Inferential, Wilcoxon, Mann-Whitney U Testing Flashcards
What is an inferential test? How does it help the researcher?
They are used to determine how significant/ strong a difference (or correlation) between variables is. It helps the researcher to determine whether the result (effect on the DV) is due to chance. If the test indicates a difference or correlation, then the alternative hypothesis is accepted, and the null hypothesis is rejected. If the test indicates no significant difference or correlation, then the null hypothesis is accepted and the alternative hypothesis is rejected.
Describe the probability and levels of significance in inferential testing.
What is common used level of significance?
Inferential tests rely on the concept of probability; the likelihood of an event occurring. The level of significance refers to the probability that the result is due to chance factors, rather than the manipulation of the IV.
P<0.05 (AKA this means that there is a less than 5% probability that the results are due to chance factors and therefore 95% confidence that the manipulation of the IV caused the changes in the DV).
What is the p-value chart?
> 0.1 - no evidence against null hypothesis
<0.001 - very strong evidence against null hypothesis
There are 2 errors in statistical testing. What is Type 1 error?
What are the two words used to describe it?
When the researcher accepts the alternate hypothesis, but the null hypothesis is true. This usually happens when the level of significance used is too lenient (eg p < 0.1).
False Positive
There are 2 errors in statistical testing. What is Type 2 error?
What are the two words used to describe it?
When the researcher accepts the null hypothesis, but the alternate hypothesis is true. The level of significance used is too cautious/ stringent (eg p < 0.01).
False Negative (TTFN)
What is the tip to remember in the Wilcoxon Test?
W- wilcoxon
O- ordinal
R- repeated measures design
D- difference
Ordinal data?
Data must be ranked from lowest to highest
What are the ranking rules?
A) do not rank 0s.
B) ignore positive and negative signs when calculating ranks.
C) any identical scores you give the next rank by adding the positions together (2nd and 3rd = 2+3=5) and divide them by the number of ranks and give them all this rank. AKA FIND THE MEAN.
D) add up ranks of positive differences and then negative differences.
E) T is the figure that is the smallest when the ranks are totalled.
How do you determine significance in a Wilcoxon test?
-Find the appropriate row on the table for either 1 or 2 tailed test.
-Find the appropriate column for the p value. P<0.05
-Find the number of participants- remember to subtract any PPS who had 0 difference.
-Select the critical value.
-The T value needs to be equal to or less than the critical value for the data to have significance.
How do you report your data at the end?
The T value of ______ is greater than/ equal to/ less than the critical value of ______. Therefore, the data does/ does not have significance and the alternative/ null hypothesis must be accepted.
What is the tip to remember in the Mann-Whitney U test?
I DO the Whitney U
I- independent groups design
D- difference
O- ordinal
How do we determine significance when reading a critical value table for the Mann- Whitney U test?
-Find the appropriate table for the p value.
-Find the Na and Nb numbers in the columns.
-Select the critical value.
-The U value needs to be equal to or less than the critical value for the data to have significance.
What do we report the results as at the end of the Mann-Whitney U test?
The U value of ______ is greater than/ equal to/ less than the critical value of ______. Therefore, the data does/ does not have significance and the alternate/ null hypothesis must be accepted.
