I have 3 data sets (eg Scores from three different Universities) all mixed up in a single excel sheet and with different number of scores. Eg Oxford sample with 4000 scores, Cambridge with 3200 scores, and Portsmouth with 1800 scores and I want to extract and trim them so that the sample sizes match the university with the lowest number of scores (in this case 1800), what would the macros look like? Please note the spreadsheet has 2 columns i.e. University and Score. If someone can also answer the same question for SPSS I will also be grateful.
Following on from David Eugene Booth 's comment, many statistical procedure allow you to weight the subsamples to take into account different sample sizes and different university populations.
Probably the most simple solution is to paste =RAND() into a new column for all your participants within the Excel sheets. Sort by this new column, filled with randomly-generated numbers, and take the top 1800 of each data-set. If you would like a permanent record of the randomization, copy & paste values over the random number column before sorting. Just as a caveat, if you're planning an ANOVA to compare the three groups, you're probably okay with unequal sample sizes so long as there's not a big homogeneity of variance violation (and, if so, a Welch or Brown-Forsythe F test might help without dropping data). Hope this helps, Blessing. ~ Kevin
I'll take a little bit different tack and ask why do you want to do that?
Best wishes, David Booth
Following on from David Eugene Booth 's comment, many statistical procedure allow you to weight the subsamples to take into account different sample sizes and different university populations.
Why do you want to do that?
Anyway, your EQUAL random samples will be biased if not weighted based on different sample sizes.
Best regards, Roberto
- Badges
- Science topic