Abstract:

As many undergraduate courses, the introductory course of management control at University of Chile includes readings assignments. The underlying assumption supporting this decision is that students who complete reading assignments will be more engaged in classroom discussion and, as a result, their learning will improve. In order to motivate students, the professor has being using reading comprehension tests. After several semesters of assessing students´ performance on these tests, the professor concluded that most of students were either not understanding or not reading the material at all.

According to the literature, peer learning can yield significant gains in academic achievement and, as a result, has being a widely adopted learning-strategy in higher education, in different contexts and disciplines (Boud, 2014). This study presents the results of introducing peer learning methodology for improving students’ reading comprehension.

In order to test the hypothesized improvement in students’ reading comprehension we used the students’ grades in reading tests during a semester as dependent variable. A total of 136 students participated in this study during 2015. Seventy-seven students from spring semester were exposed to the peer learning intervention while sixty-six students from the fall semester were used as an historical control group. Specifically, the peer learning intervention included two steps in order to mediate the way they were preparing the readings. First, students were randomly assigned to one peer. Each pair of students was responsible for jointly addressing the main issues of the assigned reading. Second, each pair of students was asked to discuss its conclusion with the rest of the students in a reading preparation class facilitated by a teaching assistant.

In order to control for the differences between semesters on the dependent variable other than those caused by the intervention, several efforts were made. First, this study only considered sections of the same course taught by the same professor. Second, there were 9 reading tests per semester, and each test included the same reading assignment across semesters. Finally, a panel of three professors of the Business School validated the similarity between the reading tests of each semester in terms of difficulty and content.

We conducted all data analysis using the SPSS® software package version 20.0, with the alpha level set at ≤0.05. First, we performed aggregated mean comparisons between groups. The average mean on reading tests of the spring semester was significantly higher than the average mean of the fall semester. In addition, the standard deviation of the experimental group was higher than the control group (1,11). Second, results form Kolmogorov-Smirnov test suggested data were not normally distributed, as the level of significance was less than 0,05 for each of the reading tests. Therefore, the non-parametric median comparison test was used to establish whether there were statistical differences between the control group and the experimental group. The results suggest that the peer learning strategy implemented improved the students' reading comprehension from one semester to the other.

Keywords:

Peer learning, student performance, reading comprehension.