3 ms·

34 is a ridiculously small sample; not very scientific. Inconclusive results that basically boil down to "boys and girls think differently". Reporting via Tim

34 is a ridiculously small sample; not very scientific. Inconclusive results that basically boil down to "boys and girls think differently". Reporting via Time magazine . . . it's surprising.

ChaitanyaSai 17y ago
Why is 34 a small number? 10-20 subjects is the typical range for fMRI studies. While this is probably underpowered, even this number is not ridiculously small. A satisfactory sample size depends on the experiment; sweeping statements don't make much sense without the context. Here's a study that looked at the statistical validity of the usual sample sizes in fMRI experiments http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6WNP-4C8NJMP-1&_user=489277&_rdoc=1&_fmt=&_orig=search&_sort=d&_docanchor=&view=c&_searchStrId=961722718&_rerunOrigin=scholar.google&_acct=C000022679&_version=1&_urlVersion=0&_userid=489277&md5=ea8c1ab88ce705734e0f594ddf0b84f4 http://www.sciencedirect.com/science?_ob=ArticleURL&_udi... 34 is NOT ridiculously small. It is good enough.
- indiejade 17y ago
  Scientists asked 34 healthy kids, ages 8 to 17, to look at pictures of 40 other boys and girls and judge how much they would like to interact with them online. The kids were asked to rate those in the photos on a scale from 0 ("not interested at all") to 100 ("very interested"). "Good enough"? 34 is a decent number for the ruse. (As the article stated, the whole experiment was a ruse). Since the sample population was ages 8-17, that gives an age range of the sample population being 10 years +/- 1. The brain can develop a lot in ten years. I like evidence. No sense in killing the messenger.
  - ChaitanyaSai 17y ago
    The age distribution is a valid question, but tangential to the subject sample size. Increasing the number of subjects to 340 still won't resolve your problems with the distribution. The statistical validity of the brain regions differentially lighting up holds even if the networks in the brain changed drastically with age. You can definitely argue the assumptions behind the study, but the methodology is pretty standard and has better-than-average rigor.

34 is a ridiculously small sample; not very scientific. Inconclusive results that basically boil down to "boys and girls think differently". Reporting via Time magazine . . . it's surprising.

ChaitanyaSai 17y ago

Why is 34 a small number? 10-20 subjects is the typical range for fMRI studies. While this is probably underpowered, even this number is not ridiculously small. A satisfactory sample size depends on the experiment; sweeping statements don't make much sense without the context. Here's a study that looked at the statistical validity of the usual sample sizes in fMRI experiments http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6WNP-4C8NJMP-1&_user=489277&_rdoc=1&_fmt=&_orig=search&_sort=d&_docanchor=&view=c&_searchStrId=961722718&_rerunOrigin=scholar.google&_acct=C000022679&_version=1&_urlVersion=0&_userid=489277&md5=ea8c1ab88ce705734e0f594ddf0b84f4 http://www.sciencedirect.com/science?_ob=ArticleURL&_udi... 34 is NOT ridiculously small. It is good enough.

indiejade 17y ago
Scientists asked 34 healthy kids, ages 8 to 17, to look at pictures of 40 other boys and girls and judge how much they would like to interact with them online. The kids were asked to rate those in the photos on a scale from 0 ("not interested at all") to 100 ("very interested"). "Good enough"? 34 is a decent number for the ruse. (As the article stated, the whole experiment was a ruse). Since the sample population was ages 8-17, that gives an age range of the sample population being 10 years +/- 1. The brain can develop a lot in ten years. I like evidence. No sense in killing the messenger.
- ChaitanyaSai 17y ago
  The age distribution is a valid question, but tangential to the subject sample size. Increasing the number of subjects to 340 still won't resolve your problems with the distribution. The statistical validity of the brain regions differentially lighting up holds even if the networks in the brain changed drastically with age. You can definitely argue the assumptions behind the study, but the methodology is pretty standard and has better-than-average rigor.