namf
Forum Elite
Topics: 80
Posts: 312
|
Hi all,
From the prior post I did on stats, PsychDr2B was wonderful in posting some definitions and questions. Unfortunately, it looks like I answered them about 20 minutes after the good Dr2B's last post (why wasn't it on my screen!?). Well, if anyone could assist then with elaborating the differences between t-tests and chi-square--especially with questions/examples--I would be very greatful indeed!
Thanks!
|
| mjl1717
Forum Hero
Topics: 955
Posts: 5,451
|
Ill attempt to start with the more difficult of the 2 under the auspices of PsycheDr2b 
1) t test uses 2 groups[ 1 interval varible and 1 nominal variable]
The Dr. believes that men who have dogs as pets have lower B.P. readings.(which is true by the way)
*men with dogs versus men without dogs(nominal)B.P. reading is an interval-Therefore simple t test
2)matched pairs t test also uses 2 groups [1 interval variable amd 1 nominal variable]
Monitoring chages in cholesterol for a group of 40 hypercholesterolemia
subjects 1 week before and 5 months after beginning a coure of anti cholesterol medication
*before and after(nominal,2groups,matched pairs), and cholesterol level
(interval) Therefore matched pairs t-test
The big difference is the simple t-test uses 2 groups/int. and nom.
The matched pairs t-test uses 2groups, linked data pairs, before and after
___________________
Smell the coffee! "Is That an Osler move??"
|
| namf
Forum Elite
Topics: 80
Posts: 312
|
Thanks for the thorough explanation of diffences between simple and matched pair t-tests, mjl1717.
However, that was not my question. My question had to do with the difference between t-tests and chi-square tests. Any ideas there--anyone else too??
Thanks!
|
| mjl1717
Forum Hero
Topics: 955
Posts: 5,451
|
Chi square only uses nominal variables
it tells us whether the proportions of observations falling in different categories differ significantly from the proportions that would be expected by chance.
A lot of times it can be set up as contingency table, but uses a critical value in a chi square table.
[N]- passing or failing USME
[N]- USMLE pass rates at 3 different med schools
* is there a relation ship between school attended and passing or failing the examination.
Hope this helps.
___________________
Smell the coffee! "Is That an Osler move??"
|
| xfwang2002
Forum Junior
Topics: 8
Posts: 47
|
Hi, mjl1717
you are an expert in statistics. i have been confused by the terms such as interval v and nominal v. could you give me some examples which type data is interval or nominal? Thanks
|
|
| |
| | | | | |
