From: Mike Wiliams [jmicha5...@aol.com]
Sent: Thursday, January 17, 2013 12:19 AM
To: Teaching in the Psychological Sciences (TIPS)
Subject: Re:[tips] my crummy knowledge of stats
You can use a conventional paired t test. Although you have dichotomous
scores that does not
om: Mike Wiliams [jmicha5...@aol.com]
Sent: Thursday, January 17, 2013 12:19 AM
To: Teaching in the Psychological Sciences (TIPS)
Subject: Re:[tips] my crummy knowledge of stats
You can use a conventional paired t test. Although you have dichotomous
scores that does not mean they are categorical. Cor
quot;
To: "Teaching in the Psychological Sciences (TIPS)"
Sent: Thursday, January 17, 2013 1:41:34 PM
Subject: Re: Re:[tips] my crummy knowledge of stats
I think step 1 would still be an answer to the question "Was the change large
enough to be considered "real" or stati
Plymouth NH 03264
==
- Original Message -
From: "Claudia Stanny"
To: "Teaching in the Psychological Sciences (TIPS)"
Sent: Thursday, January 17, 2013 1:41:34 PM
Subject: Re: Re:[tips] my crummy knowledge of stats
I think step 1 woul
Sent: Thursday, January 17, 2013 1:28:53 PM
Subject: RE: Re:[tips] my crummy knowledge of stats
My understanding of the intent of the analysis was to find items which were
most affected, not a test for an omnibus effect across items.
> - Original Message -
>
> From: "
I think step 1 would still be an answer to the question "Was the change
large enough to be considered "real" or statistically reliable."
Step 2 would be to ask which of the statistically reliable changes were
largest . . . perhaps an estimate of effect size and rank ordering of the
questions. Out
My understanding of the intent of the analysis was to find items which
were most affected, not a test for an omnibus effect across items.
> - Original Message -
>
> From: "Annette Taylor"
> To: "Teaching in the Psychological Sciences
> (TIPS)"
> Sent: Tuesday, January 15, 2013 6:2
ogical Sciences (TIPS)
Subject: Re:[tips] my crummy knowledge of stats
You can use a conventional paired t test. Although you have dichotomous
scores that does not mean they are categorical. Correct/incorrect is a
ratio scale of 1 unit.
Green/Red, Accountant/Psychologist are the type of categori
You can use a conventional paired t test. Although you have dichotomous
scores that does not mean they are categorical. Correct/incorrect is a
ratio scale of 1 unit.
Green/Red, Accountant/Psychologist are the type of categorical
dichotomies that bring in the nonparametric procedures like C
Hi
I would consider an alternative approach. For each ITEM, calculate the
percentage of students who passed that item. Then do a paired-difference test
of significance for pre vs post with items as the random factor (i.e.,
"subjects"). This will tell you whether there was an overall change.
The correct statistical test is called McNemar's Test:
http://www.medcalc.org/manual/mcnemar_test.php
It is specifically for dichotomous outcome data which is paired (repeated or
matched).
A modified Bonferroni more or less as you describe looks easy enough to do
manually, after extracting the p
Hi Annette
Perhaps McNemar's Test for significance of changes, for dichotomous data. For
each item, set up a table that looks like a 2*2 chi square but has "pretest"
and "post-test" as variables (in texts its usually labelled "before" and
"after") .
Posttest
- +
+ A B
Pretest
- C D
So e
12 matches
Mail list logo
