Authoritative Parenting and the Transmission of Religion in the Netherlands: A Panel Study

Abstract

This study was designed to explore the effect of authoritative parenting, over and above the effect of explicitly religious parenting practices, on the juvenile and adult church attendance of offspring. Data were collected as part of a panel study in which 474 Dutch respondents were questioned in 1983 as youths and in 2007 as adults. In 2007 the respondents retrospectively answered questions about how they were raised by their parents. Analyses revealed that juvenile church attendance depends mainly on parental and more specifically on maternal church attendance, whereas adult church attendance is largely an outcome of juvenile church attendance. No effects of an authoritative parenting style, that is, a simultaneous effect of responsiveness, strict control, and the granting of psychological autonomy as the three dimensions of authoritative parenting distinguished in this study, were observed. Only the dimension of strict control turned out to be a negative determinant of adult church attendance.

Notes

¹In this study we consider the transmission of religion as a one-way process from parent to child. This is actually a rather limited perspective: Nowadays, especially, bidirectionality is stressed in socialization research. In other words, not only do parents influence children but children may also influence parents (Glass et al., 1986; Kohn, 1983). Because our main interest concerns the effect of parenting style on religious transmission, however, it seems justifiable to focus on parental influence and disregard bidirectionality. We have also disregarded the influence of peers, though this may be important in shaping the religiosity of teenagers (Gunnoe & Moore, 2002). The reason for disregarding it is simply that we lack information about peer influence. Unfortunately, no questions were asked about peers at the time when our respondents were first questioned as youngsters in 1983 (Wave 1).

²We used the data set of the study on Social and Cultural Trends in the Netherlands (SOCON) in 2000 and only compared respondents in the same age group as those in the subsample (Eisinga et al., 2000). With regard to the SOCON data set, this means that only respondents born between 1957 and 1962 (N = 162) were selected.

*p < .05

**p < .01

***p < .001.

Note. ^aStandardized regression coefficient (β)

^bUnstandardized coefficient (B)

*p < .05

**p < .01

***p < .001.

Note. ^aStandardized regression coefficient (β)

^bUnstandardized coefficient (B)

*p < .05

**p < .01

***p < .001.