Regional Variation in the Secondary Sex Ratio in Sweden, 1749-1869
Johan Fellman *
Hanken School of Economics, POB 479, FI-00101 Helsinki, Finland
*Author to whom correspondence should be addressed.
Abstract
A long series of attempts have been made to identify factors influencing the number of males per 100 females at birth, also called the secondary sex ratio (SR). Earlier studies have noted regional variations in the secondary SR, and racial differences seem to exist. Statistical analyses have shown that comparisons between SRs demand large data sets because for moderate data the random fluctuations are marked and hence, reliable results presuppose national birth data. Variations in the SR that have been reliably identified in family data have mostly been slight and without notable influence on national birth registers. In this study, we analyse the regional variations in the SR in Sweden (1749-1869). We build spatial models for the regional variations in the SR and, in addition, analyse the association between the SR and the crude birth rate (CBR) and the total fertility rate (TFR). The strong random fluctuations in the SR yielded rather poor goodness of fit. Especially, the proposed models failed to satisfactorily identify counties with observed extreme SRs.
Keywords: Live births, counties, geographical co-ordinates, spatial models, stepwise regression, multicollinearity, crude birth rates, total fertility rates