3. Methods and data analysis

As mentioned earlier, we aimed to identify the ES that are more relevant for local people living near Ceiba and Arenillas, as well as to determine if the perception of such services is related to the type of user and the management of each protected area. In doing so, we selected a comparative case study design. This type of design is recommended when looking for similarities and differences between two cases [22], namely protected areas. In the issue of ES, the majority of studies have focused on quantitative studies regarding economic payments for ES [24, 25], neglecting the importance of mixing qualitative and quantitative data gathering instruments to collect data about people's perception of ES and the relevance of such perception in the management of any protected area. With these conclusions as a backdrop, we draw data from semi-structured interviews and questionnaires as detailed below:

Semi-structured interviews: A total of 32 social actors were identified, 13 in Arenillas, and 19 in Ceiba. The actors were selected according to their interaction with the protected areas either in the use, conservation or management. The social actors interviewed were grouped into the following categories: (i) Public and private managers, (ii) Municipality members, (iii) Local representatives, (iv) Agrarian cooperatives and Associations, and (v) Local Representatives.

Additional influential social actors were identified in each area and were incorporated in the study. For example, in Arenillas, school teachers were included in the study as they regularly visit the reserve with students and have knowledge about the reserve and its ecosystem. In Ceiba, this group was left out of the analysis as the teachers did not interact with the reserve.

According to the NCI website [23], Ceiba is managed with the help of local communities who are allowed to maintain sustainable agricultural practices within its territory after obtaining a permission granted by the owners of the area. Moreover, NCI claims to have helped local residents to obtain legal status and rights over the land they have been historically settled in. The other protected area under study is the Arenillas Ecological Reserve (Arenillas). This protected area was established by the Central Government and is located at south–west between the provinces of Arenillas and Huaquillas (3� 32<sup>0</sup> 20.4<sup>00</sup> S, 80� 8<sup>0</sup> 45.6<sup>00</sup> W) in the littoral region [14] (Figure 1). It holds 13,170 ha of TDF ranging from 0 to 300 m of altitude that also belongs to the hotspot "Tumbesian Region." The reserve is characterized by its high endemism and presence of species with restricted distribution, especially 55 species of birds considered unique to the dry forests of Ecuador and Peru. Arenillas is under the highest protection

Arenillas was always under the custody of the Ecuadorian government since its location between Peru and Ecuador has been strategic for the national security. In 2001, this area was declared as an "Ecological reserve" in the Ecuadorian National System of Protected Areas. This management category has the primary goal of conserving genetic material, ecological diversity as well as scenic beauties. No other activity is permitted within the reserve. According to Briceño et al. [14], the historical management of this reserve has prevented local communities from visiting the area, limiting their knowledge regarding the reserve and the

As mentioned earlier, we aimed to identify the ES that are more relevant for local people living near Ceiba and Arenillas, as well as to determine if the perception of such services is related to the type of user and the management of each protected area. In doing so, we selected a comparative case study design. This type of design is recommended when looking for similarities and differences between two cases [22], namely protected areas. In the issue of ES, the majority of studies have focused on quantitative studies regarding economic payments for ES [24, 25], neglecting the importance of mixing qualitative and quantitative data gathering instruments to collect data about people's perception of ES and the relevance of such perception in the management of any protected area. With these conclusions as a backdrop, we draw

Semi-structured interviews: A total of 32 social actors were identified, 13 in Arenillas, and 19 in Ceiba. The actors were selected according to their interaction with the protected areas either in the use, conservation or management. The social actors interviewed were grouped into the following categories: (i) Public and private managers, (ii) Municipality members, (iii) Local representatives, (iv) Agrarian cooperatives and Associations, and (v) Local Representatives.

Additional influential social actors were identified in each area and were incorporated in the study. For example, in Arenillas, school teachers were included in the study as they regularly

data from semi-structured interviews and questionnaires as detailed below:

category in the country namely Ecological Reserve [13].

resources it offers.

98 Tropical Forests - New Edition

3. Methods and data analysis

The interviews were valuable to identify the ES perceived by key social actors and to understand the reasons for these actors to prioritize some ES. The interview guide was validated through a pilot test conducted with people from the areas sharing similar demographic characteristics. The interview lasted approximately 40 min. All interviews were tape-recorded under informant's consent and later transcribed. The coding process used the categories provided by the Millennium Ecosystem Assessment for ES and further developed by Groot et al. [26]. The subcodes that emerged from the data followed the steps suggested by Saldaña [27]. By following this process, we were able to highlight the ES that were more valuable and visible to participants constructing thus an ES panel for each study area. The panels constructed were presented to each actor who was asked later to identify the five most important ES according to their perception as well as to assign a value to each of the prioritize services from one to five. Being five the most important and one the least important service.

For the analysis and interpretation of the ES, prioritized by participants in both Ceiba and Arenillas, an average per actor was calculated from the assessment of each prioritized ES. The results of each ES were then normalized between zero and one in order to facilitate their representation and comparison [28]. For comparison purposes, the values were presented as a percentage to improve the visualization of the results (Eq. (1)).

$$\mathbf{x}' = \frac{\mathbf{x} - \mathbf{x}\_{\min}}{\mathbf{x}\_{\max} - \mathbf{x}\_{\min}} \tag{1}$$

where x<sup>0</sup> = normalized value, x = value to be normalized, x\_max = maximum value to normalize, x\_min = minimum value of group value.

Questionnaires: By using the data of the latest Ecuadorian population census [29], with a standard error of 5% and 95% of confidence intervals, a sample of 240 people was established. Nonetheless, in Arenillas 24 people rejected to participate having a final population sample of 96 people, whereas, in Ceiba 20 people indicated that they did not feel like answering the questions leaving a final population sample of 100 people. In both areas, refusals were based on the grounds of scarce knowledge and acquaintance with the protected areas under study. The final sample included a total of 196 questionnaires that were randomly applied to people living in the surrounding areas of Arenillas and living within and nearby Ceiba. The survey aimed at complementing the information gathered in the interviews. Respondents were approached at their households or at public places such as recreational parks and church gathering. In Arenillas, respondents were mainly men (58%) between 18 and 30 years old (31%), and with a secondary school education (37%). In the Ceiba reserve, respondents were mainly women (57%), over 60 years old (23%), whose primary occupations were housewives (48%) and agriculturalists (37%). The questionnaire was valuable to add data from the general public regarding ES identification and prioritization. The questionnaire was validated through a pilot test conducted with people from the Arenillas reserve sharing similar demographic characteristics. Data obtained from both interviews and questionnaires were analyzed together to identify similarities and differences between survey and interview participants from both areas. The data were analyzed through descriptive statistics.
