Your LimeSurvey version: Version 5.6.23
Own server or LimeSurvey hosting: LimeSurvey Cloud
Survey theme/template: fruity
==================
I am to create an online survey in Limesurvey that is already distributed offline to households. The goal is to get those who have not yet submitted the survey offline to participate online.

However, I would like to avoid duplication. Therefore, I would like to exclude those who have already taken the survey offline based on their home address (The only personal data collected offline).

I imagine that the first thing I do is ask for the address and match it with the already known addresses. If the address is part of the "offline address list", the survey should be closed directly. Optimally, this address will not appear later in the results (saves me work later but is optional).

But how, if at all, can I implement this?

Hi,
1. How and where do you want to store your "offline list"

I find it quite difficult to compare addresses. if it would be a single word / code, ok. But addresses can often be written in very different ways and thus an automatic check is almost impossible, in my opinion.

Thank you for your replies.

How and where do you want to store your "offline list

At the moment, the offline survey results are manually entered into an Excel spreadsheet. So here are also all addresses of the participants in one column. But I don't know if I can upload a CSV file with all addresses with the cloud version of Limesurvey, for example. Maybe it is also possible to separate them in a huge code snippet with commas and place them. 

I find it quite difficult to compare addresses. if it would be a single word / code, ok. But addresses can often be written in very different ways and thus an automatic check is almost impossible, in my opinion.

I was probably not specific enough. It is about the addresses only from ONE city and the participants also come only from there. That is, the address consists only of street name + house number. The whole thing is very standardized here in Germany, which means that there is actually only one reasonable way to write the address. At least most of the duplications I will hopefully avoid this way. For example, an address has this form: Sportweg 4, Im Grund 9 oder Rosenstraße 34.

At the moment, the offline survey results are manually entered into an Excel spreadsheet.

And here you have 2-3 sources of error:
- Respondent writes the address wrong on paper or has a bad hand writing
- Data entry person reads the address wrong
- Data entry person reads the address right but types it wrong

There is just too many variables that you can't really control, that an automatic control will work reliably.

Being German myself, I can guarantee you, that even with easy addresses, things can still go wrong.

People are lazy, if you have worked on coding and analyzing of open ended questions before, you will probably know. And when answering on a mobile phone, things get even worse.

Sportweg 4 could be "sportweg 4" for example.

Im Grund 9, im grund 9, Im grund 9, etc.

So I personally would not any automatic check. Usually people are not very interested in doing a survey twice. Maybe if there is an incentive, but then I would rather check for duplicates later with a human.

Also, what about addresses that have more than one household? Can you guarantee that all addresses are only for one household?[/quote]

Beside the fact that you can't compare automatically (as @holch, I lived in "Heinrich-Heine-Straße 5"; you can write it short as "H.Heine-Str. 5" and all other variations) you will only invite people who didn't send the paper-pencil questionnaire.

In my opinion the chance that somebody who didn't answer per paper for two weeks or more now go to find it, answer it and answer the online survey is small.
But there might be people who sent the questionnaire the same day you invited for the online survey.
So these people are not in your list.

Langer Rede kurzer Sinn:
,You have to compare manually (what is not a big thing in Excel)

The reason I asked for the storing of your list:
Your LS installation is hosted at Limesurvey GmbH. So I think you have no own webspace to create a small database to store the list.

There might be away to do it.
Buzzword "autocomplete".
Upload your list as csv, ask for the address. If the respondent finds it, it is already used.

Joffm

Okay, I can see the strategic arguments. My other idea was to check for duplicates later with Excel and then delete them manually. If I have it right in my head there is a somewhat easy solution for this.

I just wanted to prevent Aunt Liesel, in her sense of duty, from filling out every survey that comes from her city. But probably the probability is indeed small.

Thanks you two!

I just wanted to prevent Aunt Liesel,

Knowing my Aunt Liesl, she would be to lazy, and she probably wouldn't be able to fill an online questionnaire anyway, so don't worry...