Re:He got 1%, we can't hire him
by disgruntledphd2 on Friday August 30, 2013 @05:20PM
(#6304291)
Dude, from my perspective, its the only sub field that is even halfway right in statistics. I have read so very many psychology papers in top-ranked journals that commit basic, stupid statistical mistakes all the time. And no-one seems to learn.
If you use a linear regression rather than ANOVA, you will often be asked to change it to an ANOVA.
I am well aware of the breadth and depth that is psychology, and most of it is poorly conducted and irritatingly bad. I actually think it has a lot to do with applying a particular experimental model of science developed for non-reflexive systems to reflexive systems, with predictably hilarious results.
Please do not take out the rest of the commentators opinions on me, it upsets me also when people bash psychology from ignorance, but I come from a place of love when I bash it, as I do really adore the subject, but feel that much of it is so very, very awfully done.
A/B testing - psychology degrees teach pretty good experimental design, but you wouldn't know it to see a lot of published research. Additionally, rigorous experimental design owes a lot more to statistics than psychology.
UX Testing: This is definitely a good area, but again I think we don't control enough for the impact of the researcher(s) - see Rosenthal, 1969 for the problem, and note the complete lack of care regarding these effects in modern psychology.
Grief management? Seriously? I don't really think psychology has added much here, but would be delighted to read some good papers that prove me wrong.
Ameliorating sexism and racism? Good intentions, but given that I read a lot of this stuff I would have to say that they are perhaps the worst for statistical sins and errors (with the notable exception of Brian Nosek).
I know very little about perception research in HUD's, do you have some good papers?
I'm not entirely certain why you felt the need to correct me, when if you look at my comments in the overall thread it can be seen that I am pretty much on the same side as you.
It is possible that I am so sick of people who commit statistical sins for career advancement purposes (something I could never do) that I may be taking it out on the field.
Incidentally, what is psychology? I would be interested to hear your definition (as long as its not waffle like the scientific study of the human mind and behaviour, which merely begs far more questions).
If you use a linear regression rather than ANOVA, you will often be asked to change it to an ANOVA.
I am well aware of the breadth and depth that is psychology, and most of it is poorly conducted and irritatingly bad. I actually think it has a lot to do with applying a particular experimental model of science developed for non-reflexive systems to reflexive systems, with predictably hilarious results.
Please do not take out the rest of the commentators opinions on me, it upsets me also when people bash psychology from ignorance, but I come from a place of love when I bash it, as I do really adore the subject, but feel that much of it is so very, very awfully done.
A/B testing - psychology degrees teach pretty good experimental design, but you wouldn't know it to see a lot of published research. Additionally, rigorous experimental design owes a lot more to statistics than psychology.
UX Testing: This is definitely a good area, but again I think we don't control enough for the impact of the researcher(s) - see Rosenthal, 1969 for the problem, and note the complete lack of care regarding these effects in modern psychology.
Grief management? Seriously? I don't really think psychology has added much here, but would be delighted to read some good papers that prove me wrong.
Ameliorating sexism and racism? Good intentions, but given that I read a lot of this stuff I would have to say that they are perhaps the worst for statistical sins and errors (with the notable exception of Brian Nosek).
I know very little about perception research in HUD's, do you have some good papers?
I'm not entirely certain why you felt the need to correct me, when if you look at my comments in the overall thread it can be seen that I am pretty much on the same side as you.
It is possible that I am so sick of people who commit statistical sins for career advancement purposes (something I could never do) that I may be taking it out on the field.
Incidentally, what is psychology? I would be interested to hear your definition (as long as its not waffle like the scientific study of the human mind and behaviour, which merely begs far more questions).