Welcome to the LimeSurvey Community Forum

Ask the community, share ideas, and connect with other LimeSurvey users!

Prefill questions/answers with external Data. E.G. SAP HCM

  • Philip_A
  • Philip_A's Avatar Topic Author
  • Offline
  • New Member
  • New Member
More
3 weeks 5 days ago #267399 by Philip_A
Please help us help you and fill where relevant: 
Your LimeSurvey version: Version 3.28.56+230404 Community Edition
Own server or LimeSurvey hosting: Hosting
Survey theme/template:
==================
Hello everyone,

I would like to know if it is possible to somehow infuse external data in my survey to prefill some questions. In this special case I would like to aid the user by prefilling the region where he/she comes from and perhaps the professional role. We work with SAP HCM at my company. The survey is planned to have a multiple choice question in the beginning or two, asking for these data, otherwhise.
My colleague said that our data team might be able to generate a link with the data out of HCM.. but how this should help is an absolute black box to me.

Does anyone of you have a clue how to realize that or can lead me to additional information? I have searched the help articles (only found something like the SGQA identifier, which doesn't seem to work that way or the RemoteControl2, which is a bit above my level of programming) and the forum but I couldn't find any fitting examples.

Thank you very much in advance and have a great week
Philip

Please Log in to join the conversation.

  • Joffm
  • Joffm's Avatar
  • Offline
  • LimeSurvey Community Team
  • LimeSurvey Community Team
More
3 weeks 5 days ago - 3 weeks 5 days ago #267414 by Joffm
Hi, I suppose it is a token based survey.
So you may use additional attributes to store these informations in the participant table.
In your survey use ExpressionScript  to prefill the questions.

Joffm

Volunteers are not paid.
Not because they are worthless, but because they are priceless
Last edit: 3 weeks 5 days ago by Joffm.

Please Log in to join the conversation.

  • holch
  • holch's Avatar
  • Offline
  • LimeSurvey Community Team
  • LimeSurvey Community Team
More
3 weeks 5 days ago #267429 by holch
There are two ways you could do this:

1. Joffm's suggestion, which I think is the cleanest and best solution: Export the data from the database and include it into the token/participant table as custom attributes (you can upload the participants via CSV file). This way you do not even have to pipe the information into the survey if you don't want to (But of course you can). Attributes from the token/participant table can be used in the survey via Expression script, just like Joffm has outlined.

2. As your colleagues suggested, you could add this information to the survey URL as variables and read these variables into questions. But I don't think that this solution is as clean and secure as the first option.

As you seem to know your respondents and thus will use a token based survey you will have to create a participant / token table anyway and by just adding a couple of custom attributes you can have all the information you need.

I answer at the LimeSurvey forum in my spare time, I'm not a LimeSurvey GmbH employee.
No support via private message.

Please Log in to join the conversation.

  • Philip_A
  • Philip_A's Avatar Topic Author
  • Offline
  • New Member
  • New Member
More
3 weeks 21 hours ago #267531 by Philip_A
Good morning holch and Joffm,

thank you very much for your fast replies! I just read about the token approach and saw that the survey can't be anonymous then, if I am right.
help.limesurvey.org/portal/en/kb/articles/information-from-survey-participants-table

But we have to do this weird stretch of - on one side having the data and it should be added by information, like the site they are working at - but at the same time the survey should be anonymous, when we get the results. So as far as I've read, the token-approach will not be applicable here, or did I read it wrong? I know this sounds weird, but as we are from Germany our data protection law is quite strong.

I've never used a tracked survey, we always just send links to anonymous ones. So perhaps I don't even see who has replied to the survey anywhere in the admin section after the participants have answered? Then it might be applicable.

Thank you for your wisdom and help
Best regards and have a nice week
Philip

Please Log in to join the conversation.

  • holch
  • holch's Avatar
  • Offline
  • LimeSurvey Community Team
  • LimeSurvey Community Team
More
3 weeks 14 hours ago #267540 by holch
As we say in German "If you shower you'll get wet".

So whenever you pass data from outside to a survey, there is some kind of possibility to breach total annonymity.

So if you add data to the survey link, you could always pass an ID that can identify the respondent. So the respondent has to trust you, that you don't do this. You are basically using individualized links, just like you would do with tokens.

Same for the token based approach. It is important to analyze data anonymously. As professional market researchers, Joffm and I have done "anonymous" surveys for years. However, we often need to be able to know who has participated (e.g. for paying incentives, include them into a prize draw, recontact them for an additional phase, quality control, etc.).

Yet, as we adhere to the ESOMAR code of conduct, we do NOT connect personal data with response data. We seperate those for analysis. So while we COULD connect them, we don't do it for the analysis.

Now, that being said, token based surveys have an anonymous mode, where the token is not included into the response table. There is a huge discussion going on, because in point of few the anonymous mode isn't that anonymous, because it allows to pass variables from the token table to the response table, just like you would pass this information on via the URL.

On the other hand, you really need to consider how anonymous a survey can be, where you know the the region and the professional role. Because depending on your database size this can already be enough to know to be able to find out who the respondent is. E.g. if you only have a few respondents from the same region and the roles are different, etc. But we can't help you there, as we do not know your sample.

