Skip to main content

Although it is empirically established that traditional religion enhances fertility, how it increases childbearing is not clear. This paper is an exploratory qualitative study investigating how religion influences decisions about intended fertility and family size. Most specifically, Michael Emerson and George Yancey ask how, if at all, do the religious understand children and family differently. Data from two online surveys with open- and closed-ended questions indicated that while the desire for children by the religious is tied to family expectations and an emotional desire for children, the desire of the nonreligious is tied to the individualistic obligations toward the new humans they bring into the world. They postulate that differing conceptualizations about the meaning of community whereby the religious focus on local communities and the nonreligious focus on global communities, lead to contrasting actions, including the higher fertility patterns of the religious. Their findings have important implications for American Christians as these findings explain more than family patterns but also provide insight into the modern culture war. Polarization between Christian conservatives and secularists in the United States may be partially tied to the differences in the communities where the two groups live. Dr. Michael O. Emerson is the provost of North Park University, and Dr. George Yancey is a professor of sociology at the University of North Texas.

Downward trends of human fertility have become commonplace among developed nations.1 This propensity has been termed “the second demographic transition,” which can be defined as the tendency of modernized societies to have persistent total fertility rates (TFR) below replacement level and the expectation that these rates will not rise above replacement level in the foreseeable future.2 Absent immigration, many developed countries do not have fertility rates high enough to replace their departing cohorts.3

Several reasons have been advanced to account for this second demographic transition. Some argue the transition is due to changing economic systems, where children are no longer the economic asset they were in premodern societies,4 and increased educational opportunities for women lead to postponed and thus decreased fertility.5 Others argue that cultural factors must be taken into consideration if we are to understand fully this transition, such as changes in the meaning of children, marriage, family, and increased value put on individual self-maximization, and psychological microprocesses.6

However, an important exception to the trend of below-replacement fertility is among the religious, defined most specifically as regular attenders of religious services and therefore regular participants in religious communities. A plethora of research finds that religiosity is positively correlated with fertility,7 with the highly religious typically maintaining fertility levels above replacement. In the United States, Christians in particular have higher fertility rate, especially if they have relatively high levels of religiosity.8 Yet how traditional religion enhances fertility is not at all clear. Most arguments for the higher religiosity/fertility link are based upon quantitative work unable to assess the micro-level processes at work within religious individuals.9 To understand fully how religion in general, and Christianity in particular, affects fertility it is important to learn how high religiosity shapes the conceptualization of childbearing. This can be particularly important in a society like the United States where Christians are shrinking as a percentage of the population since high levels of fertility should theoretically work against such a trend.

The paper is a qualitative exploratory study investigating how religion influences decisions about intended fertility and family size. We investigate possible ways the religious understand children and family differently, asking if this divergent understanding can be related to greater preferred family sizes. We focus in this study on intended childbearing. We do so because we seek to understand the thinking before children are born (once born, we would only be able access post-hoc explanations of behavior), so we more accurately discover possible ways religiosity influences the desire for children. The research includes those of non-Christian faiths to ensure adequate sample sizes, but we find general religious trends with important implications for Christians in the United States.10

Why are the Religious Fertile?

Several theorists have postulated arguments for the higher fertility of religious individuals. Hackett argues that congregational participation is a predictor of high fertility.11 He contends that there is a selection process by which those inclined to have children stay in high-fertility congregations while those inclined to remain childless leave those congregations. Reference groups may play a vital role in explaining religious fertility. Mahoney points out that importance of religion is positively related to fertility which may be tied to the specific spiritual beliefs that encourage childbearing.12 Wilcox contends that familism, or the tendency to make one’s family a central concern, is also tied to higher levels of fertility.13 Wilcox argues that familism convinces men to invest in a traditional family structure which can lead to higher levels of fertility.

These theories provide some possible explanation for the religion/fertility link. However, they only indirectly investigate the question of why the religious are more willing to engage in childbearing. To advance research on this question it would be useful to ask religious and nonreligious individuals directly how many children they want and why. By asking individuals about their intended fertility it is possible to develop novel theoretical direction with such research. Our goal is to conduct a preliminary analysis that produces potential theories about why religiosity is positively related to fertility.


Data for this study were collected from two online surveys. The second survey was conducted after analysis of the first survey raised additional questions for which we sought answers. Thus the second survey—though comprised of different respondents—was a follow-up to the first survey.

