If people are not told they are appreciated,
they will assume the opposite.
If positive reinforcement is so effective, why is it not used more commonly in business? One reason is that managers are more likely to be positively reinforced when they punish or use negative reinforcement than when they use positive reinforcement. Think back to the chapter on consequences (Chapter 4). You may recall that consequences that are positive, immediate, and certain (PIC) are much more powerful than those that are delayed (Future, as in PFU).
A manager is more likely to receive positive immediate consequences when delivering punishment than when delivering positive reinforcement. The reason? If punishment is going to work, it typically will work right away. The manager gets what she wants (usually some inappropriate behavior stops immediately), and this provides the manager with a PIC. The manager, like all of us, will tend to repeat that punishing behavior because it was reinforced. Likewise, negative reinforcement (“Do it or else!”) will result in an increase in activity, providing the manager with positive reinforcement for the use of negative reinforcement.
When that same manager uses positive reinforcement, she will have to wait until there is an occasion for the behavior to occur again to know whether the reinforcement had any effect on the performer. In this case, the consequence for the manager is a positive, future, and uncertain consequence (PFU), not a powerful consequence.
Another reason that positive reinforcement is not the predominant consequence at work is that many managers have tried to use it and have been unsuccessful. I’ve heard many times, “Oh, I tried that reinforcement stuff, and it didn’t work.” Of course, by now you know that this can’t happen. Reinforcement always works! It has to increase behavior or it isn’t a reinforcer.
What I have discovered is that when managers complain that positive reinforcement doesn’t work, they have made one of the following errors: perception, contingency, immediacy, or frequency. Let’s consider each of these.
As stated earlier, what works as a reinforcer for one person may not work for another. Many people will choose to use the reinforcers they like rather than finding what others like.
At times, we make the mistake of thinking people should want things that we want them to want. In other words, they should respond positively to the things we give them. There are many things managers think are reinforcing that are not. For example, most people assume that money is a reinforcer to everybody. Do you think that Donald Trump would find $100 reinforcing?
Many people do not find money reinforcing under certain circumstances. We have all said at one time or another, “You couldn’t pay me enough to do that” or “I wouldn’t work there for any amount of money.”
Some managers think that if $1,000 is reinforcing to someone, then $100 would be somewhat reinforcing to that person. This is erroneous thinking. There are some things that you would be happy to do for $1,000 but wouldn’t even lift a finger to do for $100. This mistake is also seen in the way we give raises and bonuses. The top performers get a 5 percent increase, and those at the level below them get 4 percent. Many outstanding performers have been heard to say in this situation, “If they think I’m going to break my back for a lousy 1 percent, they’re crazy.”
Similarly, have you ever heard someone say, “Everybody likes public recognition”? The fact is that in surveys I have done, most employees say that they don’t like it. They give various reasons, but most of those reasons center on concern about what peers might think about them or know about them. They may worry that peers think they don’t deserve the recognition. The effective manager or supervisor knows which people like public recognition and which people don’t.
The most basic element in performance management is that you must first have a reinforcer. Make sure that you use some of the ways discussed in Chapter 7 to find an effective reinforcer. Once you have a reinforcer, you are on your way, but there are still some things that could cause it to be ineffective.
This error relates to the relationship between a behavior and a reinforcer. If you can get a reinforcer without engaging in a prerequisite behavior, then that reinforcer is said to be noncontingent. If the only way you can get a reinforcer is to engage in a particular behavior, that reinforcer is said to be contingent on that behavior. For example, in the average company, an increase in fringe benefits may not increase performance because fringe benefits are provided to everyone regardless of performance. Improved fringe benefits are given whether or not a given employee’s performance increases. The fringe benefits are said to be noncontingent.
To say that hard work pays off implies that there is a reinforcement contingency between hard work and pay and/or promotions. Many people will tell you that if there is a contingency between work and pay, it is a loose one because they work hard and don’t get increases in pay.
Profit sharing is often administered noncontingently. One reason it loses its effectiveness is that there is only an indirect contingency between profit and the performance of any given individual. In many companies, if there is profit, it is shared on the basis of some predetermined formula, which is independent of a given person’s performance. Therefore, if there is a profit, everyone will share just because they are on the payroll.
Salespeople sometimes “earn” commissions not because of what they did but because they were lucky enough to have a customer who uses the company’s product. If they were not using sound sales behaviors to cause the sale to close, the commission may inadvertently reinforce behaviors that represent poor sales techniques and will lead to poor performance in the future.
