We have done this, to some extent, with a few of our surveys.



Could you not allow your respondent to go through the survey and depending upon their responses, provide one of two end of survey messages, "Sorry, you don't qualify." (fail statement) or "Congrats, you qualify!" (pass statement) ?



For example, if I had a 5 question survey:

1. What is your gender? Male or Female

2. Question 2

3. Question 3

4. Question 4

5. Question 5



At the end of the 5 question survey you have 2 separate blocks.

Each block has a Timing question and a Descriptive Text question. One block contains the "Pass" criteria. The second block contains the "Fail" criteria.



For example, the respondent gets the Fail block if they select Male on question 1. The Timing question records if they saw that screen for that block. If the respondent selects Female, they get the Pass block and the Timing question for that block records if the respondent saw the screen for that block. In either case, the respondent then proceeds to the end of the survey. In either case you have collected full data for each respondent, each just getting a different pass/fail statement (using Survey Flow branching logic).

---

@C_Bohn 's answers is spot-on. I would only add that perhaps you want to consider a grouping variable instead of a Timing questionnaire to check who saw what. For example, if they saw the "Pass" block, you can assign a "1" to a variable "group" or "0" if they saw the "Fail" block. In this case you don't even need that because you know that if gender was Male, then they did not see the Female blocks and vice versa for Females.



The group variable is usually more effective in experimental designs.



Hope that helps.

---