Page 393 - Special Topic Session (STS) - Volume 4

STS2319 Lakshman N. R. et al.
printing costs, and miscellaneous costs (e.g. storage, bags, internet, electricity,
etc.). Tablet costs are not applicable for PAPI, while the cost is $195 for CAPI.
With a lifespan of 2 to 3 years, the cost of using a tablet per day is $0.21. If an
enumerator completed 5 interviews a day, the per questionnaire cost of a
tablet comes out at $0.04. Meanwhile, PAPI data entry and printing costs were
estimated to be at $0.65 and $1.50, respectively. CAPI is not expected to incur
these costs, as these variable costs are virtually eliminated with the technology.
Finally, miscellaneous cost estimates came out to be $1.50 and $0.5 per
questionnaire for PAPI and CAPI, respectively. This brings the total fixed costs
for CAPI and PAPI at $11,500 and $6,000, respectively. Meanwhile the variable
costs for CAPI are $0.54 and $3.65 for CAPI and PAPI, respectively. Computing
for the breakeven point using the arithmetic equation stated in the
methodology results in a breakeven point of 1,769 questionnaires. In other
words, CAPI becomes more cost-effective for survey operations involve 1,769
or more questionnaires.

4. Conclusion
The study is an attempt to systematically look at the benefits of a transition
to CAPI from a developing Asia and the Pacific perspective, particularly looking
at its implications for data quality, interview and costs. The results corroborate
the literature in terms of the perceived benefits of CAPI for data quality. First,
CAPI seems to reduce the number of errors committed per questionnaire. The
reduction in error is also correlated with household characteristics, although
out of the enumerator characteristics only female enumerators tend to reduce
the number of errors. Furthermore, the findings strongly suggest that CAPI has
much shorter interview durations, which has huge implications for the
turnaround time for data to be processed from field to headquarters. Female
enumerators tend to take less time to accomplish the interviews. Finally, the
cost analysis showed that costs were in favour of CAPI for medium to large
scale surveys of more than 1800 households. This works well for statistics
offices that cater to larger national scale surveys. These findings come
together to strengthen the case for CAPI’s adoption in national statistical
systems worldwide and build a case for female enumerators to be given
priority given that they are able to accomplish less erroneous questionnaires
in lesser time than men.

References
1. Banks, R. & Laurie, H. (2000). From PAPI to CAPI: The Case of the British
Household Panel Survey.Social Science Computer Review, 18(4), 397-406.
2. Caeyers, B., Chalmers, N., & De Weerdt, J. (2012). Improving
Consumption Measurement and Other Survey Data Through CAPI:
Evidence from a Randomized Experiment. Journal of Development
Economics, 98(2), 19-33.

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