While I am behind, I am making progress on my dissertation research. I have spent a lot of time during the last week working through the process of creating and analyzing a factorial survey instrument. Unfortunately a lot of this time was in reading and thinking and not actually accomplishing anything. But I am finally testing part of the process. The picture above is from an Excel spreadsheet. This may not seem all that exciting but it demonstrates my success in negotiating the first stage in creating the instrument. This is a pilot version so does not have the same items as the final survey - which reflect some changes from the IPA interview process I mentioned in some earlier posts. This spreadsheet contains 80 randomized combinations of 7 variables. I used R for everything including writing the data to the spreadsheet. I forgot to remove the quotation marks in this version - necessary in R for string variables but easy enough to remove using find and replace (find " and replace with space; can highlight the entire sheet and do this in one transaction). Next thing, after removing those " " , is setting up the merge to create the surveys - because obviously these only show some of the words. In fact, some of these are actually reduced just for the purpose of the pilot. I have already written the code to make the 'final' survey but since I made amendments, I also had to resubmit to the IRB so I cannot give these to anyone yet.
I am going to need to make some adjustments for illogical combinations (if the activity is tennis, you would not do it alone) and take out the "null" lines. I only used "null" in one variable because I need to have some of the wording on the survey form. Some of the other null options just show blanks (or " " at present) so will leave an empty line. But still this is pretty well automated - I will insert a column for ID numbers so I'll know exactly which ones have the null from the spreadsheet. I can also change the illogical combinations using 'find' and 'replace' on the spreadsheet, so I can correct everything before feeding to the surveys. My next goal is to be able to alter the items enough to read essentially this same data (with the scores added) back into R so to do analysis that only requires me to enter a single score rather than codes for each categorical item. I expect it will take me all day today (at least) to work this out.
0 Comments
Leave a Reply. |
AuthorI am Sheryl L. Chatfield, Ph.D, C.T.R.S. I am a member of the faculty in the College of Public Health at Kent State University. I also Co-coordinate the Graduate Certificate in Qualitative Research and I am a member of the Design Innovation Team at Kent State. Archives
February 2024
Categories
|