Wednesday, May 18, 2011

Should We Refrain from Telling Our Children that They Are Smart?

We are doing our annual testing right now, which reminds me of a debate about the use and consequences of praising your children's intelligence that has been popular lately.  Sparked primarily by articles like  The Secret to Raising Smart Kids (Hint:  Don't Tell Your Kids That They Are), the theory goes that praising children's intelligence encourages them to think that their academic achievements are based on an inherent quality (intelligences) that they either have or don't have.  If they think they don't have it, they won't try; if they think they do have it, they either coast by on their natural talents or try to avoid anything that might demonstrate that they don't have as much intelligence as you think they do.

The alternative advocated by psychological research Carol Dweck, on whose studies most of this advice comes from, is to praise the amount of effort that children put into their work.  Dweck found that students who received effort-related praise continue to put more hard work into their studies and were willing to take on more challenging tasks to prove how much more hard work they could do.

It all sounds good.  But I'm always suspicious of these media-pushed, black and white, simple panaceas for educational issues.  So I did a little checking, and found an excellent article in The Chronicle of Higher Education entitled Carol Dweck's Attitude:  It's Not About How Smart You Art.  As I expected, this discussion shows that the whole situation is much more complicated than the one-page version being shared on parenting websites around the web.

The article shows that in three recent studies done to test that theory, student performance did not reflect the predictions of Dweck's framework. In one case, the researchers believed that the "hard work" mindset could be just as detrimental as the "innate intelligence" is supposed to be in Dweck's theory (that is, those who believe that their scores are due to hard work and who expect to do badly will self-sabotage their work by listening to distracting music so that they can blame their poor performance on the music, rather than on their work).

The Chronicle article also talks about Dweck's earlier work, which was focused on an outside control/deterministic viewpoint versus a self-determining perspective, When I read about this research, it seems to me that this locus of control is really what is most important, rather than the "innate intelligence" versus "hard work" perpective per se. That is, if you aren't convinced that your performance and your life is under your control, you can find "outside" factors to blame your failure on, whether that is blaming your parents' DNA contributions to your lack of intelligence, or bad music or outside life factors for your inability to do the work necessary.

Another thing to consider about this "praise the effort" system is the fact that is is diametrically opposed to the actual assessments students face in schools these days.  With all the high-stake testing that is running education today, it makes NO DIFFERENCE whether the student worked really, really hard preparing for and taking the test, or goofed off all year and treated the test as a joke.  If the hard-working students don't achieve the minimum score, they fail and are held back, and if the lazy goof-offs ace the test, they are promoted.  So I think raising students' hopes about the value of hard work could be counter-productive or misleading when the bottom line is that amount of effort doesn't count at all; only the number of right answers does.

I found a briefing paper by the Association of American Colleges and Universities entitled Greater Expectations to Improve Student Learning that I found useful about the general issue of expectations in education.  For example,  it says that Americans are particularly attached to this whole "innate intelligence" scenario, while other cultures attribute much more to hard work rather than raw ability.  I also thought Daniel T. Willingham's Ask the Cognitive Scientist column on How Praise Can Motivate--Or Stifle to be a great resource in thinking about this topic.

As is typical for me, I think the "either or" approach is wrong. So in our case, I've told my son that he is smart, because he is, and I will continue to tell him that.  I'm kind of an encouraging, "praise-y" personality anyway, and I think it would feel unnatural and inauthentic of me to withhold my assessment of his intelligence because of some research studies.

However, I also tell him that because I know how smart he is, I have high expectations for the quality of his work, and what other parents or teachers accept as "good enough" for other students may not be "good enough" for what I expect from him.  And we discuss the fact that quality is a factor of effort as well as natural talent.  We talk a lot about all sorts of gifted people and how hard they had to work to manifest their gifts, even with their natural abilities. So Michael Jordan had an incredible body for basketball--but that didn't mean that he hasn't practiced for hours almost every day of his adult life. Likewise with Mozart, or Picasso, or Einstein, or Maya Angelou or all takes work to refine and apply whatever natural abilities we might have.

Don't get me wrong--I think it is a good thing to read and consider research like this--IF you are looking at the full picture, which usually has a lot more nuances than the "sound bite" version we get in a single page in a magazine or website.  But in the end, we have to do what feels right as parents and teachers.  And usually, that is a more common sense, mixed approach than the typical distinct treatment populations of research studies.

So if you want to tell you children they are smart--go ahead!  Just don't forget to tell them that isn't the whole picture.


  1. I strongly believe all discussions of abilities AND hard work must happen in the context of particular meaningful tasks. For example, we recently discussed the need to write down intermediate steps, and compared ours. I need it because this particular side of my IQ is abysmal (short-term memory and processing). You may not need it for that task, because your abilities cover the requirements with room to spare. M needs it as an anchor to his attention and focus (because he tends to be more divergent, and the task requires convergent thinking).

    We need to match tools to our abilities and skills, on task-to-task basis. For that, we need to have meta-cognition: the knowledge of abilities and skills and tools.

  2. All definitely true.

    Maria and I had a discussion about this offline. The problem with these studies is that you have no specifics about things like the content of the tasks being studied, which could have a big bearing on thing.

    For example, I could see my son, if he were in the praised group, turning down the opportunity to take the "higher level" test. But I really don't think he would do so because he was afraid of losing his "praise;" I think he would do it because he wasn't interested in the activity and/or wasn't interested in jumping through what he perceived as being someone else's hoops. So while there may be some correlation, it is hard to discover the motivation behind those correlations.