2. Lists as Sampling Frames:

A.

List Building: Research in list building included analysis of a 4-State
study on the "snowballing technique" (using a starter list and surveying
the names to obtain additional names) as applied to specialty crops and
commodities such as grain sorghum, bees, and turkeys. Other methods
studied provided information on costs of obtaining and updating tax
assessor and ASCS lists. List duplication, as well as the frequency
and types of changes, were recorded.

B. List Use: Several mailing devices (attractive letters, cards, and informational brochures) were tested as a means of determining the association of farms with names on lists. The amount of agriculture covered by farm operators receiving mail by rural routes, city routes, post office boxes and general delivery was estimated in four States. Duplication within and between the frames was also studied.

3. Area Frame Sampling:

A. Frame Construction: Revision of area sampling frames was nearly completed for Iowa. Other work on area frames included completion of the cropland stratum in eight of the 11 Western States.

B. Sampling Techniques: For purposes of area sampling, a 4-State research project designed to study alternative ways of obtaining farm headquarter information for farm operators residing in the cities is nearly completed. Preliminary results indicate little advantage in changing current procedures in most States. In a few States where many farm operators reside in cities, a change in procedure may result in improved estimates.

4. Crop Yield Forecasting and Estimation:

A. Corn: Intensive sampling, with 80 sample units in each of two small corn fields in Maryland, produced estimates of the number of ears significantly less than the actual number. Estimates of net yield from the objective procedures were slightly less, but not significantly different from the farmers' weighted production. Shelled grain left in the fields by the harvesting process vanished at a significant rate during the first four weeks after harvest.

Immature corn plants in five fields in Maryland were studied to find observable vegetative characteristics which might be useful in improving early season forecasts of yields. This study is being continued.

B. Deciduous Fruits: Pilot surveys to test the limb sampling and fruit counting procedures developed in previous years were undertaken in 30 apple orchards in Virginia and in 23 cling peach orchards in California. Exploratory work in the application of limb sampling techniques to Oregon filberts and California almonds was begun.

C. Texas Citrus: The possibility of estimating fruit per tree using aerial and ground photography was investigated for citrus in the lower Rio Grande Valley of Texas. Other aspects of the survey were counting of fruit by limbs, measuring of fruit size by months and recording the number of fruit dropped each month. Conclusions based on the 1968-69 work include:

(1) Ground photography of citrus trees in late July or early August can
be used, with limb counts of fruit, to estimate "fruit set" per tree.
(2) Size distribution of fruit at harvest can be predicted from August
or September measurements.

42-533 O 70 pt. 3 - 2

(3) Fruit droppage can be estimated from a sample of limbs, but with less reliability than size of fruit.

(4) The possibility of counting fruit from aerial photos is still questionable. More research is needed to determine how to utilize aerial photography to estimate total fruit.

D. Cotton and Grain Sorghum: A study was made to test the relationships between measurements of plant growth and development characteristics using remote sensing techniques and yield determinants from actual field counts and measurements. The study was conducted in the lower Rio Grande Valley of Texas. The objectives were to study (a) relationship between optical densities of aerial Ektachrome infrared film to yield determinants, and (b) methods of collecting ground data needed for analysis.

Statistically significant relationships were found to exist between
certain crop yield characteristics and film optical density.

5. Aerial Photography for Estimating Livestock: A pilot operational survey for estimating livestock inventories was conducted in four Idaho counties in June 1969. This survey was similar to the exploratory survey done in California in April 1967. However, time constraints were placed on data collection and photo interpretation.

The photo survey was made to coincide, as nearly as possible, with the June 1969 Enumerative Survey. The estimates from the 1969 June Enumerative Survey and the Photo Survey will be compared from the standpoint of precision and efficiency in relation to cost.

A. Meat: The study among a nationwide sample of homemakers on their opinions about meat was completed with publication of the final report in July. The National Livestock and Meat Board cooperated with the SRS on this research, and provided a substantial portion of the funds used. Some of the more interesting findings, from USDA's point of view, concerned respondents' understanding of inspection and grading and their reactions to the quality of meat available to them. When homemakers buy meat their first consideration appears to be quality. None of the 4 meats focused on in the survey (beef, chicken, fresh pork, ham) consistently satisfied this requirement according to their testimony.

B. Orange Juice: Mobile laboratory taste tests conducted in Chicago under contract to evaluate 12 frozen concentrated orange juices varying in peel oil level within sweetness level indicated that current industry practice conforms to consumer preferences. The test juices with higher peel oil levels than those generally used were less popular with both adults and children. These taste tests were of particular interest because they provided an opportunity to appraise the extent to which SRS's panel of USDA employees, which is usually used to minimize the costs of such research, can be considered representative of "average" consumers or those in another geographic area. Experiments almost identical in design to the experiments conducted in Chicago were run with the SRS panel, though on a smaller scale. The results in our laboratory were roughly parallel to those obtained in the larger study, but were somewhat more definitive.

C. Cotton and Competing Fibers: Data collection is currently underway throughout the U.S. to ascertain consumer attitudes and opinions which affect the purchase and use of cotton and competing fibers in selected household items such as sheets, draperies, and blankets.

D. Dairy Products and Imitations:

A contract was signed recently for a

nationwide study of homemakers' attitudes and opinions about dairy
products and their imitations. This study will be conducted early in

1970.

Other Activities in the Statistical Reporting Service:

1.

