Religious ‘Nones’ Generally Have More Liberal Family Values in Areas of Greater Disaffiliation

Using quantitative studies Sarah Wilkins-Laflamme discusses how religious nones demonstrate greater photocommitment to liberal family values in areas of high religious disaffiliation, impacting upon the value divide between the unaffiliated and their more religiously committed neighbours.

It has been shown time and time again that the religiously unaffiliated (those who declare they have no religion when asked) tend to be more liberal in their views towards gender roles, sexuality and abortion compared with the religious in Western nations (Finke and Adamczyk 2008; Norris and Inglehart 2004; Putnam and Campbell 2010). One of the principal mechanisms behind this relationship between religious affiliation and conservative moral values is institutional religiosity’s role in shaping such attitudes. Individuals raised and actively participating in a religious group are more likely to adopt and adhere to this group’s beliefs about what is right or wrong (Hayes 1995; Layman 2001; Nicolet and Tresch 2009; Raymond 2011). Religious ‘nones’ on the other hand are removed from such institutions which often promote more conservative value orientations.

Yet, with an almost universal rise in rates of non-affiliation across the West (Baker and Smith 2009; Bruce 2011; Wilkins-Laflamme 2014; Zuckerman 2009), little is known on how this effect has evolved. Are the unaffiliated more liberal on average in areas with less disaffiliation, because when the unaffiliated form a small group only hard-core liberals dare to remove themselves from religion? On the contrary, are the unaffiliated more liberal in areas where they form proportionally larger groups, areas where liberal attitudes become more socially acceptable and the unaffiliated with time and new generations become even more removed from religious institutions and their beliefs? Or do contextual levels of disaffiliation have no effect on value orientation?

I attempt to shed some light on these questions with data from the 2008 International Social Survey Programme (ISSP). Full details on methods and results are available upon request. The dependent variable for the analyses, which I name the Family Values scale, was generated by means of a principal component factor analysis (one dimension generated and used as the standardized scale). The original ordinal variables included in this scale refer to a respondent’s views on pre-marital sex, sexual relations between two adults of the same sex, abortion when a defect in the foetus is likely and distinct gender roles (if the husband’s job is to earn money and the wife’s job is to look after the home and family). As shown in Figure 1, lower values on the scale indicate more liberal views (in favour of pre-marital sex, of homosexuality, of abortion if serious defect in foetus, and against traditional gender roles), and higher values indicate more conservative views (against pre-marital sex, homosexuality and abortion if serious defect in foetus, and in favour of traditional gender roles).

Figure 1: Family Values Scale

NRS blog figure 1

Figure 2: Mean Attitudes on Family Values among Unaffiliated Respondents, by Country, 2008 ISSP

NRS blog figure 2

For each country studied, Figure 2 shows the unaffiliated respondents’ mean scores on the Family Values scale. Figure 3 in turn indicates the extent to which these mean scores deviate from those of each country’s affiliated population. Unaffiliated respondents in Denmark have the least different views from those of the affiliated population (both groups scoring almost the same on average). By contrast, the unaffiliated in New Zealand have the most different views (they score 0.898 lower on the scale than the affiliated Kiwi population, or in other words are much more liberal).

Figure 3: Deviation of Mean Family Values among Unaffiliated Respondents from Mean Family Values among Affiliated Respondents, by Country, 2008 ISSP 

NRS blog figure 3

Finally, Table 1 contains selected results from separate multilevel models generated respectively for religiously committed respondents (affiliated and attending religious services at least once a month), nominally affiliated respondents (affiliated but attending religious services less than once a month) and unaffiliated respondents. The Family Values scale acts as the dependent variable, and mean regional size of the unaffiliated group acts as the main independent variable. In these models, individuals are nested within regions (counties, states or provinces) and regions are nested within countries. The models control for age, gender, marital status, child present in household, employment status and level of education.

