was done by the BOASI, from its complete 

 listings of manufacturing establishments, 

 in accordance with the pre-arranged 

 specifications covering starting point and 

 interval. All of the establishments in 

 the seventeen unclustered counties were 

 included in the sample. 



The number of primary sampling units 

 and counties included in the sample and 

 the total number of establishments from 

 which the final list of respondents was 

 selected is shown in the following two 

 tables: 



The sample, it will be noted, was 

 disproportionate by both size and region. 

 Specifically, all plants in the West were 

 over sampled, and all plants with 1,000 

 or more employees were over sampled. 



The end result of the sampling pro- 

 cess was the identification of 892 estab- 

 lishments. The sample can be considered 

 to adequately represent all manufacturing 

 establishments which had 250 or more 

 employees during first quarter 1953. 



Telephone "Screening" Interviews 



A telephone screening questionnaire 

 designated the "T" interview was designed 

 and used to establish the presence and 

 nature of feeding facilities in establish- 

 ments with 250 or more employees in 1953. 

 Of the 892 establishments identified 

 through the sampling process the "T" 

 telephone interview was completed with 

 81*5. Interviews were conducted during 

 late December, 1955 and early January, 

 1956, and the respondent was usually an 

 individual in a personnel supervisory 

 function. 



There were hi establishments with 

 which a "T" interview was not completed. 

 For the most part, the absence of an 

 interview was the result of the establish- 

 ment having gone out of business or having 

 moved out of the sampling area. 



Subsequently, 2U additional "T" 

 interviews were completed with establish- 

 ments selected on a systematic basis with- 

 in certain psu's, where problems of non- 

 cooperation in the collection of detailed 

 food schedules made this desirable. In 

 total, then, 869 "T" interviews were 

 conducted. 



Changes in Employment Size 



Almost three years had elapsed be- 

 tween first quarter 1953, at which point 

 in time these establishments were classi- 

 fied by employment size by the BOASI, and 

 the early months of 1956, when actual 

 interviewing was done. Presumably, some 

 of these plants had gone out of business 

 during the period. Others, on the other 

 hand, had grown substantially in size. 



Size of employment was also affected 

 in another way. It will be remembered 

 that the universe, and the sample, -was de- 

 fined in terms of "establishments". That 

 is, a company operating at more than one 

 location is required to submit a report 

 for the "establishment" at each location; 

 also, companies engaged in distinctly 

 different lines of activity at one loca- 

 tion are required to submit separate re- 

 ports on each "establishment" if the 

 activities are substantial in size. It 

 follows from the latter that at any 



28 



