Studies question 3rd grade retention

by David Safier

I just read that the House Education Committee has approved HB2732, which would hold 3rd graders back if their reading scores are below a set level (the bill includes a number of exceptions — some ELL students, some students with learning disabilities, etc.). So the legislative wheel is in spin.

I've noted in my Floridation of Arizona Education series how shaky the data is on the value of retaining 3rd graders. Today, I found a few studies which reinforce my concerns, including one saying any academic gains for students held back in Florida are short-lived.

Yesterday, Education Week posted a blog item, Early Retention Benefits Can Be Fleeting, Study Finds (It may be subscription only).

. . . most, but not all, studies on grade retention suggest that in the long run, the practice may not work out so well for the students. Most disheartening of all, some studies show, in fact, that students who repeat a grade in school are more likely than their nonretained counterparts to drop out later on.

The article refers to a study on the psychological effects on students being held back, which, the study says, grow more negative as the students move through school.

Some other studies talked about the academic effects of retention. The most interesting is a 2008 study, Third-Grade Retention: A Four Year Follow-Up, published by the Miami Dade County Public Schools. It says that, after a boost in reading scores for a few years, students who were held back do no better, and possibly a little worse, than equivalent students who weren't retained.

I won't get too deep in the weeds here, except to say the study takes students who were held back in the 2002-3 school year and compares them with equivalent students who weren't held back the year before, before the retention program was put into operation.

In the chart, you can see the retained students (the blue line) spiked ahead of the others (the red line), then leveled off and ended up, in 4 years, slightly behind those who weren't retained.

Interestingly, the researchers used a model created a few years before which showed the significant 2 year gain by the same students and concluded that retention successfully increased students' reading scores. The 2008 report revisited the students 2 years later and concluded that the gains didn't continue in later grades.

I've often accused G.I.'s Matthew Ladner of being obsessed with 4th grade reading scores. The reason may be, as I've suspected, the 4th grade scores serve his purpose, while more longitudinal data, following the students through later years, make the gains appear more questionable. Ladner, who has far more expertise in Florida reforms than I do, must know of these studies. But they add too many shades of gray to the black-and-white, up-and-down, good-or-bad conclusions he's so fond of reaching.