Current Projects

Search for Projects
View Projects
Project Title
by Subject/Focus Area
by Board/Committee
by Major Unit
Provisional Committee Appointments Open for Formal Public Comments
by Last Update
Meeting Information
Conflict of Interest Policy
Committee Appointment Process
More Project Information and to provide FEEDBACK on the Project

 Printer Friendly Version

Meeting Information

Project Title: Improving Data Collection and Reporting about Agriculture with Increasingly Complex Farm Business Structures

PIN: DBASSE-CNSTAT-15-14         

Major Unit:
Division of Behavioral and Social Sciences and Education

Sub Unit:
Committee on National Statistics

Mackie, Chris

Subject/Focus Area:
Agriculture; Behavioral and Social Sciences; Statistics

Improving Data Collection and Reporting about Agriculture with Increasingly Complex Farm Business Structures
February 10, 2017 - February 11, 2017
National Academy of Sciences Building
2101 Constitution NW
Washington D.C. 20148

If you would like to attend the sessions of this meeting that are open
to the public or need more information please contact:

Contact Name: Michael Siri
Phone: 202-334-3113
Fax: 202-334-3751


February 10, 2017: Open Session (Room 120)

9:00am Welcome, Introductions, Overview of Agenda
­ Cathy Kling, Chair

9:10 Sampling methods/approaches, and implications for measurement concepts and characteristics of statistics produced by NASS and ERS

Use and purpose of the Principal Operator construct. The ERS farm typology classifies farms by operators’ primary occupations, by size, and by family/nonfamily ownership. ERS will discuss the relevance of the ‘principal operator’ concept to the intent of the survey that captures financial flows along the ‘field-farm-household’ continuum. NASS will discuss how they arrived at the decisions about how to collect principal operator information on the 2017 Census.
- Linda Young, NASS (20 minutes)
- Dan Prager, ERS (20 minutes)
- Open discussion (10 minutes)

Sampling units used by NASS. NASS will provide details about the sample unit specifications for its surveys—i.e., information on exactly who/what is sampled in its operator dominant vs. operation dominant surveys. Examples will be presented illustrating how information covering complex farms is organized and stored on the list frame and exactly what components of these complex farms are sampled.

- Mark Apodaca, NASS (20 minutes)
- Open Discussion (10 minutes)

10:30 Break

10:45 Overview of methods used by statistical offices in peer countries. Among the topics that the panel will want to look at closely are: approaches for dealing with sampling and observation unit issues; how complexities such as multiple units and diverse production activities are treated; how boundaries of the “sector” are defined; how data are collected and structured (e.g., is the farm register simply a special case of a business register in terms of sampling and reporting units, and what are the alternatives). International standards and best practices—for example, relating to how do other countries' statistical agencies define and conceptualize the statistical unit that is the amalgam of farm households and businesses—will help guide the panel's consideration of optimal approaches to sampling and specifying observation units as they relate to farms.

Methods for dealing with complex statistical structures in Statistics Canada programs. This presentation will focus on the sampling/observation unit questions and will include two aspects: (1) a theoretical discussion on international best practices in statistical registers; and (2) how, on a more practical level, different agencies represent farm operations in their programs. StatCanada's practices may be compared to those of peer statistical offices such as Australia (ABS) and Eurostat.
- Gaetan St-Louis, Director of Statistical Registers and Geography Division, Statistics Canada (30 minutes)

. . . And in European programs. This presentation will outline approaches used by the European Union in general and Statistics Netherlands in particular to some of the above-described conceptual issues in their Farm Accountancy Data Network (more or less the EU's ARMS) and in Eurostat's Farm Structures Survey.
- Krijn Poppe (panel member) (20 minutes)

- Open Discussion (until lunch). Ron Jarmin, U.S. Census Bureau, and Dave Talon, BLS, may be available to comment on how key U.S. statistical agencies treat farm operations and, in the context of improving overall coherence of resulting statistical products, how this compares with NASS/ERS, and with Statistics Canada. [Some of this discussion may be held for the 4:00pm session.]

12:15pm Working lunch to continue morning discussions.

1:15 Use of Administrative (and perhaps other nonsurvey) Data; Linked Data Sources

Current uses of administrative and commercial data in the production of agricultural statistics will be described. Additionally, visions for next steps in the use of administrative sources and data linking may be discussed (e.g., how would does/could admin data tie into ARMS). Legal and other constraints aside, what uses could administrative data be put to by the agencies, from a data collection (NASS) and data analysis (ERS) standpoint?

