TheBreakingGoodProject / Dalyell-Projects

0 stars 1 forks source link

Qualtrics Implementation Question #27

Open carolineson opened 3 years ago

carolineson commented 3 years ago

Hi Dr. Scroggie @kym834 and Dr. Golumbic @yaelago !

((I believe you wanted us to post questions on the issues tab—apologies if I'm incorrect!))

I just finished up converting Challenge 1 and 3 onto Qualtrics from Redcap and just had a couple of questions:

1) Just wondering if there was an existing flow chart/map of the Redcap survey—I've just been clicking through each option and trying to capture each individual change. Specifically, does the question asking if the news is related to COVID affect any of the questions for Challenge 1? (It didn't look like it, but just making sure!)

2) Double checking that the goal was to implement the survey exactly as it was on Recap? I converted as much as possible (down to spacing between the questions), but there were a couple of differences just from what the website provides specifically:

3) Kind of in the same vein as above, but specifically regarding the introduction page and the demographic survey part of Redcap. I wasn't 100% sure how to implement the demographic survey in terms of being able to log back in and have the specific individual information that the Redcap survey asked for pre-saved (there is an option called personalised links available, but it doesn't seem very customisable in terms of what information it asks/can save, but it does keep track of which surveys and how many have been taken. It also needs the individual contacts beforehand to send the personalised links, which also does not seem efficient for our purposes). Currently, the only way I can think of is to have it at the beginning of every time you submit a challenge, which seems inefficient and frustrating

4) Also wondering if you had specific requirements/wants for how the questions themselves should be labelled/categorised in terms of how the data will be labelled when exported (not what is visible to the survey-takers)?

5) I also think my survey automatically submitted a couple of nonsensical surveys as I was clicking through for format, oopsies!! And I couldn't seem to find a way to delete the data off of Redcap!! Wanna make sure I'm not spamming the database haha

Thank you for reading through this embarrassingly long question post!!

kym834 commented 3 years ago

Hello Caroline,

Yes! Thanks for posting here. Keep it in the open so that other people can also see what is going on and contribute.

  1. No existing flow chart (logic branching) as such. I can add you to the RedCAP survey as an admin though. Then you would be able to see the branching logic without having to click through all of the options. Let me know if that would be of use. No there the COVID-19 question has no impact on anything else. We were just interested to see if any other news was making it through or if it was all COVID-19 related.
  2. It doesn't have to be exact. We still want to have the same information inputs, so try keep the questions the same. But the look and feel don't have to be exactly the same. You can play around with it and try to work out how to make the most appealing and efficient etc.
    • We can have a look at this issue a little later and try and find out what the display logic is that is affecting this feature. It would be nice to have the back button available for participants who want to use it.
    • Again we can look into the saving and come back option. Double edged sword. It might mean that participants are more likely to finish the survey if they can't save it but it could also be a determent where they don't finished can't save it so never come back to it. How to increase survey completion has been something we've been thinking about for a while now.
    • No worries. If we can't find that feature then we can't do it. All okay.
    • Yes - We basically would like it to keep offering until the participant has put in all the information they have found then to take them to the next stage. Then we aren't limiting them in anyway. Don't know if this is an option though. If not we would like you to think about your own experiences (and others from the data) and decide how many you think is required.
  3. We'll need to have a look at the functionality of Qualtrics. It may be that they have some other option that isn't exactly the same but still allows us to know who has contributed the information. We want to be able to (at some stage) have a place where you can see who has contributed what and for our own analysis to see who is participating so we want to keep collecting this info. You are right it would be frustrating to have to put that info in every time you did a challenge. That was why we had it set up the way we did in RedCAP. Let's discuss this in the workshop next week as well.
  4. Yes - This is something that we can address in the final task for the project that is about database construction. We need it to come out in a manner that it is understandable to anyone who reads it and that requires little work obtain before we share it openly with the public.
  5. No problem. We'll be able to see from the exports that they are nonsensical and can remove them before they get moved to the database :)

Not embarrassing at all! Thanks for all the thorough work with the change over to Qualtrics :)

carolineson commented 3 years ago

@kym834 Perfect!! Thank you so much for your detailed response. @dzheng918 and I will finish it up and be ready to discuss the rest on Monday!

dzheng918 commented 3 years ago

Hi Dr. Scroggie @kym834 and Dr. Golumbic @yaelago !

