Wordcraft Community Home Page
Would you trust these research results?

This topic can be found at:
https://wordcraft.infopop.cc/eve/forums/a/tpc/f/932607094/m/5360050526

June 01, 2011, 20:14
Kalleh
Would you trust these research results?
I am at a conference in Milwaukee, and here is what one researcher had up on her PPTs:

We looked at affect sizes and paired t-tests

Would you take this research seriously?
June 02, 2011, 01:42
BobHale
Well...


"affect" does have a rare noun usage (subjective emotional response) so I suppose there is a very slim possibility it was the word meant.

"paired t-tests" I have no idea what it's actually supposed to be so it's plausible that it might mean something.

Whether I'd trust the research or not would depend on the rest of the research. Smile


"No man but a blockhead ever wrote except for money." Samuel Johnson.
June 02, 2011, 01:44
BobHale
And do you remember when I told you about a slide at a whole college presentation by our quality control division?

The final slide said

"ANY QUESTION'S?"


"No man but a blockhead ever wrote except for money." Samuel Johnson.
June 02, 2011, 09:55
Graham Nice
Effect size is the term she should have used, but I can't see anything wrong with paired t-tests.
June 02, 2011, 15:26
Geoff
quote:
Originally posted by Graham Nice:
Effect size is the term she should have used, but I can't see anything wrong with paired t-tests.

Graham, I disagree. Many psychology texts refer to affect[/]as a noun, as Bob says. However, given the context, "size" is awkward. I'd think something like [i]affect intensity, or some such term, would make it clearer.


It is no measure of health to be well adjusted to a profoundly sick society. -J. Krishnamurti
June 02, 2011, 20:32
Kalleh
No, Geoff and Bob. In the field it is "effect" sizes. I am very sure of that. I realize it might just have been a typo, but it sure made me wonder about the validity of the whole study. I suppose that is superficial thinking.

There isn't anything wrong paired t-tests.
June 02, 2011, 22:52
tinman
quote:
Originally posted by Kalleh:
I am at a conference in Milwaukee, and here is what one researcher had up on her PPTs:

We looked at affect sizes and paired t-tests

Would you take this research seriously?


I have no idea what PPTs, affect sizes and paired t-tests are.
June 03, 2011, 04:39
zmježd
I have no idea what PPTs

PowerPoint slides. (It's from the file extension .ppt.)


—Ceci n'est pas un seing.
June 03, 2011, 09:39
Guy Barry
quote:
Originally posted by Kalleh:
There isn't anything wrong [with] paired t-tests.


So why draw attention to it then?
June 04, 2011, 19:13
Kalleh
Good question. I see that I italicized both of them. I guess I was just quoting her slides, but I can see where I caused confusion. Sorry about that!

Tinman, an effect size is a common term in research, and I thought maybe it was a general term as well. Maybe not. A t-test is a fairly common statistic, but again, maybe not as common as I had thought.

I actually thought PPT was an abbreviation for PowerPoint, but your explanation makes perfect sense, z.