Developments at ERS. ERS will provide information on two types of initiatives: (1) a summary of input that the agency has provided as input to the Ryan-Murray Commission—e.g., on current practices regarding passing administrative data and survey data back and forth given confidentiality constraints; and (2) some examples of ongoing work that involve linkages (e.g., between administrative databases, and between administrative data and ARMS).
- Cynthia Nickerson, ERS, and Steve Wallender, ERS (30 minutes)

Improving the Methodology for using Administrative Data Across the Agricultural Statistics System. The discussion will be broadened here to consider data collected from sources other than surveys and censuses for purposes such as maximizing the consistency of cross-sector estimates and helping to align survey data with information from non-statistical sources.
- Emily Berg, Iowa State University (20 minutes)

- Open Discussion (10 minutes)

Use of administrative data across the U.S. statistical system more broadly. Administrative data are widely used across U.S. statistical agencies. For example, IRS and SSA data are used in the construction of business lists/sampling frames, creating linked data sets, and serving other statistical/research purposes. Linkages to data from other agencies make resources such as the LEHD employee data possible. This presentation will describe what is being done now through the Census Bureau's data linkage infrastructure. Questions about what may be possible for the future may also be addressed.
- Amy O'Hara, Center for Administrative Records Research and Applications (30 minutes)

Generic Statistical Business Process Model (GSBPM). The GSBPM offers a model of the steps involved in a statistical activity—i.e., the set of business processes needed to produce official statistics. It is a tool that encompasses administrative data as well as survey-based sources, and includes aspects that are applicable to our work.
- Greg Peterson, StatCAn/panel member (15 minutes)

- Open Discussion (10 minutes)

3:15 Break

3:30 Advancing uses of new data and technology

NASS uses GIS and remote sensing data for improving data collection, reducing respondent burden, and for use in research. How does the survey cycle and the availability of data such as the Cropland Data Layer (CDL) impact when and how such data can be used? Also, what are the legal issues involved with using this data?
- Lance Honig, NASS (20 minutes)
- Open discussion (10 minutes)

4:00 Conceptual questions and statement of task issues of scope for the study

This presentation will consider approaches for measuring food and agriculture as a supply chain that cuts across current industry definitions. Dividing the economy into "farm" and "non-farm" sectors is a useful distinction for reporting, but may be less so for data collection and analysis. Various sectors bleed into one another in markets for labor, farm supplies, and downstream processing, marketing, and retailing. For example, in the case of contract labor from an off-farm employer, is the employer in "farming" or labor services management? How does this factor into the NIPA?
- Brent Hueth, Richard Dunn (30 minutes)

- Open Discussion, beginning with Ron Jarmin, U.S. Census Bureau; Dave Talon, BLS; and [Carrie Litkowski and Kurt Kunze (BEA estimates of proprietor income) and Patrick Canning (ERS, on food dollar expenditure series, especially the part going to the farm) may also be able to attend] to comment on the presentation and to help respond to the following types of questions:
? How do other statistical agencies measure/differentiate agricultural and non-ag sectors for purposes such as measuring productivity by sector, constructing the NIPAS, and for conducting the census of manufacturing and various surveys, and for construction of business lists.
? How does statistical coverage of the economy get the overall and sector specific measures right?
? How should agencies be collaborating to take advantage of data from different sources to capture overlaps? How does current definition of “farm” for purpose of data reporting impact various estimates?
? Details about MOU with BEA for farm proprietor income. Also, for their state measures.

5:00pm Adjourn--panel in recess

February 11, 2017: Closed Session (Board Room)

This session is closed to the public in its entirety.

Closed Session Summary Posted After the Meeting

The following committee members were present at the closed sessions of the meeting:
J. Arbuckle
Norman Bradburn
Richard Dunn
Allen Featherstone
Joe Glauber
Brent Hueth
Ani Katchova
Catherine Kling
Doris Mold
Jean Opsomer
Greg Peterson
Krijn Poppe
Daniel Sumner
James Wagner
Jeremy Weber

The following topics were discussed in the closed sessions:
The panel reviewed the extensive presentations from day 1, discussed their draft report outline, and made plans for future plenary and subgroup meetings.

The following materials (written documents) were made available to the committee in the closed sessions:

Date of posting of Closed Session Summary: July 14, 2017