Page 405 - Special Topic Session (STS) - Volume 2
P. 405
STS507 Katherine Jenny T. et al.
the same problem from different perspectives leads to innovative applications.
Often, these teams provide practical opportunities for methodologists to learn
about data and data collection and for subject matter experts to learn about
alternative methodologies. From an administrative perspective, these teams
can help with succession management planning, especially when junior staff
are included. Lastly, they provide justification for the production procedures
under the umbrella of data-driven decision making.
Of course, there are equally real disadvantages. The limited scope in
research can lead to missed requirements, which can be revealed as
unexpected results in implementation testing or in production. Delaying
decisions until implementation can preclude having sufficient time for careful
investigation, and quick decisions are made for convenience based on
anecdotal justification, with no alternatives tested. Having two separate teams
increases management challenges as well, as appropriate leaders need to be
recruited and team members struggle with competing duties (and on
occasion, motivation and morale challenges).
When the end-product is a theoretically solid and operationally viable
system, this approach is a success. It certainly was in the case study presented
in this paper. The two-phase team approach has been used for other 2017
Economic Census applications such as determining and implementing a
variance estimation method for product estimates (Thompson and Thompson
2018) and for developing standard response rates (Lineback, Oliver, and
Willimack 2012). Certainly in these examples, the advantages outweighed the
disadvantages, with workable solutions and buy-in as well as shared
understanding of implemented methods. And of course, the imperfect
solutions provide plenty of exciting research ideas and opportunities for the
next Economic Census.
References
1. Ellis, Y. and Thompson, K.J. (2015). Exploratory Data Analysis of Economic
Census Products: Methods and Results. Proceedings of the Section on
Survey Research Methods, American Statistical Association.
2. Fink, E.B., Beck, J.L. and Willimack, D.K. (2015). Data-Driven Decision
Making and the Design of Economic Census Data Collection
Instruments. Proceedings of the FCSM Research Conference.
3. Garcia, M., Morris, D.S., and Diamond, L.K. (2015). Implementation of
Ratio Imputation and Sequential Regression Multivariate Imputation on
Economic Census Products. Proceedings of the Section on Survey
Research Methods, American Statistical Association.
4. Lineback, F., Oliver, B., and Willimack, D.K. (2012). Developing Response
Metrics for the Economic Census. Proceedings of the FCSM Research
Conference
394 | I S I W S C 2 0 1 9