I’ve never said the words “good job” to my toddler. I was lucky – I stumbled on Alfie Kohn’s book Punished by Rewards early enough that I was able to break the habit before my daughter had really done anything much that might be construed as requiring a “good job.”
I’m going to be absolutely transparent here and say that this episode draws very heavily on Alfie Kohn’s book Punished by Rewards, which – along with one of his other books, Unconditional Parenting, are a cornerstone of my approach to parenting. If you have time, you should absolutely buy the book and read it yourself. But assuming you don’t have the time for 300 pages of (really, very good) writing plus a hundred more of notes and references to explain why both physical and verbal rewards are just as harmful to your children as punishing them, this episode will help you to get to the crux of the issue much faster. I’ll also get into the research that Kohn draws on, as well as relevant research that’s been published since the book came out in 1993.
Kohn’s thesis is that saying “good job” is really no different than punishing your child, since rewards are essentially the same thing – stimuli designed to elicit a response. He argues that while this approach is actually quite effective in the short term, not only is it not effective in the long term but it doesn’t mesh well with the kinds of relationships that many of us think or say we want to have with our children.
Birch, LL., Marlin, D.W., & Rotter, J. (1984). Eating as the ‘means’ activity in a contingency: Effects on young children’s food preferences. Child Development 55, 432-439. Retrieved from: https://www.jstor.org/stable/1129954?seq=1#page_scan_tab_contents
Brummelman, E., Tomaes, S., Overbeek, G., Orobio de Castro, B., van den Hout, M.A., & Bushman, B.J. (2014). On feeding those hungry for praise: Person praise backfires in children with low self-esteem. Journal of Experimental Psychology 143(1), 9-14.
Condry, J. (1977). Enemies of exploration: Self-initiated versus other-initiated learning. Personality and Social Psychology 35(7), 459-477.
Dweck, C. (2006). Mindset: The new psychology of success. New York: Ballantine.
Eisenberger, R. & Rhoades, L. (2001). Incremental effects of reward on creativity. Journal of Personality and Social Psychology 81(4), 728-741. DOI: 10.1037//0022-35184.108.40.2068
Gottfried, A.E., Fleming, J.S., & Gottfried, A.W. (1994). Role of parental motivational practices in children’s academic intrinsic motivation and achievement. Journal of Educational Psychology 86(1), 104-113.
Gray, P. (2016). Children’s natural ways of educating themselves still work: Even for the three Rs. In D.C. Geary & D.B. Berch (Eds.), Evolutionary perspectives on child development and education (67-93). Cham, Switzerland: Springer International Publishing.
Jeffery, R.W., Drewnowski, A., Epstein, L.H., Stunkard, A.J., Wilson, G.T., Wing, R.R., & Hill, D.R. (2000). Long-term maintenance of weight loss: Current status. Health Psychology 19(1 Suppl.), 5-16. DOI: 10.1037//0278-6133.19.1(Suppl.).5
Kazdin, A.E. (1982). The token economy: A decade later. Applied Behavior Analysis 15, 431-445. Full article available at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1308287/
Kohn, A. (1993). Punished by Rewards. Boston: Houghton Mifflin. (Affiliate link)
Kohn, A. (2001). Five reasons to stop saying “Good Job!”. Retrieved from: http://www.alfiekohn.org/article/five-reasons-stop-saying-good-job/
Pomerantz, E.M., & Kempner, S.G. (2013). Mother’s daily person and process praise: Implications for children’s theory of intelligence and motivation. Developmental Psychology 49(1), 2040-2046.
Rietzschel, E.F., Zacher, H., & Stroebe, W. (2016). A lifespan perspective on creativity and innovation at work. Work, Aging and Retirement 2(2), 105-129.
Schwartz, B. (1982). Reinforcement-induced behavioral stereotypy: How not to teach people to discover rules. Journal of Experimental Psychology: General 111(1), 23-59.
Hello, and welcome to the Your Parenting Mojo podcast. Today’s episode is called “Do you punish your child with rewards?”
I’ve never said the words “good job” to my toddler. I was lucky – I stumbled on Alfie Kohn’s book Punished by Rewards early enough that I was able to break the habit before my daughter had really done anything much that might be construed as requiring a “good job.” But she started at a fairly high-quality (as far as I can tell) daycare a couple of weeks back and even though it seems like she’s been home for half of that time with hand, foot and mouth and earache she was playing with her Lego by herself over the weekend when all of a sudden she said to herself “Good job, Carys.” The director of her daycare assures me that the teachers don’t say “good job” to the children and that she has probably picked it up from another parent or child in the room. I know that it’s my interactions with her that will have a far greater impact on her than her interactions with her teachers, but the incident reminded me that not everyone thinks about this in the same way that I do. So I want to take this opportunity to look at the research on how rewards may not be the foundation for the kind of relationship we want to build with our children.
