Hypothesis Testing

Question 1

What are steps to hypothesis testing?

Answer 1

Develop research ideas
Review the empirical literature
Develop research hypothesis
Determine which TYPE of research is most appropriate
Get IRB approval
Set alpha level based on acceptable level of risk
Collect Data: Which stat and why?
Compare Significance with preset p value

Question 2

What different types of research?

Answer 2

Qualitative/Quantitative
Nonexperimental/Experimental
Descriptive/Inferential Statistics

Question 3

Set alpha level based on acceptable level of risk

Answer 3

a. p<.05 – least strict, less than a 5% chance of committing a Type I error
b. p<.01 – more strict, less than a 1% chance of committing a Type I error
c. p<.001 – VERY strict, less than a .01% chance of committing a Type I error
d. The stricter our alpha level, the less likely we are to make a Type I error, but the greater our chance of making a Type II error

Question 4

Different types of stat data?

Answer 4

a. Frequencies (Freida) – descriptive statistics that give us general information about the demographic variables (No DV)
Cronbach’s alpha Correlations
T-tests

Question 5

What is a T-test?

Answer 5

(Tanya “Two timing Teetest”) – inferential statistics that check for differences between groups

Question 6

What is a correlation?

Answer 6

(Cora Lations, MFT) – inferential statistics that look for relationship (+/-) between variables

Question 7

What is Cronbach Alpha?

Answer 7

Cronbach’s alpha (Coach Cronbach) – reliability check of our measure (inferential)

Question 8

What are Frequencies?

Answer 8

descriptive statistics that give us general information about the demographic variables (No DV)

Question 9

What are the steps to compare significance with present p value?

Answer 9

a. If p<.05, we reject the null hypothesis. WHOOPPEE! We have significance!
b. If p is NOT less than .05, we do NOT reject the null…. Rats! There is no significance
c. Assess the likelihood that we might have made a Type I or Type II error

Question 10

What are Type 1 -errors?

Answer 10

– rejected the null, but should NOT have rejected it
1. Could have significance, but feel that the statistics are WRONG or that an error was made in the research and we should TRUST our results

Question 11

What are Type 11 error?

Answer 11

did not reject the null, should have rejected it
1. Could have NO significance, but we were very close OR we feel that revising the study slightly will yield the results we expected