out, “To say that [only] thought guides action is an abbreviated statement of convenience rather than a conferral of agency of thought,” because “if people acted. . . on the basis of informative cues but remained unaffected by the results of their actions, they would be insensible to survive very long.” As a final summary statement, it can be said that the theory of reinforcement, like learning in general, is not perfect and still needs development. However, it can also be said that reinforcement does serve as an excellent theoretical foundation and evidence-based guiding principle, and the implementation issues need to be overcome by effective behavioral performance management.


The terms rewards and reinforcers are often used interchangeably and loosely, but in behavioral performance management have very precise definitions and usage. An often cited circular definition of reinforcement says that it is anything the person finds rewarding. This definition is of little value because the words reinforcing and rewarding are used interchangeably, but neither one is operationally defined. A more operational definition can be arrived at by reverting to the laws of behavior. Specifically, reinforcement in behavioral management is defined as anything that both increases the strength and tends to induce repetitions of the behavior that preceded the reinforcement. A reward, on the other hand, is simply something that the person who presents it deems to be desirable.
       Reinforcement is functionally defined. Something is reinforcing only if it strengthens the behavior preceding it and induces repetitions. For example, a manager may ostensibly reward an employee who found an error in a report by publicly praising the employee. Yet on examination it is found that the employee is embarrassed and chided by coworkers, and the error-finding behavior of this employee decreases in the future. In this example, the “reward” of public praise is not reinforcing. Besides clearing up differences between reinforcers and rewards, behavioral management also requires making the distinction between positive and negative reinforcers.


There is much confusion surrounding the terms positive reinforcement, negative reinforcement, and punishment. First of all, it must be understood that reinforcement, positive or negative, strengthens the behavior and increases the probability of repetition. But positive and negative reinforcers accomplish this impact on behavior in completely different ways. Positive reinforcement strengthens and increases behavior by the presentation of a desirable consequence. Negative reinforcement strengthens and increases behavior by the threat of the use of an undesirable consequence or the termination or withdrawal of an undesirable consequence. Figure 12.1 briefly summarizes the differences between positive and negative reinforcement and punishment. Giving recognition and attention to an employee for the successful completion of a task could be an example of positive reinforcement (if this does in fact strengthen and subsequently increase this task behavior). On the other hand, a worker is negatively reinforced for getting busy when the supervisor walks through the area. Getting busy prevents or terminates being “chewed out” by the supervisor.
           Negative reinforcement is more complex than positive reinforcement, but it should not be equated with punishment. In fact, they have opposite effects on behavior. Negative reinforcement strengthens and increases behavior, whereas punishment weakens and decreases behavior. However, both are considered to be forms of negative control of behavior. Negative reinforcement is really a form of social blackmail, because the person will behave in a certain way in order not to be punished. A clearer understanding of punishment will help further clarify how it differs from negative reinforcement.

Figure 12.1


Unfortunately, about the only reinforcing function that traditional monetary reward systems such as base-pay techniques provide is to reinforce employees for walking up to the pay window or for opening an envelope and seeing their paycheck or direct deposit stub every two weeks or every month. These traditional pay plans certainly have come up short of having the intended impact on improving employee performance at all levels. Yet, despite the problems with traditional pay approaches, money is still a very important reward to employees at all levels. For example, former GE head and now with his wife giving advice to managers in an online column, Jack Welch recently quipped,

You surely have seen how effective money is in lighting a motivational fire—even in employees who claim money doesn’t matter to them. Plaques gather dust. Checks can be cashed. And employees know the difference in their bones.

Recent analyses of the research studies also conclude that money contingently administered can have a positive effect on employee behavior.However, there are even shortcomings with merit pay

1. Unless a merit raise is at least 6 to 7 percent of base pay, it will not produce the desired effects on employee behavior.

2. Beyond a certain point, increases in merit-raise size are unlikely to improve performance.

3. When merit raises are too small, employee morale will suffer.

4. Cost-of-living adjustments, seniority adjustments, and other nonmerit components of a raise should be clearly separated from the merit component.

