Disability Awareness Training and Inclusive Local Governance: Experimental Evidence from the Philippines

The intervention is a disability awareness and sensitivity training workshop for Local Government Unit staff in Antique Province, Philippines. The training is designed to improve understanding of disability rights, inclusive local governance, respectful engagement with persons with disabilities, accessibility, reasonable accommodation, and practical strategies for making local government programs and services more inclusive. The training will be delivered online to staff in municipalities assigned to the treatment group. Municipalities assigned to the comparison group will not receive the training during the main study period but will be offered the training after the study has concluded. The intervention is non-clinical, low-risk, and intended as professional development for local government staff.

2026-07-01

2026-07-17

Primary Outcomes (End Points)

The primary outcomes are survey-based measures of Local Government Unit staff members’ self-assessed knowledge, attitudes, and self-reported practices related to disability inclusion.

The main primary outcomes are:

Self-assessed knowledge of disability rights and inclusive governance, measured using Likert-scale items on knowledge of Philippine disability legislation, the responsibilities of the Persons with Disability Affairs Office, the medical and social models of disability, practical ways Local Government Units can implement disability-inclusive programs, and the Philippines’ international commitments to disability rights.
Attitudes toward disability inclusion, measured using Likert-scale items on the participation of persons with disabilities in governance, perceptions of persons with disabilities as community leaders, comfort facilitating participation, and the importance of disability inclusion in local development planning.
Self-reported inclusive workplace practices, measured using frequency-based items on whether staff have consulted persons with disabilities or disabled persons’ organisations, facilitated participation in consultations and decision-making, advocated for accessibility improvements, supported employment inclusion, or included persons with disabilities or their representatives in official committees or planning groups.

Objective knowledge will be analysed as an additional outcome using multiple-choice questions coded as correct or incorrect, with “Not sure” coded as incorrect.

Outcomes will be measured at baseline, approximately 2–3 months after the intervention, and approximately 9–12 months after the intervention. The primary analysis will compare outcomes between municipalities assigned to the training group and municipalities assigned to the comparison group.

Each primary outcome will be constructed as an index from pre-specified survey items. Items will be coded so that higher values indicate more favourable outcomes. Self-assessed knowledge and attitude items use Likert-scale responses from Strongly Disagree to Strongly Agree. Self-reported practice items use frequency responses from Never to Always.
Where multiple items measure the same outcome domain, items will be coded numerically, standardised using the baseline comparison-group mean and standard deviation, and averaged within each domain. Higher index values will indicate higher self-assessed knowledge, more inclusive attitudes, or more inclusive self-reported practices. Objective knowledge questions will be analysed separately as binary correct/incorrect items.

Secondary outcomes will include additional survey-based measures related to perceived barriers, institutional readiness, training exposure, and possible mechanisms of impact.

The main secondary outcomes are:

Perceived constraints and barriers to disability inclusion, measured using Likert-scale items on budget constraints, lack of technical knowledge or guidance, leadership support, competing local development priorities, and staff workload.
Training exposure and participation, measured using follow-up survey items on whether respondents attended the online disability training and the proportion of the training attended.
Mechanisms of impact, measured using follow-up survey items on whether the training improved understanding of disability rights, increased confidence in implementing inclusive practices, influenced perspectives through interaction with persons with disabilities, provided practical examples, encouraged discussion about inclusion, and affected how disability inclusion is discussed or applied within the respondent’s office.

Secondary outcomes will be measured in the follow-up surveys approximately 2–3 months and 9–12 months after the intervention. Some secondary outcomes, such as perceived constraints and barriers, are also measured at baseline.

Secondary outcomes will be constructed from pre-specified survey items. Items will be coded so that higher values indicate greater perceived barriers, greater training exposure, or stronger agreement with proposed mechanisms of impact, depending on the outcome.

Perceived constraints and barriers will be constructed from Likert-scale items ranging from Strongly Disagree to Strongly Agree. Training exposure will be measured using indicators for whether the respondent attended the training and the proportion of the training attended. Mechanism outcomes will be constructed from Likert-scale items included in the follow-up surveys.

Where multiple items measure the same secondary outcome domain, items may be coded numerically, standardised using the baseline comparison-group mean and standard deviation where applicable, and averaged within domain. Effects on individual component items may also be reported to aid interpretation.

This study uses a two-arm cluster-randomised controlled trial design across the 18 municipalities of Antique Province, Philippines. Municipalities will be randomly assigned to either a treatment group or a comparison group, with approximately 9 municipalities assigned to each arm. Local Government Unit staff in treatment municipalities will be offered an online disability awareness and sensitivity training workshop during the main study period. Local Government Unit staff in comparison municipalities will not receive the training during the main study period, but will be offered the same training after the study has concluded. Survey data will be collected from eligible Local Government Unit staff before the intervention and at follow-up. The primary analysis will compare outcomes between staff in municipalities assigned to the treatment group and staff in municipalities assigned to the comparison group.

Not available

Randomisation will be conducted by computer by the research team. The 18 municipalities in Antique Province will be randomly assigned to either the treatment group or the comparison group, with 9 municipalities assigned to each arm. Randomisation will take place before the intervention is delivered.

The unit of randomisation is the municipality. This is a cluster-randomised trial, with Local Government Unit staff nested within municipalities.

Yes

18 municipalities.

Approximately 360–540 Local Government Unit staff members.

Approximately 180–270 Local Government Unit staff members in treatment municipalities and approximately 180–270 Local Government Unit staff members in comparison municipalities.

Power calculations account for the cluster-randomised design, with 18 municipality-level clusters, 9 municipalities assigned to treatment and 9 to comparison, and approximately 20–30 respondents per municipality. The main outcomes will be analysed as standardised survey indices. Assuming 80 percent power, a two-sided 5 percent significance level, an intra-cluster correlation in the range of 0.01–0.05, and adjustment for baseline values of the relevant outcome, the minimum detectable effect size for the main standardised outcomes is expected to be approximately 0.25–0.35 standard deviations. The realised minimum detectable effect will depend on the final number of respondents, response rates, the observed intra-cluster correlation, and the predictive power of baseline outcome measures.