Focusing students' attention on the relevance of science improves educational outcomes

In the discussion about improving science education, the focus often falls on large-scale programs. However, some studies are suggesting that even short-term, small interventions in classrooms can have significant effects on student achievement. In the 4 December 2009 issue of Science (okay, I'm a little behind in my reading...), Hulleman and Harackiewicz investigate the efficacy of short writing assignments on student interest and achievement in science.

It's generally well accepted (and logical) that students with an interest or stake in the material they are learning perform better, work harder, and retain more of what they learn. However, it can be difficult to identify causal relationships between various interventions and student success, because many programs that do produce good results include a lot of different components, not all of which are focused on increasing student interest in the topic.

Hulleman and Harackiewicz hypothesized that activities that draw students' attention to the relevance of instruction to their everyday lives would lead to an increase in achievement. They also hypothesized that the increase in achievement would be largest among students who started out with low expectations of their own abilities--that is, students who think they will do poorly in science would benefit more from these interventions than students who already think they will do well.

To test these hypotheses, the researchers performed an experiment involving 262 high school students. The students attended two different high schools, were taught by seven different teachers, and were enrolled in three different science courses (biology, physical science, or integrated science). Nearly all (92%) of the students were ninth-graders (the remaining 8% were tenth-graders). About half (52%) were female. The majority (66%) were white; 15% were African-American, 12% were Asian, and 8% were Hispanic. The experiment spanned a single academic semester, but data were also collected on about 40% (100/262) of the students' grades for the following semester.

The students were randomly divided into two groups (one experimental group and one control group). Slightly more than half (136/262) of the students were in the experimental group. The teachers did not know which group each student had been assigned to. Each group was assigned to write between one and eight short essays over the course of the semester. Students in the control group wrote summaries of the content they were learning in the class. Students in the experimental group wrote essays describing the relevance of what they were learning to their own lives. (It was not clear whether that context was provided to the students or whether they were expected to identify it themselves.)

At the beginning of the semester (before the experiment), the researchers surveyed students to learn their expectations of success in the course and their interest in science. They also surveyed the students at the end of the semester to reassess their level of interest in science and to learn their future plans for careers in scientific fields. Students' grades during the semester were also monitored.

The results suggest that even this type of simple, short-term intervention can significantly improve student results, at least for those students who start the class with low expectations. Students with low expectations who were in the experimental group had significantly higher grades--"nearly two-thirds of a letter grade", to be exact--than those in the control group. Low-expectation students in the experimental group also showed significantly increased interest in science compared to those in the control group.

Among students with high initial performance expectations, there were no statistically significant performance or interest increases from the intervention. In other words, students who thought they'd do well in the class generally had the same grades (and post-class interest in science), regardless of which group they were in.

Although these results are encouraging--they suggest that teachers can increase the likelihood of student success, especially for the most at-risk students, through a few simple writing assignments--the sample size was small, and relatively homogeneous. Larger studies involving more students, especially minority students, would increase the reliability of the results. Until then, though, these results certainly suggest that these kinds of assignments can't hurt, and could be very helpful.

Hulleman, C., & Harackiewicz, J. (2009). Promoting Interest and Performance in High School Science Classes Science, 326 (5958), 1410-1412 DOI: 10.1126/science.1177067