Good after noon.
Am enquiring, why do I get all data for filter questions in the next analysis if I applied skips. for instance in this question, I expect only analysis for those who said yes in the next question but even data for those who said no is included. How do I manage this when using form builder. Thanks
You can test the skip in form preview or with the Online validator. You can also check it in the table view (or in Excel) using there a filter (“no”) on the first question.
1 Like
@higenyi, if you look at table 1, you have 6 cases for Yes. Now going to table 2, you have 3 cases at the top that has no label. Which refers to as a skip cases. Now if you come down, you will see 1+1+1+1+2 which is 6. So I would say the skip has worked well.
1 Like
That is true however, I expect to produce statistics at 100% of only those who said yes. But in this case the percentage analysis includes those who responded no. I want to filter out those who said no. Thanks
@higenyi, it seems like you are using SPSS for analysis. You may need to filter before running the table to get the desired output.
Thanks for the feedback
You can define the ‘’ empty string as missing value in SPSS, for all tabulations. See official SPSS doumentation or here SPSS Guide: Missing Values.
1 Like
The reason why you are getting all data for filter questions in the next analysis even though you applied skips is because skips are only applied when respondents click the “Next” button. If a respondent clicks the “Back” button or exits the survey early, they will still be included in the analysis for all questions, even those that they skipped.
Welcome @warnersmith323,
Do you refer to Collect or to Enketo.
quote=“LN, post:2, topic:43309”]
Going backwards and jumping do not trigger recomputation in general. I believe there is an exception for filtered choice lists, though, which are recomputed when displayed.
[/quote]
See for more info:
But any data not “relevant” will at least be cleaned off on finalisation/submission. Before, as far as I know, data no more relevant normally become hidden and are no more included in referencing calculations.
Maybe @Kal_Lam and @Xiphware or @LN can you add more info, please, when and how relevant is treated in Collect and in Enketo.
