New findings from old data
Arong, J.R.
Options for Policy and Practice 2(2): 1-12
1976
ISSN/ISBN: 0115-2033 PMID: 12263496 Document Number: 460758
Secondary analysis or reanalysis of data from previous research studies has many advantages. One of these is that data from such studies can identify findings which may be useful in policy formulation, provided the variables are relevant to the problems at hand and can be manipulated. This paper discusses the results of a secondary analysis of data in the Norinyl Acceptor Study which originally evaluated the suitability of the contraceptive pill Norinyl for Filipinio women. The study interviewed 1950 clients from rural health units in the Tagalog region from July 17 to November 23, 1973. In the reanalysis, the statistical significance of variables was determined using analysis of variance and factor analysis. The reanalysis revealed some interesting characteristics of women who go to family planning clinics. Family planning clients were found to exhibit varied family planning behavior patterns which correlate highly with age, number of children, and educational background. Age and number of children appeared to be negative factors in family planning decisions, particularly in the decision to consult family planning clinics. The social, cultural, and psychological reasons for this finding should be studied further. There are also signs of decreasing interest in family planning as both age and number of children increase; these should also be seriously considered in the programming of future family planning campaigns. Factor analysis identified 4 general groups of highly related side effect complaints (psychological complaints, psychosomatic complaints, skin irritations and muscular cramps) manifested at varying levels by different types of women. Medical explanations for this phenomenon are possible, but the myths perpetuated by women regarding the dangers of family planning should not be ignored. The nature and causes of these myths should be studied. Almost half of the original sample of the study dropped out at different stages of the research period. Use of transformation techniques should enable the identification of a sample of dropouts and hence the determination of their characteristics. This paper shows that data reanalysis has its place in policymaking and administration of family planning programs.