2. General Information: *Note: These items were tested in German. These are English translations of the original German wordings.*
3. Introduction: In the following, we are interested in whether you would agree to your activities and content being stored and evaluated in anonymous form by scientists at individual Internet services.
   [Im Folgenden interessiert uns, ob Sie damit einverstanden wären, dass Ihre Aktivitäten und Inhalte bei einzelnen Internetdiensten von Wissenschaftlern in anonymer Form gespeichert und ausgewertet werden.]
4. Question Text: How likely is it that you would consent to such use of your information in the following services?
   [Wie wahrscheinlich ist es, dass Sie Ihr Einverständnis für eine derartige Nutzung Ihrer Daten bei den folgenden Diensten geben würden?]
6. Answer Categories

   Very unlikely [Sehr unwahrscheinlich]

   Rather unlikely [Eher unwahrscheinlich]

   Undecided [Unentschlossen]

   Rather likely [Eher wahrscheinlich]

   Very likely [Sehr wahrscheinlich]

   I do not use it [Ich benutze es nicht]

   • Findings for Answer Categories
   • Recommendations for Answer Categories

---

7. Findings/Recommendations for Multi-Item Scale:

   • Findings for Multi-Item Scale
   • Recommendations for Multi-Item Scale

   1. Findings for Multi-Item Scale: Do the test persons have problems understanding and/or answering the question?
      
      Three test persons (TP 04, 09, 10) answered the question with "very unlikely" for all four Internet services, although they did not use them or any of them (TP 04, Twitter) (and consequently could or should have selected "I do not use"). However, the answers of these test persons were not necessarily incorrect, due to the fact that all three argued that they would not give their consent even if they used the services:
      • "I'm generally more the type who doesn't upload anything of myself on Facebook, Instagram and so on. I'm there simply to look. But I still find it a bit unpleasant when all the things on Facebook are evaluated somehow, for example, that you also get a lot of personal advertising. That's why I wouldn't give my consent for the scientists to get this data. [...] I don't use Twitter, but if I did, I wouldn't give my consent for the data to be passed on here either.” (TP 04)
      • "So, actually in general, I don't give anything away, I'm not for it.” (TP 09)
      • "I don't use them all, for privacy reasons.” (TP 10)
      How willing are the test persons to share their (Facebook) data?
      
      Three test persons stated that they did not use Facebook. Of the remaining seven test persons, the majority (n = 5) considered it "rather" or "very unlikely" that they would share their Facebook data. The reasons given were the privacy of the content on Facebook (TP 01), a general rejection of sharing personal data (TP 04), a lack of information about what exactly is done with the data (TP 05), and a general distrust of the storage and use of social media data (TP 09, 10):
      • "I think that very private things are discussed on Facebook, that this is a network in which people tend to regulate or discuss private things. And I personally think that these are not topics that concern research or evaluation.” (TP 01)
      • "Because somehow, in the background, you don't know how it all interacts, what really happens with this data. [...] I would rather not make my data available to anyone in such a direct and general way.” (TP 05)
      Two test persons justified their indecisiveness or tendency to be willing to share their data by saying that they were resigned due to the fact that their data would already be evaluated anyway:
      • "I don't really know. I basically have no problem with it at all, my data is all tapped anyway. So really, it doesn't matter. I have nothing to hide.” (TP 02; Answer: “rather likely”)
      • "In general, I would say 'unlikely'. But all these services and end devices are so tightly networked and collect so much data. That's why I now think it doesn't matter.” (TP 03; Answer: “undecided”)
      In general, the willingness to share social media data for the test persons did not differ between the services mentioned. With the exception of test person 01, all persons gave identical answers for the services they use (e.g. TP 05: “That's unlikely for all four 'more likely than not‘.” ). Test person 01 thought it "very unlikely" to share her Facebook data, "somewhat unlikely" to share her Instagram and WhatsApp data, and "somewhat likely" to share her Twitter data:
      • "So with Instagram and WhatsApp, it's more or less the same as with Facebook, that I think it's very private content. And with Twitter, I don't think that's really the case. Personally, I only use Twitter to follow certain news agencies and certain politicians. And that's why I wouldn't have any problems with it now.”
      • INT: “But you still gave a different answer on Instagram and WhatsApp than on Facebook, so you made another gradation. So do you see a difference there as well?”
      • "I see a bit of a difference there. With WhatsApp in particular, I think that if you want to evaluate something, you'd rather measure the usage times, I'd say, or the frequency in general, which I personally don't have that much of a problem with. So, I think with WhatsApp you can analyze less the content. And with Instagram, I think it's the same thing, that it's not so much about the content, but rather about the frequency or the use in general.”
      What do the test persons understand by "scientists"? External scientists or also internal ones at the services mentioned?
      
      The term "scientist" was interpreted very heterogeneously by the test persons. While the majority of the test persons associated it with external, independent scientists at universities or institutes (SP 02, 03, 04, 06, 07, 08), others thought primarily of persons employed by the Internet services mentioned (SP 09, 10) or of employees in (advertising) agencies (SP 01, 05) who evaluate corresponding data:
      • "I assume that these are people who evaluate this data for some marketing agencies or for other, I would say, election campaigns or similar. That these are people who evaluate these data and then sell them on or evaluate them for more commercial purposes.” (TP 01)
      • "I was thinking of external scientists, at the university for example.” (TP 02)
      • "Yes, scientists are good, but I don't know what kind of scientists they are. [...] What it says, Facebook, Instagram, Twitter, WhatsApp. These are all companies that I don't trust. Because their servers are in the USA, more or less, so anyone can access them. I have nothing to hide, but it's kind of unpleasant that anyone can then access this data, or just the scientists. If someone writes to me privately or you ask me, I have no problem with that. But I don't have a problem with everyone being able to access it as they please.” (TP 09)
      Due to the low variance in the test persons' answers to question 15, no statement can be made as to whether the interpretation of the term "scientist" had an influence on the answer to the question. Test persons 02 and 03, who would be most willing to share their data, mainly thought of external scientists, but they did not justify their answers with a perceived integrity of scientists, but rather with the fact that these data were already stored and analyzed anyway and that it therefore made no difference whether scientists additionally analyzed them.

   1. Recommendations for Multi-Item Scale: Question: If the question is intended to capture the general willingness to share social media data, regardless of whether respondents currently use the individual Internet services, the question can be left as it is. Otherwise, we recommend that this question be preceded by a filter question asking whether the respondents use these four Internet services. Question 15 should then be asked only for those services that are used by the individual respondents.
      
      Response format: No changes recommended. When using an upstream filter question, the answer category "I do not use" should be removed.

---

1. Cognitive Techniques: Comprehension Probe, Category Selection Probe, Specific Probe

Items