Does redshirting help?

Any parent with a child born near the school entry age cutoff faces a dilemma – should they let their child start school a little early, or a little late? In the US, the practice of holding one’s child back a year is known as ‘redshirting’.

I have a little personal experience with this. My birthday is August 3, and the NSW cutoff was August 1. My parents opted to send me to school with the earlier cohort, but at the end of primary school, they decided that they didn’t think I was socially ready to enter high school, so they had me repeat grade 6.

Did they make a mistake in failing to hold me back from starting school? If a new paper from Norway is to be believed, the answer is no. At age 18, children who start school at a younger age have higher test scores than those who do not. On the other hand, girls who start school early are also more likely to be teenage mums, a factor that some parents may want to bear in mind. Overall, the paper seems to indicate that redshirting is generally a bad idea, but the effects – to the extent we can observe them – are reasonably small.

Too Young to Leave the Nest: The Effects of School Starting Age
Sandra E. Black, Paul J. Devereux, Kjell G. Salvanes
Does it matter when a child starts school? While the popular press seems to suggest it does, there is limited evidence of a long-run effect of school starting age on student outcomes. This paper uses data on the population of Norway to examine the role of school starting age on longer-run outcomes such as IQ scores at age 18, educational attainment, teenage pregnancy, and earnings. Unlike much of the recent literature, we are able to separate school starting age from test age effects using scores from IQ tests taken outside of school, at the time of military enrolment, and measured when students are around age 18. Importantly, there is variation in the mapping between year and month of birth and the year the test is taken, allowing us to distinguish the effects of school starting age from pure age effects. We find evidence for a small positive effect of starting school younger on IQ scores measured at age 18. In contrast, we find evidence of much larger positive effects of age at test, and these results are very robust. We also find that starting school younger has a significant positive effect on the probability of teenage pregnancy, but has little effect on educational attainment of boys or girls. There appears to be a short-run positive effect on earnings of beginning school at a younger age; however, this effect has essentially disappeared by age 30. This pattern is consistent with the idea that starting school later reduces potential labor market experience at a given age for a given level of education; however, this becomes less important as individuals age.

This entry was posted in Economics of Education, Economics of the Family. Bookmark the permalink.

24 Responses to Does redshirting help?

  1. Cathy says:

    It is interesting that the negative effect on girls is noted, as there seem to be a number of child rearing experts telling us that it is better to start boys at school later (supposedly, they are less mature at the same chronological age).

  2. Patrick says:

    I was furious to have to hold my son back because the ages have shifted since I was at school (and they are now absolute rules, it appears). I know some people who have done it because they want their kids to be ‘leaders’ or whatever but frankly I don’t believe that better impressing their prep class is going to enable them to swing a corporate tax team, a project team, a boardroom, a conference or anything at all their way later in life.

    Better to have as much experience under their belt as possible, I would say.

  3. conrad says:

    The rate of teenage pregnancy — now there’s an important factor that obviously deserved to come before earnings in the list. It’s surprising how pervasive US culture is. I wonder why people don’t start reporting the almost as equally useless becoming a sporting star too, which the media seems to think works against younger kids (perhaps there’s data too).

  4. Kevin Cox says:

    It is good for some kids and bad for others and you have to look at the individual child to know who will benefit and who will be hurt. The fact that you can find a statistically significant number of teenage pregnancies or whatever is irrelevant because how do we know how many Andrew Leighs we destroyed by holding them back. Allow a judgmental choice within some arbitrary range (say three months) would seem to be a good enough compromise and keep track of the progress of those children where a choice was made and see how they all perform and then try to understand what were the characteristics of the child that led to good results and bad outcomes to inform future decisions.

    However there are more pressing social issues on which to spend resources.

  5. Sinclair Davidson says:

    My parents were advised to keep my sister back a year but didn’t and later argued that they regretted the decision. Mrs D and I kept our son back a year and we have been very happy with that decision. It wasn’t a decision made lightly and we consulted widely with educational experts etc.

  6. Thinking in old ways says:

    A non firewalled version of the paper is at

    An interesting result:

    1) It suggests either the brain plasticity phase is over – or that the argument for early intervention is a little overstated. If young brains respond best then one would expect to see some evidence of this in the results of those who started youngest. (Although the fact that Norwegian kids start school at age 7 might mean that all the action in this regard is already over and done with. The paper suggests that in most cases the kids were at home prior to starting school with only a small proportion in formal child care so it suggests that as far as formal education/social patterning is concerned they are on an equal footing.)