In any way, I think "ethical" anonymity is probably overall stronger than "technical" anonymity, because later generally has flaws in one or the other way, while if your team is commited to the ethics of anonymity, even if technically you COULD breach anonymity, in practice you won't do it.

Read a bit about Survey participants in the manual. I don't really know those help pages and how useful they are. I always refer to the manual.
www.limesurvey.org/manual/Survey_participants
www.limesurvey.org/manual/manual/Partici...Anonymized_responses

So you can see that the data from the participant table is disconnected from the response table. But there is a loophole (which I harshly critize because it makes the anonymous mode not so anonymous), but in your case can be helpful.

You cann pass data from the token table to your survey and store it there with something like {TOKEN:ATTRIBUTE_1} saved to a hidden question.

Here you can read about the discussion of this "feature":
bugs.limesurvey.org/view.php?id=18253

I answer at the LimeSurvey forum in my spare time, I'm not a LimeSurvey GmbH employee.
No support via private message.

Please Log in to join the conversation.

  • holch
  • holch's Avatar
  • Offline
  • LimeSurvey Community Team
  • LimeSurvey Community Team
More
3 weeks 14 hours ago #267542 by holch
By the way, I have just seen this:
Code:
Version 3.28.56+230404 Community Edition

You are running a 1,5 year old version of Limesurvey, who's branch is EOL for a while now. This means your installation hasn't received any update, bug fix or security fix since 04.04.2023.

Not sure if you should worry much about data protection and anonymity... :-)

I am sure no judge will consider this "state of the art software" in terms of GDPR/DSVGO in case of a data breach.

I answer at the LimeSurvey forum in my spare time, I'm not a LimeSurvey GmbH employee.
No support via private message.

Please Log in to join the conversation.

  • Joffm
  • Joffm's Avatar
  • Offline
  • LimeSurvey Community Team
  • LimeSurvey Community Team
More
3 weeks 11 hours ago - 2 weeks 6 days ago #267545 by Joffm
Short addition, this behavior of LimeSurvey, which is also shared by many other commercial products (which do not have this "anonymous" mode), is fully in line with the GDPR.

Long before the regulation was passed, these very points were discussed in numerous discussions between the German market and opinion research associations (ADM, BVM, DGOF. ASI) and the legislative bodies, so that this now fully complies with the GDPR.

The basic requirement is, of course, that the researcher acts in accordance with the professional rules and submits to the ESOMAR code of conduct, in particular the German declaration of the ADM.

You may read  here
[url] www.adm-ev.de/standards-richtlinien/ [/url]

Joffm

Volunteers are not paid.
Not because they are worthless, but because they are priceless
Last edit: 2 weeks 6 days ago by Joffm.
The following user(s) said Thank You: Philip_A

Please Log in to join the conversation.

  • holch
  • holch's Avatar
  • Offline
  • LimeSurvey Community Team
  • LimeSurvey Community Team
More
3 weeks 11 hours ago #267546 by holch

is fully in line with the GDPR.


While I agree that this is not an issue in terms of GDPR (the standard text presented by Limesurvey might be though, as it suggest a different level of anonyity than it can provide, but that is a different topic - my main concern is less with the "feature" itself, but rather how it is presented), but sometimes GDPR might be not far enough for the respondent.

While I am not overly concerned with market research surveys, other surveys e.g. internal employee satisfaction surveys, etc. might be a different beast.

While I am not worried too much if someone at a market research institute knows that I prefer color X over color Y, I might not be fully honest with my responses in a employee survey if I fear that someone might be able to identify my answers given and I might have to fear repurcusions from my superiors.

So we as market researchers are usually another level of protection, as we are generally independent and do not pass on personal data of respondents to the client. Therefore we tend to see no major issues here.

But when looking at internal surveys, this independent instance is missing. Therefore I generally recommend to anyone to do surveys that might be "controversial" with an indepentend and trustworthy third party, to reduce the fear of respondents in terms of privacy.

I answer at the LimeSurvey forum in my spare time, I'm not a LimeSurvey GmbH employee.
No support via private message.

The following user(s) said Thank You: Philip_A

Please Log in to join the conversation.

  • Philip_A
  • Philip_A's Avatar Topic Author
  • Offline
  • New Member
  • New Member
More
2 weeks 6 days ago #267554 by Philip_A
Hehehe yeah, "If you shower you'll get wet". True ;)

Thank you very much for your elaborate answers again, you two. I agree to what you say and with the differentiation between "ethical" anonymity and "technical" anonymity. That's on spot. I will have to get in the loop with our data security team.

And thanks as well for the hint regarding the outdated version. I will ask our IT colleagues to have a look at that. (Big company processes  ).

Keep the amazing work up, you helped me and us a lot here and I highly appreciate it! The internet is a better place with you <3

Best regards and have a wonderful week
Philip

Please Log in to join the conversation.

Moderators: tpartnerholch

Lime-years ahead

Online-surveys for every purse and purpose