tag:blogger.com,1999:blog-6894866515532737257.post2268036879926933338..comments2024-03-27T01:01:09.785-07:00Comments on Probably Overthinking It: Are first babies more likely to be late?Allen Downeyhttp://www.blogger.com/profile/01633071333405221858noreply@blogger.comBlogger9125tag:blogger.com,1999:blog-6894866515532737257.post-75677264734246204082013-08-28T07:20:15.131-07:002013-08-28T07:20:15.131-07:00Very interesting! When you ask if time of birth a...Very interesting! When you ask if time of birth affects the data set, you might be asking one of<br />two questions: (1) Is there a difference between babies born in summer or winter, for example, or (2) if so, does this difference explain a substantial part of the effect I wrote about.<br /><br />I haven't checked either one, but my subjective prior probability for (1) is 5%. There might be a difference in gestation period between summer and winter babies, but I doubt it.<br /><br />And my prior for (2) is less than 1%. Even if there is a seasonal effect, it could only explain the first baby effect if first babies are substantially more likely to be born during particular seasons.<br /><br />But if your subjective priors are higher than mine, you should check it out!Allen Downeyhttps://www.blogger.com/profile/01633071333405221858noreply@blogger.comtag:blogger.com,1999:blog-6894866515532737257.post-16171312815018528622013-08-23T20:06:31.298-07:002013-08-23T20:06:31.298-07:00Just as nature of birth (natural vs induced) could...Just as nature of birth (natural vs induced) could affect the data set, it may be interesting to find if time of birth (as in season during brith, summer or winter) may have an influence of affecting this data set. For example we may be able to infer that winter babies are likely to be late compared to summer babies ?S M Kripanidhihttps://www.blogger.com/profile/15610323065207473561noreply@blogger.comtag:blogger.com,1999:blog-6894866515532737257.post-77388539202044278802013-01-08T08:06:55.589-08:002013-01-08T08:06:55.589-08:00Good question. I just ran a quick experiment and ...Good question. I just ran a quick experiment and posted the results here:<br /><br />http://allendowney.blogspot.com/2013/01/are-first-babies-more-likely-to-be-late.htmlAllen Downeyhttps://www.blogger.com/profile/01633071333405221858noreply@blogger.comtag:blogger.com,1999:blog-6894866515532737257.post-11307875585754392592013-01-08T00:14:09.776-08:002013-01-08T00:14:09.776-08:00While interesting, I can't help but think you ...While interesting, I can't help but think you need to compare the first and others for the same woman. While may be unlikely it could still be that a tendency exists for a woman's second, third, etc, child comes earlier. Unknownhttps://www.blogger.com/profile/10921076653725377247noreply@blogger.comtag:blogger.com,1999:blog-6894866515532737257.post-72234846705218952612011-04-02T17:03:36.406-07:002011-04-02T17:03:36.406-07:00Actually, the medical intervention I was primarily...Actually, the medical intervention I was primarily thinking of was induction, which makes up a significant portion of birth events - possibly more than c-sections (see, http://www.chron.com/disp/story.mpl/breaking/6538913.html) and is subject to non-medical election. <br /><br />If election of induction was uniform across birth order, it wouldn't have a significant effect on the lateness theory.<br /><br />It would be interesting to see if there is any correlation between the timing of a first baby and whether or not the mom has a second baby. Not that I'm trying to generate more work for you, I just find these things fascinating.Richhttps://www.blogger.com/profile/13503921340078070514noreply@blogger.comtag:blogger.com,1999:blog-6894866515532737257.post-2959078907255925372011-04-02T15:52:52.310-07:002011-04-02T15:52:52.310-07:00Could be scheduled C-sections, but I doubt it is b...Could be scheduled C-sections, but I doubt it is because of other interventions, which tend to be later (40+).<br /><br />The dataset does have other medical information about the births, so I could investigate your theory.<br /><br />As for gender skew due to parental choices, I did a bunch of work on that, but in the end I didn't find anything that held up to statistical scrutiny.Allen Downeyhttps://www.blogger.com/profile/01633071333405221858noreply@blogger.comtag:blogger.com,1999:blog-6894866515532737257.post-78777034656823889052011-04-02T14:26:01.544-07:002011-04-02T14:26:01.544-07:00Interesting. Does the dataset have any informatio...Interesting. Does the dataset have any information on the nature of the births? It seems that the spike at 39 would be in part due to medical intervention after 38 weeks (which ends up getting pegged as the "due date" and creates a mindset that 39 is "late" even though 38-40 is normal). Elective C's are usually done at 39 weeks, iirc.<br /><br />So, it might be that fewer first babies are the subject of medical intervention (induction or c-section) due to desire to push for a "natural" birth process for the first child and more >1st babies are the subject of elective C's or other interventions, reducing the >39 dataset for the >1sts.<br /><br />I recall the discussion Ted mentioned - and that there was a skewing issue caused by parental choices - which actually created interesting stats not about birth gender, but family size based on gender balance. That is, there was an uneven distribution of bb, gg families were more likely to go for a third child than bg, gb families. Or something like that.Richhttps://www.blogger.com/profile/13503921340078070514noreply@blogger.comtag:blogger.com,1999:blog-6894866515532737257.post-49578986523628396052011-02-07T07:13:41.786-08:002011-02-07T07:13:41.786-08:00Yes, I hope so. I had a model that worked well fo...Yes, I hope so. I had a model that worked well for the 2002 dataset, but when I tested it on earlier data (also from the NSFG) it all came crashing down. I will try to get back to it for a future post.Allen Downeyhttps://www.blogger.com/profile/01633071333405221858noreply@blogger.comtag:blogger.com,1999:blog-6894866515532737257.post-64364985762887611262011-02-07T07:07:04.817-08:002011-02-07T07:07:04.817-08:00Quite a while ago, I remember discussing another c...Quite a while ago, I remember discussing another childbirth-related question with you: correlations between the sexes of siblings. Given that your first child is a boy, what's the probability that your next child will be? (And so forth -- given the sexes of the first n children, what can you say about the (n+1)st?) I seem to recall that you had a data-based answer to this. Future blog post?Ted Bunnhttps://www.blogger.com/profile/12230509214302717664noreply@blogger.com