I have conducted pre-tests to find out two brands that will be used later for my experiments. The brands (twelve brands in two separate versions of pre-tests, N= 36) are evaluated based on familiarity, attitude, brand credibility and brand innovativeness (existing scales used). The idea is to find out two brands that are similar in familiarity, attitude and innovativeness perception but differ in brand credibility perception. Now my question is how shall I identify the two brands? Shall I do multiple t-tests for each of these criteria ( familiarity, attitude,BC, BI) for two brands each? or shall I conduct an ANOVA to identify two brands? or any other method should be applied. Your answer is much appreciated.
You can do ANOVA but it will not point you to the exact brands which are similar/different. Thus you need to do something called post-hoc analysis in ANOVA . This feature is available in SPSS. Following link might be useful
https://www.youtube.com/watch?v=ehOWuOcyejI
Thanks everyone for the answer..is there any paper or book chapter that may provide guidance to T-test procedure? Thanks in advance...
Normally, T-test is used to compare the two population means. In your case score can be made to measure the familiarity, attitude and innovativeness perception for both brands and then independent t-test may be applied to see the similarity . Same thing can be done for brand credibility perception.
T-test can be conducted using SPSS.
Dear Mahmud Bhai, I am completely agree with Mr. Abhay Tiwari. For your case particularly will provide better result than any other test.
When sample size is small and population variance unknown T-test is appropriate used to compare the two population means . Furthermore, For two independent samples either t or F can be used. We know that F-test is similar with t–test (since F=t2).
Since, you have small sample size, you can use t test ..for two brands
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude