Don't make me turn this car around...
Here's an example of some left vs. right studies (the first on this thread, I believe):
http://www.webmd.com/news/20000908/d...y----sexuality
If we look at how comparisons are made, we can see the authors have a perfectly workable understanding of statistics. Let's peek at some short examples:
Compared with heterosexual participants, gays had 39% greater odds of being non-right-handed
Here we COUNT the number heterosexual and homosexual participants, and find a 39% higher rate of lefties among the latter group. People are countable, and numbers of people are comparable. 39% makes sense. Note that they could just as well have counted the number of artists.
Left-handedness is more common in men than in women, and tests of fetuses (using ultrasound) show 92% sucking their right thumbs,...
We COUNT the fetuses, both those who suck their right and left thumbs. Of the total counted, 92% are sucking their right thumb, 8% their left. Fetuses are countable, and numbers of fetuses are comparable. 92% is a useable statistic.
...an analysis of 13 identical and fraternal twin studies -- ranging from 1933 to 1985 -- showed that only 76% (of the expected 100%) of identical twins are both left-handed.
13 studies were tallied, and the total number of left-handed members of twin sets were COUNTED. Of the left-handed twins, the number who also had a left-handed sibling-twin was COUNTED, and was 76% of the first group. Twins are countable, left-handed twins are countable, and sets of twins with differeng left/right-handedness are countable. 76% is statistically meaningful.
Other research has shown that left-handed people have fewer offspring,
Offspring were COUNTED for both left- and right-handed people. The number for left-handers was smaller within their sample. Parents are countable, offspring are countable. "Fewer" describes this simple relationship. A percentage would have been more specific, but not more "correct."
Left-handed people have been shown to have more disorders involving the nervous system, autism, stuttering, cerebral palsy, mental retardation, epilepsy, and schizophrenia. They also tend to have some minor physical anomalies -- like low-set ears and mismatched fingers
Occurances of these disorders were COUNTED. Either you have the disorder, or you don't. The lefties in the sample group had a higher total of disorders. "Shown to have more" is quite clear in its meaning.
...within the 10% of the population that is left-handed...
People were COUNTED and grouped into left- and right-handers. The first group was 10% the size of the total sample group. 10% of a number of people is completely valid.
There are some examples of statistics being used properly. Nowhere did they give a quantitative comparison of creativity, because they understood it has no statistical merit. Even if one used a test to guage the creativity of people, they could only statistically compare test scores, not "creativity." ie: "Left-handers scored on average 27% higher on the McSpaggo-Clemmer Test for Creativity."
|
|
Bookmarks