In the first survey, almost two thirds of the respondents (65%) were gathered though Amazon Mechanical Turk while the rest of the respondents were supplied by Survey Monkey.14 All respondents in the second survey were found through Survey Monkey. In both surveys we limited the respondents to only those who attended church at least weekly or no more than once a year, are under 45 years of age, and have no children. Amazon Mechanical Turk is a service whereby individuals perform online tasks for small amounts of money. We sent a survey that eliminated unqualified candidates and only allowed those who were qualified to complete the entire survey. Survey Monkey worked a bit differently in that it locates respondents and sends them to our survey. We paid Survey Monkey, and the company in turn compensated their respondents. We used screens to eliminate those who did not qualify by age and whether they had children. However, we still had to screen by religious attendance in the actual survey. Yet this additional screening helped to make Survey Monkey more efficient, which is why we used it exclusively for our second survey. The respondents gathered through Survey Monkey are significantly more religious, older, more female, wealthier, and more likely to be married. The samples differ from each other, but since we are interested in a qualitative explanation of the religious effects on fertility decisions, having respondents from two different populations provides us the opportunity to see if our results are due to the unique nature of one of these populations.

One of the major limitations is the lack of randomness in our sample. Of course research does not always require random samples and much research has been conducted with samples of university students who are much less diverse in key demographic factors (such as education and region) than our sample. Our research should be seen as an exploratory-grounded theoretical approach that helps us discover possible distinctions between the religious and nonreligious that may account for fertility difference between groups. Both Amazon Mechanical Turk and Survey Monkey draw samples by offering individuals money for their time. Survey Monkey offers to provide money to non-profits, which may help account for different demographics between the two samples.15

To assess fertility intentions it is necessary to ask questions only of those who do not yet have children; yet, it is fair to speculate about how not having children may bias our results. We suspect that those without children are limited in their knowledge about the costs of children and that many of them have self-selected themselves into our sample due to a potential disregard for the prospects of raising children. As such it is likely that our sample will ironically be both more pessimistic about children and ignorant about the difficulty of raising them. Regardless of the particular bias of using a childless sample, we acknowledge that this bias, as well as the other biases discussed above, will influence our results. Thus there is limited generalizability due to the nonrandom nature of our samples. Yet, patterns emerging from our work can illustrate possible explanations that should be subject to future assessment. After exploring our results we will venture to theorize about the social processes taking place. Future work with more representative samples can be applied to test our assertions.


In the first survey we found 393 individuals of which 100 we categorized as religious. By that we mean that they attended a religious service at least once a week. We asked the respondents whether or not they wanted children. If they said yes then we provided an open-ended question asking why they wanted children. We also asked them why they wanted the particular number of children they indicated. If they said they did not want children then we provided an open-ended question asking why. For all respondents we provided open-ended questions about the perceived costs and benefits of children as well as potential influence of their friends, education, gender roles, and religion on their fertility decisions.

We used an open coding strategy whereby codes were identified in answers to the open-ended questions. After 100 respondents a set of codes emerged that was applied to the rest of the sample. Codes were not mutually exclusive and some of the answers included several codes. The graduate student’s coding of the first 100 students was compared to the coding of one of the authors. The percent of codes that matched between the two coders ranged between a low of 83.2% match to several that matched 100%. The average match between all codes was 95.9% with a standard deviation of 4.2%. In the 4.1% of the cases where there was not coder agreement, the code assigned by the senior researcher was used. The respondents in the sample are divided into religious (attend church at least once a week) and nonreligious (attend church no more than once a year). Once the codes were set we examined if certain codes were more common among the religious population.

A similar strategy was used with the second survey. In that survey there were only two open-ended questions. An initial question asked the respondents if they liked children. If they stated “yes” then they were asked why they liked children. If they stated “no” they were asked why they did not like children. The answers to both open-ended questions were coded. The coders’ agreement on each particular code ranged from 85.1% to 99.6% with a mean of 97.5% and a standard deviation of 3%. The code of looking at life from a child’s perspective had the lowest percent of agreement, but that was an outlier as the next lowest percent of agreement for a code was 95.3%. We found that the coding for both coders for “child’s perspective” was positively correlated to whether the respondent is religious. We do not obtain different results due to this outlier code even as we compared the codes given to respondents in our religious group to those given in our nonreligious group.

