Ethnic Harmony in the Aftermath of a 30 Year Civil War in Sri Lanka

We produced two documentary movies (one from the Tamil perspective and the other from the Sinhalese perspective) in collaboration with the Journalism Unit, Faculty of Arts at University of Colombo, Sri Lanka. The scripts of the documentary videos will be produced under the supervision of Jagath Pathirage, who is a Senior Lecturer in Anthropology at University of Colombo. The documentaries have few layers of narration. Firstly, it gives a brief background picture of the Tamil and Sinhalese cultures, traditions, cuisines, and historical values (most of this videography is done during the Sinhalese and Tamil New Year season in mid-April). Secondly, it sheds some light into the livelihoods of people from both ethnicities and the hardships that they have been experiencing due to various hurdles such as war and displacements (this part of the filming is done in the absence of a script to capture the ethnographic nature of the context through face-to-face interviews). Finally, the documentary attempts to understand their desired future endeavours (this part of the filming is done in the absence of a script to capture the ethnographic nature of the context through face-to-face in- terviews). The run time of this documentary is approximately 30 minutes.

2023-08-01

2023-08-31

Prejudicial Attitudes, Compassion, Trust and Collaboration

We evaluate the effectiveness of two documentary movies by employing a cluster randomized control trial in the districts of Jaffna and Hambantota in Sri Lanka. The rationale of selecting these districts is that Jaffna is the home for the highest Tamil population as a percentage of total district population which is 98.88%, while Hambantota is the home for the highest Sinhalese population as a percentage of total district population which is 97.07%. Jaffna and Hambantota districts could be highly vulnerable towards ethnic prejudice due to the limited ethnic spread. Hence, we anticipate that the intervention will be highly effective in these two districts. Then, based on the administrative division codes provided by the Department of Census and Statistics, Sri Lanka, we randomly select 40 villages into treatment group and 40 villages into control group from both districts. Moreover, we intend to carry out Trust Game based on (Berg et al, 1995), Ultimatum Game as lab-in-the-field experiments.

Randomization done in office by a computer

Randomization is done at village level.

Yes

80 villages

1800 individuals including spillover respondents.

We randomly select 40 villages into treatment group and 40 villages into control group from both districts. Treatment Groups: 20 villages in Jaffna, N = 300 Tamils and 15 randomly selected households per village and 20 villages in Hambantota, N = 300 Sinhalese and 15 randomly selected households per village. Control Groups: 20 villages in Jaffna, N = 300 Tamils and 15 randomly selected households per village and 20 villages in Hambantota, N = 300 Sinhalese and 15 randomly selected households per village. Moreover, 600 untreated individuals [who are both Tamils (N=300) and Sinhalese (N=300)] who live in treated households in both districts will be surveyed at both baseline and endline to assess any spillover effects of the intervention. (N = 300 from treatment villages in Jaffna district and N = 300 from treatment villages in Hambantota district)

Due to the absence of a pilot study, we calculated the effect size (δ) based on previous literature. (MDE ((Maiti et al, 2022))= 0.27, intra-cluster correlation coefficient (Rho) to be 0.12 (Jansen et al, 2022) and we intend to reach 15 households (one participant per household) per cluster. Thus, at a 5 percent significance level (α) with a minimum of 80 villages (80 clusters) in both districts, we generate a power of 80 percent and a sample size (n) of 1200 (15×80).