![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Bad Social Science
Sep. 30th, 2005 10:42 pmToday I read some sad things about social research.
If you do a study with male subjects but no female subjects, that proves that you think that females do not have the characteristics you are studying. (I thought it might mean you were working in a boy's school or something, and it was just easier to find males.)
Any study that doesn't involve both genders is highly biased and therefore totally wrong because the two genders are obviously completely different in every way. (I thought that being from the same species might give you some similarities.)
All thirty-year-old research is invalid because things have completely changed in the last thirty years. (I thought our genetics had been relatively stable over that time period.)
If a study is done in one culture, and six states are discovered, and then the study is replicated in other cultures, where five of the same states were found in all of the cultures, plus an additional one, this proves that all cultures are so different that you can never understand what's going on when you switch from one to another. (I thought it showed there was a suprising similarity.)
Any famous guy in your field before you did your study must have influenced you in some way. (I thought that maybe only some of them had.)
If someone studies several phases of a trait, then one consequence of the research is that the trait is very important, and if only everyone had it, the world would be a better place, so we should teach people to have the trait. (I thought the consequences were much narrower.)
If someone can't because of why, then if they were trying, you should. (I'd always thought that if someone can't write well enough for you to be fairly confident that you know what they're saying, then maybe they can't get credit for what they were trying to say.)
I've decided I'm never going to score social science tests again unless they are desperate because I really don't like reading stuff like that. (Especially no the bad of deciphering stuff.)
If you do a study with male subjects but no female subjects, that proves that you think that females do not have the characteristics you are studying. (I thought it might mean you were working in a boy's school or something, and it was just easier to find males.)
Any study that doesn't involve both genders is highly biased and therefore totally wrong because the two genders are obviously completely different in every way. (I thought that being from the same species might give you some similarities.)
All thirty-year-old research is invalid because things have completely changed in the last thirty years. (I thought our genetics had been relatively stable over that time period.)
If a study is done in one culture, and six states are discovered, and then the study is replicated in other cultures, where five of the same states were found in all of the cultures, plus an additional one, this proves that all cultures are so different that you can never understand what's going on when you switch from one to another. (I thought it showed there was a suprising similarity.)
Any famous guy in your field before you did your study must have influenced you in some way. (I thought that maybe only some of them had.)
If someone studies several phases of a trait, then one consequence of the research is that the trait is very important, and if only everyone had it, the world would be a better place, so we should teach people to have the trait. (I thought the consequences were much narrower.)
If someone can't because of why, then if they were trying, you should. (I'd always thought that if someone can't write well enough for you to be fairly confident that you know what they're saying, then maybe they can't get credit for what they were trying to say.)
I've decided I'm never going to score social science tests again unless they are desperate because I really don't like reading stuff like that. (Especially no the bad of deciphering stuff.)