CITIZEN-SDSS: Guidelines for conducting household-level interviews in mosaic forest landscapes of the Philippines

CITIZEN-SDSS PROJECT: Guidelines for conducting household-level interviews in mosaic forest landscapes of the Philippines Melvin Lippe, Cecilia Mangabat, Prince Asare, Renz S. Come, Rattiya Lippe Authors affiliation Melvin Lippe Johann Heinrich von Thünen Institute, Institute of Forestry, Leuschnerstraße 91, 21031 Hamburg, Germany Cecilia Mangabat College of Forestry and Environmental Management, Isabela State University, Cabagan, Philippines Prince Asare Johann Heinrich von Thünen Institute, Institute of Forestry, Leuschnerstraße 91, 21031 Hamburg, Germany Renz S. Come Department of Forest Science, Faculty of Forestry and Environmental Science, Visayas State University, Baybay, Philippines Rattiya Lippe Johann Heinrich von Thünen Institute, Institute of Forestry, Leuschnerstraße 91, 21031 Hamburg, Germany Suggested citation: Lippe, M., Mangabat, C., Asare, P., Renz, S. C., & Lippe, R. (2025). CITIZEN-SDSS PROJECT: Guidelines for conducting household-level interviews in mosaic forest landscapes of the Philippines (Version 1). Zenodo. https://doi.org/10.5281/zenodo.17233788 Project consortium Thünen Institute of Forestry Leuschnerstr. 91 21031 Hamburg | Germany Forest Foundation Philippines Makati City Philippines Isabela State University Cabagan, Isabela province Philippines Visayas State University Baybay, Leyte province Philippines Funding Velux Stiftung Kirchgasse 2 CH-8001 Zürich | Switzerland Grant ID: 1881 Contact and Homepage E-Mail: [email protected] Homepage: https://www.thuenen.de/en/institutes/forestry/projects-1/citizen-sdss- Household survey________________________________________________________________________________________ Table of Contents 1 Background CITIZEN-SDSS Project 1 2 Objective of household surveys 1 3 General conventions household questionnaire (HHQ) 2 3.1 Selection of study sites and Barangays for conducting HHQ 2 3.2 Reference administrative boundaries study landscapes 2 3.3 Barangay 2 3.4 Barangay selection 2 3.5 Sampling using household lists and the total no. of interviewed HHs 3 3.6 The layout of the questionnaire 4 3.7 General guidelines when interviewing the respondent 4 3.8 Training and pre-testing 5 3.9 Data backup, sanitation and data entry 5 4 Definitions and terminologies 5 4.1 Household 5 4.2 Household head (HHQ section A3) 6 4.3 Georeferenced photos (HHQ section A3) 6 4.4 Forest ecosystem services (HHQ section B2) 6 4.5 Land degradation (HHQ section D) 7 4.6 Land use assets (HHQ section D) 7 4.7 Family labour (HHQ section E) 7 4.8 Household income (HHQ section E) 8 4.9 Income from agriculture and plantation activities (HHQ section E1) 8 4.10 Wage employed activities (HHQ section E4.1) 9 4.11 Income from off-farm self-employed/ business activities (HHQ section E4.2) 9 4.12 Other income sources (HHQ section E4.3) 9 5 Major land cover and land use types considered 9 5.1 Forest types 9 5.2 Timber plantations 10 5.3 Cropland 10 5.4 Pasture 10 5.5 Agroforestry 10 5.6 Fallow 10 5.7 Shrublands 10 5.8 Grassland 10 5.9 Wetlands 10 6 References 11 Household survey________________________________________________________________________________________ 1 1 Background CITIZEN-SDSS Project The remaining Philippine forest landscapes cover around 24% of the country's total land area. These landscapes are characterised by a high level of endemism, yet they are simultaneously threatened by deforestation and forest degradation, which has long-lasting effects on the ecosystem services they provide to society. Naturebased solutions (NBS) encompass a wide range of measures, such as forest conservation, reforestation, and restoration, and can support local livelihoods through community woodlots, agroforestry, and other initiatives, including living fences and riverbank vegetation. NBS have the potential to safeguard the Philippines' remaining forest landscapes and support national wood demand, which is currently met by industrial timber plantations and imports. The CITIZEN-SDSS project aims to inform decision-making to support the development of long-term land use and forest management strategies that utilise nature-based solutions to sustain local livelihoods and ecosystem services. The project focuses on four study sites and their surrounding provinces - see for further details on study landscapes: Asare and Lippe (2025). The project combines participatory approaches and various forms of stakeholder interaction, such as key informant interviews, household interviews and focus group discussions, with geographic information systems (GIS) and spatiotemporal modelling. The latter can also be referred to as spatial decision support systems (SDSS). The project develops methodological approaches to make projections on long-term sustainable forest and land use management pathways, supporting local livelihoods and fostering nature-based solutions at Barangay level. This is important because decision-making processes require contextsensitive approaches that are also grounded in scientific rigour. 