Redshirting: Article XIX

Here's another from the journal "Psychology in the Schools" (Vol 39(4), 2002, pp 441-457), which unfortunately isn't available free online. However, this article is available here, for a price of $30.

"Winning the Battle and Losing the War: Examining the Relation Between Grade Retention and Dropping Out of High School"

This paper, again, is a survey of other papers, trying to build support for its thesis by examining multiple other published research.

Abstract:

A systematic review of seventeen studies examining dropping out of high school prior to graduation demonstrates that grade retention is one of the most powerful predictors of dropout status. The discussion addresses the discrepancies among the perspectives of many educational professionals regarding the effectiveness of grade retention and deleterious long-term correlates. The transactional model of development is presented, which emphasizes developmental trajectories over time, in order to facilitate the interpretation of the association between grade retention and school withdrawal. Educational professionals, teachers, researchers, parents, and policymakers considering the efficacy of grade retention are encouraged to consider the implications of these findings.

So far, I've seen articles about high-school drop-outs, but this is the first one that mentions a lower level of post-secondary (ie. college) education as well:

From the first page (p 441)

In a review of retention research spanning the last 100 years, Jimerson (2001a) concluded that the results of research published during the past decade examining the efficacy of grade retention on academic achievement and socioemotional adjustment are consistent with the converging evidence and conclusions of research from the remainder of the century that fail to demonstrate that grade retention provides greater benefits to students with academic or adjustment difficulties than does promotion to the next grade. Moreover, results of recent longitudinal retention research suggests that children who are retained are more likely to drop out and less likely to attend post high school educational programs (Jimerson, 1999).

From page 443, in the "Results" section:

Barro and Kolstad (1987) provided a report on who drops out of high school that discusses many important contemporaneous and early predictors. The results indicate that early grade retention increased the risk of dropping out by 30% to 50%. In the examination of indicators of progression through school, the authors suggest, ". . . that grade retention has a more nearly independent status than performance as a determinant of failure to complete school"

Read that last sentence again. It is saying that a kid who does poorly academically, but who has never failed a grade or been retained, is less likely to drop out than a kid who does well in school but who has been retained. That to me is a stunning result.

Here's another study that was a comparison of Latino to other-white and black kids. In the end, they said the same thing, race mattered less than retention:

Fernandez, Paulsen, and Hirano-Nakanishi (1989) compared male and female high school dropouts among several Latino groups, in addition to non-Hispanic White and Black youth. [...] The Latinos as a group were more likely to be retained than non-Hispanic Whites and Blacks and had the highest dropout rate. Overall, it was found that retention emerged as a powerful predictor across all groups. Moreover, results of multivariate analyses demonstrated that subjects who were retained were more likely to drop out, independent of all other significant variables, which included math achievement, grades, and nuclear family responsibilities. Because the respondent's grades and test performance were statistically controlled, the authors suggest ". . . this pattern is strong evidence that students who have been grade delayed tend to drop out because of the age disjuncture between themselves and their peers and the consequent lack of fit between the respondent's peer group and classmates" (Fernandez et al., 1989; p. 37). Fernandez and colleagues (1989) conclude, "Regardless of race and ethnicity, scholastic performance and grade delay affect students' decisions to remain in school or drop out" (p. 47).

All together now: "Well, duh!"

And another: it's not about educational attainment, it's about being too old for the grade level:

Grissom and Shepard (1989) reevaluated three large data sets and consistently found that students who were retained, regardless of socioeconomic level, were placed at risk for dropout. Analyses that controlled for student background and gender revealed that low academic achievement alone could not account for dropping out. Utilizing causal modeling techniques, it was found that retention was the most significant predictor for high school dropout for these students.

This article has a table running 6 pages, summarizing 17 different studies of the topic. Some put numbers on the statistics. Like these:

* 78% of high school drop outs had been retained

* Kids retained in K-4 were 5 times more likely to drop out, and kids retained later than 4th grade were almost 11 times more likely.

* Repeating a grade during elementary reduced pursuit of post-secondary education by 85% (unfortunately, I can't find that article, it is listed as: Brooks-Gunn, J., Guo, G., & Furstenburg, F.F. (1993). Who drops out of and who continues beyond high school? A 20-year follow-up of Black urban youth. Journal of Research on Adolescence, 3, 271–294.)

* 27% drop out rate for non-retained students [! - one in four for non-retained students? Yikes] and 69% for retained, 94% if retained twice.

* Retained kids are 4 times less likely to get a diploma or GED.

That's interesting: retention rates are increasing, more boys than girls are retained, leading to higher drop out rates among boys and few college-bound boys. Could something like retention during kindergarten explain some of the disparity in college attendance between men and women, with college campuses now leaning 60/40 to women?

It's amazing that decisions made 13 years earlier can lead to the current college gender imbalance.