2. Methodology
To bolster the validity and credibility of findings, a robust
triangulation protocol was implemented (Denzin, 2017):
1.
Source triangulation
Comparing data from survivors, institutional actors,
and community groups.
2. Methodological triangulation
Cross-verifying insights from the desk review, interviews, and focus groups.
3. Analyst triangulation
To support analytical rigor and minimize the risk of bias
from a single-analyst process, an additional internal
review step was applied during the analysis. As part
of this step, AI tools were consulted to review already
human-drafted findings for general alignment with the
Terms of Reference, inception report, and an anonymized research matrix. The purpose was to help flag
any potential gaps, inconsistencies, or areas where
themes could be more clearly grouped. Qualitative
inputs were referenced only through non-identifiable
codes. No raw data, personal data, or sensitive information was uploaded or processed. Any observations
generated through this step were carefully reviewed,
interpreted, and rephrased by the analyst before being
reflected in the findings.
2.6 Ethical considerations
and data protection
The research was governed by the core ethical principle
of “do no harm,” adhering strictly to the WHO guidelines
for research on violence against women (WHO, 2016).
Informed consent was obtained from every participant.
The process, conducted in Arabic, clearly explained the
study’s purpose, the voluntary nature of participation, the
right to decline any question or withdraw at any time, and
the detailed measures in place to protect confidentiality.
All researchers received specialized training in gendersensitive and trauma-informed interviews to minimize the
risk of re-traumatization. Interviews were conversational,
allowing participants to control the depth and pace of
disclosure.
17