2 Objective of household surveys The objective of the household survey is to collect information about the socio-economic profile of a household and its land use assets. Moreover, the survey aims to reveal households’ perceptions on ecosystem services and land degradation, the role forests play in supporting local livelihoods at the Barangay-level, and to further assess how the Covid-19 pandemic influenced natural resource extractions. In order to achieve this goal, household surveys are conducted in 6 villages for each of the four study landscapes of CITIZEN-SDSS (see section 3.4 for more details). For data collection purposes, a structured questionnaire is used, commencing in a face-to-face situation, interviewing the household head as the primary respondent, and in case of his/ her absence, interviewing a household member (older than 18 years) to replace the household head. The latter one is then defined as a secondary respondent. Furthermore, key informant interviews (KII) are conducted with the relevant Barangay captain or deputy to obtain information on the general demographic composition of a study Barangay, to collect information on local farm gate prices of the most relevant crops and non-farming products, among others. The following sections provide instructions and explanations for enumerators in using the household questionnaire for conducting face-to-face interviews. This includes information on terminologies and conventions for the purpose of the household questionnaire (HHQ) and the CITIZEN-SDSS project. The document builds on several completed projects and their training guideline documents, and questionnaires, namely the Landscape Forestry in the Tropics (LaForeT) project 1 and the related studies of Ojeda et al. (2020) and Wiebe et al. (2022), the Poverty Environment Network (PEN) project 2 and the related studies of Angelsen et al. (2011) and prototype questionnaires and guidelines of PEN (2008), and the Thailand Vietnam Socio Economic Panel (TVSEP) 3 , 1 https://www.la-foret.org/ 2 https://www2.cifor.org/pen 3 https://www.tvsep.de/en/ Household survey________________________________________________________________________________________ 2 and the related prototype questionnaires and guidelines of TVSEP (2024a, b, c). Templates of the HHQ and KII are accompanying the guideline document and the open repository DOI (Zenodo, OpenAgrar. 3 General conventions household questionnaire (HHQ) The questionnaire is designed in a code-based tabular format. Thus, for almost all questions, a code list is presented from which answers have to be chosen or ticked by the enumerator. As it is, however, impossible to cover all answers via code lists, code option “90” allows the enumerator to add further items named “90=others, please specify”. In such a case, the enumerator has to write down “90” plus the missing/ new code item. In general, it is important to try not to read out codes in front of the interviewed respondent, but instead to have a conversation where codes are discussed with the respondent. It is the enumerator's task to “encode” the respective answer to the HHQ. In this way, it should ensure the respondent’s attention during the interview, and avoid interview fatigue. 3.1 Selection of study sites and Barangays for conducting HHQ The selection of the study area is a critical decision. The area is often chosen due to certain features of interest for the research project, which, in the case of CITIZEN-SDSS, refers to land degradation, nature-based solutions (e.g., agroforestry, forest conservation, reforestation), and ecosystem services provided by forests and locally prevailing other non-forest land use systems. Due to financial and time constraints, it is usually not possible to include all Barangays of a certain municipality for household interviews. Thus, a selection is necessary that follows representativeness and context variability in key variables across the selected Barangays. Given the overall project scope of CITIZEN-SDSS, 4 study sites were selected, with: • Municipality of Penablanca and Balasig Watershed, being located in Region 2 Cagayan Valley, and • Municipality of Silago and the City of Maasin are located in Region 8 Eastern Visayas. 3.2 Reference administrative boundaries study landscapes For the purpose of study consistency, we rely on the administrative boundaries used at the level of municipality, independently of the administrative boundaries published by NAMRIA, DENR or other provincial or regional government agencies. When overlaying the different administrative boundaries in a GIS, inconsistencies among the various sources and agencies may occur and often become clearly visible. In such a case, it is not the task of the project to find agreement among the identified administrative boundary inconsistencies. Hence, the project relies on the administrative boundaries that are provided by the respective local government unit (LGU) or watershed boundary, given the research scope of CITIZEN-SDSS on LGU, respectively. 3.3 Barangay For purposes of CITIZEN-SDSS research, a Barangay is defined as the lowest administrative unit in a study area, normally under the jurisdiction of a Barangay captain and other officials, such as the Barangay secretary. 