Historically, tree planting was primarily implemented to achieve economic functions; more recently, environmental and social goals have also been emphasised. This is due to the realisation that schemes operate in a socioecological system; hence, understanding and fulfilling local social (community) interests is imperative.

We conducted 13 focus group discussions and 40 interviews with communities at 13 Ethiopian tree planting scheme sites to evaluate their perceptions of the landscape challenges, scheme governance processes, and to identify areas for improvement. We analysed the survey responses qualitatively and scored the level of community satisfaction with the governance of each scheme.

The results indicated that the communities understood and felt the impact of the landscape challenges while being optimistic about tree planting schemes' potential to avert these issues. However, the communities differed markedly in their satisfaction with the scheme's governance, indicating that the degree to which schemes met their goals was related to the community's satisfactory engagement. Most respondents discussed issues about proper community consultation, engagement, institutionalisation, and capacity building, starting from the input phase to the scheme governance process.

We concluded that communities' perceptions are valuable at all scheme phases and should be employed to improve a scheme's governance, outputs, and impacts. We hope the results will help to encourage local communities' participation in steering and refining the governance of inclusive and appealing tree planting schemes with sustainable ecological and socioeconomic outcomes.

Case study selection and engagement with community representatives

In identifying relevant tree planting schemes, we noted no consolidated register of schemes in Ethiopia, nor do all schemes appear in scientific literature. Therefore, we used various methods, including checking organisational websites, networks and Google search engines to identify (donor and implementer) scheme websites. We also conducted informal consultations to identify suitable schemes by visiting various organisations (including the government Forestry Commission, United Nations Food and Agriculture Organisation, and the Ethiopian Civil Society Coordination Office), networking events, and offices of programs that coordinate and implement tree planting schemes across Ethiopia. In these visits, we collected data about the schemes and their activities, details about their donors, partners, location of implementation, and their overseeing organisations. The selection of the schemes was based on their (1) inclusion of a tree planting activity, (2) implementation in communal areas, (3) status as either ongoing or phased out in the last decade, and (4) distribution over broad geographic areas. The selection did not consider the location, species planted and scale of schemes. We identified 33 schemes, from which we selected 13 (Figure 1), with the remainder (20) excluded as they were in contextually comparable areas to those already selected or in geographic areas with security issues limiting our ability to access the communities. We created a profile for each selected scheme (See .csv format data below).

After selecting the 13 tree planting schemes, we communicated with the implementing organisations to gain first-hand information from the managers and scheme documents about the community and their involvement in the scheme governance. Then, we confirmed that schemes worked with community representatives (from different social strata: women, men, elderly, youths, farmers, livestock herders, landless, local government administrators, religious leaders, and 2 to 3 village heads) elected by the community for the scheme governance process. We undertook focus groups with  10–14 representatives per scheme. We selected these representatives, considering they reflect the social group they represented in their thoughts and ideas, which could provide insight into the community's concerns and attitudes. We also selected three households from each community by randomly walking through villages to three local households, which are adjacent (within about 100 m from the tree planting site) and those that are distant (about 400 meters away from the planting site). The household members were selected to supplement the focus group discussions as the focus group discussants are, in most cases, involved with the scheme; it was assumed that the household members would provide a check to see if the focus group discussants represented the community in their responses.

Scheme context

The selected schemes were implemented in three agroecological zones (warm/humid, cool/subhumid and cool/semi-arid; Table 1). Ten schemes were initiated primarily to fight landscape degradation, and the other three were to conserve the natural forest habitat by integrating sustainable utilisation of non-timber forest products. The implementers were from local to international and governmental to intergovernmental organisations. The local communities selected to be part of these schemes comprise 500 up to 12000 households. The schemes received grants of between USD 0.24 and 10 million from various sources (Table 1).

Focus group discussions and interviews

Between January and April 2022, we undertook focus group discussions with community representatives. At the beginning of the sessions, which lasted 90–120 minutes, we explained the purpose of the research, and participants agreed and signed the informed consent form along with the confidentiality assurance forms while also giving permission to record the process. We facilitated the focus groups by ensuring inclusive and focused discussions and, as observers, capturing group dynamics and contextual insights. During the facilitation, we encouraged diverse perspectives adapted to group dynamics.

After each focus group discussion, we interviewed three randomly selected community members. These interviews lasted 30–45 minutes and used the same guiding questions (Supporting information S1. Guiding questions list) as the focus groups.

In the focus group and the interview sessions, we stimulated the discussions and supplemented the qualitative responses by initially asking respondents to score their satisfaction with the scheme phases, input, throughput, and output phases (van Tatenhove, 2013). To help with the scoring, we presented and explained the five-score Likert scale categroies and the associated numeric values (Highly satisfied (5), Satisfied (4), Neutral (3), Dissatisfied (2), and Highly dissatisfied (1)) to respondents, thereby gathering responses in writing from each participant. In cases where there are outlying responses, their reasoning was taken as 'quotes' for the qualitative data to identify the diversity of responses.

