This post originated from an RSS feed registered with Java Buzz by Brian McCallister. Original Post: Motivation and Rewards Feed Title: Waste of Time Feed URL: http://kasparov.skife.org/blog/index.rss Feed Description: A simple waste of time and weblog experiment Latest Java Buzz Posts Latest Java Buzz Posts by Brian McCallister Latest Posts From Waste of Time

Dale Asberry started a tempest in a teapot when he posted that to better motivate employees companies need to do away with rewards, bonuses, etc. People don't like this idea at all. The problem is that all the scientific research available backs him up. I don't have references off the top of my head, but I did my grad degree in learning theory, and motivation is a big thing there -- I read way more and did more research than I care to remember. Short form is that introducing extrinsic rewards (a bonus, for example) for actions destroys any intrinsic motivation to perform the action.

As most people^H^H^H^H^H^H Americans don't believe in science, much less statistics, I'll try to relate it via anecdotes ;-) Chances are that if you read my blog you enjoy programming. I suspect most readers enjoy it so much that they go home and do it at night and give away the fruits of that labour. This is intrinsic motivation. Now, domain specifiers become a big deal when you are immersed deeply in a domain. Writing code for object-relational mapping is fun (one domain), writing web application is painful (a different domain). Five years ago writing web applications was fun (okay, we are looking at a case study now, me). Then I started getting paid to write web applications. Now typing a certain four letters used as an acronym for a common transport protocol makes me grit my teeth.

There are a number of reasons I might tell you for this, it could be because I do it so much I get tired of it, it could be that it is because I need variety in what I do, and I do web apps all day, it could be any number of things. These may even be decent labels for the factors that might be identified via a good study. The single biggest factor in a controlled experiment will be that I have been externally rewarded for writing webapps, quite nicely so, but still externally rewarded. During the day I don't do it because I want to, but because I get a pay check. Five years ago I didn't get paid to write web apps, I'd go home at night and... write web apps. I loved it. Once you start getting paid for something, it becomes work. Soon you expect to get paid for it. It is something you do for money, not for fun. Hobbies, sports, etc may be harder, more labor intensive, etc -- but you do them for the fun of doing them

Okay, back to science, as I believe in science, even if many do not =) In controlled experiments on many scales, offering external rewards for behavior previously performed because of intrinsic motivation destroys the intrinsic motivation. You can replicate this easily, particularly if you have children. Choose an activity your child enjoys, which you want to encourage her to do more of. Tell her that you are glad she does this, and want to encourage her to do it. To help encourage her you will reward her (candy, congratulations, obvious pride, etc, all work). It is critically important that you really do want to encourage this behavior. Children will detect manipulation immediately. Now, for the next two weeks reward your child for this behavior. Then, after two weeks, stop. At first the behavior will pick up in frequency as the child wants the reward then it will dwindle when it becomes apaprent the reward isn't coming (classic Skinn 1000 er behaviorism), then the behavior will stop completely and will almost never occur again without an offer for the reward, and may not even then. This is a terrible experiment ethically as it will stop good behaviour, but it illustrates the point in a way you will never forget.

Now, people are screaming across blogspace that they bloody well want to get paid well for their work. This is valid. The pay is not there to encourage behavior though. I cannot recall his name, but another gent in industrial research put together a simple but very good theory to explain this. I call is simple because I think it doesn't cover all the bases, but it is definately accurate for what it does cover. he discusses a second factor in employee motivation, which is a sanitation factor. If you perceive that you are not being proeprly compensated, adequated equipped, or workign with people who just plain annoy the heck out of you, you will be unhappy, and it will kill motivation. Preventing these problems will not create motivation, but not preventing these problems will destroy motivation (writing object relational mapping software isn't fun if I have to do it while someone holds a lighter to my toe). Real motivation needs to be intrinic, and the environment needs to be set to allow that to flourish.