5. Smaller percentage raises given to employees at the higher ends of base-pay ranges are demotivating.

In other words, both the traditional base- and merit-pay plans have some problems.

           The “New Pay” plans (e.g., pay for performance at both the individual and group levels, paying for customer and/or employee satisfaction, pay for knowledge, skill pay, competency pay, and broadbanding) have overcome many of the problems. For example, a large study sponsored by the American Compensation Association was able to place a dollar value on the positive impact of pay-for-performance plans. The value of the performance improvement translates into a 134 percent net return on what is paid out to employees (excluding the costs associated with training, communications, and consulting), or, for every $1 of payout, a gain of $2.34 was attained.
        In terms of basic research, a field experiment conducted by Stajkovic and Luthans in the biggest credit card processing firm in the world found the following:

1. A traditionally administered pay-for-performance plan (i.e., announced through normal channels in terms of the amount of pay that would be received for various levels of performance) did increase performance by 11 percent; but
2. The same plan that was implemented through the behavioral performance management approach discussed next (i.e., specifying the critical performance behaviors that would lead to monetary consequences) had a significantly higher 32 percent increase in performance.In other words, because the performance behaviors strengthened and increased, the theory and principles of reinforcement explain that money can indeed be a powerful reinforcer. Importantly, money may not be a reinforcer when administered through the traditional pay plans, but when made contingent on identified performance behaviors as in behavioral performance management, money can be a powerful reinforcer.
            The same could be said for the very expensive benefit plans in the organizational reward system. Flexible benefit plans and those that depend on performance may have better intended results. Instead of benefits taking on an entitlement mentality, an increasing number of firms (18% according to an American Compensation Association survey) are making the amount and choice of benefits dependent on employee performance. For example, under Owens-Corning’s “Rewards and Resources Program,” workers get to clearly see how their work is reinforced with extra pay in the form of more benefit choices.


Money is the most obvious organizational reward, but the nonfinancial rewards are receiving increased attention. In fact, one comprehensive review of surveys that ask the value employees place on various rewards found that nonfinancial rewards were ranked much higher than financial ones. For example, one study of 1,500 employees in a wide variety of work settings found personalized, instant recognition from managers as being the most important of the 65 types of rewards evaluated. However, more than half of these same employees reported that they seldom, if ever, received such personal recognition from their managers. Also, a staffing company reported that the number-one reason employees give for leaving companies is the lack of praise and recognition. Also in the same Stajkovic and Luthans research study cited previously, it was found that both social recognition (24%) and Performance Feedback (20%) had a significantly higher relative performance increase than did the traditionally administered pay for performance (11%). Finally, in a recently conducted study in the fast-food industry, we (Peterson and Luthans) found that financial incentives initially had a bigger effect on profit, customer service, and employee retention outcomes, but, over time, except for employee retention, both financial and nonfinancial incentives had an equally significant impact. In other words, there is little doubt that the nonfinancials can be very powerful, but are often overlooked as a reinforcer in behavioral performance management.
     Table 12.2 summarizes some of the major categories of nonfinancial rewards. Notice that even though these are considered nonfinancial, they may still cost the organization. This is true of the consumables, manipulatables, and visual and auditory rewards. The job design category is a special case and is usually not, but could be, considered as an organizational reward. On the other hand, the social recognition and attention and performance feedback categories are relatively easy to apply in behavioral performance management, cost nothing (except perhaps for preparing some of the performance feedback), and may be even more powerful than the cost-based nonfinancial rewards. These two are major reinforcers and deserve special coverage.

Social Recognition and Attention
Informally providing contingent recognition and attention (and praise, if genuine) tends to be a very powerful reinforcer for most people. In addition, few people become satiated or filled up with this; no one “suffers” from too much genuine recognition. However, similar to monetary reinforcers, social reinforcers should be administered on a contingent basis to have a positive effect on employee performance. For example, a pat on the back or verbal praise that is insincere or randomly given (as under the old human relations approach) may have no effect or even a punishing “boomerang” effect. But genuine social reinforcers, contingently administered for performance of the target behavior, can be a very effective positive reinforcer for most employees and improve their performance. The added benefit of such a strategy, in contrast to the use of monetary rewards, is that the cost of social reinforcers to the organization is absolutely nothing.
       Importantly, this informal social recognition based on a valued person’s (e.g., boss, peer, subordinate, friend, spouse, etc.) attention and appreciation may have not only a bigger impact as a reinforcer in behavioral management than money, but also than formal recognition programs. Unlike valued social recognition and attention, formal recognition programs, especially over time, can easily turn into being phoney, not valued by the recipient, or go against group and/or cultural norms. As Luthans and Stajkovic noted:

