Tuesday, July 07, 2009

Implications of academic redshirting

I just realized that there is another advantage to redshirting kindergartners--the practice of delaying the start of first grade by one year, either by enrolling a child in an extra year of preschool or an extra year of kindergarten. The rationale given is that the child will be bigger and more apt to be a star athletically, academically and socially.

This practice is an arms race in upper-middle class families, especially for boys. Those families can afford the extra year of preschool, childcare, or a stay at home parent.

Furthermore, the actuarial implications of this are significant. I have read studies that estimate the delay of one year to the earning years for an entire population without changing retirement age or benefits produces a 3% shortfall. Add a "gap year" between high school and college and we are in serious financial jeopardy. (No cheap digs about living in a state issuing IOUs, please.)

Anyway, back to my aha! moment today.

The multiple choice tests to identify "gifted" kids compare a kid's score relative to a nationwide sample of kids in the same grade. A borderline kid with an extra year of schooling under their belt has a relative advantage. Those older kids are more likely to be males and from upper-middle class backgrounds.

You can buy your way to a gifted kid. That is sooooo unfair.

  • Parents who back this practice like to ask, "Do you want your child to be a leader or a follower?"

    My answer has always been, "Neither". I would be surprised if anyone was able to lead Iris. I would be equally surprised if anyone actually follows her.

  • BTW, the birthday cutoff for kindergarten enrollment vary by states. In some states, it can be as early as June. In others, notably California, the child need only to turn 5 by Dec 2 of the kindergarten year. If no kids were redshirted, and births are distributed evenly throughout the year (they aren't), the kindergartners in CA are half a year younger than their cohorts in the early cutoff states. That accounts for some of the differences between low and high scoring states on national tests.

  • I realized this because Iris is attending the Center for Talented Youth (CTY) camp this summer. Look at their testing requirements for Grades 2-6.

    "Achievement at the 95th percentile or higher on one or more areas of a nationally normed standardized test."

    When we received Iris' scores in the mail, I learned they were comparing her against kids in the 3rd grade, not kids the same age. Our school district does the same thing for GATE identification. She's a November baby and she had skipped a grade, pitting her against kids 1-2 years older. Luckily, she made the cut.

    But what about a bright kid from south Los Angeles whose parents put her in school at the earliest opportunity to save on babysitting costs? What if she barely missed the cut? Should she be denied the opportunities given those whose parents could afford the extra year?
The Lengthening of Childhood:
New England Public Policy Center Working Paper No. 08-3
by David Deming and Susan Dynarski, Harvard Kennedy School, Harvard University
You can download a pdf of the entire paper.
Forty years ago, 96 percent of six-year-old children were enrolled in first grade or above. As of 2005, the figure was just 84 percent. The school attendance rate of six-year-olds has not decreased; rather, they are increasingly likely to be enrolled in kindergarten rather than first grade. This paper documents this historical shift. We show that only about a quarter of the change can be proximately explained by changes in school entry laws; the rest reflects "academic redshirting," the practice of enrolling a child in a grade lower than the one for which he is eligible. We show that the decreased grade attainment of six-year-olds reverberates well beyond the kindergarten classroom. Recent stagnation in the high school and college completion rates of young people is partly explained by their later start in primary school. The relatively late start of boys in primary school explains a small but significant portion of the rising gender gaps in high school graduation and college completion. Increases in the age of legal school entry intensify socioeconomic differences in educational attainment, since lower-income children are at greater risk of dropping out of school when they reach the legal age of school exit.
Steve Sailer notes in Redshirting:" A Kindergarten Arms Race that:
The urge to redshirt is also found among state legislators. From 1975 to 2000, 22 states raised the minimum age at which a child is allowed to start kindergarten by moving the cutoff birthrate back earlier in the year. This has the effect of making younger children redshirt involuntarily. Only one state cut the starting age, and that by a mere two weeks.
Somewhat similarly, states have been moving to raise the age of their students so that their students will achieve higher average scores on national tests. A 1999 California law changing the minimum birthdate from December to September explained, "Comparisons between California pupils and pupils in other states on national achievement tests in the later grades are likely to be more equitable if the entry age of California pupils is more closely aligned to that of most other states."

