ABSTRACT
Research about fertility has focused in the main on studying separately the influence of communication networks and social capital on reproductive behavior, but it has rarely tried to integrate both network properties theoretically or analytically. We therefore discuss a general model of purposeful behavior that perceives individuals’ subjective perceptions of the utilities of different courses of action to be affected by structures of interpersonal influence. Resources needed to realize desired goals are furthermore shaped by exchange relationships that build social capital. These considerations are empirically applied to explanations of the intentions of 758 Polish men and women ever to have a first, second, or third child. Personal networks are especially relevant for the considerations to have a first or second child. The intentions of childless respondents are positively influenced by network partners who are at a similar stage of their reproductive biographies or who have already taken the step of having a first child. However, respondents with one child have a higher probability of intending to have a second child the more they have access to fertility-related social capital.
Acknowledgements
We are grateful to the journal's two anonymous reviewers and to Paola Di Giulio for their helpful comments. We also thank Susan Backer for English editing. An earlier version of this paper was presented at the 37th World Congress of the International Institute of Sociology, July 2005, Stockholm, Sweden.
Notes
1Our argumentation does not consider the ‘Theory of Planned Behavior’ (Ajzen 1991). One outstanding characteristic of this theory is its tightness. Therefore, an integration of social networks in this theory needs substantive preparatory considerations of how personal relationships matter for attitudes, subjective norms, and perceived and actual behavioral control. This would go beyond the scope of this article.
2The survey was supported by The State Committee for Scientific Research (KBN), Grant No. 1 H02F 00419, the Narodowy Bank Polski, Credit Bank.SA w Warszawie, Bank – PKO BP. SA, ING Nationale Nederlanden Polska, and Powszechny Fundusz Emerytalny. See Fratczak and Peczkowski (Citation2002) for more detailed information about the survey.
3As the multivariate analyses aim to explain the fertility intentions of individual respondents, the coefficients’ standard errors rest on robust Huber-White estimators, which consider the fact that respondents are clustered in households.
4Due to space limitations, only a very general description of the variables will be provided. See Bühler and Fratczak (Citation2005) for a more detailed presentation, especially of the control variables used in the analyses.
5In order to avoid that interpersonal influence and social capital are primarily related to relationships between couples, marriage partners are not considered.
6As undecided respondents may or may not have a first or another child in future (Morgan Citation1981), they can neither be used in an ordinal logit as an intermediate category nor as an individual category in a multinomial logit.
7See Bühler and Fratczak (Citation2005) for the descriptive statistics of the variables used in these models.