A formal recognition award such as the “Golden Banana” at Hewlett-Packard or “Employee of the Month” given at many companies can initially be a reinforcer, but over time may cross the line and become an empty reward and be perceived even in a negative light. The first few Employee of the Month recipients may be very deserving instances that everyone would agree with, but over time selections become more and more controversial and subjective, usually resulting in selecting less-qualified or not-qualified employees. At this point company politics often come into play and those who truly deserved the recognition feel betrayed. In this case, the program would actually produce negative effects (e.g., “rewarding A while hoping for B”). Also, from a (collectivistic) cultural values and individual differences standpoint, although everyone may like to be recognized for their efforts and achievements, not everyone likes to be singled out in the public way that usually goes along with formal recognition.

With the increasing use of teams, there is also recent evidence that they may be providing social reinforcement to their members that yields organizationally desirable outcomes. For example, in the American Compensation Association research study cited earlier, team suggestion plans, under the umbrella of an organizational performance reward plan or operating independently, were found to be particularly powerful contributors to organizational success. Importantly, the team suggestion plans, which typically used nonfinancial rewards, outperformed the individually based plans, which typically used financial rewards, by 4 to 1. For example, the average value per idea adopted from team suggestion plans using nonfinancial rewards was an impressive $46,200 for a major airline, $14,500 for a manufacturer, $19,344 for a newspaper, and $19,266 for a bank.

Performance Feedback
There is little question that despite the tremendous amount of data being generated by today’s advanced information systems, individuals still receive very little, if any, feedback about their performance. People generally have an intense desire to know how they are doing; they engage in feedback-seeking behavior. Even though feedback has been found to be complex in research studies, it is generally accepted that feedback enhances individual performance in behavioral management. A comprehensive review (30 laboratory and 42 Field Experiments) concluded that performance feedback had a positive effect. Also, as cited earlier, the Stajkovic and Luthans study found that, although not as high as contingently administered money and social recognition reinforcers, the performance feedback intervention still yielded a highly significant 20 percent performance improvement. Importantly, this was significantly higher than the traditionally administered pay for performance (11 percent). As a general guideline for behavioral management, the performance feedback should be as positive, immediate, graphic, and specific—thus, the acronym PIGS—as possible to be effective.
     Despite the recognized importance, there is still disagreement among scholars as to whether feedback per se is automatically reinforcing or too simplistic. For example, after reviewing the existing research literature on feedback, one researcher concluded that its impact is contingent on factors such as the nature of the feedback information, the process of using feedback, individual differences among the recipients of the feedback, and the nature of the task. One study, for instance, found that self-generated feedback with goal setting had a much more powerful effect on technical or engineering employees than externally generated feedback with goal setting. Also, another study found subjects rated specific feedback more positively than they rated nonspecific feedback and preferred feedback that suggested an external cause of poor performance to feedback that suggested an internal cause.
     An argument can also be made that “actionable feedback” (feedback that leads to learning and appropriate results) is more effective than just critical, negative feedback. And the source of the feedback seems important as well. Not only are the amount and the frequency of feedback generated by a source important, but also the consistency and usefulness of the information generated, as a study found. Individuals viewed feedback from formal organizations least positively, from coworkers next, then from supervisors and tasks, with the best being self-generated feedback. Feedback from multiple sources may be most effective, and the 360-degree feedback systems (the individual is anonymously appraised not only by the boss but also by subordinates, peers, and sometimes customers) can be automated on a software system to provide more timely, objective, and less-costly feedback. Also, studies have found that choice of reward interacting with feedback had a positive impact on task performance in a laboratory exercise, but workers in highly routine jobs who received positive feedback did not improve their performance. Despite these qualifications and contingencies, a general guideline regarding performance feedback is that it can be a very effective reinforcer for behavioral performance management.

