README
======

Dataset title:
Survey principles for responsibility distribution

Creators:
Erik Persson
Kerstin Eriksson
Åsa Knaggård

Responsible organization:
Lund University

See the Researchdata.se catalogue entry for this dataset for complete metadata, including the dataset description, recommended citation, contact information, and information about how to obtain the dataset. 

Description: 
File "Survey answers" is an Excel file that contains the data from a survey on attitudes towards different principles for responsibility distribution among the adult population of six Swedish municipalities 2020-2021. It contains 510 entries (i.e. answers from 510 respondents). 
File "Variable info" is an Excel file that contains information about how to interpret the data from "Survey answers".

Essential information for dataset reuse.
----------------------------------------

The data were collected through a questionnaire survey distributed in Swedish to 6 000 members of the public aged 18 and up in six municipalities in southern and western Sweden. A regular letter with a link to an online questionnaire, open from 1 December 2020 to 31 March 2021, was used to invite the respondents. The letter also included information about the project and the questionnaire, including information about privacy and legal issues. All respondents had to provide written informed consent before they answered the questionnaire. To remind the respondents about the survey, two additional letters were sent out. This resulted in 510 completed questionnaires, equalling a response rate of 8.5 percent. The Sunet Survey questionnaire tool was used.  Approval for the project was granted by the Swedish Ethical Review Authority, Etikprövningsmyndigheten (protocol code 2019–06124).  

Out of the 510 participants, 53% were males. The average age was 52 years, with a median age of 53 years. Nearly 61% lived in a household with two adults, and 70% of the households had no children. Over 67% of participants owned their residence (either a self-owned property or a condominium), while 31% were renters. More than 75% resided in major cities, and 65% had attained university-level education. 

To assess potential sampling bias, we use statistics from the governmental agency Statistics Sweden. The sample demonstrates some noteworthy biases, many of which were anticipated. One of the most pronounced biases is the high average age (52 years) compared to in the country at large (41.4 years). Several of the other biases are related to the age of the respondents. For example, the proportion of respondents residing in self-owned homes is higher (67%) in our sample compared to that of the general population across the six municipalities studied (55%). 

Additionally, the proportion of households without children in the sample in higher than in the general population of the studied municipalities, 70% compared to 49%. The timing of the questionnaire distribution (around the Christmas period), combined with the effects of the COVID-19 pandemic (such as parents staying home with children, also with only mild symptoms, and some children attending school online) may have contributed to the lower response rate among households with children.  

Less directly related to the age of the respondents was the educational level. Although the statistics are not entirely comparable, the sample exhibited a substantially higher proportion of individuals with university-level education compared to the general population across the six municipalities. 

Regarding political sympathies, only a small proportion of participants indicated that they would vote for the right-wing party, Sweden Democrats, which differs markedly from the results of a national poll conducted earlier that year (6% in our sample versus 19% in the national poll). Conversely, the share of respondents expressing support for the Green Party was also notably higher than in the national poll (8% compared to 4%). This bias was anticipated, as it is reasonable to expect that individuals sympathetic to the Green Party demonstrate a greater-than-average interest in climate change issues, and thus a survey about climate adaptation. Similarly, the low representations of sympathizers with the Sweden Democrats among respondents is unsurprising, given that many of the party’s voters are sceptical towards global climate change.  

 The study presented in this paper was primarily based on the following question from the questionnaire (translated from Swedish by the authors): 

“How do you think responsibility in general should be distributed? Rank the alternatives from 1 (most important) to 5 (least important). 

1. Responsibility should be shared equally. [Equal shares] 

2. The more one stands to gain from a problem being solved, the more responsibility one should take. [Gain] 

3. The more guilty one is for the emergence of the problem, the more responsibility one should take. [Guilt] 

4. The more able one is to solve a problem (in terms of money, knowledge, influence, etc.), the more responsibility one should take. [Ability] 

5. How responsibility is distributed does not matter as long as the problem is solved. [Effectiveness]” 

The reason for including Effectiveness among the ethical principles for distribution is that according to consequentialist ethics, only the outcome, not the distribution as such, is ethically relevant. Utilitarianism, for instance, claim that our only moral duty regarding distributions, be it of responsibility or of resources, is that it results in as much good as possible while other aspects of the distribution are irrelevant. In this paper we name the corresponding distribution principle ‘Effectiveness’. 

In the questionnaire, the questions were stated in Swedish to minimize the risk for misinterpretations by the respondents. The short-names for the principles used in this paper (in square brackets in the list above) were not included in the questionnaire. 

 The question described above was number 14 (Q14 in the questionnaire). The previous questions (Q1-Q13) covered personal and socio-economic characteristics of the respondents (age, gender, education, etc.). Details about the questions and respons alternatives can be found in File 2. The rest of the survey (Q15-Q28) covered questions concerning climate change, climate adaptation, and experience of flooding.   


The data is saved as an Excel file (.xlsx).

 

Dataset organization
--------------------

Excel file named "Survey answers": Respondent answers
Excel file named "Variable info": Information About the questions and which alternative answer the numbers in File 1 represent.


Datafile descriptions
----------------------

Excel file named "Survey answers": Respondent answers
Excel file named "Variable info" Information About the questions and which alternative answer the numbers in File 1 represent, i.e. an explanation of each variable (column) in "Survey answers"


Reference information
---------------------

This data set is a subset of a larger data set. The results of an analysis of other parts of the data set has been published in:
Persson E, Knaggård Å, Eriksson K. 
Public Perceptions Concerning Responsibility for Climate Change Adaptation. 
Sustainability 2021;13(22):12552. 
doi: 10.3390/su132212552. 

The sample in our study is biased in some ways. Those who choose to respond to our questionnaire are on average older, has a higher education level, weaker sympathy for the right-wing party and stronger sympathy for the green party compared to the national average. However, since the preferred distributions where the same in all age groups, over all education levels and political sympathies, these biases may not play a big role for the overall result.  

Another potential problem is that our sample is relatively small (510 respondents), and that the response rate is below 10 percent. This calls for more studies with bigger samples.  

