Catalog of Holdings

Study Report

Study Number: LA-021-020-1-1-United States-NCARO-1977

Subject Area: Public Opinion on Political Matters, Political Participation

Bibliographic Citation: Harris business and product safety survey, number 7690, January 1977.  [machine-readable data file] / Louis Harris and Associates, Inc.  [principal investigator(s)] / Chapel Hill. NC: University of North Carolina. Institute for Research in Social Science; Social Science Data Library  [distributor].

Access Status: Access limited to UW-Madison campus

Documentation: 1 volume hard copy.

Abstract: Harris national surveys are based on a national sample of the civilian non-institutional population of the U.S. (Alaska and Hawaii, however, are not represented in the sample.) Interviews are conducted with randomly designated respondents in a minimum of 100 different locations throughout the country. The schedules (completed questionnaires) are edited and coded in New York. The coded questionnaires are keypunched and the data tabulated by standard computer equipment. There are six of these national samples, based on the 1970 census information of the population of each state in the country, and of the population resident in standard metropolitan areas and in the rest of the country. These population figures will be updated by intercensal estimates produced annually by the Bureau of the Census. The sample locations are selected biennially to reflect the cumulative changes in the country's demographic profile. The national samples are stratified in two dimensions -- geographic region and metropolitan (and non-metropolitan) residence. Stratification insures that the samples will reflect, within one percent, the national proportions of the constituent strata. Within each stratum the selection of the ultimate sampling unit (a cluster of adjacent households) is achieved in a series of steps, technically called multi-stage cluster sampling. First states, then counties and then minor civial divisions (cities, towns, townships) are selected with probability proportional to census estimates of their respective household populations. Maps of the selected civil divisions are obtained and are partitioned by segments containing approximately the same number of households. This is generally done in New York, but for the smaller civil divisions, segmenting is generally performed in the field. One of the segments in each civil division is included in each survey. The sample is rotated from study to study. The sets used in one study generally are adjacent to those used in the next study. Interviewers contact a designated number of households within each segment. At each household the respondent is chosen by means of random selection pattern, geared to the number of adults of each sex who live in the household. For larger surveys utilizing the national samples, more than one national sample may be employed. This avoids having too many respondents in one cluster. In general, the maximum number of respondents in one cluster is 16. For political surveys screens are employed that eliminate unqualified persons. In general, 75% of the person contacted pass such screens; thus, if 16 contacts were made at a cluster point, 12 interviews would be obtained at the point. When the completed interviews are received in New York, a subsample of the respondents are re-contacted to verify that the data has been accurately recorded. Occupation, industry, and 'open-ended' opinion questions are coded. The information contained in the coded questionnaires is then transferred to punch cards to permit computer processing and tabulation of the data.

Media/File Reports:

7211 (CD-ROM)
7214 (CD-ROM)
7215 (CD-ROM)
7402 (CD-Rom)
7403 (CD-Rom)
9002 (USB Flask Drive)
9003 (USB Flask Drive)
9004 (SSCC network drive)