to Pretest Database Pretest Database
Project Title:Comparative Study of Electoral Systems (CSES) Module 5 (English Version)
  2. General Information: *Note: The item was tested in German. This is an English translation of the original German wording. The translation does not correspond exactly to the wording in the English CSES source questionnaire.*
  3. Question Text: Some people think that the government should cut taxes even if it means spending less on social services such as health and education. Other people feel that the government should spend more on social services such as health and education even if it means raising taxes. Where would you place yourself on this scale where 0 is “Governments should decrease taxes and spend less on services ” and 10 is “Governments should increase taxes and spend more on services”?
    [Manche Leute meinen, dass die Regierung Steuern senken sollte, auch wenn dies bedeutet, weniger für Sozialleistungen, z.B. im Gesundheits- oder Bildungsbereich, auszugeben. Andere Leute meinen, dass die Regierung mehr für Sozialleistungen wie z.B. im Bildungs- oder Gesundheitsbereich ausgeben sollte, auch wenn dies bedeutet, dass die Steuern erhöht werden. Wo würden Sie sich selbst auf der folgenden Skala einordnen, wobei 0 bedeutet „Regierung sollte Steuern senken und weniger für Sozialleistungen ausgeben“ und 10 bedeutet „Regierung sollte Steuern erhöhen und mehr für Sozialleistungen ausgeben“?]
  5. Answer Categories 0 Governments should decrease taxes and spend less on services [Steuern senken und weniger für Sozialleistungen ausgeben]

    1

    2

    3

    4

    5

    6

    7

    8

    9

    10 Governments should decrease taxes and spend less on services [Steuern erhöhen und mehr für Sozialleistungen ausgeben]

    Refused [Verweigert]

    Don't know [Weiß nicht]

    1. Recommendations: Question: We recommend to shorten and thereby to simplify the question text. In addition, the relationship between an increase/decrease in taxes and cuts/expansions of social services should be made clearer. Finally, we recommend to replace the term “Sozialleistungen” with “Gelder für öffentliche Ausgaben” in order to avoid that respondents primarily think about the support for people in need (such as Hartz IV recipients) when answering this question and not about government spending in general. The wording “Gelder für öffentliche Ausgaben” seems to be closer to the wording in the English source questionnaire (“social services”). Hence, we recommend the following wording:
      “Now thinking about taxes and social services: Do you prefer that the government cuts taxes even if this means spending less on social services such as health and education, or do you prefer that the government should spend more on social services even if it means raising taxes? Where would you place yourself on this scale where 0 is “Government should decrease taxes and spend less on services”, 5 is “Taxes and spending on social services should be kept as they are”, and 10 is “Government should increase taxes and spend more on services”?
      ["Nun zum Thema Steuern und öffentliche Ausgaben: Sind Sie dafür, dass die Regierung die Steuern senkt, auch wenn damit Gelder für öffentlichen Ausgaben, z.B. im Gesundheits- oder Bildungsbereich, gekürzt werden oder sind Sie dafür dass die Regierung mehr Geld für öffentliche Ausgaben bereitstellt und dafür die Steuern erhöht? Bitte ordnen Sie sich auf der folgenden Skala ein, wobei 0 bedeutet „Regierung sollte Steuern senken und öffentliche Ausgaben kürzen“, 5 bedeutet „Steuern und öffentliche Ausgaben sollten so bleiben wie sie sind“ und 10 bedeutet „Regierung sollte Steuern erhöhen und mehr Geld für öffentliche Ausgaben bereitstellen“?“]
      Answer categories: In order to avoid that respondents use the middle category („5“) as an evasive option (“don’t know”), we recommend to explicitly refer to this scale point as reflecting the status quo:
      “Taxes and spending on social services should be kept as they are."
      ["Steuern und öffentliche Ausgaben sollten so bleiben wie sie sind."]
      In order to further simplify the response process, it could also be worth considering to adapt the numeric scale points to the content of the verbal labels, resulting in a bipolar scale ranging from -5 to +5 with 0 as middle category: “0” as midpoint of the scale means “Taxes and spending on social services should be kept as they are”, “-5” means “Government should de crease taxes and spend less on services” and “+5” means “Government should increase taxes and spend more on services”. In this case, the bipolar scale corresponds more directly to the content of the question.
      In addition, answering this question might be further simplified for respondents if the scale is shortened from 11 categories to 3 answer categories, namely:
      − I prefer increasing taxes and spending more on social services.
      [Bin für Erhöhung von Steuern und öffentlichen Ausgaben]
      − I prefer decreasing taxes and spending less on social services
      [Bin für Senkung von Steuern und öffentlichen Ausgaben]
      − None of that, taxes and spending on social services should be kept as they are.
      [Nichts davon, Steuern und Ausgaben für Sozialleistungen sollten gleichbleiben]
  2. Cognitive Techniques:Information image/link to cognitive pretesting Specific Probing, General Probing, Comprehension Probing, Confidence Rating
  3. Findings for Question: More than a third of the respondents (n=6) answered with a number on the right side of the answer scale, therefore stating that they favour an increase in taxation coupled with an increase in social services. The answers of three respondents (ID 01, 04, 16) fell on the left half of the scale, declaring that they would prefer tax cuts along with cuts in social services. Four respondents chose the middle category of 5 (ID 02, 08, 09, 14), two respondents “don’t know” and one respondent refused to answer (ID 11).