I’m going to be absolutely transparent here and say that this episode draws very heavily on Alfie Kohn’s book Punished by Rewards, which – along with one of his other books, Unconditional Parenting, are a cornerstone of my approach to parenting. If you have time, you should absolutely buy the book and read it yourself. But assuming you don’t have the time for 300 pages of (really, very good) writing plus a hundred more of notes and references, this episode will help you to get to the crux of the issue as it relates to your children. I’ll also get into the research that Kohn draws on, as well as relevant research that’s been published since the book came out in 1993. I should say as a side note that Kohn spends quite a few words on how rewards are detrimental in a workplace environment as well and I’m going to ignore that here, so if that interests you then you should definitely go and buy the book.
Now I’d like you to close your eyes for a minute (as long as you’re not driving or operating heavy machinery) and think of some words that describe the kind of relationship you want to have with your children. Go ahead; think for a few seconds and then I’ll tell you my words.
There are really only three words that I use to describe both my relationship with my daughter now and the one I hope we continue to have as she gets older. The first two are “unconditional love” and the third is “respect.” I believe that love is a necessary but insufficient ingredient in a parent-child relationship. And respect is the missing part of that.
Did you use similar words or concepts? Or did you use words and phrases like “conditional love” and “coersion” and “control”?
If your words were more like mine than like the others, even if they weren’t exactly the same as mine, then my thesis statement to you is that if you use the phrase “good job” when you talk to your child, then you’re aiming to create one type of relationship with your child but using tools that are more likely to create an entirely different kind of relationship.
Now I don’t want to come off all ‘preachy’ here. I don’t want to give you the impression that I’m some kind of parenting know-it-all. I’m not. I just got lucky enough to read a really good book early enough in my daughter’s life for it to really make a difference for our relationship, and I want to share what I learned with you.
So, here we go. Why do we say “good job?” At this point it really seems to be a sort of cultural verbal tic. It seems like the vast majority of parents in the U.S. say “good job” to their children several if not many times a day. On the face of it we probably think we’re encouraging our children and helping them to develop skills. But what if we dig a little deeper? What are we trying to achieve with praise? If you think about it, it’s a classic behaviorist approach. The central idea in behaviorism is that we as humans don’t make choices; we just respond to stimuli. Rewards act as a positive stimulus and punishment is a negative stimulus, so we’ll do more of an action in response to a reward and less of an action in response to punishment. If you do say “good job” to your child, think about some of the reasons that cause you to say it. Are any of them situations when you’re trying to mold your child’s behavior? Do you ever say “good job sharing!” or “good job putting your toys away!” or “good job listening!”? If so, you’re trying to control your child’s behavior using rewards. Alfie Kohn’s central thesis is that this really is *no different* than using punishment to control your child’s behavior since rewards and punishment are essentially the same thing – stimuli designed to elicit a response. He argues that while this approach is actually quite effective in the short term, not only is it not effective in the long term but it doesn’t mesh well with the kinds of relationships that many of us think or say we want to have with our children.
First, let’s examine whether rewards are effective. Say you want your child to clean his room and you offer M&Ms in exchange. Will he clean his room? Probably! The reward worked! But what happens when you run out of M&Ms? Most of the behaviors we try to reward are behaviors we want the child to keep doing in the long-term. And what happens when the rewards run out? The behavior stops. The child is no longer interested in cleaning his room. Alfie Kohn cited research to support his point and a variety of more recent research rewarding everything from weight loss in adults to rewarding children for eating an undesirable food in exchange for a desirable food has no long-term efficacy at changing the target behavior (and it shouldn’t be a surprise to you if you listened to the episode 7, which was called “Help! My toddler won’t eat vegetables” that using ice cream as a reward for eating broccoli actually serves to increase the child’s dislike of broccoli). Kohn summarizes the research using a fabulous quote from a 1977 paper by John Condry of Cornell University, and I’ll quote a bit more than Kohn does:
“All in all, the evidence described [in Condry’s paper] suggests that task-extrinsic rewards (by which he means arbitrary rewards that are the opposite of a task-intrinsic reward, which is a task that produces a reward just by virtue of completing it), when they are used to motivate activity, particularly learning, have widespread and possibly undesirable effects. These extend to effects on the process, as well as the products, of the task activity and to the willingness of the subject to undertake the task at a later date. It is difficult to summarize this material adequately, but in general, compared to non-reward subjects, subjects offered a task-extrinsic incentive choose easier tasks, are less efficient in using the information available to solve novel problems, and tend to be answer-oriented and more illogical in their problem-solving strategies. They seem to work harder and produce more activity, but the activity is of a lower quality, contains more errors, and is more stereotyped and less creative than the work of comparable nonrewarded subjects working on the same problems. Finally…subjects are less likely to return to a task they at one time considered interesting after being rewarded to do it. The facts appear true of a wide range of subjects doing a wide range of tasks. Attempting to account for them all is a formidable challenge.”