Table 12.2


In this first step the critical behaviors that make a significant impact on performance (making or selling a product or providing a service to clients or customers) are identified. In every organization, regardless of type or level, numerous behaviors are occurring all the time. Some of these behaviors have a significant impact on performance, and some do not. The goal of the first step of O.B. Mod. is to identify the critical behaviors—the 5 to 10 percent of the behaviors that may account for up to 70 or 80 percent of the performance in the area in question.
        The process of identifying critical behaviors can be carried out in a couple of ways. One approach is to have the person closest to the job in question—the immediate supervisor or the actual jobholder—determine the critical behaviors. This goes hand in hand with using O.B. Mod. as a problem-solving approach for the individual manager or a team. Its advantages are that the person who knows the job best can most accurately identify the critical behaviors, and, because that person is participating, he or she may be more committed to carrying the O.B. Mod. process to its successful completion.
           Another approach to identifying critical behaviors would be to conduct a systematic behavioral audit. The audit would use internal staff specialists and/or outside consultants. The audit would systematically analyze each job in question, in the manner that jobs are analyzed using job analysis techniques commonly employed in human resource management. The advantages of the personal approach (where the jobholder, immediate supervisor, and/or team makes a vital input into the audit) can be realized by the audit. In addition, the advantages of information from those closest to the action and consistency can be gained.
        Regardless of the method used, there are certain guidelines that can be helpful in identifying critical behaviors. First, only direct performance behaviors are included. A team’s lack of commitment and teamwork or someone’s “goofing off ” all the time is unacceptable. Only direct performance behaviors such as absenteeism or attendance, tardiness or promptness, or, most importantly, doing or not doing a particular task or procedure that leads to quantity and/or quality outcomes play the major role in O.B. Mod. Something like goofing off is not acceptable because it is not operationally measurable. It could be broken down into measurable behaviors such as not being at the workstation, being tardy when returning from breaks, spending time at the water cooler, disrupting coworkers, playing computer games or surfing for personal reasons, and even Socializing with coworkers face-to-face or with others online. However, for a behavior to be identified as a critical behavior appropriate for O.B. Mod., there must be a positive answer to the questions: (1) Can it be directly measured? and (2) Does it have a significant impact on a performance outcome?
     Most organizations do not have problems with their technology or the ability or training of their people, but they have many behaviorally related performance problems. Functional behaviors (those that contribute to performance goals) need to be strengthened and accelerated in frequency, and dysfunctional behaviors (those that detract from, or are detrimental to, performance goals) need to be weakened and decelerated in frequency. As in the initial step of any problem-solving process, these behaviors must be properly identified, or the subsequent steps of O.B. Mod. become meaningless for attaining the overall goal of performance improvement.


After the performance behaviors have been identified in step 1, they are measured. A baseline measure is obtained by determining (either by observing and counting or by extracting from existing records) the number of times that the identified behavior is occurring under existing conditions. Often this baseline frequency is in and of itself very revealing. Sometimes it is discovered that the behavior identified in step 1 is occurring much less or much more frequently than anticipated. The baseline measure may indicate that the problem is much smaller or much bigger than was thought to be the case. In some instances, the baseline measure may cause the “problem” to be dropped because its low (or high) frequency is now deemed not to need change. For example, attendance may have been identified in step 1 as a critical behavior that needed to be improved. The supervisor reports that the people “never seem to be here.” The baseline measure, however, reveals that on average there is 96 percent attendance, which is deemed to be acceptable. In this example, the baseline measure rules out attendance as being a problem. The reverse, of course, could also have occurred. Attendance may have been a much bigger problem than anticipated.
        The purpose of the baseline measure is to provide objective frequency data on the critical behavior. A baseline frequency count is an operational definition of the strength of the behavior under existing conditions. Such precise measurement is the hallmark of any scientific endeavor, and it separates O.B. Mod. from more subjective human resource management approaches, such as participation. Although the baseline is established before the intervention to see what happens to the behavior as a result of the intervention, it is important to realize that measures are taken after the intervention as well. Busy managers may feel that they do not have time to record behavioral frequencies objectively, but, at least initially, they should record them in order to use the O.B. Mod. approach effectively. Most measures, however, can be taken from existing archival data (e.g., quality, sales, and productivity numbers) that are gathered for other purposes and can be easily obtained for this measurement step of O.B. Mod.