In addition to our open-ended questions we also asked the respondents in the second survey a series of closed-ended questions intended to assess their attitudes toward family and the community. On a 1 to 5 point scale respondents were asked whether having a happy family, having children, success at work and having money is important. They were also asked if life’s purpose was to develop their own talent, serve others, both or neither. On a 1 to 5 point scale they were also asked whether they agreed that they enjoyed having guests, enjoyed sharing what they have, do not like to lend things to people (even good friends), do not mind giving rides to those without a car, do not like having anyone home when they are not there, and enjoy donating to charities. Finally, we asked them questions to document the number of social and community organizations where they have membership. These key closed-ended questions allowed us to assess if family importance, generosity, societal participation are important for shaping attitudes toward children and childbearing.


Just as it has been observed in larger society, religious individuals in both surveys indicated a higher desire for children. In the first survey the answers given to the question of how many children were desired indicated an estimated desired16 TFR of 2.2 for those who attend church at least once a week17 compared to 1.2 for those who do not attend church more than once a year. The second survey produced a TFR of 2.5 for weekly church attenders compare to 0.8 for those attending yearly or less. Although we do not have a probability sample in either survey, our respondents do indicate the same propensity of the larger society for the religious to desire significantly more children. This provides us an opportunity to explore possible answers for this intended fertility difference.

Religious Definition of Family

Our major finding centers on how the religious and nonreligious have distinct definitions of family, which in turn explains important differences in the fertility desires between the religious and nonreligious. For the religious, children are more central to how they define their families. It is not surprising that, based on analysis of the open-ended responses, the religious are more likely to see children as God’s blessing (5.3% v. 0%: p < .01) and their faith compels them to desire children (21.1% v. 1.5%: p < .001). However there are more than religious justifications. The religious also are more likely to state that one of the benefits of children is that children bring happiness into their lives (30.0% v. 18.8%: p < .05). They were more likely to mention loving children in any of their answers we well (41.1% v. 28.4%: p < .05).18 The way some religious respondents talk about their love for children is insightful:

As a Catholic, I am open to God’s plan for my life and that is especially true in terms of family planning. I love kids and would love a big family, but we will make the decision whether to have one more kid as it arises. (Female, Age 18-25)

They make the purpose of your life complete. Carry on your family. Makes family complete. (Male, Age 18-25)

I love kids. And new life is so beautiful. Looking into the eyes of an infant, you’re clearly looking at a miracle. Watching them grow into wonderful people and helping them along

the way is what I’ve always wanted to do. (Female, Age 18-25)

Religious individuals talked of loving children as a way to complete their family and/or to help fulfill their duties in a family. Because children are such a central part of how they defined their families it is not surprising that when asked why they want their stated number of children, religious individuals were much more likely to state that they want a large family (10.5% v. 1.0%: p < .001).19 This is not merely an individualistic decision for some as the religious were also more likely to desire children due to spouse and partner influence (7.4% v. 2.5%: p < .05).20

For religious individuals the notion of family is incomplete without the inclusion of children. Several respondents indicated the degree to which their notion of family centered on having children.

…children are a part of the marriage and family life cycle. If God blesses you with them, that is good. However that is not always the case and if adoption is a viable option that would be good too (Female, Age 26-35)

For me being a father completes the concept of masculinity. The roles of leadership, provider, and other masculine concepts is mainly played out in the raising of children and family life. (Male, Age 18-25)

The family is the fundamental unit of society. Marriage is between a man and a woman and children add to that family unit. Children are pure and come from God’s presence. It is then the parents’ opportunity to raise, teach, and nurture those children. By having children, parents can further their legacy and better the world by teaching those things which are right and good. (Male, Age 26-35)

The religious vision of family, where kids are essential, undoubtedly help produce a higher fertility desire among religious relative to nonreligious individuals. We also found evidence of an increased fondness toward children among the religious. Some of these responses indicate that this fondness may be tied to the realization of achieving the expected child-centered family.

Non-Religious Definition of Family

The nonreligious are less desirous of children than the religious, and when they do desire children they want fewer. Our findings suggest the non-religious do not define families as being centered on children to the same degree as religious individuals. Even among those that desired children, children were less likely to be seen with the same affinity that religious indicated in their answers. Children were not the goal of the nonreligious but rather raising them was the key component to how they defined their desire for children. Thus, when asked about the benefits of children the nonreligious were more likely to state taking care of children (13.2% v. 5.3%: p < .05), have the ability to give to someone else (27.0% v. 16.8%: p < .1) and watch children mature (16.2% v. 6.3%: p < .05) as fertility benefits. In comparison to the religious, the focus of the nonreligious is more on the obligation they have for their children and less on whether they gain emotional affection from them. The comments of some nonreligious respondents reveal their sense of obligation toward the prospect of having children:

I want to be able to share my life with a child to love, teach, care for & be a good example for another contributing human being into our society/world. (Female, Age 36-45)

I want to make an impact on the world in some small way and I think having children is the surest way to do that. (Female, Age 18-25)

Helping them through life the best I can and trying to raise a better person than I turned out to be. (Male, Age 18-25)

When our nonreligious respondents wanted children they were more likely to enunciate altruistic motivations. They envisioned having children as an opportunity to teach the next generation, to provide good humans for society and to pass on the lessons they have learned. Furthermore, we picked up a quantity versus quality distinction whereby the nonreligious conceptualize having fewer children as an opportunity to raise more “quality” children. Among respondents wanting children the nonreligious desired fewer children,21 were less concerned about the child’s role in the family, more concerned about the child’s personal development and more concerned about how their children would improve society.22 To meet the needs of the children, a focus on having fewer children to take care of rather than increasing one’s joy with more children is a key distinction between the religious and nonreligious in our sample.

A simplistic way to envision this difference is that the religious family exists to produce children. There are other functions of the family, to be sure, but producing children is central for religious individuals. For the nonreligious the family is a tool to raise children. This indicates the need for more focus on the children one begets. When this distinction is seen, it becomes easier to see why the religious may have higher levels of affection toward children. That affection is tied to their satisfaction in meeting a cultural desire to use the family to produce children and the centrality of that purpose. We had not anticipated this difference in expression of affection toward children by the religious and thus had not asked our respondents about the origin of that affection in our original survey. To investigate this difference fully, we conducted our second survey designed to look at why the religious have a higher emotional affiliation to children and how that affiliation is related to the importance they place on families.

Affection of Children and the Religious

The second survey is designed to investigate why respondents state their affection for children and if it is connected to their family attitudes. Since we had more success finding religious individuals through Survey Monkey, we decided to use only Survey Monkey audience for this survey. With this audience our final sample consisted of 276 respondents of which only 47 were religious. In addition to religious respondents having a higher estimated desired TFR,23 they also are more likely to state that they liked children (83.0% v. 55.5%: p < .001). While liking children is strongly correlated with the number of children desired (r = .429), whether the respondent is religious is slightly more strongly correlated with the number of children desired (r = .466).

Respondents provided several reasons why they liked children. Reasons that often came up were that children were fun, the respondent enjoyed seeing life from a child’s perspective, and they wanted to teach children and children can teach us.24 These three reasons were given more by religious respondents than by nonreligious respondents, but only when we look at the entire sample. In all three reasons significant differences between religious and nonreligious individuals disappear when we eliminated those who do not like children from our sample. But the basic propensity of religious individuals to enjoy children due to seeing life from a child’s perspective still remains in the sample of only respondents, religious and nonreligious, who liked children.

We see how these differences manifest themselves in the open-ended answers of religious respondents. Our religious respondents indicated a wish to interact with the general character of children and in that interaction developed their emotional affection of children:

They are high-energy. Teachable. Have relatively unformed views of the world which I would not expect an adult to have, but can challenge my ideas. They ask why a lot. If something is not fun they will tell you. (Female, Age 18-25)

I would like to be a positive influence on them. They are fun to be around, creative and have new perspectives on the world. They are unhindered by adult cultural habits, comparatively. They like to think the best of everyone (well, most do anyway). (Male, Age 18-25)

I enjoy educating children whenever the opportunity arises. Helping them is very important to me, and I try to enlighten them on subjects and concerns that are not necessarily learned in a school environment. It’s also fun just to listen to them and play a simple game with them too. (Male, Age 36-45)

Religious individuals, more than nonreligious individuals, want to be in the company of children both to teach them and to experience their nature. They desire an atmosphere where interaction of children is commonplace. This implies less sophisticated conversation and communication then one would expect at an academic organization, business or other career-oriented atmosphere. Instead of these adult-oriented environments, respondents enunciate a longing for a family culture with individuals of varying ages and life experiences to teach and learn from each other as well as to connect through primary relationships.

