Experimental Study on the Implementation of Raskin Coupons

SurveyMETER was again under the mandate of The Jameel Poverty Action Lab (J-PAL) and the National Team for the Acceleration of Poverty Reduction (TNP2K) to conduct the Indonesian Household Socio-Economic Survey (SUSETI-RASKIN) in early December 2013 until the end of January 2014. It was an endline survey following the midline implemented in April-May 2013 and the baseline in September-December 2012. Previously set forth that a total of 3 study rounds were ever intended to conduct for 15 months.

The field implementation of the endline survey was postponed for a month from the plan due to many changes and additional question variables that required a fairly time-consuming process of discussion and concept maturation, said research coordinator, Cecep Sukria Sumantri. Some of these variables occurred in the household and the community questionnaires, including the addition of the Social Protection Card (KPS) and issues with the quality of distributing Raskin both to recipients in terms of the amount received, location of purchase, payment method, and well-timed of distribution, so there existed several variables issues regarding the Raskin Coupon.

“Unlike the baseline and midline, the main point of experimentation from the endline survey is to see whether the Raskin Coupon will have an impact on the effectiveness of the Raskin distribution process or not,” he explained.

Survey Preparation

The series of endline field survey activities itself commenced with the preparation of manuals and questionnaires in mid-October, then conducting the pilot test in Central Lombok district, West Nusa Tenggara on 14-16 November, followed by the Training of Trainers (ToT) on 22-24 November, afterward the Training of Enumerators (ToE) was held in parallel from 25 November-1 December: in Yogyakarta, for the Central Java, while in Bandar Lampung for Lampung and South Sumatra enumeration areas.

Cecep said the team composition in the endline survey was exactly the same as in the midline and the baseline ones. However, the implementation at the endline survey would take a little longer since additional household respondents occur for 2 to 4 households per enumeration area, hence 13 households would be respectively interviewed in all enumeration areas. (JF)