Once the performance behavior has been identified and a baseline measure has been obtained, a functional analysis is performed. A functional analysis identifies both the antecedents (A) and consequences (C) of the target behavior (B), or, simply stated, an A-BC analysis is performed. As discussed under behavioristic learning theory and operant conditioning, both the antecedent and the consequent environments are vital to the understanding, prediction, and control of human behavior in organizations. Remember that in an operant approach, cognitive mediating processes do not play a role. Such an omission may detract from the comprehensive understanding of organizational behavior and the analysis of modeling and self-control processes, but for pragmatic application, an A-B-C functional analysis may be sufficient. In the A-B-C functional analysis, A is the antecedent cue, B is the performance behavior identified in step 1, and C is the contingent consequence. Table 12.3 identifies some of the As, Bs, and Cs for attendance and absenteeism. A review of absenteeism found work unit size, worker responsibility, and organizational scheduling to be three potential antecedent influences that could be used to improve employee attendance, and feedback, rewards, and punishers to be effective attendance control procedures.
     This functional analysis step of O.B. Mod. brings out the problem-solving nature of the approach. Both the antecedent cues that emit the behavior, and sometimes control it, and the consequences that are currently maintaining the behavior must be identified and understood before an effective intervention strategy can be developed. In this step, the question may be asked as to whether the employee can do the identified performance behavior if his/her life depended on it. If the answer is “no,” then there may be an “A” problem (i.e., equipment, training, even expectations) that must be attended to. However, this is usually not the case. The human resources of an organization can do the identified critical performance behavior if their lives depend on it, but they are not doing it. Then this becomes a “C” problem. They know how to do the desired performance behavior and there is all the necessary support to do it, but there are not reinforcing consequences and therefore the behavior is not occurring. This lack of reinforcing consequences is the major problem and challenge facing behavioral management. The accompanying Application Example: Functional Analysis in Action gives the functional analysis of a supervisor’s problem of his workers’ taking unscheduled breaks.

OB in Action: Functional Analysis in Action

In an actual case of an O.B. Mod. application, a production supervisor in a large manufacturing firm identified unscheduled breaks as a critical behavior affecting the performance of his department. It seemed that workers were frequently wandering off the job, and when they were not tending their machines, time—and irrecoverable production—was lost. When a baseline measure of this critical behavior was obtained, the supervisor was proved to be right. The data indicated that unscheduled breaks (defined as leaving the job for reasons other than to take a scheduled break or to obtain materials) were occurring in the department on a relatively frequent basis. The functional analysis was performed to determine the antecedent(s) and consequence(s) of the unscheduled-break behavior.

   It was found that the clock served as the antecedent cue for the critical behavior. The workers in this department started work at 8 A.M., they had their first scheduled break at 10 A.M., and they had lunch at noon. They started again at 1 P.M., had a break at 3 P.M., and quit at 5 P.M. The functional analysis revealed that almost precisely at 9 A.M., 11 A.M., 2 P.M., and 4 P.M., a consistent number of workers were leaving their jobs and going to the rest room. In other words, the clock served as a cue for them to take an unscheduled break midway between starting time and the first scheduled break, between the first scheduled break and lunch, between lunch and the scheduled afternoon break, and between the afternoon break and quitting time. The clock did not cause the behavior; it served only as a cue to emit the behavior. On the other hand, the behavior was under stimulus control of the clock because the clock dictated when the behavior would occur. The consequence, however, was what was maintaining the behavior. The critical behavior was a function of its consequences. The functional analysis revealed that the consequence of the unscheduled-break behavior was escaping from a dull, boring task (that is, the unscheduled-break behavior was being negatively reinforced) and/or meeting with coworkers and friends to socialize and have a cigarette (that is, the unscheduled-break behavior was being positively reinforced). Through such a functional analysis, the antecedents and consequences are identified so that an effective intervention strategy can be developed.