Of course, even if its test scores go up, California won't necessarily be doing a better job of educating its children. It's just doing what educators all over have known for years. The easiest way to have your students score better on measures of educational performance is to start with students who score better on the measurements even before you try to educate them. Requiring students to be older when they start at the front end of the educational process is an attractive, if sneaky, approach to getting seemingly more impressive results out the back end.

This is true both with kindergarteners and professional students. For example, master of business administration programs are largely evaluated on the starting salaries of their graduates. Back in the '70s, business schools realized that the higher the salary a student earned in his last job before entering the MBA program, the higher his likely starting salary upon graduating.

So, elite business schools virtually stopped taking new college graduates. Entering MBA students at the famous Wharton School at the U. of Pennsylvania now average a full six years of work experience. Those spectacular near-six-figure starting salaries that MBA schools trumpet are less amazing when you realize that it's not uncommon for the average student at a top school to have left behind a job paying $65,000 or more.


  1. And in full disclosure, you should note that we moved across the island so that you could start 1st grade early too.

  2. My daughter went to first grade early but I could not in good conscience make her do kindergarten when she could already read! The school said she was too young at 5, but she had no trouble and graduated from both high school and college with honors. She is happy now because she got out of college two years ago and got a job which she loves - if I had held her back as suggested she would have gotten out just in time to job search during the great recession.

  3. Precisely. I think that some of the difference between CA and some Asian countries stem from the age difference. They also weed out kids and send them to vocational schools. Nothing wrong with vocational schools, done right--but they aren't reporting true national averages.

    Mom says she encountered pressure to have me repeat first grade when she enrolled me for school in Kansas, an early birthday cutoff state. I am so glad she held her ground.

  4. This was fascinating. There are so many things we haven't even started thinking about yet, and so many ways we feel disconnected from other parents in our economic class. We did get a bit of a surprise when the mother of one of Pumpkin's "classmates" at day care said she was thinking of moving her child to a more "academic" center. I had to struggle to hide my surprise. Pumpkin is 2! I mostly want her to spend her days playing.

    I want to raise a child who grows into a happy adult. I'm not sure that the strong, almost exclusive, focus on academic success some families seem to have is the way to do that. I certainly don't want to teach my child that it is OK to game the system just to get ahead. And I don't really want a new crop of "leaders" who were raised that way.

  5. Anonymous13:33

    You raise a good point about comparing age-to-age instead of grade-to-grade. (Age-to-age is indeed what they do for the athetics equivalent of “gifted kids.”) But… I see possibilities for intrinsic unfairness with doing age-to-age. Except for really unusual kids, one doesn’t figure out how to factor a polynomial on one’s own. It’s not a very hard thing to do, but someone, either a teacher or a parent, shows you. If in your school district you learn how to do that in eighth grade, then a seventh grader (say, one trying to place into a gifted program) taking the SAT will miss those questions, while the eighth grader has a shot at getting them. (I’m leaving out of this comparison those kids whose circumstances are fortunate enough that they could learn this at home. Sorry, Iris!) So, while there is some unfairness in comparing an 11-year-old 7th grader against a 12-year-old 7th grader, surely there is at least as much unfairness in comparing the math scores of 12-year-old 7th grader against a 12-year-old 8th grader. There are plenty of kids who are a little older than the median in their grade through no fault of their own, and through no hypercompetitiveness on the part of the parents.

    I’m not fully convinced of the risk of there being an “arms race” among parents. The phrase suggests there is no end in sight. But truancy laws would I think keep parents from waiting more than 12 months minus epsilon, no? I could be wrong about this. For the record, my kids are both on the young half of their grade’s distribution.


  6. You can actually download sample questions for the SCAT test from

    They are not strongly dependent upon what kids learn in school. For instance, the more you have read, the better understand word analogies. An older kid has had more time to read.

    In the computerized test, I think the older kids would get through the test faster, and then be offered up more of the harder Q, which lead to higher SCAT scores.

    And that is another thing that bugs me about some computerized tests. You can't skip hard Q and come back to them later. You have to pick an answer before you are offered the next Q. If you guess incorrectly, then they offer you up easier Q. Then you get stuck and can never get to other hard Q that you might have been able to solve.

  7. My son struggles still with what I believe are age-related issues compared to his same grade peers - he's young for his grade.

    He would certainly have had a better time socially had he been a year older in early grades. I'm sure this is what parents who redshirt recognize.

