What is the best way to help children who have problems with reading comprehension? There are dozens of reading programs out there and you would think we would know the answer to this question by now. But the shocking thing is that until recently we didn’t know much at all about which ones really work. Educational researchers have been reluctant to conduct experiments in which children are randomly assigned to condition, to test the effectiveness of educational programs. (Thomas Cook, for example, published an article with the title, “Why Have Educational Evaluators Chosen Not to Do Randomized Experiments?”) Experiments are hard to do, but they are the gold standard in evaluating whether a program works. Fortunately, educational researchers are beginning to use this approach more.
Researchers in England, for example, studied fourth graders who had difficulties with reading comprehension. They randomly assigned the kids to receive one of two interventions over the course of 20 weeks. One program focused on text comprehension, teaching the kids strategies such as how to clarify unknown words and how to figure out the global meaning of a text. The other focused on spoken language; the tutor engaged the child in a dialogue that dealt with vocabulary, figurative language, and narratives. Other kids were randomly assigned to a control group that did not take part in either program.
Before reading any further, which of the two programs do you think worked the best? If you are like me, you might guess that the first (text comprehension) program would be most effective, because after all, it attacks the core of the problem. But this is why we need to do experiments instead of relying on common sense (at least mine): The second (oral comprehension) program worked the best, leading to the biggest gains in reading comprehension, even 11 months after the program ended. Now that we know what works, this program can be widely applied.
Some readers might be a little queasy about including a control group of kids in this study who didn’t get any help with their reading difficulties. Is it ethical to keep these kids out of programs that might help them? The tradeoff is that if don’t include control groups, we will never know which programs actually work, and risk exposing thousands of kids to interventions that are ineffective or worse, do more harm than good. This has in fact happened, not only with educational interventions, but programs designed to prevent alcohol and drug abuse, prevent teenage pregnancy, reduce violence, and so on (I discussed several such examples in my book Redirect). Note that this ethical dilemma is no different than the one faced by medical researchers who conduct experimental trials of the efficacy of new drugs. Some patients are randomly assigned to a placebo group that gets no treatment, so that the researchers can find out whether the new drug really works. Why should we have different standards for social, psychological, and educational interventions?
(The study on reading comprehension was by Paula Clarke and colleagues, published in the August, 2010 issue of Psychological Science).
[Originally posted on PsychologyToday.com]