Table 12.3



The first three steps in an O.B. Mod. approach are preliminary to this action step, the intervention. The goal of the intervention is to strengthen and accelerate functional performance behaviors and/or weaken and decelerate dysfunctional behaviors. There are several strategies that can be used, but the main ones are positive reinforcement and punishment–positive reinforcement.

A Positive Reinforcement Strategy
Positive, not negative, reinforcement is recommended as an effective intervention strategy for O.B. Mod. The reason is that positive reinforcement represents a form of positive control of behavior, whereas negative reinforcement represents a form of negative control of behavior. Traditionally, and to a large extent still today, organizations depend on negative control. People come to work in order not to be fired, and they look busy when the supervisor walks by in order not to be punished. Under positive control, the person behaves in a certain way in order to receive the desired consequence. Under positive control, people come to work in order to be recognized for making a contribution to their department’s goal of perfect attendance, or they keep busy whether the supervisor is around or not in order to receive incentive pay or because they get social recognition/attention and feedback for their good work. Positive control through a positive reinforcement intervention strategy is much more effective and longer lasting than negative control. It creates a much healthier and more productive organizational climate.

              A positive reinforcer used as an O.B. Mod. intervention strategy could be anything, as long as it increases the performance behavior. Most often money is thought of as the logical, or sometimes the only, positive reinforcer available to managers using this approach. Money is potentially a very powerful reinforcer, but it often turns out to be ineffective because it is not contingently administered as a consequence of the behavior being managed. Besides money, positive reinforcers that are also very powerful, readily available to all behavioral managers, and cost nothing are the social reinforcers (attention and recognition) and performance feedback. These reinforcers (money, recognition, and feedback) can be and, as has been demonstrated through research, have been used as an effective O.B. Mod. strategy to improve employee performance. In fact, the most comprehensive evidence shown in Figure 12.5 indicates that when these three reinforcers are used in combination in the intervention, they produce a stronger (synergistic) effect and probability of success than any of the reinforcers used by themselves, the sum of the individual effects, or the combination of any two of the interventions.

A Punishment–Positive Reinforcement Strategy
There is little debate that a positive reinforcement strategy is the most effective intervention for O.B. Mod. Yet realistically it is recognized that in some cases the use of punishment to weaken and decelerate undesirable Behaviors cannot be avoided. This would be true in the case of something like unsafe behaviors that need to be decreased immediately. However, as was pointed out earlier, so many negative side effects such as hate and revenge accompany the use of punishment that it should be avoided if at all possible. Punished behavior tends to be only temporarily suppressed; for example, if a supervisor reprimands an associate for some dysfunctional behavior, the behavior will decrease in the presence of the supervisor but will surface again when the supervisor is absent. In addition, a punished person becomes very anxious and uptight; reliance on punishment may have a disastrous impact on employee satisfaction and commitment and create unnecessary stress.
           Perhaps the biggest problem with the use of punishment, however, is that it is very difficult for a supervisor to switch roles from punisher to positive reinforcer. Some supervisors and managers rely on a negative approach so much in dealing with their associates that it is almost impossible for them to administer positive reinforcement effectively. This is a bad situation for the management of human resources because the use of positive reinforcement is a much more effective way of changing employee behavior. If punishment is deemed necessary, the desirable alternative behavior (for example, safe behavior) should be positively reinforced at the first opportunity. Use of this combination strategy will cause the alternative desirable behavior to begin to replace the undesirable behavior in the person’s behavioral repertoire. Punishment should never be used alone as an O.B. Mod. intervention. If punishment is absolutely necessary, it should only be used in combination with positive reinforcement of the desirable alternative behavior.