I've been putting up Challenge 2 and the registration survey, and I have a couple of questions about the registration aspect:

  1. From testing things out on Qualtrics, it seems to me that the registration survey would have to be a separate survey from the rest of the challenges. For an automated email to be sent, the entire survey must be complete as it is a post survey option. I've tried to put it in the one survey by changing the conditions for email triggers, however the email was not sent successfully.

    Screen Shot 2021-09-11 at 11 34 23 am
  2. At the end of the registration survey, would you like a custom “end of survey” message saying something like “An email with your personalised link to the E$$ENTIAL MEDICINE$ challenges has been sent” rather than the default which is:

    Screen Shot 2021-09-11 at 10 43 39 am

    If so, is there a particular message you would prefer?

  3. I have tested the registration survey on a separate trial survey just to see how it works using my own email as the sender. Which email should I use as the sender for the actual registration survey? And would you like to automate a response report to the sender after each survey response or not? This was the result:

    Screen Shot 2021-09-11 at 10 46 57 am

    I used the exact format of the email I was sent by RedCap. For the “head here” I linked to the GitHub page that provided on the RedCap email. I’m still confused about constructing a personalised link so I left that blank. Would a different email format be preferred, or is the original one fine?

  4. I'm still unclear on how to create a personalised link that is able to record a respondent's progress. Like Caroline said, it seems as though respondent's would have to submit the registration survey over and over again each time they do a challenge.

Thank you for your time!

carolineson commented 3 years ago

Hey Daisy/everyone, I was thinking maybe for the end of survey message (if our supervisors don't have something specific in mind), what if we put up links to all the Breaking Good socials? I know the survey itself briefly mentions the Breaking Good blog/website in the Circle of Life challenge, which is really good to look through for more information about Breaking Good, but it might be nice to add in the links to the Twitter and Facebook and maybe even the Github since there seems to be a lot more activity on those channels for the people who want to know more!

dzheng918 commented 3 years ago

Oops sorry, accidentally closed the issue there. That sounds like a great idea @carolineson , I'll get onto that soon.

carolineson commented 3 years ago

Hey @dzheng918 I found some videos that might be helpful for the demographics/introduction survey!! Not 100% what we're trying to do, but thinking we can adapt something similar! https://www.youtube.com/watch?v=v-qIls9GEpY https://www.youtube.com/watch?v=0hoHRm6VYKs

I've also figured out the dropdown menu on Qualtrics if you want to send/tell me what you were trying to implement!

dzheng918 commented 3 years ago

Hi @kym834 @yaelago , I was wondering if you would be able to give permission to distribute the survey "E$$ENTIAL MEDICINE$ - Dalyell Showcase" survey. I noticed that I have been granted permission to the distributions tab for the "E$$ENTIAL MEDICINE$ - Registration" survey but not the main survey. It'd be great to have access to the distributions tab as I need to customise the distribution link to embed data from the registration survey, thanks!

kym834 commented 3 years ago

@dzheng918, done. You should now be able to see the distributions tab in the EM survey as well.

It goes without saying but please only distribute for testing purposes until we have finalised the surveys :)

dzheng918 commented 3 years ago

Yes of course @kym834

Update on the registration survey

I tried many different features in the last week and none of them had worked including personalised links, unique IDs and authenticators. This morning I called Qualtrics Support and they said that:

Accordingly, that would mean we wouldn't be able to use a personalised link or unique code. As an alternative, I provided an anonymous link in the email to allow for multiple attempts, however I have stored the email of the respondent collected from the registration survey as External Data Reference, and have embedded the data in the main survey.

From the Data and Analysis tab on the main survey, we are able to identify the respondent by email using the External Data Reference column as such in the image below:

Screen Shot 2021-09-24 at 12 17 02 pm

I'm currently unable to combine the responses from the same respondent (i.e. same email address) so that may lead to some inefficiencies with data analysis. As of now, that's the best solution I can think of, but I'll definitely keep looking. It'd be great if someone could run the whole thing and test it out, just because I've run it so many times that my brain's a little confused 🤦‍♀️.