Table 1: Macro-Level Effect of Mean Proportional Size of the Unaffiliated Group on Family Values, among the Religiously Committed, Nominally Affiliated and Unaffiliated, 2008 ISSP

Model 1: Religiously committed

Model 2:

Nominally affiliated

Model 3: Unaffiliated

Mean no religion

(macro level: region)







N = 9,107 (Model 1). N = 18,232 (Model 2). N = 8,937 (Model 3). N. of region gr. = 244. N. of country gr. = 27. Multiple imputation estimates. †= p ≤ 0.10; * = p ≤ 0.05; ** = p ≤ 0.01; *** = p ≤ 0.001.

As we can see from the coefficients in Table 1, the unaffiliated are on average more liberal in regions with larger proportions of unaffiliated, compared with the unaffiliated in areas with smaller proportions. When we move between regions with 0% to 100% non-affiliation, we see an estimated decline of 0.176 on the Family Values scale among the unaffiliated. This significant decline is also present among the nominally affiliated, but not among the religiously committed. The larger differences in family values between the religiously committed and the unaffiliated in regions where the unaffiliated group is larger is confirmed in another model with the full sample and a cross-level interaction between the effect of being religiously committed on family values and the size of the unaffiliated group (results not shown here).

To summarize, the unaffiliated are more liberal when it comes to family values in regions where they form larger groups. This creates a greater divide between the unaffiliated and the religiously committed in such areas. With most regions seeing an increase in their numbers of religious ‘nones’, this indicates that if they are not already characterised by this greater divide, they may be so in the near future. Since an individual’s value orientation tends to influence her/his political choices as well as other behaviour, these greater divides could be leading to greater social cleavages between the actively religious and the unaffiliated in many Western societies.


Baker, Joseph O. Brian and Buster G. Smith. 2009. None Too Simple: Examining Issues of Religious Nonbelief and Nonbelonging in the United States. Journal for the Scientific Study of Religion 48 (4):719-733.
Bruce, Steve. 2011. Secularization: In Defence of an Unfashionable Theory. Oxford: Oxford University Press.
Finke, Roger and Amy Adamczyk. 2008. Cross-National Moral Beliefs: The Influence of National Religious Context. Sociological Quarterly 49(4): 617–652.
Hayes, Bernadette C. 1995. The Impact of Religious Identification on Political Attitudes: An International Comparison. Sociology of Religion 56(2): 177-194.
Layman, Geoffrey. 2001. The Great Divide: Religious and Cultural Conflict in American Party Politics. New York: Columbia University Press.
Nicolet, Sarah and Anke Tresch. 2009. Changing Religiosity, Changing Politics? The Influence of ‘Belonging’ and ‘Believing’ on Political Attitudes in Switzerland. Politics and Religion 2: 76-99.
Norris, Pippa and Ronald Inglehart. 2004. Sacred and Secular: Religion and Politics Worldwide. Cambridge University Press, New York.
Putnam, Robert D. and David E. Campbell. 2010. American Grace: How Religion Divides and Unites Us. New York: Simon & Schuster.
Raymond, Christopher. 2011. The Continued Salience of Religious Voting in the United States, Germany, and Great Britain. Electoral Studies 30: 125-135.
Wilkins-Laflamme, Sarah. (2014). Towards Religious Polarization? Time Effects on Religious Commitment in US, UK and Canadian Regions. Sociology of Religion 75(2): 284-308.
Zuckerman, Phil (ed.). 2009. Atheism and Secularity [2 volumes]. Santa Barbara, CA: ABC-CLIO.

Sarah Wilkins-Laflamme is a DPhil student in sociology at Nuffield College, University of Oxford. She completed her Bachelor’s and Master’s degrees in sociology at the University of Ottawa. In Fall 2014, she will begin a postdoctoral fellowship at the Centre d’études ethniques des universités montréalaises (University of Montreal) and at the Université du Québec à Montréal. Her research interests include sociology of religion, quantitative methods, ethnic, linguistic and cultural minorities, secularization and religious voting.

This entry was posted in Discussion, NSRN Blog, Research Questions and tagged , , , , , , . Bookmark the permalink.

1 Response to Religious ‘Nones’ Generally Have More Liberal Family Values in Areas of Greater Disaffiliation

  1. Sarah Wilkins-Laflamme: if anyone has any questions or would like more details related to the study, feel free to e-mail me at

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s