Figure 12.5


A glaring weakness of most human resource management programs is the absence of any systematic, built-in evaluation. A comprehensive analysis of the evaluation of human resources programs concluded that the traditional approach has been “to review a program with one or two vice presidents at the corporate office, various managers in the field, and perhaps a group of prospective trainees. It continues to be used until someone in a position of authority decides that the program has outlived its usefulness. All of this is done on the basis of opinion and judgment.” Such haphazard evaluations have resulted in the termination of some effective programs and the perpetuation of some ineffective ones. In either case, there are severe credibility problems, and today all programs dealing with people, whether they are government social service programs or human resource management programs, are under the pressure of evaluation and accountability. Human resource managers no longer have the luxury of just trying something new and different and hoping they can improve performance. Today there is pressure for everything that is tried to be proved to have value. As in the case of the validity of selection, training, and appraisal techniques, systematic evaluations of all human resource management techniques should have been done all along.
           O.B. Mod. attempts to meet the credibility and accountability problems head on by including evaluation as an actual part of the process. In this last step of the approach, the need for Kirkpatrick’s well-known four levels of evaluation (reaction, learning, behavioral change, and performance improvement) is stressed. The reaction level refers simply to whether the people using the approach and those having it used on them like it. If O.B. Mod. is well received and there is a positive reaction to it, there is a better chance of its being used effectively. In addition, reaction evaluations are helpful because (1) positive reactions help ensure organizational support, (2) they can provide information for planning future programs, (3) favorable reactions can enhance the other levels of evaluation (learning, behavioral change, and performance improvement), and (4) they can provide useful comparative data between units and across time.
        The second level of evaluation is learning, which is especially important when first implementing an O.B. Mod. approach. Do the people using the approach understand the theoretical background and underlying assumptions and the meaning of, and reasons for, the steps in the model? If they do not, the model will again tend to be used ineffectively. The third level is aimed at behavioral change. Are behaviors actually being changed? The charting of behaviors started in step 2 of the O.B. Mod. process gives objective data for this level of evaluation. The fourth and final level, performance improvement, is the most important. The major purpose of O.B. Mod. is not just to receive a favorable reaction, learn the concepts, and change behaviors. These dimensions are important mainly because they contribute to the overriding purpose, which is to improve performance. “Hard” measures (for example, data on quantity and quality, turnover, absenteeism, customer complaints, customer satisfaction, employee grievances, safety, length of patient stay, number of clients served, sales revenue, and rate of return on investment) and scientific methodology are used whenever possible to systematically evaluate the impact of O.B. Mod. on performance.


There is a considerable body of research that has evaluated the effectiveness of behavioral performance management in general and the five-step O.B. Mod. approach in particular. It has been widely applied in manufacturing as well as in nonmanufacturing, service-oriented organizations. In addition to the direct application of O.B. Mod. as described, considerable basic research has been conducted on operant and social learning and social cognitive variables in experimental psychology. For many years and in very recent times, a number of studies have assessed the application of the behavioral management approach to improving employee performance in a number of different areas. The following summarizes some of these areas:

1. Employee productivity. Most applications by far have focused on performance output. The considerable number of research studies clearly indicate that employee productivity or task completion is positively affected by behavioral management techniques. The performance improvement is for both quantity and quality of employee output and cuts across virtually all organizational settings and all intervention techniques.

2. Absenteeism and Tardiness. This is probably the second-biggest area of application. Studies that have examined this area have typically used small monetary incentives or lottery Incentive Systems for attendance or promptness and/or punishers for absenteeism or tardiness. One extensive search of this literature found very positive results. The six most sound methodological studies reported an 18 to 50 percent reduction in the absence rate and a 90 percent reduction in the frequency of tardiness. One study found a positive, causal impact that an O.B. Mod. program had on the attendance of employees in a bank.

3. Safety and accident prevention. Most organizations, especially manufacturing firms and others in which dangerous equipment is used, are very concerned about safety. However, because accidents occur at such a relatively low frequency, most studies have focused on reducing identifiable safety hazards or increasing safe behaviors (for example, wearing earplugs, which went from 35 to 95 percent compliance according to one study; wearing hard hats; and keeping the safety guard in place on dangerous equipment). A review of the research indicates the considerable success that behavioral management techniques have had in these areas. Some actual company examples are Boston Gas, where employees without accidents are eligible for lottery drawings; Virginia Power, where employees can win from $50 to $1,000 for safe work habits; Southern New England Telecommunications, which gives gift coupons to employees without accidents; and Turner Corporation, a New York–based engineering and construction firm, where employees can earn company stock if they meet safety goals. All these companies report lower accident rates through the use of a behavioral management approach. Southern Fineblanking, a 225-employee metal stamping plant in South Carolina, reported a 33 percent reduction in accidents after implementing a behavioral management program aimed at safety, and the average cost per injury decreased from $1,400 to $500.