Correlational analysis indicated that the same characteristics of those who liked children are also connected to religious individuals. Using our closed-ended answers we found that those who liked children prioritized having a happy family (r = .169), serving other people (r = .253), having guests in their home (r = .204), and lending items to others (r = .152). We also found that religious individuals also prioritized having a happy family (r = .156), serving other people (r = .124), having guests in their home (r = .122), and lending items to others (r = .116). These similarities, along with the reasons why individuals have affection toward children, help paint a picture of how religiosity may interact with affection to increase fertility desires. Individuals who like, and thus want, children prioritize their families above their careers, seek to give to others, and enjoy looking at life from the perspective of children. They prioritize individualized service to others in a willingness to entertain guests and lend items to friends. In this sense the religious not only prioritize adding more children to their family but also value aiding others in their local community. In our sample, these are also the characteristics of individuals with relatively high religiosity. It suggests a more traditional lifestyle mindset whereby children are a vital part of the family religious individuals want to create. Children are not only seen as additions to this lifestyle to be trained by adults but also as those who are teachers of adults, bringing a fresh perspective to the family as some of our religious respondents suggest:

Children keep a person young. Their innocence also provides a great perspective to the closed-mindedness of many adults. (Male, Age 18-25)

I like their unique viewpoint on life and their love of fun. I also enjoy being able to influence kids in a positive way. (Female, Age 36-45)

For the religious the family is a community and children play an important role in that community. They are not only the receivers of care and love but they are also providers of youth and a needed perspective for that family. But this community is not necessarily limited to the family as those wanting children also desire to serve others and invite guest. Thus they also display a higher priority to share with those in a local community. They do not limit their commitments to their immediate families but appear to have a strong interest in helping those in their own social networks. This indicates a focus away from larger social and institutional issues and toward relationships in the context of a more localized community.

This is in contrast to nonreligious respondents. As noted, our survey showed that they have a greater focus on preparing individuals for the larger society. The nonreligious concentrate less on serving those in their immediate social circles. We did not ask the respondents about larger societal issues and thus have no way to document this attitude directly, but it is telling that the non-religious focused upon raising good children for society. This implies a concern for a larger society outside of merely their social circles and a more encompassing vision of their community. Indeed one of the reasons given by some of the nonreligious for not having children was due to overpopulation – a reason that fits with the notion that they are more concerned with larger social goals than immediate family desires.

Differing Religious and Non-Religious Communities

Our preliminary analysis suggests that differing attitudes toward family and community is a potentially powerful explanation for why religiosity is positively correlated with childbearing. Religious individuals appear to have a vision of family that focuses on primary relationships and envision their immediate family as their most important community. Their focus in on individualized service which may partially distract them from prioritizing issues beyond their local community. The nonreligious who choose to have children focus on producing good individuals for the larger society. The prioritization of the local family over the larger society would not only account for why more religious individuals want children in the first place but also why they want more children. A large family increases the local community which is important for the religious while more children for the non-religious may place a strain on their ability to train their children to be good citizens. The most powerful distinction between religious and nonreligious fertility desires is linked to differential values which may lead to distinct answers to quantity versus quality concerns about children.

Many nonreligious find alternative communities to their local nuclear families. Some scholars argue that those with traditional religiosity are less likely to be able to deal with modern global issues.25 Furthermore, Armfield and Holbert argue that the nonreligious are more likely to take advantage of internet resources to make connections with like-minded individuals from different regions and foreign countries.26 This should not be surprising since early humanist literature emphasizes a concern for the global society.27 On the other hand, previous research indicates that religiosity generally does not encourage political activism.28 Furthermore, theological conservatives tend to have lower levels of education.29 It is likely that political activism and education exposes nonreligious individuals to ideas about the larger society and allows those individuals to invest their energy into working for national and global issues rather than into a nuclear family. Thus it is not surprising that conservative Christian theology is inversely related to belonging to social activism groups such as environmentalist groups and feminism.30

Religious/nonreligious differences in fertility can signal other important social distinctions. Individualism has been argued as an important factor explaining decreasing fertility.31 Individualism is tied to family formation since decisions surrounding that formation are linked to the desire to live self-fulfilling lives. Whereas historically it was a given that a person would attempt to engage in the childbearing process, in modern society one can choose to not have children with little or no social costs. Therefore those who decide that children are too costly can forego parenthood. However, in religious communities there are religious values that de-emphasize individualism and promote child rearing.32 Our findings about the focus of religious groups on child-centered families and local communities also reinforce Hackett’s arguments about the importance of reference groups in shaping Christians’ fertility decisions.33 In Christian communities, particularly conservative Christian communities, the focus is not on individuals’ desires as much as building the local religious community. Child-centered families are seen as part of that community.