    2) While I am sure that Kevin is correct in “It is good for some kids and bad for others” I have my doubts as to whether you can simply “look at the individual child to know who will benefit and who will be hurt”. In some cases you might be able to pick up something around maturity of social interaction – but then we do not know if this is the factor that drives the different outcomes.

    3) I am not sure why people think that teenage pregnancy is “irrelevant” or that concern about it is some evidence of the persuasiveness of US culture.

    4) “because how do we know how many Andrew Leighs we destroyed by holding them back” – I thought this was in essence examined – the study compares the results of those that start early and those that start late.

    5) A question remains whether, other than the initial earnings result which can be put down to time in workforce, the result is telling us much about the actual age of starting school or about the relative physical/social development of kids and how less developed kids progress when mixed with more developed kids, that is, up to some age, being almost 12 months younger than others is a (slight) issue. I suppose though the point is that the main finding is that it doesn’t really seem to matter much anyway – although having higher income up to age 30 and an extra year’s income might be of value in lifetime earnings.

  7. Jennifer says:

    Although I have chosen to send both my children to school young, rather than “redshirting them”, I have to wonder about causation. I haven’t read the study, but if there was any parental choice involved, wouldn’t the children who were sent younger be more likely to be more intelligent to begin with? Intelligence and ability to cope with starting school are far from being the same thing, but they are probably positively correlated.

  8. Thinking in old ways says:

    No parental choice was involved. The whole study is based on “the administrative school starting rule in Norway – children born in December start school a year earlier than children born in January, with a December 31 cutoff.”

  9. Gavin Findlay says:

    Andrew, an interesting parellel to this is whether the age a child starts a musical instrument affects their progress in the medium term or their ultimate proficiency. As you may know, the Suzuki string teaching method, where children begin as young as four and learn alongside a parent, is famous for producing seeming prodigies among primary age children. In my experience, however, the difference in proficiency between these and traditionally taught players lessens starting later pretty much disappears by the late teens. (Uh oh, another potential research topic…)

  10. Curaezipirid says:

    Of course the best outcome depends on the home social environment, in contrast to the school social environment!!

    One child I know is a small very dark skinned Tamil orphan who was adopted by a wealthy lonely white Aussie single mother with old money in her family. The son in the family is much older, and neither older brother or mother are much interested in a small black child’s perspective. She begged to go into school as soon as possible, and is the youngest and smallest child in her class, and thriving.

    Another child I know is a boy with two younger brothers, a stay at home mum, and whom is still now, at sixteen, more interested in the science of how toy cars manovoure through sand in the sand pit, than in pencils and paper, but he is in his final year of high school now, (after starting young and then moving interstate from ACT to QLD), and would have been best served starting a year later. However, the bright side of the story for him, is that if the school system can not cater to his needs, he still has time to fit in a tertiary entrance course at TAFE before starting at Uni, without being older than the peers he’ll be at Uni with. The only reason it would have been best for him to stay at home another year, is because he had a very happy home life, in which his learning was well structured and his emotional safety always assured.

  11. spog says:

    There is a separate finding that kids sent a year early develop an unhealthy interest in earned income tax credits.


  12. ChrisPer says:

    I was not started early, but put up a grade after starting because I could read at year 6-7 level already.

    I fell down badly in upper school and university because I never had to learn to work – my reading level meant I coasted my whole school life. That did not lead to the success at university level my early advantage might have predicted.

  13. Patrick says:

    There is a separate finding that kids sent a year early develop an unhealthy interest in earned income tax credits.

    Anecdotally, this is clearly correct, since Milton Friedman was schooled at a young age (and so was I).

    ~ ~ ~

    my reading level meant I coasted my whole school life.

    I found that my 3rd grade schooling from the US served me just fine in Australia until about 7th grade! From a maximising academic potential perspective I definitely should have had stricter teachers and a more accelerated curriculum, but then again, I might have had less fun, and I never went into anything so accumulated knowledge-intensive as engineering, scholarship, etc in any case.

  14. reason says:

    I wonder how they controlled this study. Isn’t there a danger of self-selection of those who are held back being on average less mature?

    My experience was similar to yours (I was a young starter, went to Europe and back picking up a year and then repeated a year). We held both my daughters back diliberately (both were younger than the compulsory starting age but would have been accepted). It was a strong argument for us that the kids would always be younger their classmates, and in Germany, where I live there is evidence that kids who start school younger are less likely to get into Gymnasium when they make the cut at the end of the 4th Year. I remember reading evidence that football players are at an advantage if they are born earlier in the year (as they are slightly larger as juniors). Maybe it depends how competitive the school environment is during the early years.

