Change in Composition versus Variable Force as Influences on the Downward Trend in the Sex Ratio at Birth in the U.S., 1971-2006
Display Full Text |
Download Full Text
231.60 kB PDF file
In the United States, the sex ratio at birth exhibits a downward trend between 1971 and 2006. Prior literature implies, but has not made explicit, the difference between changes in variable force and changes in composition as two important, but very different categories of factors affecting trends. Indeed prior studies examining the trend in the sex ratio at birth have empirically related change in variable force but not change in composition to the trend. An objective of this work was to make this distinction explicit. Also, this work undertook to measure composition change for mother’s education, race, marital status, and age and relate their trends to the recent downward trend in the U.S. sex ratio at birth.
This research used all birth certificates from 1971 to 2006; in total, 125,578,596 birth certificates from 36 separate data files. I used the individual birth certificates for each year to construct a record for that year. Each record contains the sex ratio and measures of variable composition for my four predictors. Thus, my data set has year as the unit of analysis and is comprised of 36 total records. I desired an intuitively meaningful measure of composition. Thus, composition was measured by dichotomizing each of the 4 composition variables and using the percent in the category that had the corresponding lower sex ratio. Nevertheless, a more elaborate procedure was used to arrive at the categories; the procedure strove to exclude as much as possible the confounding of variable force and composition. The sex ratio at birth equals 100 times the male births divided by the female births in a year. Analyses used SAS’s autoregression procedure to derive regression coefficients and their standard errors with correction for autocorrelation.
My analysis of the effects of composition change on the 1971 to 2006 trend in the sex ratio at birth demonstrated the following: 1) the trend in education and, to a small extent, race composition would be expected to exert an upward pressure, while the trend in marital status and age composition exert a downward pressure on the trend in the sex ratio at birth; and 2) the net effect of compositional change in these four variables – these offsetting pressures – was to account for the trend in the sex ratio at birth. The downward trend in the sex ratio from 1971 to 2006 was certainly affected by composition change.
Document number: ucin1292363042
Permalink: http://rave.ohiolink.edu/etdc/view?acc_num=ucin1292363042
This ETD has been downloaded 101 times (through March 2013)
© 2010, all rights reserved.
This open access ETD is published by
University of Cincinnati and OhioLINK.
Refworks