4. Sales performance. Sales managers and Trainers have traditionally relied on internal motivation techniques to get their salespeople to improve their performance. For example, one behavioral performance management consultant tells about a company that gave its sales personnel a typical high-powered, multimedia training program, which supposedly taught them effective selling skills. However, when the enthusiastic trainees finished the program and actually tried the things presented to them in the program, they received little, if any, feedback or reinforcement. Within a few weeks the enthusiasm began to wane, and, most important, actual sales performance began to decline. In other words, even though these salespeople had probably acquired effective selling skills during their training, the environment did not support (reinforce) the use of these skills. A behavioral performance management approach, in which important selling behaviors such as customer approach, suggestive statements, and closing statements are identified, measured, analyzed, intervened in, and evaluated, would be an alternative to the motivationskill- teaching approach. A comprehensive review of the behavioral approach to sales in restaurants, retail stores, wholesale establishments, and telephone operations found considerable success. When a combination of antecedent and consequence intervention strategies was used, dramatic improvements were shown in areas such as wine and dessert sales, average customer transactions, customer assistance, sales forecasting, sales-call frequency, sales of telephone services, and airline reservations. A study of fast-food restaurants also found that antecedent prompts (“Can I get you some fries with that?”) significantly increased consumer purchases, and another more recent study in the same industry indicated a significant increase in customer service. The successful application of O.B. Mod. to the selling, absent-from-the-workstation, and idle-time behaviors of clerks in a large retail store was also found.

Although these results are not exhaustive and do not always reflect the exact O.B. Mod. model outlined in this chapter, they are representative of the very extensive application of the behavioral performance management approach. In addition, both comprehensive qualitative and quantitative (meta-analytic) reviews strongly support the findings.


As cited throughout, the specific O.B. Mod. model has been directly tested by basic research and has been found to have positive performance results in both manufacturing and service organizations (retail, restaurants, banking, and hospitals). The O.B. Mod. approach has also “gone international” and has been shown to have a positive impact on the performance behaviors and output of Russian factory workers, Russian retail clerks, and most recently South Korean telecommunication employees.
        The two Stajkovic and Luthans meta-analyses mentioned earlier and, especially, the one that specifically reviewed O.B. Mod. studies (as opposed to the latest one on all behavioral management studies) and is utilized in the principles, examined the relationship between O.B. Mod. as defined here and task performance. The overall result of the O.B. Mod. meta-analysis was, on average, an impressive 17 percent increase in task performance (as indicated, it was 16 percent for all behavioral management studies). Further analysis revealed that O.B. Mod. had a stronger average effect in manufacturing firms (33%) than in service organizations (13%), but the O.B. Mod. approach was highly significant in both. The difference in application effectiveness of O.B. Mod. between manufacturing and service organizations was explained as

(1) the definition and accurate assessment of performance outcomes; and (2) the nature of the employee behaviors and work processes involved in the delivery of performance outcomes. The first point refers to the difference between the definition and measurement of the more vague and complex service organization performance outcomes (e.g., customer satisfaction, return business) versus tangible performance outcomes (e.g., productivity and quality) in manufacturing organizations. The second point refers to the difference between specifying service delivery employee behaviors and processes that go into making a tangible product. Service performance behaviors and outcomes are more complex and less identifiable than those found in manufacturing organizations.

So, although O.B. Mod. may be more difficult to apply in service than in manufacturing organizations, it still works in both, and the challenge is to make it even more effective in service applications. The overall implications of these findings from the meta-analyses are that behavioral management systematically applied through steps such as the O.B. Mod. model can help meet the performance improvement challenges facing today’s and future organizations.