It’s hard to top the completeness of that quote, so I’ll only say that more recent research supports this point as well. One study found that even though parents think that rewards are a good way to motivate children, in fact parent’s providing task-extrinsic rewards results in lower levels of intrinsic motivation to complete academic studies.
Now because I neither want to blindly follow one path without considering all the others nor blindly lead *you* down one path without considering the others I will say that when I searched the literature for “the effect of task-extrinsic rewards on children” several studies did pop up which found that people can be more creative if they are rewarded, both in terms of the amount of creative ideas they have and the quality of those ideas. But a study that reviewed both sides of the literature summarized the findings nicely: “What appears to be important is not so much whether people are working for a reward, but rather the degree to which these rewards (and other extrinsic motivators) make people feel controlled by external factors – when this is the case, intrinsic motivation, and in turn, creative performance, indeed suffer. So when people are rewarded for creative performance, this stimulates creative performance. When people are rewarded for “performance” or simply for doing a task, however, creativity is not stimulated and is even likely to be inhibited.” So I guess the question to ask yourself is how much of the time you’re offering rewards to your child you’re doing it to stimulate creative behavior. If the answer is “most of the time” then perhaps you’re on the right track. (And, to be clear, when I think about the times I’ve heard parents saying “good job” there’s usually no element of creativity involved in whatever the child was just doing.) But if you’re mostly using rewards to try to increase intrinsic motivation or to increase compliance with your wishes or to increase your child’s liking for broccoli or basically anything other than increasing creativity, then perhaps not so much.
Alfie Kohn tells us four reasons why rewards fail, and I’m going to go through them each individually. His first reason is that “Rewards Punish,” by which he means that rewards and punishment are not opposites, as we’ve been lead to believe, but that they’re actually two sides of the same coin. If you go back to the research I just described on people’s performance decreasing on a task when those people feel controlled by the reward, and we realize that the vast majority of the time we are using rewards to control behavior and not to stimulate creativity, then Kohn argues that “it is likely that the experience will assume a punitive quality over the long run, even though obtaining the reward itself is usually pleasurable.” Rewards also punish because children don’t always receive them. Many parents promise a far-off treat like a visit to the circus on Sunday, and use the threatened withdrawal of that treat to induce compliance from now until Sunday. And what is the withdrawal of the treat, even though it hasn’t been experienced yet? A punishment. It’s the ‘do this, get that’ situation that creates the control of people’s behavior, which is why we tell children about potential rewards beforehand rather than surprising them with the reward after the fact (which, according to research that Kohn fails to cite, is less destructive than rewards people are told about beforehand). We all get demoralized when something we counted on doesn’t come through. Children are no exception, and they – like we – see the deliberate withdrawal of rewards as a punishment.
Kohn’s second reason why rewards fail is because they rupture relationships. Where people who have comparable status try to compete for rewards, it sets up an environment of “if you succeed, I must fail.” Hardly the type of environment that fosters teamwork. The school where I got my first master’s degree refuses to give letter grade assignments for precisely that reason – so no student’s work can be compared to anyone else’s, or graded on a curve – when no student can be “better” than any other, the conditions are set to promote teamwork and cooperation. But what about situations where relationships are not equal? The parent-child relationship is not an equal one, because the parent holds most of the power. The parent probably also wants to create the kind of safe environment where the child feels he can come and ask for help if he needs it, but if the parent is also the doler-outer of rewards, the child won’t be in a collaborative relationship. She will be trying to get the parent to approve of what she’s doing so she can get the reward she was promised, and not only is that kind of relationship detrimental to the sort of risk taking that children need to do to learn and grow, but it’s detrimental to the relationship based on unconditional love that we say we want to have with our children.