Enumerative Surveys for Other Agencies: The Statistical Reporting Service, as the fact-gathering Agency for the Department, conducted various enumertive surveys for other agencies on a reimbursable basis.

The June Enumerative Survey served as a vehicle for collecting data needed by the Economic Research Service on irrigation, cropland not cropped, and farm population.

A survey was conducted for the Wage and Hour and Public Contracts Division of the Department of Labor to provide information on the need for partial overtime exemptions available to agricultural processing industries under the Fair Labor Standards Act. The survey results will be published by the Department of Labor in their January 1970 report to Congress. A summary of the first part of the survey was published in their 1969 report.

2. Washington Data Processing Center: The Center provides automatic data processing services on a government-wide basis. The most important activities during fiscal year 1969 were (1) the installation of the Remote Access Com-, puting System (RAX), and (2) the installation of an IBM 360-50.

The RAX time-sharing service was started on October 1, 1968. At the end of the fiscal year 34 agencies were using the Center, 16 within the Department of Agriculture and 18 outside the Department.

Plans were developed for a second type of time-sharing service called Remote Job Entry (RJE). The Center entered into agreements to provide the RJE service to the Forest Service and the Foreign Agricultural Service in their Washington offices.

Owing to an increase in workload, a System 360/50 was ordered and installed in November 1968. With this added capacity, the Center was able to handle the increasing workload and offer the additional service of RJE to customers. Training in data processing continues to be a major effort of the Center. The Center completed its third year of the Intern Program with nine graduates. Customers are turning more and more to the Center for many of their special training needs.

PASSENGER MOTOR VEHICLES

The 1971 Budget Estimates propose the replacement of four passenger motor vehicles. All passenger motor vehicles of the agency are located in the field at the various state offices. These vehicles, as well as those obtained for use throughout the year from the General Services Administration, are used by professional statisticians, supervisory enumerators, and other field employees in conducting enumerative surveys and related work pertinent to the preparation of the crop and livestock estimates. The use of common carriers in this type of work is not satisfactory in that it is frequently necessary to visit, during a short period of time, numerous farms and various commercial establishments within a specified geographical area. The four sedans proposed for replacement during 1971 are necessary to meet the transportation requirements inherent in carrying out the agency's program in the States of Kansas, Wisconsin, Missouri and Wyoming. The vehicles proposed to be replaced will have passed the minimum replacement standards of 6 years of age or 60,000 miles prescribed by the General Services Administration.

Age and mileage data for passenger motor vehicles on hand as of June 30, 1969 are indicated below:

Mr. WHITTEN. Dr. Trelogan, we are glad to have you and, of course, Dr. Paarlberg representing the Secretary. We started some years ago to be sure we had a policy man here, so some of the civil servants wouldn't get in hot water trying to answer policy questions.

Dr. PAARLBERG. We might get in hot water.

Mr. WHITTEN. You are in the same job I am in. It is a hot water job. Dr. Trelogan, we are always glad to have you and your associates, as well as Dr. Paarlberg, and we would be glad to have your general

statement.

GENERAL STATEMENT

Dr. TRELOGAN. Thank you, Mr. Chairman.

It is a genuine pleasure to get a chance to visit with you again. Before I begin with the testimony, I would like to introduce two members of our staff who haven't appeared before the committee before. First, I would like to introduce William Kibler, who is in charge of the research and development branch and Mr. Raymond Long, who is in charge of the Washington data processing center.

Mr. WHITTEN. We are glad to have you with us and we trust that you will see that the record carries a biographical sketch. Dr. TRELOGAN. Very well.

(The biographical sketches follow :)

WILLIAM E. KIBLER

Mr. Kibler grew up on a general farm in Laurens County, Ga. He attended the University of Georgia where he received a MSA degree in agricultural economics and statistics. He began his service with USDA in July 1951 with the Bureau of Agricultural Economics as a statistician, in crop reporting work. After 9 years service in USDA field offices in Georgia and North Carolina, he transferred to the Washington office of the Statistical Reporting Service.

As head of the statistical methods group in SRS he directed research projects for the promotion and improvement of statistical methods employed by USDA in preparing crop and livestock estimates. In February 1967, he was appointed chief of the research and develoment branch of SRS where he directs the agency program of research and development in theoretical and applied statistical methods, and provides consulting and advisory services on statistical methodology and techniques for other USDA agencies.

Mr. Kibler received a fellowship from the Woodrow Wilson School of Public and International Affairs at Princeton University in 1967. He spent a year (1967– 68) at that graduate institution studying economic development and modernization as a Princeton fellow.

He is a member of the USDA Graduate School faculty in the Department of Mathematics and Statistics and serves as a consultant in statistics for the special programs department. He holds memberships in the American Association for the Advancement of Science, the American Statistical Association, and the International Association of Agricultural Economics.

RAYMOND J. LONG

Mr. Long was born in Brooklyn, N.Y., August 16, 1928. He holds a bachelor of science degree from the University of Louisville and the degree of Juris Doctor from the Catholic University of America.

His Government service began with 9 years in the U.S. Army which he left in 1957 to join Ford Motor Co. as a production control statistician. In 1959, he joined the Navy Department as a computer programer and advanced to chief of the programing branch when he left in 1962 for a position as systems analyst with Datatrol Corp.

In 1965, Mr. Long returned to Government service as chief of the information processing systems staff at the U.S. Department of Commerce. While at Commerce, he rose to the position of Director of the Automatic Data Processing Division for Domestic and International Business.