Page 389 - Special Topic Session (STS) - Volume 4

STS2319 Lakshman N. R. et al.
Of course, another practical point that is considered in the study is whether
CAPI is even a cost-effective move.

2. Methodology
Experimental Design
A randomized experiment was designed to identify the effects of CAPI
vis-à-vis PAPI. This was implemented during the third quarter of 2017 in the
Ho Chi Minh enumeration phase of the LFS with field operations conducted
during the first fifteen days of July, August, and September 2017, respectively.
From each enumeration area, 30 households were randomly selected, from
which 15 were randomly assigned to CAPI, while the remainder was assigned
to PAPI. The experiment enlisted separate enumerators for CAPI and PAPI. The
main respondent for the survey was the household head, although some
household members were requested feedback on certain questions. The CAPI
arm of the experiment was implemented using CSPro for Android, which is a
free platform developed by the US Census Bureau and widely used among
national statistical systems in Asia and the Pacific.
A total of 174 enumerators were recruited, with 73 of them interviewing
for CAPI and 101 assigned to PAPI. To the extent possible, the enumerators
were randomly assigned to each group, which comes across in the similarity
of enumerator characteristics shown in Table 1. Across both modes, over 60%
of the enumerators were males. The average ages of enumerators for the two
groups are similar. Moreover, more than three fourths of the enumerators
across both methods had reported completing a university degree. The
average number of years of enumerator experience was between 5 to 6 years.

Table 1. Enumerator Characteristics
Method
DESCRIPTION CAPI PAPI Total
Freq % Freq % Freq %

Total number of 73 42 101 58 174 100
Enumerator
Sex
Male 51 69.8 63 62.38 114 65.52
6
Female 22 30.1 38 37.62 60 34.48
4
Average Age 34.9 38.41 36.94
2
Educational Attainment
Lower secondary school 0 0.00 2 1.98 2 1.15

378 | I S I W S C 2 0 1 9