Three stakeholder groups, one decision
The team gathered perspectives from potential tourists via an online survey, from residents of three villages via semi-structured interviews, and from local government unit (LGU) officials via in-depth interviews. Fifteen officials were interviewed across three municipalities in Bohol and Leyte, five in each, with interviews averaging 58 minutes. Data collection ran from January to May 2026 under an approval from the university’s Research Ethics Committee.
Results across all three groups were combined into a weighted scoring matrix to identify which municipality had the highest ecotourism potential. The preprint is posted open access on Preprints.org and has not yet been peer reviewed.
Valence coding inside the thematic analysis
The LGU interviews were the qualitative strand run through Evidano. Rather than open thematic coding alone, the team applied valence coding, classifying each informant’s responses as positive, neutral, or negative based on identified words and phrases, with behavioural observations from the interviews used as a complementary criterion.
Each valence was then scored (positive 2, neutral 1, negative 0), which let the study compare LGU support directly against resident interest and tourist demand within the same scoring framework. The authors cite a published National Chengchi University case study on AILYZE for interview analysis when describing the tool.
How Evidano fit the workflow
Interviews in two languages, coded to a defined scheme.
- 115 key-informant interviews conducted
Officials with at least three years of public service and direct influence over the initiative, interviewed in Cebuano or English.
- 2Interviews recorded with informed consent
Explicit consent to record was obtained from every respondent before each interview.
- 3Thematic analysis run in Evidano
The qualitative insights from the in-depth interviews were analysed in the platform.
- 4Valence coding applied
Responses were classified positive, neutral, or negative, then fed into the scoring matrix.
“Insights from the IDIs with the representatives of the local government were analyzed qualitatively through thematic analysis in AILyze software.”
“In order to discretely categorize the responses, we utilized valence coding, wherein informants' responses were classified and labeled as positive, neutral, or negative based on a group of identified words/phrases suggesting such sentiments.”
“Behavioral observations noted during the interviews were also used as a complementary criterion in the classification of responses.”
Valence coding, a codebook, or an AI-suggested structure: the analysis follows your design.
The study’s interviews were conducted in Cebuano and English.
Findings were compared across three municipalities within one framework.
Approved by the University of San Carlos Research Ethics Committee, with informed consent from every respondent.