Take a look at the results from a new study out of the University of Notre Dame and then go visit The Wall Street Journal for details. The differences are small, but still interesting and I’m curious to find out what readers make of this…
Children born in the winter months already have a few strikes against them. Study after study has shown that they test poorly, don’t get as far in school, earn less, are less healthy, and don’t live as long as children born at other times of year. Researchers have spent years documenting the effect and trying to understand it.
But economists Kasey Buckles and Daniel Hungerman at the University of Notre Dame may have uncovered an overlooked explanation for why season of birth matters.
Maybe. But on a personal note, I’m a May baby and it happens my older brother was born in January. He’s the smart one. And while personal anecdotes do not serve as evidence, correlations don’t necessarily either.
September 23rd, 2009 at 8:24 pm
http://xkcd.com/552/
September 23rd, 2009 at 8:39 pm
The increased numbers suggest that May pregnancies are more planned. That could have a lot to do with the education numbers, though it goes in the other direction (education => May birth).
September 23rd, 2009 at 9:03 pm
I think you still need to consider school calendars. These researchers have no explanation as to why different socio-economic groups have a seasonal pattern to childbirth.
Maybe some teens had a romance that peaked by the end of the school year.
Certainly college age and graduate students might specifically plan a baby’s birth around classes and finals or thesis exams. Summer births would work. Even workers seeking a maternity or parental leave would probably plan the event as best they could around employment demands.
September 23rd, 2009 at 9:49 pm
How about if there is a difference but there are actually eight, quite different reasons that cause the difference? How would they figure those out?
Looking for a simple answer to figures made up of so many different stories and life histories is probably going to result in an illusion for an explanation.
I’m looking at these kinds of things and thinking there are more pressing things for economists to be looking at. Like why so few of them saw trouble coming in the economy when the banking regulations were gutted in years past. I suspect THAT correlation would be a lot easier to figure out, pretty much related to one ideological fashion, which favors rich bankers and the such.
September 23rd, 2009 at 10:20 pm
A previous study found that professional athletes had birth dates preferentially clustered in the month that would have made them the oldest or near oldest on their teams as children. Thus they were more frequently the star child These means that many equally talented people never had a chance for a basically arbitrary and silly reason.
I think it would be beneficial for our society to see that children were provided with the best opportunities possible. Maybe the fix would have multiple causes, but some could turn out to be something fairly easy, like handing out Vitamin D tablets in the wintertime.
I don’t think that conducting these studies would in any way impede proper regulation of the banking system.
September 24th, 2009 at 1:05 am
Can you say “statistically not significant”?
September 24th, 2009 at 4:55 am
Gaythia, having a number of school teachers in my family and hearing how a lot of their colleagues already have a bias against the youngest children in their classes “based on what science says” I can almost guarantee that instead of the prescription for remedies, this is going to be taken by many people as a sort of neo-astrological determinant of fate. There’s a history that would seem to indicate that tendency, the difference in “IQ” testing which was originally taken as diagnostic, so that amelioration could be undertaken to what became essentially a social class assignment.
Even if this kind of study was finding an actual correlation between birth month and real life phenomena, that’s not any evidence that their analysis of the data isn’t too simplistic due to lack of detail in the information they have. What if the results they got for the people with birthdays in January were due to a number of discrete and unrelated factors, how would you design remedial measures? I would guess that some of those factors could be due to things like school year or that its easier for teenagers to find places to have sex at one time or another. Maybe the scheduling of proms has something to do with it. But its unlikely that you’re going to find one single reason for it because the reasons for them having unprotected sex are probably quite varied.
Whether or not these kinds of studies would impede the regulation of the banking system might be a simpler thing to study, there being so many fewer economists than teenagers who have children and their professional activities being, perhaps, less varied. If you couldn’t figure out a correlation between the known factors of their documented professional lives, you’re going to have a much harder time with the far more varied and very spottily documented lives of the general population.
But maybe it’s true that economists time might not be well spent on looking into this. I would trust people outside of economics to look into the failures of their profession more than economists, unless they were skeptical of the theories of capitalism and neo-classical economics. I suspect that’s the cause of the failure of economists to see what so many of us did when Phil Graham was gutting banking regulations, their theoretical models overrode what history had shown to have been the real life result of letting people with an economic interest in theft and plunder their ” Freedom to Choose”. J. K. Galbraith once said that an economist should have a thorough knowledge of history which seems to have given away to theoretical fantasy in that profession. And it’s interesting how often, as in the case of “gender science” how those fantasies seem to benefit those with the most money and privilege.
September 24th, 2009 at 9:22 am
Definitely insignificant correlations. Both me and my sister are January babies and we both graduated Magna Cum Laude with degrees in Math. We’re also primarily left handed, with both parents being right handed. I guess we simply defy the odds. If there’s anything significant, I’d say it may be the result of decreased exposure to the sun in winter months, resulting in decreased vitamin D production and possible differences in cognitive development.
September 24th, 2009 at 3:03 pm
Inconclusive results within the margin of error? I tend to think so.
September 24th, 2009 at 4:24 pm
This analysis violates everything I know about science, namely, that the plural of anecdote is data, that correlation is causation, and that they laughed at Galileo.
September 24th, 2009 at 6:21 pm
I can’t help but wonder what result mining techniques were used. On the other hand I don’t care to waste my time to find out. Anyone out there care to find out or explain why there appears to be a somewhat reproducible pattern over what are very small numbers?
September 24th, 2009 at 11:45 pm
I wonder what kind of sound time series statistical analysis were employed by these Notre Dame researchers. Seems quite suspect to me.
September 25th, 2009 at 1:43 am
One obvious possible explanation is that planned babies could be conceived uniformly year-round, and the number of “unintentional”/accidental conceptions experiences an upward fluctuation in late spring to early summer due to well known seasonal sex drive patterns (especially in colder states, where winter clothes give way to cleavages and mini-skirts starting around April).
September 25th, 2009 at 11:21 am
When Sheril calls her brother the “smart one”, is anybody else reminded of Simon Tam?
September 28th, 2009 at 6:45 pm
I think there is some misunderstanding here about WHY this result has any importance. The Wall Street Journal article helps explain it a little bit. The basic issue is this: it has sometimes been proposed in various studies that “month of birth” be used to drive some exogenous change in some variable, so that we can identify causal effects of that variable on some other variable. For example, it has sometimes been proposed that “month of birth” may be interacted with data on state laws on when you can leave school to identify “causal” effects of education on later outcomes such as earnings. We know more education is correlated with higher earnings, but is this relationship totally causal? The month of birth and school leaving laws interact to produce differences in what grade you will be able to leave school. If month of birth is uncorrelated with anything else — that is, if we can assume that those born in each month are a random sample of the population, with no tendency to have unobserved characteristics that vary systematically — then the correlations of month of birth with educational and earnings outcomes has the potential of relatively simply identifying causal effects of education on earnings. But if those born each month are NOT a random sample of the population, then things are more complicated.
Therefore, the issue is not the inherent importance of month of birth as something to investigate. The issue is the degree to which the authors’ finding makes it more difficult for social scientists to use month of birth as a source of “natural experiments” involving more important topics.
As for statistical significance, it should be noted that the authors are using huge samples.
December 15th, 2009 at 1:08 pm
[...] graph shows a clear seasonal trend and brings to mind the study on birth season that came out earlier this year from the University of Notre Dame. So perhaps we have yet another [...]