I am trying to perform the cell-weighting procedure on SPSS, but I am not familiar with how this is done. I understand cell-weighting in theory but I need to apply it through SPSS. Assume that I have the actual population distributions.
Hello Eltion,
The answer depends on how you've entered the data. If you enter the data at the cell level, then just add one more variable to the data set, representing the intended target weight. Then, from main menu bar, choose Data / Weight cases... and complete the subsequent dialog box by noting the name of the added (weighting) variable.
However, not all procedures in SPSS work with the weight cases command. Check the help menu for more specifics.
Good luck with your work.
Hi David,
Perhaps I did not express my questions clearly.
The data is entered at the cell level, and I know how to apply the weights. In this particular case, my concern is with calculating the cell weights. Any advice on that?
Hello Eltion,
If you know the population values, then the cell weights should be proportional to those values. Let's take a simple example of a population with 3 subgroups:
Subgroup A, 500 cases
Subgroup B, 700 cases
Subgroup C, 2000 cases
The weights would be any three numbers in proportion to 500:700:2000. For example, if you wanted the cell values to represent the entire population, then use those specific values. If you wanted cell values to represent a sample of 1/3 of the total population, then weights of 167/233/667 would do the trick (and so on).
Good luck with your work.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS