COHORT ANALYSIS

• NCJ Number: 63302
• Author(s): N D GLENN
• Date Published: 1977
• Length: 72 pages
• Corporate Author: Sage Publications, Inc
• Sale Source: Sage Publications, Inc
• Publication Type: Statistics
• Language: English
• Country: United States of America
• Note: ONE OF SAGE'S SERIES IN QUANTITATIVE APPLICATIONS IN THE SOCIAL SCIENCES

Annotation

ELEMENTARY PRINCIPLES OF COHORT ANALYSIS ARE PRESENTED FOR THE PURPOSE OF FACILITATING SIMPLER ANALYSIS WITH SURVEY SAMPLE DATA; USE OF COHORT ANALYSIS IN TWO STUDIES ILLUSTRATE THE PRINCIPLES.

Abstract

COHORT ANALYSIS IS A RESEARCH METHOD FOR ANALYZING CHANGES IN GROUP MEMBERS THROUGH USE OF A SET OF STATISTICAL TECHNIQUES AND WAS DEVELOPED BY DEMOGRAPHERS, PRIMARILY FOR THE STUDY OF FERTILITY. TYPES OF CHANGES INCLUDE BEHAVIOR OR ATTITUDES OF A COHORT, CHANGES ATTRIBUTED TO THE AGING PROCESS OF THE COHORT, AND CHANGES WHICH ARE ASSOCIATED WITH THE EVENTS OF A PARTICULAR PERIOD IN TIME. COHORT ANALYSIS WAS USED IN THE PRESENTATION OF CHANGING SURVEY RESPONSES TO THE QUESTION OF ADMISSION OF MAINLAND CHINA TO THE UNITED NATIONS, AND IT STUDIED NON-CATHOLIC RESPONSES OVER TIME TO THE POSSIBILITY OF A CATHOLIC PRESIDENT. SINCE ANY GROUP IS A COHORT, ANY GROUP THAT EXPERIENCES CHANGE CAN BE A SUBJECT FOR COHORT ANALYSIS. RECENTLY, IT HAS BEEN ADAPTED TO THE STUDY OF ATTITUDINAL AND BEHAVIORAL PHENOMENA AND ESPECIALLY TO THE STUDY OF POLITICAL PARTY IDENTIFICATION. COHORT ANALYSES WITH SURVEY SAMPLE DATA ARE USUALLY 'SECONDARY ANALYSES;' THAT IS, THEY UTILIZE AVAILABLE DATA ORIGINALLY COLLECTED FOR SOME OTHER PURPOSE, SUCH AS PUBLIC OPINION POLL INFORMATION. FOR THIS REASON, COMPARABILITY PROBLEMS MAY RESULT DUE TO CHANGES IN THE MEANING OF WORDS OR PHRASES OVER A PERIOD OF TIME. IN ADDITION, RANDOM SAMPLING ERRORS MAY OCCUR. IF POSSIBLE, ANY COHORT ANALYSIS SHOULD BEGIN WITH CONSTRUCTION OF A 'STANDARD' COHORT TABLE IN WHICH THE INTERVAL BETWEEN DATES FOR WHICH THERE ARE DATA ARE THE SAME AS THE INTERVALS USED TO DELINEATE THE COHORTS. REGARDLESS OF THE PURPOSE OF THE ANALYSIS, THE RESEARCHER WILL USUALLY APPROACH THE DATA WITH CERTAIN HYPOTHESES WHICH PREDICT CERTAIN PATTERNS OF VARIATION, OR LACK OF VARIATION, IN THE COHORT TABLE. FOR EXAMPLE, IT MAY BE HYPOTHESIZED THAT CONSERVATISM WILL INCREASE WITH THE AGING PROCESS. ANALYSIS OF COHORT ATTITUDES OVER A NUMBER OF YEARS WILL EITHER PROVE OR DISPROVE THIS HYPOTHESIS. NOTES, REFERENCES, AND TABULAR INFORMATION ARE PROVIDED IN THE PAPER. (LWM)