Make a list of reinforcers and rewards in your organization, and state the behavioral contingency in the following format:
You can get … if and only if you …
To use a previous example, “You can get a bonus if and only if you are on the payroll.” If you make such a list, you will be surprised at how few reinforcers are contingent on productive, quality, and safe behaviors. You will be enlightened and reinforced at the same time because you will immediately see solutions to some of your problems.
I explained the importance of immediate reinforcement earlier. If someone said to me, “I don’t have time to reinforce,” my response would be, “In that case, you’d better reinforce immediately.” The longer you wait after the behavior, the less effective the reinforcement will be. The error is delaying the positive reinforcement. If you have a limited amount of time to spend in the activity of reinforcing, spend it reinforcing desirable behavior while the behavior is occurring.
Most leadership research is based on what leaders say they do. Dr. Judi Komaki (1986) is one of very few people who have done research on what leaders actually do. What Dr. Komaki has found is that the most effective leaders, managers, and supervisors do not necessarily reinforce more often than the ineffective ones. What they do is reinforce while people are performing. To be able to do this, they spend more time in the work area observing behavior. Where do you think the ineffective supervisors spend their time? You guessed it—in their offices.
The only behavior a supervisor can reinforce in his office would be whatever the person is doing in the supervisor’s office. Ironically, most supervisory duties today take the supervisor out of the workplace and into an office or a meeting room.
When reinforcement is immediate, you know what you are reinforcing because it is happening before your eyes. The longer reinforcement is delayed, the less sure you are of which behavior is actually being reinforced because people are continuously behaving.
The greatest advantage of teams (and one that is almost always overlooked) is that team members can provide immediate reinforcement to each other. If team members are taught to deliver positive reinforcement, they are in the best position to make it immediate. In most cases, they are also in the best position to know whose behavior merits reinforcement.
One positive reinforcer will not change your life. “I reinforced him, but he didn’t change” is an oft-stated complaint. It is not that managers don’t understand the need for reinforcement and recognition; it is that they don’t have an understanding of the amount of reinforcement that is needed to create peak performers and a high-performance organization.
A case can be made for the fact that a little reinforcement is better than none at all, but in the amounts given in many organizations, it is not much better. To give you a perspective on frequency, in The Technology of Teaching, B. F. Skinner states that it may take as many as 50,000 reinforcers to teach competence in basic math—roughly the first four grades.
This works out to more than 70 reinforcers per hour per student. Although getting the correct answer in math is a reinforcer, the delay between doing a math problem in a traditional classroom and knowing whether it is correct could be several days. Computer-aided instruction helps because the computer lets the student know immediately if the answer is correct, but the number of problems to be solved per class is not enough for students to gain fluency in the math operation tested. Also, the number of attempts at verbal reinforcement varies widely from teacher to teacher, but the median number found in several studies is about six per hour. When you understand this relationship of reinforcement to learning, you can understand the problems we are experiencing in the schools today.
Students simply do not get enough reinforcement to maximize learning. I would not want to suggest that the number of reinforcers that adults need to acquire a new skill or make substantial improvement in an existing way of doing something is equal to what a child needs in school. However, it certainly gives you a perspective on what it takes to make significant changes in behavior. With this frame of reference, everybody should understand that an occasional reinforcer at work will make only a small difference in performance.
The way that many organizations think about frequency may be seen in annual performance appraisals, annual recognition dinners, quarterly bonuses, employees of the month, and so on. This low frequency of reinforcement will have little or no impact on organizational performance. Positive reinforcement needs to be a daily affair.
Years ago, when I first started working in industry, I was often asked, “What is the proper balance between positive reinforcement and punishment?” This question was sometimes phrased as, “How many ‘atta-boys’ does it take to erase one ‘You really screwed that up!’” In researching the literature, I found two studies that I think have some bearing on the question.
Madsen (1974), in training classroom teachers, found that teachers who had positive reinforcement-to-punishment ratios of 4:1 or better had good discipline and high achievement in their classrooms. Below these ratios, they had problems. Stuart (1971), in a study of delinquent children, found similar ratios with parents of nondelinquent and delinquent children. Parents with no delinquent children had higher ratios of positive interactions than parents of delinquents. Hart and Risley (1995) found that vocabulary development in young children was dramatically accelerated in families in which the ratio of positive to negative interactions was high (approximately 6:1).
White (1975) and Thomas (1968) studied naturally occurring ratios in classrooms and found ratios in the range of 1:2. Self-reports by supervisors and managers revealed ratios of 2:1 at best. The ratios reported are probably more positive than the naturally occurring rates because the rates were taken as a part of a class in performance management. Few would disagree that there is a lot of room for improving the ratio of positive reinforcement to punishment in most organizations.
