Project "Best_FDM" - Questions on the Corona pandemic, religion, and household income (English Version)

This question is about being able to evaluate groups in the population with, for example, high, middle or low income, Therefore, we would like to know: What is the total average monthly net income of your household?
We assure you that your answer will not be evaluated in connection with your name.

[Bei dieser Frage geht es darum, Gruppen in der Bevölkerung mit z. B. hohem, mittlerem oder niedrigem Einkommen auswerten zu können. Daher möchten wir gerne wissen: Wie hoch ist das durchschnittliche monatliche Nettoeinkommen Ihres Haushalts insgesamt? Wir versichern Ihnen, dass Ihre Antwort nicht in Verbindung mit Ihrem Namen ausgewertet wird.]

(INT: Read out the explanation!)
Average monthly net income of your household means the sum resulting from wages, salary, income from self-employment, pension or retirement. Please also add income from public aid, income from renting and leasing, assets, housing allowance, child benefit and other income and then deduct taxes, social security contributions and (if you are self-employed) business expenses.

[(INT: Erläuterung vorlesen!)
Unter durchschnittlichem monatlichem Nettoeinkommen Ihres Haushalts ist die Summe zu verstehen, die sich aus Lohn, Gehalt, Einkommen aus selbstständiger Tätigkeit, Rente oder Pension ergibt. Rechnen Sie bitte auch die Einkünfte aus öffentlichen Beihilfen, Einkommen aus Vermietung und Verpachtung, Vermögen, Wohngeld, Kindergeld und sonstige Einkünfte hinzu und ziehen Sie dann Steuern, Sozialversicherungsbeiträge und (wenn Sie selbständig sind) Betriebsausgaben ab.]

_____________ Euro
Don't know [Weiß nicht]
No answer [Keine Angabe]

We recommend adding a note to the question that an exact amount does not necessarily have to be given. A possible wording could be integrated into the question text as follows:

Therefore, we would like to know: How much [...] of your budget in total? If you do not know the exact amount, please estimate. We assure you [...]

In addition, (as currently planned in the Best_FDM project) in case of non-response to the question (item non-response) or if respondents select the categories "Don't know" or "No answer", a follow-up question with closed answer format should be asked. For this follow-up question, the number of answer options should be kept as low as possible, on the one hand to keep the cognitive effort on the part of the respondents low, and on the other hand to reduce the sensitivity of the question. If necessary, the follow-up question could be formulated in several variants with different granularity, so that questionnaire developers can decide between variants with rather differentiated or rather rough answer categories, depending on their analysis interests.

Cognitive Techniques:

Specific Probing, Process Probing

Intention of the question:
Question 8 is designed to capture respondents’ monthly net household income.

Findings:
As shown in Table 6, ten of the twelve test subjects named a monthly amount in euros. The remaining two respondents did not want to answer the question (no answer) or could not answer it (don’t know).
The respondent who refused to answer justified this by saying that stating the income was too private for her and that she would only answer the question - if at all - as a closed question with very rough answer categories:

• TP: "This question is too private for me, I wouldn't answer it. [...] If it was broader, I would find it better. If you don't have to define it exactly, then I would answer the question."
• INT: "What income ranges would you find most useful here?"
• TP: "For example, below 100,000 euros or above 100,000 euros."

The respondent who could not answer the question justified this by saying that she did not know exactly the income of the other household members and was not sure who to count as a household member and who not:
"Now I have to ask a quick question about that. My father has a girlfriend who doesn't live here but is here very often. Does that count then? [...] I would simply include her now because she is already here often. But she has her own house. [...] I know my income, but not my father's. I really don't know at all how much he earns. I could only make a rough estimate. But then there's the question of whether the girlfriend counts as well. Because then I would say I don't know. [...] I would state 'don't know' here now."

1. How do the test persons go about answering? What income do they consider?

Three test persons named income ranges instead of exact amounts. The reason for this was due to monthly fluctuating amounts due to the subjects' self-employment and/or individual household members:

• "It's difficult, of course, because my mother is self-employed. So am I. It's very variable. [...] I'll try to estimate that, I can only give an approximation."
• "I can't say exactly, just roughly. It's my husband's net income for one, then my self-employment income and our rental income."
• "So, an exact overview of my husband's finances I don't have. I guess. [...] I added my net income and rental income from a condominium. Then I also estimated what my husband earns from his self-employment and what he gets for his property."

The remaining seven subjects who reported their income gave a (more or less) exact amount. They considered only net salaries and government benefits (unemployment, child support) and stated that they had no other sources of income:

• "With child benefit, exactly 1,076 euros. That's our Hartz 4 plus our child benefit."
• "I sort of calculated my wages and my friend's unemployment benefits."
• "I only thought of my monthly salary, since I have no other sources of income. I also don't receive benefits or anything like that."
• "My partner's salary and my salary. No other income at all. So only from dependent work."

2. Is the definition of monthly net income understandable (especially in CATI mode)?

All but one of the test persons indicated that the explanation of what is meant by average monthly net income was "rather" or "very understandable." Subject 05 rated it as "rather incomprehensible," but noted that its content was important to answering the question:
"That was a lot of things that were mentioned. But clearly, you have to mention that, of course, so that people consider all [relevant] things."

Occasionally, interviewers reported being interrupted by respondents' spontaneous answers while the question was being read aloud. In these cases, the explanation was read aloud again or after the respondents' answers were noted, and in no case did this lead to a change in the spontaneous answer. The interview process was also not negatively affected by the interruption.

3. Do the test persons prefer a closed question?

Finally, the test subjects were asked whether they would prefer to answer a closed question with different income ranges instead of the open-ended question. The majority of test subjects (n = 9) preferred a closed question and justified this mainly on the grounds that it would make it easier to answer, especially if the amounts varied from month to month:

• "I find the ranges useful. It's easier to sort yourself in there. Here [with the open-ended question] you have to be more precise about the amount."
• "I find it easier to have different answer categories."
• "Sure, I named a range. With [the open-ended question], you don't have one hundred percent accurate answers either. [...] I find it better with answer categories, because then you can name an area. It's easier."
• "I think the [closed question] is better, where you can put in a category. Because sometimes it does vary."

Social demographics

Net household income