THERE IS A VERY DETAILED LIST TO FOLLOW, IF YOU DO NOT KNOW OR UNDERSTAND HOW TO READ REGRESSION OUTPUTS on SPSS PLEASE DO NOT ACCEPT.
1. the pages that need to be written are on the FINAL HONORS THESIS DRAFT.
the sections that need to be written/filled are
– abstract (half page)
– data and methodology ( just a section)
– dependent varible
– independent varible
– binary logic regresion of traffic stop data
– results
– conclusion
2. the item titled “lousiville kentucky ky-” is the data set with all of the cases listed. you would need spss to view it.
3. the item listed regressions 01 are the results form the binary logistic regressions ran. use these to write the results
4. the last screen shot are the varibles and what they were coded, use this to discuss the “dependent” snd “independent” varible sections
5. For the results section and to analyze the data the P- valuse would need to be “.01” becssue of the size of the data set. use this below for reference on how to write it.
– Okay…here is what I suggest (along with my friend who is a wizard statistician).
Basically the curse of monster sized data is that trivially small effects can be found to be “significant” with very large sample sizes. You are seeing everything hit significant because of your data size, 110k cases or so.
This is cleary an issue,but you don’t have to toss your data. There are three things you should do however.
1) make the alpha smaller (i.e. the standard for p-value) – .01
But, you can’t just talk about it’s significant like you could with smaller data.
2) You will need to interpret the effect size alongside the hypotheses and make arguments about whether the finding is conceptually meaningful given the effect size.
3) make sure to include a few sentences about how this data set is large so relying on just the p values would be problematic.
———————–
if you have any questions PLEASE REACH OUT.