GESIS Pretest Database: Documentation Questions Details - Project „LoneCovid“ – Questions on social networks and their evolution throughout the COVID-19 pandemic (English Version)

General Information:

*Note: The item was tested in German. This is an English translation of the original German wording.*

Introduction:

In the following, we'll show you all the possible connections in your social circle.

[Im Folgenden werden wir Ihnen alle möglichen Verbindungen in Ihrem sozialen Umfeld zeigen.]

Question Text:

How likely is it that these two people will make plans to do something together or talk to each other without your involvement?

[Wie wahrscheinlich ist es, dass sich die beiden Personen unabhängig von Ihnen verabreden, um etwas miteinander zu unternehmen oder um sich zu unterhalten?]

Answer Categories:

Not likely [Nicht wahrscheinlich]
Rather likely [Eher wahrscheinlich]
Very likely [Sehr wahrscheinlich]

Recommendations:

The test persons interpreted the question in the same way and focused primarily on whether the pairs of people shown knew each other at all and, if so, whether they liked each other and would make plans to meet (rather than, for example, just running into each other by chance). The only difficulty in answering the question arose from the three-point response scale, which could not be clearly displayed using the NetworkCanvas software. This problem could be solved, for example, by labeling the fourth response option “These people do not know each other.” This would also make it possible in the data analysis to distinguish between pairs of people who do not know each other at all (and therefore cannot go on a date) and those who do know each other but do not want to go on a date.

Cognitive Techniques:

Process Probing, Difficulty Probing

Findings for Question:

Findings:
How do the test persons approach answering the question?
For Question 7, all possible pairs of people were displayed from the six individuals selected in Question 2. For the vast majority (72.7%) of the pairs displayed, the test persons stated that it was “not likely” that these people would arrange to meet with one another independently of themselves. This was mostly justified by the fact that the people came from very different circles and either did not know each other at all, or, even if they did know each other, would not specifically arrange to meet because, for example, they might not like each other or did not have each other’s contact information:

“I’ve been wondering if these people even know each other. For example, my son knows my friends by name, but he doesn’t know their phone numbers or addresses and wouldn’t be able to contact them at all. Or my friends don’t know my coworkers, so they don’t meet or talk to each other.” (TP01)
“[I was thinking] that some people don’t know each other at all, or have no connection with each other, or don’t even like each other. For example, my son doesn’t like my friend at all and would never hang out with her.” (TP02)
“I was thinking about the relationship between these two people. [NAME] is my best friend, and [NAME] is my daughter. They probably wouldn’t do anything together. If they ran into each other on the street, they’d talk and get along just fine, but they wouldn’t make a point of getting together.” (TP05)
“Whether people even know each other. A lot of them don’t know each other at all. And how well they get along with each other.” (TP09)

In contrast, pairs of people whom the test participants rated as “rather likely” or “very likely” to go on a date with each other usually came from the same social circles (e.g., family or friends):

“Relatives naturally keep in touch with each other much more often than my friends and my relatives do.” (TP04)
“[NAME] and [NAME] are partners.” (TP05)
“I used ‘very likely’ for people who are partners or belong to the same family.” (TP08)

Were there any difficulties in answering the question?
The software used (NetworkCanvas) did not allow for a three-point response scale for this question type, which is why a fourth response option without text was displayed (see screenshot). One participant (TP09) was confused by this fourth response option and recommended that interviewers point this out during the survey:

TP09: “I see here that the fourth option is just a line. What does that mean?”
INT: “Unfortunately, there's no other way to program it.”
TP09: “I'd explain the line thing before people answer anything.”

In addition, this participant and another test person (TP07) suggested using a four-point response scale with the options “very likely,” “rather likely,” “rather unlikely,” and "very unlikely,” as this would simplify answering the question and thus allow for a fourth, substantive response option:

“At first, I was a little confused by the choice of words. I think ‘very likely’ and ‘rather likely’ are well-chosen, but I often stumbled over ‘not likely’ because I think ‘rather unlikely’ would be better, so that the degrees of likelihood can be compared.” (TP07)
“I would use a four-point scale here, just as before.” (TP09)

With the exception of test person 07, who had difficulty with the three-point response scale, all test persons reported that they found answering the question “rather easy” or “very easy.”

Question Topic:

Society & social affairs/ Relationships

Construct:

Social connectedness