Hello community,
I hope everyone is doing well during this weird time.
I have already read kbobk092, drfreeman and kip_ferguson's question and it gave me a rough idea of what I have to do but I do not feel like their posts have answered my question or solved my issue to my fullest satisfaction, or maybe I did not completely understand the process? Anyway, here it goes:
As part of my Master Thesis, I am doing a correlational research on relationship satisfaction & optimism/pessimism for which I will invite couples to participate. I have sent my ethics application to the board but it was rejected because of how I was planning to pair the participants and identify as a couple, there was a privacy issue about sharing/providing email addresses without consent for identification.
My original idea was for participants to create a code word of their choice to simplify the identification process later on, e.g. partner A chooses the word "Australia2016" and then partner B would somehow (via a trigger?) receive the survey and put in this code word somewhere so that I can then later pair both participants according to their code word. I think the idea was great, because it meant complete anonymity, but my supervisor said this was too much burden for the participant.....
As this is a global research, I will set up the questionnaire in Qualtrics and post it on Social Media etc. to recruit couples. I have read about creating individual IDs which I can do when I, of course, message my friends and ask for help, but given that I need to find more than 300+ couples, I will not be able to do that for everyone. I do not have much control over who answers these questionnaires once I post the ads, and do not think I will have the time to manually go on a search for partner B within my data set after partner A has answered the survey. I am trying to find a time efficient way that allows me to simplify the process.
How can I pair participants without emailing every single partner A, and make sure it can be achieved by posting the advertisement only so I don't have to follow up hundreds of people?
What I would like to achieve: Partner A clicks on the survey link -> completes survey -> nominates partner -> survey gets sent to partner B -> they fill out the questionnaire -> Qualtrics shows me paired participants / couples
Any help would be very much appreciated.
Thank you in advance, I look forward to reading the responses.
Best regards from Australia!
Pairing participants for couple research
I don't know how to address the privacy issue you allude to, but if you are triggering the second survey using the 'Distribute survey' task (Actions tab) you can add the Response ID (and all survey responses too) from survey #1 as embedded data fields in survey #2.
https://www.qualtrics.com/support/survey-platform/actions-module/xm-directory-task/#SendingEmails
Hey bmcgrath,
Thank you for your response.
Generally, this is a good idea, but it does not work because of the privacy issue. The ethics board said I cannot use or store Partner B's email address without their consent but it's tricky to ask for it in the first place. Plus, I don't have anyone's email address in the first place.
OK, I see. If you can't email the second participant, I think you'll have to present a url to the first participant for them to give to the second on their own? You can present in a static content question the anonymous url, but with a parameter added using piped text. I don't know if there's a way to pipe in the response ID that Qualtrics generates (there might be), but you could instead use the old school random number generator at the very least.
- In the survey flow, create an Embedded Data field where some id field is set equal to a random number.
- At the end of the survey, present a url to the participant to provide to the second participant using yoursurveyurl.qualtrics.com?ref={idfield} just by typing in the parameter field and using piped text for the id.
- You know two surveys are linked if 'id' in survey one matches 'ref' in survey two. You could also create a third embedded data equal to either of those, so you only have to use the one field.
- Maybe add display logic to the instructions with the second link so it only displays if 'ref' is empty.
Again, there might be a better way than the random number generator, but that type of set up could work as an approach.
Leave a Reply
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.