Public acceptance of policy instruments to reduce forest loss: Exploring cross-national variation in East Africa

Aloyce Hepelwa - University of Dar es Salaam, Tanzania

Michael Ndwiga - University of Nairobi, Kenya

Bosco Okumu - African Population and Health Research Centre

Hailemariam Teklewold - Environment and Climate Research Center, Policy Studies Institute, Addis Ababa, Ethiopia

Peter Babyenda - Makerere University. Kampala, Uganda.

Matilda Ntiyakunze - Ardhi University, Dar es Salaam

Jesper Stage - Luleå University of Technology

Daniel Slunge - University of Gothenburg, Sweden, Environment for Development

Show less..

We collected data in five East African countries (Kenya, Tanzania, Uganda, Ethiopia and Rwanda) focusing on citizens’ perceptions on forest loss-reducing policy instruments along with the role of socio-economic factors on these perceptions. The questionnaire included questions that asked about their opinion about a ban or tax on cutting trees in public and community forests, and a ban or tax on using charcoal. The survey was performed under informed consent. A survey company based in Kenya was recruited to collect the data. The questionnaire was composed in English and then translated into the following languages: Kenya—Swahili and Somali; Tanzania—Swahili; Uganda—Luganda and Runyanoke; Rwanda—Kinyarwanda and French; and Ethiopia—Amharic, Tigrinya, Oromo and Somali. These translations were performed by native-speaking translators recruited by the company. The interviews were conducted by 26 experienced enumerators and 5 supervisors using Computer Assisted Telephone Interviews (CATI), and all responses were recorded with Kobo Toolbox software. Before conducting the interviews, the enumerators completed a two-day training session on the topics in the questionnaire and various techniques for collecting data using the CATI method. A pilot study was conducted in January 2022 with 200 respondents in each of the five focal countries to test the reliability and content validity of the questionnaire. Additionally, the pilot study enabled refining the questionnaire with feedback from both the enumerators and respondents. The company used its existing national databases of respondents involved in earlier investigations to recruit survey respondents in each of the five countries. Screening questions were used to recruit samples that were representative of the adult population in terms of age, gender and area of residence in the five countries. In total, 7,622 respondents were contacted. Following three reminders, a total of 4,766 responses with complete answers (63% response rate) were collected during March 17–28, 2022, in the five countries as follows: Ethiopia, 950; Kenya, 959; Rwanda, 991; Tanzania, 981; and Uganda, 885. Since questions regarding trust in institutions can be sensitive, we allowed respondents to opt out by answering "don't know". In the data, these responses are treated as missing values. As a result, we currently have 312 missing values for the question regarding trust in institutions. Respondents took between 10 and 23 minutes to complete the survey, with a mean completion time of 16 minutes. Research approval was received from the National Commission of Science, Technology and Innovation (NACOSTI) in Kenya, and the survey company possess national research permits for each of the five focal countries. The present data description is related to data descriptions "Public acceptance of policy instruments to reduce plastic pollution in East Africa" and "Public acceptability of policy instruments for reducing fossil fuel consumption in East Africa". The three data descriptions are subsets of the same main data collection, and are part of the Environment for Development (EfD) catalog in the Swedish National Data Service. Each data description with its corresponding dataset contains only the relevant dependent variables for a particular research study. In particular, this dataset does not have questions q1, q2, q3 and q4, q5, q6. Dependent variables for this study are q7, q8, q9, q10. Missing data points are marked with the value 98. Show less..