3.4 Barangay selection CITIZEN-SDSS relies on a stratified sampling approach for selecting a total of 6 Barangays per study (Municipality or Watershed) using land cover information from NAMRIA (2020), the most recent census data (PSA 2024), and the administrative boundaries of the selected municipalities of Maasin, Silago and Penablanca. For the case of the Balasig watershed, the hydrological boundaries defined by topographic features are used instead, which incorporates Barangays of the Municipalities of Cabagan and Tumauini. Subsequently, the following calculation was done for each Barangay: 1) Calculate the area of forest and cropland per Barangay in hectare (ha), Household survey________________________________________________________________________________________ 3 2) Calculate the share of forest and cropland of a Barangay (ha) in relation to the total area of the Barangay (ha), and convert into %, 3) Use the results of (2.) to "sort" the barangays into three categories, namely: a. "have cropland area but no forest area" (Category CROPLAND, CL), b. "have more cropland area than forest area per total Barangay area" (Category CROPFORESTLAND, CFL), c. "have more forest area than cropland area per total Barangay area" (Category FORESTCROPLAND, FOCL). Based on this categorisation, 2 barangays are selected randomly per land cover category and study landscape, totalling 6 Barangays per study site, and 24 Barangays for all landscapes together (Table 1). Table 1. List of selected Barangays for each study landscape based on the predefined land cover categories (CROPLAND: CL, CROP-FORESTLAND: CFL, FOREST-CROPLAND: FOCL) Study region Study landscape Name of Barangay Land cover category Cagayan Valley Balasig watershed Municipality of Cabagan Union CL Pilig Alto CL Masipi East FOCL Municipality of Tumauini Camasi CFL Cumabao CFL DY Abra FOCL Penablanca (Municipality) Buyun CFL Cabbo CL Callao CFL Dodan CL Minaga FOCL Nanguilatan FOCL Eastern Visayas Maasin (City) Bactul I CL Baugo CFL Cagnituan FOCL Hantag CL Lunas FOCL San Jose CLF Silago (Municipality) Catmon FOCL Katipunan FOCL Tuba-on CLF Salvacion CLF Sap-ang CL Sudmon CL 3.5 Sampling using household lists and the total no. of interviewed HHs The selection of households within a preselected Barangay is carried out using random sampling. This step is essential to ensure that the study sample is representative of the total population under investigation, defined here as all interviewed households across the four study sites. Random selection not only strengthens the validity Household survey________________________________________________________________________________________ 4 of conclusions drawn from the data but also facilitates subsequent statistical analysis by minimising bias and enhancing the reliability of comparisons. For this purpose, household lists with the name of the household and, if available, the names of the other household members are required for a random selection of the target households per Barangay. For each land cover category (CL, CFL, FOCL), a target of 30 households is chosen, which requires interviewing a total of 90 households per study site. Altogether, a total of 360 households were interviewed across the four study sites. To account for potential household dropouts or cases of unavailability during the survey period, the number of households randomly selected per Barangay was set at 20% above the target sample size. This oversampling strategy ensured that enumerators could replace unavailable households with preselected alternatives, thereby maintaining the intended sample size while preserving randomness and representativeness. 3.6 The layout of the questionnaire The questionnaire contains different layout features to guide the enumerator: • Each question is continuously numbered, and each answer has to be noted in the provided cells or tables. Continued numbering of questions later facilitates the entry of HHQ data into Excel data sheets. • Bold text refers to particular remarks for enumerators and should not be overseen. • Coordinates of the household location have to be either recorded in the format of “minutes, degrees, seconds” for Northing and Easting using either a handheld GPS device or a smartphone with a GPS recorder app. • Code lists are provided for almost all questions, and they guide the enumerator in the conversation with the respondent, as well as to check whether certain issues were not (yet) mentioned by the respondent. • Time frames: Many questions refer to a particular time frame, e.g., “during the past 12 months” or “in 2024”. Enumerators must ensure that the time frames are clearly communicated and understood by the respondent. • Amounts and percentages: In some cases, it may be useful to first rely on local units (e.g., bags) rather than on the desired units of kilograms, hectares or m². If one uses local units, it is important to note such a calculation in the particular section of the HHQ. If respondents have difficulty providing amounts, enumerators should try to get a good estimate before skipping the question completely. 