Some people have misunderstood the 4:1 rule. It does not mean that if you reinforce four times, you have to punish something. The ratio can be higher than 4:1. In a high-performing organization, punishment may be infrequent indeed, but it is unrealistic to expect that there will be no occasion for negative consequences in any organization over a period of time. If you have a work environment where there are never any negatives, it probably means that you are positively reinforcing some poor performance.
Check your own ratios. Keep a three-by-five-inch card, and tally all your attempts at reinforcement and punishment each day. If the ratio is less than 4:1, work on increasing your attempts at positive reinforcement. The card will be an antecedent for looking for opportunities to reinforce. Remember to reinforce only behavior or performance that warrants it. Many improvements have been reported in quality and productivity when supervisors simply increased their daily frequency of contingent positive reinforcement.
When there is too little reinforcement to go around, people will compete with each other to get it. Competition for significant reinforcers and rewards, financial or social, can generate behavior that is incompatible with the team-oriented work environment that most organizations are trying to promote. Infrequent reinforcement promotes the kind of “political” behaviors with which we are all familiar: blaming others, covering your rear, and even sabotaging the initiatives of others.
People should not be exerting effort to compete for reinforcement within your company. If you want employees to compete, you should focus them on external competitors. Because we have all grown up in a competitive society, we assume that internal competition is an acceptable model for business. I suggest that internal competition isn’t a good model. A company should focus employees on performing better than their competition and not on their teammates.
In obviously competitive circumstances (such as sports), competition within the team becomes visibly counterproductive very quickly. When a basketball player takes a shot when he should have passed the ball to a teammate, everybody on the team and in the stands knows it. In football, when a teammate steps in front of the intended receiver to catch a pass, everyone clearly sees it. At work, it is not as easy to see behaviors that cause others to fail or to have a more difficult job.
We certainly want our employees to be competitive. But we want them competing within our industry, not within our company. We don’t want people to fight for reinforcement. We want reinforcement to be available for all who earn it.
When you think about the principles described in this chapter, it becomes obvious that employee of the month (EOM) and other such recognition programs violate every rule of effective positive reinforcement. I will describe later how to fix EOM and other recognition and reward programs, but for now, let me just say that EOM not only doesn’t reinforce the behaviors you want, but it may in fact be punishing them.
If you have such a program in your organization, think about the following: Does the award go to the best performer every month? Does the performance of the recipient improve after he or she receives the award? Can you identify the specific behaviors that the award is reinforcing? Do you know if the actual award is something that the recipient likes? If you answered no to any of these questions, cancel the EOM program now. If you answered, “I don’t know,” take a closer look and be prepared to alter your approach,
The “No But” Rule. When supervisors and managers are asked, “Do you tell people that they are doing a good job?,” they practically always say that they do. When you ask employees, “Do your supervisors tell you that you do a good job?,” they practically always respond, “They don’t.”
My analysis of this is that managers actually say the words, “You did a good job” or “I appreciate what you did.” However, they don’t stop there. Frequently, they go on to say something like “but you could have done ….” As is often taught in sales training, “Don’t talk yourself out of a sale!”
A complete example may sound something like this: “You did a good job compiling the final project report, but it would have been even better if you had included the original documents as an appendix.” From the perspective of the supervisor making this statement, he has taken the time to reinforce the behavior of “compiling the final project report” and has mentioned a little helpful hint: “Next time include the original documents as an appendix.” Although well intended, this statement probably would be more punishing than reinforcing.
John Domenick, a performance management consultant, said it well: “Good intentions are terrible things to waste.” When the supervisor added the word “but,” he turned a positive statement into a negative with a single word. You see, “but” is a verbal eraser when used in an attempt to praise. It erases everything that preceded it. The reinforcement is gone. The criticism remains.
The president of a nationally known dairy products company once told me that he received a call from the chairman of the board of his company’s holding company. The call started off, “John, I was just looking at your financial results from last year. You set a record. It was the best the company has ever done.” At this point, the president said that he was feeling great. However, the chairman went on to say, “but if you had had a better handle on your inventory during the second quarter, you could have done even better.” John said that he was devastated immediately and got angry later. I’m quite sure the chairman is certain he complimented John for a good job. If asked, he would probably say that John knows how much he appreciates his work. I know that John did not feel either complimented or appreciated. Do not use the occasion for praise as an opportunity to prompt or instruct.
