2. General Information: *Note: The items were tested both in English and German.*
   
   Online Probing & Cognitive Interviews
3. Question Text: English version: Are you able to choose or change...
   
   German version: Können Sie sich Folgendes aussuchen bzw. ändern?
5. Answer Categories

   Yes

   No

   Don't know

   Refusal

   • Findings for Answer Categories
   • Recommendations for Answer Categories

---

6. Findings/Recommendations for Multi-Item Scale:

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

   1. Findings for Multi-Item Scale: Online Probing:
      Across all countries, self-employed respondents are more likely to answer that they can choose or change their order of tasks, methods, and also speed of their work. However, the majority of all respondents agree with these items, indicating positive decision latitude.
      In question 18a, Item A and Item B were probed to get a better understanding of how respondents interpret both items and whether the interpretations differ between employees and self-employed.
      
      Cognitive Interviews:
      This question was asked in the cognitive online pretest to assess differences between employed and self-employed. It was decided that these differences and further related issues should be taken up in more depth in the face-to-face cognitive interviews.
      As in the online pretest, most respondents answered positively to all three items, with only two respondents not choosing “yes” for each statement in Germany and three to four in Poland. In almost all cases, these were employed respondents.
      Comprehension and differentiation of items
      The understanding of the second item and the term “methods” was explored. The German translation into “Vorgehen” (literally “procedure”) is rather similar to the translation of the first item “Reihenfolge” (order).
      For both the German and Polish respondents, differences in comprehension don’t arise from employment status as such (i.e. employed versus self-employed, or typical versus atypical employment situation), but from the individual work settings. Associations are commonly related to temporal, organizational or procedural aspects, or referring to order.
      Respondents gave definitions of the term “methods of work” with ease and give in-scope examples:
      • “maintaining competence” (DE03)
      • “determination of the daily routine and rhythm by/for yourself” (DE04)
      • “creating a priority list and executing them according to importance” (DE05)
      However, respondents often confound the dimensions in their explanations. This means that they demonstrate a correct understanding of the term methods, but usually add aspects that refer to the order of their tasks and/or their pace of work:
      • “I can decide which steps are necessary. I can decide how these steps are carried out. And I can decide when these steps are carried out” (DE01)
      • “Well, the job that I do in my company – most people even where I work don’t know how to do that. So that gives me the possibility to choose how I work. I can say, this takes so and so long, because no one can check that” (DE06)
      In sum, “methods of work” is clearly a related concept to order of tasks and also pace of work, with some respondents differentiating these aspects better than others.
      
      Answer Options
      Both in Poland and in Germany, one respondent refused to answer item B, saying that they couldn’t decide. These respondents wanted to have an answer option “partially” or “half/half”. These respondents explain why they would need an “in-between” category:
      • “Sometimes it is imposed by employer, the predefined standard, conditions and I cannot change that, but sometimes I can decide on what I believe is better for the group and adjust to conditions we have” (PL02)
      • “Sometimes it is imposed by employer, the predefined standard, conditions and I cannot change that, but sometimes I can decide on what I believe is better for the group and adjust to conditions we have” (PL02)
      In Poland, nearly half the respondents spontaneously reacted to the questions saying it is hard to give a categorical, binary answer, as situations vary. Nine respondents in Poland (PL02, PL03, PL05, PL06, PL08, PL09, PL11, PL14) explicitly state that a scale should be used to give a differentiated answer.
      
      • “A scale would better reflect reality” (PL12)
      In Germany, ten of 16 respondents state that the answer category is very good, good, ok, or indicated that there was no category missing. Three respondents (DE03, DE11, DE16) explicitly missed a third category “in parts”. Another three respondents (DE02, DE05, DE14) would prefer to give open answers to this question.
      In sum, most respondents choose “yes” as an answer category because they feel that this answer is more appropriate when they feel that they can “sometimes” or “often” influence their order or tasks, methods of work or pace. Giving “no” as an answer is only done when they have “absolutely no say” in this regard.
      
      Other findings
      One respondent continually refers to both his main paid job and his additional job while answering the question (DE14). This is an example of respondents forgetting to focus on the main paid job only. Another respondent (DE05) double-checks with the interviewer whether she should refer to her main paid job.
      

   1. Recommendations for Multi-Item Scale: Question: No changes recommended.
      
      Answer categories: We recommend using a 5-scaled answer option, ranging from “always” to “more than half of the time”, “about half of the time”, “less than half of the time”, “never”. This way, it remains possible to differentiate between respondents whose work gives a high and low level of autonomy.
      

---

1. Cognitive Techniques: Specific Probing. (OP) Paraphrasing, General/Elaborative Probing, Specific Probing. (CI)

Items