    However, as he gets older and the percentage age difference between him and his same grade peers decreases he seems to have a better and better time. On the whole I predict greater ultimate success for him based on my anticipation of a better age to intellect match in higher grades yet to come.

  8. I have friends who have redshirted their children, usually their sons, and it may have given them a better start in school, but I remain unconvinced that it has helped them long term.

    On the other hand in must be rather prevalent because my grandson meets the age requirement to begin preschool in the fall but his parents are encouraged to hold him back as his birthday is close to the cut off. They (and I) disagree, because he is already well ahead of his peers verbally and in a few other parameters, although not necessarily socially.

  9. I was another who could read at kindergarten age, so went to a (no longer extant) private school for kids born in January. (The Toronto cut off is 31 Dec.) My birthday is in May, so that wasn't an issue.
    I skipped another year due to a contracted high school, so ended up at university at 17 when most of my Ontario classmates were 19.
    My sister also ended up compressing her schooling and starting university two years younger than most of her classmates.
    I don't recommend that either - it just isn't good socially.
    I have to ask my elementary teacher friends if this "redshirting" practice is common in Toronto. I really had not heard of it.
    (Now I am wondering if senior kindergarten is compulsory, so it can't happen here.)
    Thanks for the thought-provoking post as usual!
    Lisa in Toronto

  10. Eric, you are right that this arms race cannot go on ad infinitum. But I have seen kids born in March and April being redshirted.

    Lisa, you cannot do a controlled experiment with life. For instance, my mother skipped many grades and did not like that socially, especially when she hit college way young.

    My birthday is in October, making me one of the youngest in my grade. When the school district suggested that I skip two grades in middle school(which would have put me on track to enter college at 15), she would have none of it. The district countered with one grade. She turned them down again.

    She and my high school vice-principal in charge of discipline got to know each other very well.

    When I asked her for advice about whether her granddaughter should skip a grade, her answer was, "YES! As many grades as she wants!"

    We had a lengthy discussion and our joint conclusion is that there is no ideal solution. Some kids will be different, no matter what you do. We do the best that we can to make her feel comfortable, understood and loved.

  11. I found an interesting discussion at http://www.wcer.wisc.edu/news/coverStories/pros_cons_holding_out.php

    "All groups who are overage-for-grade, whether they have been redshirted or retained, have higher participation in special education services for learning disabilities, cognitive disabilities, and emotional disabilities. In fact, delaying entry or promotion was associated with higher likelihood of exceptional educational needs (EEN) services, with redshirts being 1.89 times more likely to participate in EEN programs."

  12. We know a child 10 days younger than my daughter. The other family chose to enroll their child in kindergarten a year later than Iris, even though their child's birthday fell a few weeks before the Dec 2 deadline.

    Their child is bored beyond belief in school and has tuned out the teacher. Now, they are begging the teacher to give their child more challenging work to do. I've read that is a common complaint with the older 'super-kindergarteners'. If the teacher complies, then they either have to teach two classes at once, or they have to direct their teaching towards the more advanced students. That ramps up the academic content of kindergarten and first grade and puts some of the kids even further behind. Which makes people hold back their younger kids in an escalating cycle.

    Please don't create a behavior problem in the classroom. It harms every child in the classroom, including yours.

  13. Thanks for putting this information and your thoughts together. I came across your blog while looking for some good reading on this issue, as my younger daughter (also named Iris!) has an August birthday and so many people around us seem to be holding their kids back.

    I find it a little bit mind-boggling to think about this trend. We are racing towards what, again? First to complete school? Most advanced kid in the class wins at life? Hm.

    I hate the thought of my Iris being the youngest in the class and well over a year younger than the oldest kids. More than that, though, I hate the thought of her being bored out of her mind academically.

    The facts about gifted program testing & the prevalence of this practice among white males are downright creepy.

    And then the article that badmomgoodmom linked to (er, the bit she quoted, anyway -- I haven't followed the link yet) sort of makes me wonder if this practice really as insidiously common as people say it is. Perhaps we and various reporters and such are just whipping one another into a frenzy. Maybe the kids getting held back really mainly have legitimate issues.

  14. Whoops - badmomgoodmom is *you*! Sorry -- I'm new here. I meant the last article you linked to in the comments, from the wisc.edu site.

    Thanks again. I'll be back.


Comments are open for recent posts, but require moderation for posts older than 14 days.