What is the best approach for creating pre- and post-intervention surveys with randomization?
Hi all,
True Qualtrics novice here. I have a survey-based project with the following features:
* There will be a survey emailed to a group of potential participants to evaluate their current knowledge and practices (pre-intervention survey)
* The respondents to the pre-intervention survey will be randomized into experimental and control groups (is this possible using Qualtrics?)
* The experimental group will receive an educational intervention
* Both the experimental and control groups will receive a post-intervention survey three months later with similar questions
* I would like all respondents to have an anonymous identifier assigned but I would also like to be able to compare pre- and post-intervention surveys from single participants (as well as report the whole of each group). If needed, I can store email addresses of respondents but I would prefer anonymity
Is this all feasible? Can I build this into a single project using a different block for each survey or must it be multiple projects? Can Qualtrics help with randomizing half of the respondents into each group?
I would sincerely appreciate any guidance!
+2Hi,
This is one of the nice things about Qualtrics, is that if you can imagine it, between the documentation and the community, you'll most likely find a good solution. So with that being said, here is a potential solution.
I would recommend using two surveys, one for pre- and one for post-intervention.
In the pre-survey, you would want to create an embedded data field called "group" or "experimental_group" or something that you know what it means with two possible values that indicate your experimental group and your comparison group. (documentation: https://www.qualtrics.com/support/survey-platform/survey-module/survey-flow/standard-elements/embedded-data/) You would then create a randomizer in your survey flow that would have the two embedded data values ("experimental" and "comparison") of the one embedded data field ("group"). (documentation: https://www.qualtrics.com/support/survey-platform/survey-module/survey-flow/standard-elements/randomizer/) So, you would have an embedded data field "group" = "experimental" and underneath that embedded data field you would have all the blocks for the experimental group. You would then do the same for the comparison group. The hierarchy would look something like this:
- Randomizer (evenly present 2 items)
-- Embedded data (group = experimental)
--- blocks for experimental group [* Does this include the intervention?]
-- Embedded data (group = comparison)
--- blocks for comparison group
- Other blocks that either group would respond.
- End survey
Qualtrics has also created documentation on linking repeated-measures surveys. (https://www.qualtrics.com/support/survey-platform/common-use-cases-rc/pulling-data-into-a-second-survey-longitudinal-surveys/) When combined with creating anonymous IDs, I think this will get you on the right path: https://www.qualtrics.com/support/survey-platform/common-use-cases-rc/assigning-randomized-ids-to-respondents/
I know I just supplied a bunch of documentation, but I have been impressed with most of Qualtrics' support pages. If you are still having trouble, the Community is here to help.
Good luck!
Leave a Reply
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.