3.7 General guidelines when interviewing the respondent General principles when interviewing a respondent can be summarised as “listen, observe, and let the respondent talk”. This further has to include: 1) Properly introduce yourself to the respondent and other attending household members before starting the interview. 2) Put the respondent at ease before starting the interview. Ask easier questions first, e.g., names of family or household members, before starting with the more difficult ones. Show genuine interest in the respondents. 3) In order to make the interview a pleasant experience and get good results, you should always be polite to the respondents and show interest in their stories. 4) A little small talk helps to get the conversation going. Especially when your respondents are nervous or shy, take a little time for a chat, so that everybody feels more relaxed and comfortable. 5) Do not ask questions simultaneously. 6) Do not read out codes in front of the interviewed respondent, but instead try to have a conversation where codes are discussed with the respondent. It is the enumerator to “encode” the respective answer to the HHQ, not the respondent him-/herself. Household survey________________________________________________________________________________________ 5 7) You may use indirect questions for sensitive aspects, e.g., income, ownership, disability, age, marital status, educational attainment, etc. 8) Be conscious of the time you conduct and the time spent on the interview. 9) Be patient. You will often encounter respondents who are unsure of the answer to certain questions. Explain to the respondents that there is no need to be insecure if they do not know the exact answers. 10) Try to get reliable estimations, e.g., by using a shorter recall period; otherwise, mark “99 = no answer”. 11) If an interview is difficult, resist the urge to skip questions and keep going through the questionnaire step by step, even if people do not know certain answers. 12) Completeness is a key criterion for high-quality information. So, use the questionnaire as a structured instrument to gather a complete data set. 13) Be critical. Sometimes respondents get bored during the interview and will just give you any answer to rush through the interview. If a respondent seems suspicious, use follow-up questions to check for consistency. 14) Be aware that some activities in the questionnaire may not be completely legal. A good relationship with the respondent may help to get some answers, but do not expect to obtain “the full truth and nothing but the truth”. 15) Sensitive questions: Keep in mind that respondents are free to skip certain questions if these are sensitive for them. While you should ask follow-up questions in some cases, you should not push too hard if your respondent is obviously uncomfortable. 16) Do not abuse respondents’ hospitality. 17) Do not interpret in front of the respondent. 3.8 Training and pre-testing Training and pre-testing are important components of conducting successful HHQs, as they help to cross-check and test questionnaire settings under real-world situations, as well as to check whether certain questions and code lists are clear to the enumerator and respondent. In the case of CITIZEN-SDSS, several days of training were conducted at ISU and VSU, and at least two pretesting rounds took place, testing the HHQ with a total of 10 respondents who do not belong to any of the study areas. 3.9 Data backup, sanitation and data entry 1) Data backup: Each HHQ should back up twice: 1.) Taking photos of each HHQ page in the evening of an interview day, 2.) Uploading photos to the provided Google Drive folder after returning from a survey. 2) The information from the completed questionnaire should be checked by the enumerator directly after the interview for completeness and relevant immediate recall. The questionnaire has to be then given to an enumerator colleague who reviews the HHQ and checks all entries. As a next step, the responsible ISU/ VSU study leader will review the entries of the HHQ and approve data entry. 3) After data entry is completed, each HHQ questionnaire has to be labelled and uploaded to the provided Google Drive folder. Labelling has to be done as, e.g., HHQID_MonthDayYear_FirstName_FamilyName 4) Data entry should be done on a regular basis and after returning from a field survey. It should NOT be delayed towards the end of the overall interview period to ensure information consistency, and to allow for recall respondents in case of missing information or unclear entries. 4 Definitions and terminologies 4.1 Household A household is defined as a “group of people (normally family members) living under the same roof, and pooling resources (labour and income)”. Johann Heinrich von Thünen-Institut Bundesallee 50 38116 Braunschweig Germany www.thuenen.de Text and Tables: ©2025 the author(s), Thünen Institute. This is an open access publication distributed under the terms and conditions of the Creative Commons Attribute on 4.0 International (CC BY 4.0) license (https://creativecommons.org/licenses/by/4.0/).