Mann-Whitney U Test and Alternative Non-Parametric Tests in SPSS
HTML-код
- Опубликовано: 7 сен 2024
- This video demonstrates how to conduct a Mann Whitney U Test in SPSS. The alternatives to the Mann Whitney U Test are also reviewed (Kolmogorov-Smirnov Z, Moses extreme reactions, and Wald-Wolfowitz runs).
I have been looking at videos and much of them are confusing. This, however, was amazing. So simple yet so clear! Thank you!
These examples help put the information into real terms. Thank you for these practice examples.
Thank you soooooooooo much, Dr. Grande! You saved my life today, or my Thesis rather!
Can you make a video on how to correct for covariates in non-parametric tests like the Mann Whitney U test? And how to do Quade's test (non-parametric ANCOVA)?
Hello Todd, thanks for the supportive video. I have found it very important for my analysis, I have a question though. I run mann-whitney test to compare the annual income difference between two groups i.e Forest user groups and Non-forest user groups. I have a sample size of 190 individuals from each group. After I run the test I found the following results.
Mann-Whitney U=10078.500
Wilcoxon W= 28223.500
Z= -7.453
Asymp. Sig. (2-tailed)= .000
However, I am not certain whether my results are certain or not and if I can compare the above two groups using mann-whitney test. Any suggestion is appreciated.
Thanks a lot.
First of all, thanks for so great video. I'd like to ask you one question: When I change my values from String to Numeric and set values to them on the data view tab I get just dots instead of the values that I assigned to them. Could you please help me?
Hello again, I have another question about the sign of results of ks test and wilcoxon rank sum test. As in your example that the first is negative while the second is positive. how to interprete it because at the first glance, it seems contradictory. I have the same question for the result of t-test. I don't know how to explain it. Thanks a lot!
Is there a non parametric test corresponding to MANCOVA ?
Dear Dr, I'm not MPhil or Ph.D. student but I would simply like to ask that if I want to find the relation between risks factors (independence Variables) such as age (classified into various ranges), gender, education, and independable variables like knowledge, attitude or practice score, then what type of test will I need to use? If I want to see the effect of education on knowledge of patients towards their disease?
Assuming that your data is normally distributed, you might wanna go with a linear regression with knowledge of patient level as dependent and dosage etc as independent. There you will see the sig. value also called p-value to test Null Hypothesis and if the value is below .05 then you consider your model significant and there is certainly an affect of variable on dependent
where is the interpretation for mann whitney u test.....apart from that you said everything..
Thanks Todd
I found ur video very interesting
I hope I could have ur E mail to contact u and get ur help in my clinical research statistics