The third reason rewards fail is because they ignore reasons. As with all good behaviorist approaches it essentially ignores what’s happening between the ears – or even assumes that nothing is happening between the ears, because people just respond to the stimulus of the reward. The behaviorist says that humans are nothing more than what they do. Change what they do and you’ve dealt with the problem. But if we are willing to say that our children do have something going on between the ears and that there might be reasons why they behave in certain ways, then why do we use an approach that ignores this? Kohn cites a mother who wrote to him to challenge his view on behavioral manipulation who said “If I can’t punish or reward my children, what do I do when my almost three year old wanders out of her room again and again at bedtime?”. Kohn says that Behaviorist A might use consequences to deal with this: “If you’re not back in that bed by the time I count to three, young lady, you won’t be watching television for a week!”. Behaviorist B favors rewards: “if you stay in bed until morning for the next three nights, honey, I’ll buy you that teddy bear you wanted.” The nonbehaviorist wonders why anyone would propose a reward or punishment without seeking to understand why the child is out of bed in the first place. Maybe she is hungry or not tired or just wants to know what’s going on downstairs. This is not to say that the parent needs to cave to the child’s every demand – far from it. But if all we do is offer a reward without understanding the reason why the child is doing what they’re doing, we can’t ever really fix the underlying problem.
The fourth reason rewards fail is that they discourage risk-taking. Kohn’s principle here is that when we are working for a reward, we do exactly what is necessary to get it and no more. Most of us must have had this experience in our school careers – we ask what material will be covered on the test so we can be sure we will study just enough to get the grade we want, and not a moment more. I still see it in myself today – my Masters in Psychology essays are graded, and I tend to cite research that supports my point so I can get the essay done and start the next one. These podcast episodes are not graded. I do the research primarily because I want to support my daughter’s development, but also because I want to share this knowledge with you. When I identified the research on rewards boosting creativity that seemed to conflict with Alfie Kohn’s work, I considered not including it because it didn’t directly support his – or my – point. But I wanted to fully understand the issue, and I wanted you to fully understand it too, so I looked for research that cites the conflicting results until I found a plausible explanation. There’s no reward for me for doing this, except maybe you think I’m a bit more credible now than you did before. My knowledge improved in the absence of a reward, not because of one. The same holds true for children. Why would you take a risk if taking the risk could impact your chances of getting a reward? Why would you write a paper on a topic you don’t know very well but would be interested in learning about, compared to a topic you already know well if you’ll likely get a better grade by writing about the material you know? And once you’ve completed an assignment for a teacher and received a good grade for it, do you attempt a completely different approach to the next assignment if you think the new approach might be better? No, you do not. Barry Schwartz of Swarthmore College showed that if you’re trying to get a reward, once you find the formula that appears to dispense the reward then you’re more likely to keep using that formula, even if the reward is later taken away. Alfie Kohn summarizes the findings of at least ten studies which collectively indicate that 1. The bigger the reward, the easier the task that people choose; 2. When the rewards stop, those who received them earlier continue to prefer to do as little as possible and 3. Easier tasks are selected not only in situations where rewards are offered but by people who are, as a general rule, more reward-oriented.”
Think about those summer reading programs that are designed to motivate children to read while they’re not in school Two of my local libraries (which are in different counties, so have different programs) offer these types of programs (one of them says that “even babies can join!). Children track their time spent reading to earn badges and other prizes. If I was part of this program, I’d probably be tempted to read for just enough time to meet the criteria for winning a prize. But worst of all, the fine print for one of the programs says that “all prizes are while supplies last” – so the child might do the required reading and not get a prize! This is not to say that the child might not do far more reading than is needed and some children may well do this – but if he does, he’ll be swimming against the mindset that rewards, by their very nature, try to introduce. Wouldn’t we rather have a child who loves to read for the sake of reading, not to get a badge? Alfie Kohn summarizes these ideas as follows: “Do rewards motivate people? Absolutely. They motivate people to get rewards.”
So rewards aren’t a way to build a real relationship with another person. And what is praise? And what is the phrase “good job”? They are both a type of reward. Kohn has an article that draws on his book Punished by Rewards – the article is called “Five reasons to stop saying “Good job!.” I’ll summarize these points bringing in research from the book. The first of the five reasons is that rewards manipulate children. We don’t offer praise because we’re interested in our child’s motivations or the inner workings of her mind; we offer them because they’re a quick way to gain compliance. Children will do almost anything to gain our attention and approval, and praise manipulates this tendency four our own benefit – to get our children to do what we want them to do.