We used semi-structured guiding questions (Supporting information S1. Guiding questions list) to initiate, direct, and prompt the discussions with the focus group discussants and household interviewees. The guiding questions are selected to understand the communities' overall perception of their surroundings, including the key topics, such as (1) the communities' perception of the land degradation challenges which prompted the tree planting, (2) the importance of tree planting schemes, (3) the satisfaction with the governance phases (input, throughput, and output), and (4) the approaches communities propose for the amendment of current and design of future schemes. The guiding questions and the research method for this study were reviewed and approved by the Ethics Committee of Coventry University under approval number P115392.

Analysis

We analysed the data qualitatively by transcribing the recorded interviews and focus groups verbatim. Transcripts were initially coded in NVIVO Version 13 (Santo-Tomás Muro et al., 2022) using an inductive process (logical process based on experiences, observations, and facts (Sauce & Matzel, 2017)) to evaluate and collate findings from each of the focus groups/interviews into themes related to the guiding questions (Fleming et al., 2019; Metcalf et al., 2015). We stratified the responses during the transcription from the different social groups (female, male, elderly, youth, farmers, livestock herders, landless, local government administrator, religious leaders, village heads from 2-4, and people with disabilities. An inductive approach (where we collected and observed data to form some patterns and generalisations) was used to examine the coded data in more detail and to organise themes relating to the guiding questions. The analysis involved an iterative process that linked the different perceptions of study participants by reading transcripts and comparing them with relevant literature and their interpretations (Drury et al., 2011). Findings are supported with illustrative quotes, the source of which is identified using the type of data collected like 'focus group' or 'Interviewee', using social group respondents represented like 'elderly' or 'female' and the scheme number where we conducted the focus group or interviews like 'Scheme 1'.

We assumed that the community considered the tree planting scheme satisfactory if and when it fulfilled the three phases of successful participation during the governance process (van Tatenhove, 2013; Yitbarek et al., 2025). These include the input phase, which is when schemes engage the community during the initiation and planning process; the throughput phase is when schemes improve the quality and transparency of their governance by engaging communities in the decision-making process during the intervention and monitoring and evaluation process; and output phase is when schemes deliver the community sought and relevant outputs, outcomes, and impacts (van Tatenhove, 2013; Yitbarek et al., 2025). Accordingly, in evaluating the community's satisfaction with the study schemes, we averaged the respondents' satisfaction scores for each scheme site and the three scheme phases of input, throughput, and output (complete detailed data from the Dryad Digital Repository https://doi.org/10.5061/dryad.2jm63xt10 (Yitbarek et al., 2025)).

References

Drury, R., Homewood, K., & Randall, S. (2011). Less is more: The potential of qualitative approaches in conservation research. Animal Conservation, 14(1), 18–24. https://doi.org/10.1111/j.1469-1795.2010.00375.x

Fleming, A., O’Grady, A. P., Mendham, D., England, J., Mitchell, P., Moroni, M., & Lyons, A. (2019). Understanding the values behind farmer perceptions of trees on farms to increase adoption of agroforestry in Australia. Agronomy for Sustainable Development, 39(1). Scopus. https://doi.org/10.1007/s13593-019-0555-5

Metcalf, E. C., Mohr, J. J., Yung, L., Metcalf, P., & Craig, D. (2015). The role of trust in restoration success: Public engagement and temporal and spatial scale in a complex social-ecological system. Restoration Ecology, 23(3), 315–324. https://doi.org/10.1111/rec.12188

Santo-Tomás Muro, R., Rodríguez Romero, E., & Sáenz de Tejada, C. (2022). Methodological Design and Application of NVIVO for a Perceptual Analysis of Green Infrastructure in the Periphery of Madrid (pp. 82–96). https://doi.org/10.1007/978-3-031-04680-3_6

Sauce, B., & Matzel, L. D. (2017). Inductive Reasoning. In J. Vonk & T. Shackelford (Eds.), Encyclopedia of Animal Cognition and Behavior (pp. 1–8). Springer International Publishing. https://doi.org/10.1007/978-3-319-47829-6_1045-1

van Tatenhove, J. P. M. (2013). How to turn the tide: Developing legitimate marine governance arrangements at the level of the regional seas. Ocean & Coastal Management, 71, 296–304. https://doi.org/10.1016/j.ocecoaman.2012.11.004

Yitbarek, T. W., Wilson, J. R. U., & Dehnen-Schmutz, K. (2025). Investigating tree planting in Ethiopia and the extent to which scheme implementation aligns with good governance practices. Journal of Environmental Management, 373, 123475. https://doi.org/10.1016/j.jenvman.2024.123475