The Sandwich. The so-called sandwich method of correcting performance is probably the most widely taught concept in basic supervisory training. A variation of the “no but” rule, the sandwich is a negative between two positives. It goes like this:
Positive: “Thad, you are one of the best employees I’ve got.”
Negative: “But if your attendance doesn’t improve, I’m going to have to terminate you.”
Positive: “You know, you have more talent in your little finger than most people have in their whole body. I would hate to lose you.”
Some psychologists say that this method preserves one’s self-esteem in the process of correcting. I know of no experimental data that demonstrate this fact. I think that it helps the punisher by making the feedback easier. At best, it is confusing to the performer. After all, there were two positive comments and only one negative. In addition, when do you think Thad will hear positive statements like this again? Probably not until he is in trouble again.
Sandwiching is not a good practice. Criticism should be short and to the point. You should be very clear regarding which behavior must increase or stop, what will happen if it does, and what to do instead. Positives should be saved until there is some improvement in performance.
Do not pair positive reinforcement with punishment.
Our example would have been better handled like this: “Thad your attendance has put your employment in jeopardy. If you miss one more day during the rest of this quarter, you will be terminated. I hope this will not be necessary. If there is anything I can do to help you, let me know.” Later that day, when he is doing his job, you could approach Thad and say something like, “Thad, I hated to have to get on you this morning. You’re one of the best employees I’ve got.”
While this may appear to be a trivial difference, I assure you it will produce a very different result. Because showing concern is usually positive to most people, the pairing of concern with good work probably will act as a positive reinforcer for the productive performance. If concern is paired only with problem performance, you will likely get more problem performance.
A plant manager once asked me if I would talk to one of his department supervisors to see if he was “cut out to be in management.” The plant manager thought that because I was a psychologist, I would be able to analyze the supervisor’s management potential. As a matter of fact, the management staff had already decided to terminate this supervisor because the general foreman didn’t think he was management material.
I explained that it would be difficult to make a recommendation just by talking with the supervisor. I suggested that if the manager was serious about making an informed decision, he should put the supervisor through the scheduled performance management training and see how he responded. I would then have some data about his effectiveness and could help the plant manager make a better judgment.
As was discovered in class, the young supervisor was too positive. That is, he attempted to reinforce everyone, violating the contingency principle many times a day. When his reinforcement attempts didn’t work, he started using the negative management style of most of his peers. Since he didn’t like managing that way, his heart was not in it, and he had become totally ineffective.
When he learned about contingency, immediacy, and frequency, the performance of his shift changed dramatically, almost overnight. I don’t know the end of this story because it hasn’t been written yet, but the last I heard was that the young supervisor was then a plant manager.
Most managers using atta-boys, pats on the back, and other positive management techniques typically commit some or all of the reinforcement errors mentioned in this chapter. They are usually well-meaning individuals who are not aware of the precision required in the effective delivery of positive reinforcement. Because of this, they are usually ineffective in bringing out the best in people.
However, these managers are usually predisposed toward being positive. When they learn to avoid these errors, they are able to increase their personal effectiveness dramatically and in a short period of time.
Some supervisors claim that they follow all the rules discussed in this chapter, and it still doesn’t work. This brings up the relationship issue. If the person on the receiving end of your reinforcing attempts doesn’t like or respect you, telling the employee that you appreciate her work will not work. If the employee knows that you say one thing to her face and another, or the opposite, thing behind her back, then telling her that she is doing a good job may have the opposite effect on her performance and, if I may use the term, her attitude.
My mother once told me that my brother, a high school teacher, had received a nice letter from the assistant county superintendent about his work. Later I asked him why he had not told me about it. He said, “It didn’t mean anything. He sent that same letter to all the teachers in the county.”
For all new supervisors or managers, the first task is to pair your presence with positive reinforcers. I was an artillery officer stationed in Korea, and we had a high turnover of officers because rotation back home was looked toward from the first day. Unfortunately, many of the officers, from their first day, wanted to show everyone that they knew more than the troops did. These officers had a very difficult time. However, a few of them spent the first few days getting to know each soldier personally, asking about their families and where they grew up and went to school and asking about their job in the battery. In almost every case, they did not act like a know-it-all but like a student, having the soldier teach them something. As you can imagine, the first ones spent much time in the battery commander’s office explaining some problem or other, whereas the latter always had troops “at their back.”
When you do positive reinforcement correctly, it makes hard work fun while accomplishing much. Done wrong, it makes easy tasks difficult and accomplishes little.