The second reason is that they create praise junkies. When you say “I like the way you did (whatever) or “Great sharing!”, you’re imposing your judgement on the child’s actions. And I should say that English parents actually manage to one-up Americans in this regard, and not in a good way – while American parents are far more likely to say ‘Good job!,” a phrase which as far as I recall at least – doesn’t really exist in England, English parents are far more likely to say “Good Girl!” or “Good boy!”, a judgement not just of the action just completed but of the child as a person. I even read a psychology thesis recently that described a series of simple experiments done with children in a laboratory setting and each time the child completed a required action the experimenter’s script said “Good girl!”. I wonder whether the experimenters considered the impact that the judgement of the child had on the outcome of the study. Children who are praised become more tentative in their assertions and answers to questions because they’re so used to saying something to see if it elicits praise. Saying “good job” doesn’t reassure children; it makes them dependent on our opinion of them.
Thirdly, when you say “good job” you steal a child’s pleasure in having done the activity for themselves. Many American parents want their children to develop intrinsic motivation, which means they want children to love to learn for the sake of learning, or to love to climb for the sake of climbing, or to share a toy because the *want* to share a toy, not because the parent is either telling them to share in the moment or because the child hears the parent’s voice in her head saying “Good sharing!”. So what we want children to do is come to us and say “I did it!,” not “was that good?”
Fourthly, saying ‘good job’ can cause children to lose interest in the task at hand. Just as using ice cream as a reward for eating broccoli causes children to like broccoli less, offering praise for activities like sharing actually causes children to become less generous. The next time you find yourself on the verge of saying ‘good job,’ ask yourself – is this something you want your child to do again? And do you want them to do it without being praised next time? Then don’t praise them this time.
Finally, saying ‘good job’ reduces achievement. Research more recent than Kohn’s book backs this up – one study found that the more mothers used person praise like “You’re so smart!” and ‘you’re a good kid!” the more children avoided challenging work in school six months later, which previous studies have suggested undermine children’s achievement. And research from 2014 found that person-praise is particularly punishing to children with low self-esteem, who then feel more ashamed following a failure because they seem to have attributed the failure to their inherent ability rather than to their needing to develop a particular skill. This effect was, of course, the subject of Carol Dweck’s famous book “Mindset” which I read a year ago not long after I started my blog that now almost never gets updated. I read it from a parenting philosophy mindset but what I wasn’t expecting would be that it would also alter the way I see my own abilities. I’d always thought of myself as a person who couldn’t draw. I thought there were some skills that were just innate, and drawing was one of them, and I didn’t have that skill. But Dweck’s book shows some before and after self-portraits drawn by people who went through a one week-long drawing course, and the difference was astonishing. I decided to see if I can draw and it turns out that by gosh I can! I’m never going to be the world’s greatest artist, but simply by shifting my mindset from “drawing is an ability that I don’t have” to “drawing is something that can be learned and I’m going to learn how to do it” enabled me to learn the basics of drawing technique. If I wasn’t so busy researching papers for school and podcast episodes for you I’d probably have made more progress in recent months but such is life. The point is that focusing on judgements of the self restricts your (and your child’s) abilities. Creating a growth mindset, where many abilities are simply waiting to be learned with the application of some hard work opens up a world of possibilities to your child – and to you if you want – that mere praise alone can never do.
So imagine our child just shared a toy with another child voluntarily. Or put his shoes on by himself. Or brought a picture she drew over to show you. The phrase “good job!” is on the tip of your tongue, but you want to try not saying it. What do you do? Alfie Kohn has three suggestions. Firstly, consider saying nothing. You don’t *have* to reinforce the behavior. Praise simply isn’t necessary and, as we’ve already seen, might do more harm than good. Secondly, say what you saw – without judging. “You shared your truck with Suzie.” “You put your shoes on by yourself.” “You used a lot of blue in that picture!”. Consider saying something about the effect that the action has on the other person. “Suzie looks really happy playing with the truck.” That way the child learns to focus on the outcome of her action and its effect on other people, not on your judgement of her action. Finally, ask questions. Ask what part of the drawing your child likes best, or why he picked certain colors. These kinds of questions show your interest while nourishing the child’s interest in drawing.
So why don’t we just stop saying ‘good job’? Aside from the fact that’s become a habit, what are we afraid of? Are we afraid that children will stop obeying us, and that their true lazy selves will be revealed? Hardly! If your child is older and has been raised on praise and grades then you have a more difficult task ahead of you. But if you have a toddler or preschooler then it may not yet be too late to just stop saying it. In the absence of instruction, children have inherent curiosity and motivation to learn. If we allow their curiosity and playfulness room to grow in an environment rich in self-educational opportunities and with guidance from parents that doesn’t revolve around rewards OR punishments, children can and will learn academic skills and social skills simply because they want to. Because they can’t help but learn. It’s how our brains are wired.
Thanks for listening – if you’d like to read more on this topic, do check out the references for this episode on YourParentingMojo.com; search for “Do you punish your child with rewards?”