Go to the NASS Home Page U.S. Department of Agriculture
National Agricultural Statistics Service
Research and Development Division

Office
of the Director
Census and Survey Research Branch
Data Quality | Statistical Methodology
Geospatial Information Branch
Area Frame | Spatial Analysis

GRAPHICS RESEARCH 

Interact with State and County Data from the 1997 Census of Agriculture

This research is divided into two parts based on the type of data displayed and the purpose of the display. One part involves displaying micro-level data in a way to aid analysts in identifying unusual data values. Another part involves displaying macro-level data as an analysis tool, with the added benefit of exposing patterns in the data that typically remain unnoticed using traditional methods. 

With the potential use of an automated edit and imputation system, analysts would have minimal contact with the micro-level data. However it is important for them to understand how the data are distributed in order to identify anomalies or changing trends. This project would explore the use of graphical displays, such as scatter plots, box plots, and dot plots to achieve this. Such display methods could be part of, or supplemental to, the Interactive Data Analysis System (IDAS), which is a SAS-based system developed by statisticians in NASS. Succinct training materials on the interpretation of these graphics may be needed for analysts to get maximum benefit from the graphics. 

With large volumes of data to review, such as with the Census of Agriculture, analysts need quicker, easier, and more efficient ways to understand inherent patterns or structure of aggregate level or summary data. To achieve this, this project explores how such data can be more effectively displayed and disseminated using specialized "row plots", time series plots, and concepts such as small multiples and panel displays. The ability for NASS analysts to create these graphics dynamically and for their own customized applications via the Intranet will also be explored. The research can be extended to allowing data customers an enhanced ability to view, analyze and interact with summary data via the Internet. Historically, such customers have had access to these data, primarily in tabular form only, with some limited graphics displays.

Interact with State and County Data from the 1997 Census of Agriculture