Shotgun weddings and the meaning of marriage in russia

Abstract

This article examines the factors commonly thought to influence first marriage entry in industrialized countries, formulating a theoretical framework that identifies the ways in which these factors are modified by Soviet conditions. Using survey data collected in Voronezh, Russia, I utilize event history techniques to evaluate the influence of these factors on marital entry and timing among never-married pregnant women. Results indicate that education exerts a strong positive influence on marital entry. Household wealth and the baptismal status and religiosity of the respondent also influence marital entry.

Acknowledgements

I gratefully acknowledge Alaka Basu, Blair Cohen, Miriam King, Alastair McAuley, and Brian Silver for their comments on earlier drafts. A very early version of this article was presented at the Population Association of America in 1992. This work was funded by a training grant sponsored by the University of Michigan Population Studies Center and the National Institute of Child Health and Development, and a National Research Service Award (grant number HD08089 F32) sponsored by the National Institutes of Health.

Notes

1 The low earnings decile ratios during the Soviet period, roughly estimated to be between 2.8 and 3.4 from 1965 to 1989, suggested that incomes were not widely dispersed, but hovered under a low ceiling. See Atkinson and Micklewright 1992, Table UE1; Braithwaite and Heleniak 1990; and Rabkina and Rimashevskaia 1978, p. 20.

2 This point was often made in the interviews conducted by Hansson and Liden (1983) in the 1970s.

3 Statute 20 in Osnovy zhilishchnovo zakonadel'stva Soiuza SSR i Soiuznykh respublik 1981, p. 13. It is unclear how long this statute was in effect. It is present in the 1986 Moscow codes (Osnovy zhilishchnovo zakonadel'stva Soiuza SSR i Soiuznykh respublik 1986).

4 In low mortality countries like the former Russian Republic, children-ever-born is a good approximate for sibship size.

5 I employed the Chow test to see if the logistic regression should be run separately for the three age groups (see Pindyck and Rubinfield 1991, pp. 115–119). The Chow test indicates whether the slope of the marriage function is continuous or has a structural break at a given age. Since the Chow test revealed that the slope was continuous, it was not necessary to partition the sample.