  15. Spiros says:

    It’s interesting how there are all these strongly held opinions based on nothing more than people’s own experiences, or that of their children.

    Which of course proves nothing.

    Why are girls who are held back more likely to get knocked up? Here’s my two cents worth: parents of younger girls in a given school year (those who aren’t held back) are more likely to be afraid that there daughters will get sexually exploited by older boys and so they are more likely to fit them out with contraceptives.

    Whaddayez reckon?

  16. Spiros says:

    That of course should be “their” daughters.

  17. MsLaurie says:

    More anecdote – I went to school early, and was always the youngest by at least six months, and often more. It was never a problem for me, except in first year Uni, when everyone else was 18 and going into bars, and I was still 17!

    My brother went to school ‘on time’, but struggled, and ended up repeating Grade 6, which did him a world of good.

    My conclusions? It does depend on the child, and kids should be held back, or jumped a grade (possibly at another school to avoid the social stigmas) as required.

  18. reason says:

    I know the language is ambiguous, but I think the effect on teenage pregnancy is the other way around. (A “positive” effect means an increase – it is mathematically positive not morally.)

  19. christine says:

    If I understand the variation in the paper correctly from the abstract, they’ll still not be able to distinguish between an absolute starting age and a relative starting age effect, I think? (ie it could be good to start early if everyone else does too, but not if you’re starting early relative to everyone else.)

    Given all the discussion of personal needs, perhaps the better policy recommendation would be that a bit more flexibility is needed? Except I think the problem with that is that then people get worried about whether there’ll be different enrolment patterns by socio-economic status, causing a bigger education gap between kids from richer vs poorer families (speaking roughly). But anyway, maybe what needs studying is flexible vs inflexible starting age systems, rather than systems with starting age at 6 vs 6.5 or whatever. Don’t think I’ve really seen anything on that at all.

  20. backroom girl says:

    Well, when it comes to personal experience, I can see you all and raise you one. I started school in the year I turned 5 (April birthday). Then as a result of moving back and forward between States in Australia and being a clever kid I managed to skip a year as well, with the result that I was still 16 when I finished Year 12.

    Everything had gone along fine until then. I was always top of the class in most things, albeit at a fairly small school, and having few other talents I was probably never destined to be popular. It was when I started university at the age of 16 that things started to come unstuck, no doubt because no matter how clever I was, I really wasn’t old enough to be there. So I dropped out after two terms in order not to fail and went into the workforce for a couple of years before returning to uni when I was coming up to 20.

    But thankfully, I never ever developed an unhealthy interest in earned income tax credits 🙂

  21. Patrick says:

    But thankfully, I never ever developed an unhealthy interest in earned income tax credits 🙂

    Well, BG, I dunno. I think it must depend on the definition of ‘unhealthy’, because I’d say you definitely developed a vastly greater interest than the other 99.99 per cent of the population 🙂

  22. backroom girl says:

    OK, Patrick – I’ll admit to a healthy interest in discussions about earned income tax credits and certainly in the broad field of social policy that includes earned income tax credits, but I would certainly never advocate them as a solution to any Australian problem. And that’s probably ’nuff said.

  23. GavinF says:

    I made a decision to hold back my eldest son, after family breakdown and 6 months in a pre school in Rome, from going in to kindy (part of primary school in the ACT) half way through the school year at age 5 yrs 3mths. Instead he spent an extra six months at pre-school. It clearly benefitted him emotionally, and when he started the following year he was physically and academically at the top of his year (and remains there, although his academic performance has slipped to Bs in some areas as his primary interests are sport, science, and music).

    I found this initial advantage helped his confidence greatly and stabilised any remaining emotional issues. From this experience I would suggest that a study of this kind would benefit from measuring social integration and confidence as well as academic perfomance.

  24. Kevin Cox says:

    It struck me in reading these submissions that noone has questioned the measure of age. Everyone is assuming that the years start ticking at the stime of birth. Surely a better starting point for measuring the effect of age should be that your life for this purpose starts at time of conception.

    Again I will come back to the point that having decisions based on the individual’s characteristics rather than that based on a statistic will be almost always better. I understand we do not know all the factors to be considered but making sensible judgements is a skill we are pretty good at and a fixed age for starting school is an administrative convenience rather than anything else. All the randomised trials will never change that fact.

Comments are closed.