Day: October 5, 2020

Here is a research study from the Fordham Institute that should give all elementary teachers and administrators something to think about. I will say as a preface that I know nothing of the Fordham Institute and I tend to trust studies I find published in scholarly journals more than technical reports. This is because the review process for journal publication requires the review of an article by other scholars who might identify flaws I have not considered. Still, the results of this study seem solid and the deeper question is likely why the results obtained happen. The paper itself (a pdf available for download from the link I provide) may be longer that many want to take on, but the executive summary is easy enough to consume and consider.

The study makes use of a longitudinal design following more than 18,000 k-12 students focused on the development of reading comprehension skills. The basic idea was to predict (a regression statistical approach) differences in 5th grade reading comprehension scores from variations in classroom time focused on different subject areas over the K-5 time span.

Aside from the focus on the study just the data on how student time is allocated and how it varies from school to school is interesting. Over this time span, the average daily academic time is 302 minutes. ELA (language arts) takes up consistently the most time (120 minutes) with math coming in a distant second. ELA time spent is more than the total of all other academic areas combined if math is excluded.

Even with this solid generality, time in different subjects does vary so the regression model attempts to use the time spent to predict variations in reading comprehension in fifth grade based on these variations (controlling for some other variables such as K reading assessments).

The study found that the only variable predicting reading comprehension differences was the time spent in social studies. Note that this includes variations in time spent in ELA. In addition, the researchers broke the data down by SES quartiles and found that this outcome was consistent across the bottom three quartiles, but not in the most affluent quartile.

The interpretation. The authors conclude that social studies offers the best opportunity to develop general knowledge and comprehension itself ends up highly predicted by what one already knows about a content area. I agree with this explanation based on other information I have read. One of the most persuasive study (actually mentioned in this report) involves the description of the play by play from a baseball game. Learners were differentiated based on reading skill and baseball knowledge. So, think of this as four groups – high knowledge/high reading, high knowledge/low reading, low knowledge/high reading, and low knowledge/low reading. In the study high knowledge/low reading demonstrated better comprehension than low knowledge/high reading. If you think about the importance of background knowledge in understanding, this should make sense.

Before I reached their explanation, I wondered about science. Why was time spent studying science not a predictor? The authors thought of this too and proposed that science is more about specialized and not general knowledge.

What about the SES results? Again, I found their explanation quite credible. They suggest that kids from more affluent families simply have far more opportunities to learn about the world – develop general knowledge. Students with fewer opportunities are more dependent on schools to provide background and just focusing on the strategies of reading as a skill does not provide this exposure.

The authors argue that the allocation of school time so heavily to traditional reading misses out on the opportunity to develop general knowledge important to both reading comprehension and life. The authors argue for diversion of some time (they suggest 30 minutes a day) from ELA activities to social studies. As I read this, I thought about the potential of Newsela as a way to do this. Newsela offers reading material on many topics with each article available at multiple reading levels. This would seem the type of activity relevant to both reading skill development and knowledge development.

Back to my general concern. Longitudinal studies lack the manipulated controls of the best research. Hence, longitudinal research is by design correlational and without random assignment to treatments more prone to misinterpretation because of direction of causality or biasing variables. The student attempted to control for such possibilities, but this type of bias always remains a possibility. On the other hand, longitudinal research has a certain type of validity in taking a long term perspective based in actual experiences and while expensive or at least demanding the results are appealing.

I think it will be interesting over the next months to see if this study attracts some attention. It is unfortunate that our present point in time has us focused on so many other issues.

Adam Tyner and Sarah Kabourek. Social Studies Instruction and Reading Comprehension: Evidence from the Early Childhood Longitudinal Study. Washington D.C.: Thomas B. Fordham Institute (September 2020). https://fordhaminstitute.org/national/resources/social-studies-instruction-and-reading-comprehension.