RIP Marshmallow Test - full entry

The marshmallow test is wrong and I should have known it.  Now, my husband and I need to rethink some of our parenting priorities. The original news about the marshmallow test was that four year olds who were able to resist eating a marshmallow for 15 minutes in order to earn a second marshmallow scored disproportionately better in their SATs many years later. The parenting guidance was to help your child learn to delay gratification and to control their impulses. I wrote about this a few months ago.

There’s nothing wrong with either of those lessons, but they’re not actually what appears to have

Boy with marshmallows on his fingers

happened with the marshmallow test.  First, the sample size of students who were tracked 15 years later was very small and not statistically significant.  According to Po Bronson’s research on the marshmallow test, it was actually only 35 kids who did the classic test—17 boys and 18 girls.  The marshmallow test is apparently a better predictor of children who would later be diagnosed with ADHD, but not of future success.


When I was in college, I took a statistics class with Prof. Ed Rothman (one of my thesis advisors) called The Art of Scientific Investigation.  One might call it “educated skepticism of pop-statistics”.   For example, he asked on the first day of class how many of us had read that 50% of US marriages end in divorce.  Everyone raised their hand. That statistic, like the analysis of the marshmallow test, is both right and wrong depending if you ask the right question.  50% of all marriages in the US apparently do end in divorce BUT people who marry multiple times count multiple times.  The insightful question is what percentage of FIRST marriages end in divorce.  As of the late 1980s, only about 20% of first marriages ended in divorce. I should have known to ask about the sample size of the marshmallow test. Sorry Prof. Rothman.

Regardless of the research, my husband and I believe that delayed gratification and self-control are good skills to learn. But when I heard Po Bronson speak recently about what skill really helps children succeed, it was a blinding insight of the obvious.  Creativity.  Make-believe. Problem solving.  All early forms of executive function. When I look around at my colleagues at work and my friends, it’s easy to recognize the most productive – they’re both creative and communicative.  Now the puzzle is thinking about how to encourage creativity. I know our school is focused on these capabilities as well.

At home, creativity often happens by accident (maybe most of the time).  My 8-year old was playing with Lego and happened to grab my camera.  Later that week, I uploaded images and discovered a filmed Lego battle that he had scripted and recorded while I worked in the room next door.  If I had asked him to make me a Lego movie, he would have been stumped, but leaving the Lego and camera near each other was enough to light a spark.  We celebrated his creativity by publishing the video on YouTube, Facebook and to the grandparents. 

What do you do to encourage and capture creativity?


This post was originally contributed to SVMoms.