    Those respondents (ID 11, 13, 15) who were not able to or did not want to choose an answer on the given scale, believe that taxes should be lowered, but that this should not be accompanied by fewer social services. Rather, taxes should be redistributed differently (“The taxes should simply be lowered and the missing money should be taken from other sources, i.e. by reducing the Bundeswehr's missions abroad”, ID 11; “Tax revenues should be redistributed differently. But that’s not expressible here.”, ID 13). This opinion however cannot be expressed on the given answer scale, as it assigns a direct link between increases or decreases in taxes and social services, respectively.

    A further problem in answering this question is that respondents cannot express if they want taxes to only be raised (or lowered) for certain segments of the population (such as an increase in taxation of the rich). This causes three respondents (ID 02, 08, 09) to choose the middle category of 5, therefore answering the question incorrectly:
    • “For me it is not about lowering or raising the taxes, but about redistribution. It is about closing the gap between rich and poor. Therefore, 5 is the middle value of this question. I understand it like this: I can’t really give an answer to this question, that’s why I chose 5.“ (ID 08))
    • “It depends. If you’re paid an hourly wage of € 8,50, you can’t pay a bunch of taxes. If the wage is higher, you could also pay more taxes.” (ID 09))
    Similarly, respondent 03 draws attention to this issue: She explains that her answer of 10 is only valid “if the right taxes are raised, like the inheritance tax or the luxury task. So those who have more money pay more taxes”.

    All in all, the question seems to be phrased in a too general manner to truly and adequately reflect the complex nature of the topic. The respondents’ differentiated opinions cannot be expressed through the given answer options.

    Additionally, it was analyzed whether the respondents primarily thought of taxes used for social services directed at people in need (such as the socially disadvantaged) or of taxes used for the general public (e.g. for financing the healthcare system) while answering. Nine respondents primarily thought of aid for those in need and understood the term “social services” to encompass unemployment pay (Hartz IV), retirement pensions/old-age poverty as well as child or housing benefits. The remaining seven respondents thought of taxes used for the general public and interpreted the term „social services“ to refer to things such as financing the healthcare system or investments into the education system. A total of seven respondents did not think of social services in the field of healthcare or education, even though these were explicitly mentioned in the question text.

    Finally, the respondents were asked about the length of question text and the width of the answer scale. Two respondents (ID 05, 11) found the answer text to be too long, while the other respondents did not. Concerning the width of the scale, six respondents stated that they could just as well have expressed their opinion on a scale with fewer response options (like a 5-point scale) and three respondents (ID 05, 07, 14) remarked that they would have preferred this. On the other hand, four respondents (ID 04, 06, 09, 10) prefer the used scale which allows them to state their opinion in a more nuanced manner.
  2. Question Topic: Politics/ Attitudes, appraisals, & ideologies
  3. Construct: Attitude towards distribution of income