Gender and reading achievement

iStock_000020536048XSmallA recent study challenges earlier research by demonstrating that reading achievement levels are more variable than previously thought. Former studies indicated that relative reading achievement remains largely unchanged as children progress through elementary school.

However, new data from a study by Linda M. Phillips, Stephen P. Norris and Agnes M. Maynard, University of Alberta, and Wendy C. Osmond, St. James Elementary School (Port aux Basques, Newfoundland), reveals that reading levels are variable and that about half of students change relative reading levels during elementary school.

In particular, boys who tended to read at lower levels than girls in the primary grades were about equal in reading achievement in the upper elementary grades.

Reading achievement followed for six years

Phillips et al. tracked the reading achievement of 187 boys and girls for six years. Compared to earlier researchers, they studied three times as many children from different schools for a longer period of time while attended more carefully to gender differences. These researchers sought to determine the relative reading achievement of boys and girls from first through sixth grades and to find out to what extent students’ relative standing remained stable over time.

Complete data was available for 87 boys and 100 girls. All students spoke English and came from similar socioeconomic and ethnic backgrounds. Children receiving special education services were excluded. All children in the study were exposed to the same basal reading program. This eclectic program provided opportunities to learn words as wholes, to use context and pre- and post-reading questions to aid comprehension.

Teachers both read to students and encouraged them to read and write on their own. Children were not given any remedial instruction outside of class and there were no early literacy interventions. As a matter of district policy, no children were retained in grade. Children were tested with the Gates-MacGinite Reading Tests to evaluate their achievement in decoding words and in reading comprehension at the end of every year.

Significant increases and decreases

Although there were no children whose achievement moved from below-average to above-average or vice versa during the elementary years, approximately half the students’ achievement level rose or fell significantly. In addition, there were systematic relationships between gender and reading only during the primary years.

In the first, second and third grades, higher proportions of boys than girls were in the below-average reading category, and lower proportions of boys than girls were in the average and above-average categories.

Performance at the end of fourth grade, however, showed no systematic relationship between gender and reading achievement. While the proportion of boys remained about the same in the below-average category, the proportion of below-average girls doubled. The differences between the proportion of boys and girls in the average category decreased in the upper elementary grades. In the above-average category, little change occurred between the primary and upper elementary grades.

Therefore, parity between boys and girls came about mostly through the deteriorating reading performance of some girls and the rise of some boys from the below-average to the average group.

30% probability of change in reading level

In summary, for children who were below average in first grade, there was about an equal chance for them to be either below average or average by sixth grade. Similarly, there was about an equal probability for students who were above average in first grade to either remain above average or drop to average by sixth grade.

Overall, there was a 30 percent probability of children changing category between the first and sixth grades without any remedial intervention. Fourth grade appears to be a transition point at which children’s increasing or decreasing achievement tends to accelerate.

For those who began as below-average readers, the results are encouraging — about half will improve to average readers. However, for girls these results are discouraging. There were more girls than boys in the below-average range in sixth grade, and a higher proportion of girls declined from average in first grade to below average in sixth.

Examining all children who were in the average range in first grade, there is no way to predict, on the basis of this data alone, those who would fall below average over the next few years. Phillips et al. contend that there is no sound basis for a policy decision to intervene for some of these children and not others. This data reveals that, without help, half of the children in the below-average range will remain below average in sixth grade. Phillips et al. state that schools should intervene with all children in the below-average range and with those whose relative achievement level drops from average to below average between first and second grades.

Fourth grade pivotal year

This study challenges evidence from earlier research. There was a much higher probability than expected that children below average in first grade would be in the average range by the sixth, as well as a significant probability for average students to become above average. However, there is almost an equal probability for students to drop to a lower level during these years. And there was evidence of an important transition during fourth grade. During that year the differences in achievement between boys and girls disappeared. Girls lost their advantage over boys.

Phillips et al. state that it is important to learn whether declines and improvements evident after first grade could have been predicted from performance in kindergarten. Since remediation works best when it begins early and is strategically directed, research in this area is needed.

“Relative Reading Achievement: A Longitudinal Study of 187 Children From First Through Sixth Grades” Journal of Educational Psychology Volume 94, Number 1, March 2002 Pp. 3-13.

Published in ERN May/June 2002 Volume 15 Number 5

Leave a Reply

  • (will not be published)