Assessing Knowledge of Autism Spectrum Disorders (ASD) in Tanzania; Results from an Intensive Four-Day Training on ASD for Special Educators

Background: Autism Spectrum Disorder (ASD) is a worldwide phenomenon. There are however large disparities in provision of service for children with ASD around the world (Elsabbagh, 2012). In Africa, many countries have small number of professionals dedicated to working with children with ASD. Recommendations to address this include increased engagement with community stakeholders, increase in access to information about ASD, and specific trainings on ASD in community settings (Ruparelia et al., 2016).

Objectives: An examination of results from a pre and post-test measure of ASD knowledge given during two ASD trainings for special education teachers in Tanzania.

Methods: The 20-hour, four-day training occurred in two special needs units in primary schools in Dar es Salaam (Dar) and Arusha, Tanzania in July and August 2016. A total of 79 special education teachers attended the training (Dar, N=39; Arusha, N=40). The training was run by five clinicians from the U.S based non-profit, OMPACO in partnership with special needs units in each location. Topics covered during the training included etiology, characteristics, assessment, instructional and communication strategies, behavior management and sensory processing. Learning outcomes of the training were assessed using a pre-post test methodology with a 19-item measure, the “Autism Knowledge Survey” based on items taken from the “Autism Survey” (Schwartz & Drager, 2008). All items were translated into Swahili and then back translated into English for accuracy.

Results:  Reliability results were fair (Cronbach alpha= .428). Results from paired-samples t-test indicated a significant difference in scores in Dar, t (31) = 3.22, p< .001 and in Arusha, t (37) = 7.46, p< .001 with effect sizes that were medium in Dar (Cohen’s d= .569) and large in Arusha (Cohen’s d= 1.209). An analysis of variance between pre and post-test scores was significant for location, F (1, 67) = 15.97, p< .001 with a larger increase in scores on the measure in Arusha.

Conclusions: There was an increase in ASD knowledge on the measure in both locations with a significantly greater increase in scores in Arusha. Reasons why were hypothesized to be multifactorial and included the presence of a translator in Arusha, differences in perception regarding collaboration and partnership between OMPACO and the sponsoring schools, longer travel times for teachers to reach the training site in Dar (teachers came late or missed days of the training), and differences in the facilities provided for the trainings.The low internal consistency result may be because the measure only assesses one construct (characteristics of ASD). It is not possible to compare the alpha value to that of the original measure as no reliability statistics were reported. The reliability finding can be framed within a broader discussion about the need for culturally reliable and valid measures of ASD knowledge. Improved measures that assess different domains of knowledge related to ASD would better reflect learning in different content areas covered in a multi-day training. This could in turn increase specialized knowledge in multiple content areas to key stakeholders working in the field of ASD in Tanzania.