Hall points to another potential cultural difference that may help to shape fertility difference with his discussion of the risk society.34 Marriage and childbearing are seen as risks to self-fulfillment and in modern society that risk can be managed by bypassing either or both options.35 Such decisions may lower the anxiety tied to such risks. Rising individualism can be tied to higher preoccupation with reduction of risk.36 If this is accurate then the social expectations tied to a child- centered family formation, which is connected to lower levels of individualism, can help to mitigate this anxiety. A “fertile” Christian community can be desirable in countries undergoing a second demographic transition.

The Notion of Community and Fertility

Our premise is that differences in fertility by religiosity are tied to larger cultural distinctions. Thus, we postulate that the driving force behind different notions of fertility are important value differences between religious and nonreligious individuals in how they view the nature of families and define their vital communities. The child-centered families of the religious fit well in their definition of community as local and their downplaying of individualism. This is in contrast to a nonreligious definition of family where children are to be nurtured for the larger global community but only if an individual desires to do so. With the local family as the focus of religious individuals, having more children is seen as useful for increasing the number of individuals within the nuclear family and increasing the pleasure they gain from the family.

These differing notions of community reveal more than mere reproductive differences. Previous research has suggested that the United States is bifocal with secular progressive and religious conservative components.37 Our findings build on these arguments by connecting reproductive differences to divergent understandings about family and community. We envision this conflict as more than mere political disagreement, but rather it is a fight between those with distinct understandings of families and communities. In the United States this conflict generally plays itself out between the majority religion—Christianity—and more secular elements in our society. The nonreligious community is built upon individualism and global concerns. Lower fertility is tied to choices that minimize risks and maximize quality per child. These individuals may be more likely to meet their needs for community and relationships in political and social activism than the conservative Christian groups. Much of this activism may not be as tightly linked to their vision of individualized families, but it is natural that they would also promote activism that supports an individualistic, global communal vision.

Christian communities, particularly more conservative communities, are more likely to be tied to local families and communities. The risks of childrearing and marriage may be reduced by local support and the child-centered families encouraged by religious values. While clearly many conservative Christians are very politically active, their activism tends to focus on issues that may enhance an atmosphere for larger families, such as the pro-life movement.38 Debates over reproductive freedom illustrate contrasts between those who value child-centered families with those who envision children as optional. More recently the battle over same-sex marriage also features contrasting visions of family structures. The value of child creation within Christian families is not compatible with sterile same-sex unions, but those unions fit well in an individualist framework for families. Much of what are called social issues are reflections of differing visions of families and communities.

Previous research has suggested contrasting values between the religious and non-religious. For example, Haidt pointed out a possible contrast between what we might simplify as conservative religious individuals and progressive nonreligious individuals.39 He argues that progressives tend to form a morality highly based on norms of fairness and care for others. This morality enables them to conceive of a higher degree of a global concern than their more conservative counterparts. Conservatives are not devoid of fairness and care values, but they also include in their morality a mix of elements such as loyalty, authority and sanctity. These elements make it more important for conservatives to pay attention to their family and local social network as much, or more, than the global community. Another group of researchers note the emphasis that religious individuals place on local traditionalist family structures.40 Other scholars have argued that the religious use private family structures to maintain their values in spite of the influences from larger culture.41

To be clear, our work does not directly test any of these religious/nonreligious differences in social values. But what our work does suggest is that these differences may be related to a distinct notion of community and family that reinforces patterns of thought among Christians and secular individuals. Social networks have been discussed as a possible reason for religious fertility effects. Christians with more traditionalist perspectives and ties to local communities are likely to find their values on what constitute family and community reinforced by others in that local community. Those with more secular notions of a globalist community may find reinforcement of their beliefs, even if not in the local community, due to connections they have online or through other means of long-distance communication. In this way modern technology has made the development of non-local community more plausible and may have aided in the rise of modernist family structures.


In this work we have found that the divergent “community” orientations of the religious and non-religious produce vitally important differences in desired family size. Deeper understanding of these differences will be of much importance in the future. Beyond the particular implications our observation about different values of community offers possible pathways for societies seeking to reverse shrinking population trends. Many of these countries are dominated by the single non-religious community that struggles to maintain its population. However, the development of ideals valuing local community may play a role in reversing population declines.

In the United States there are many fewer concerns about shrinking populations, in part due to the persistence of Christian religiosity. However, this theory is also useful for its implication that we have dual communities that often struggle to live in the same society. We have speculated that the polarization between conservative Christian and progressive secularists in the United States may be partially tied to differences in the communities where the two groups live. Whereas changing notions of community may be useful in many European countries for understanding stagnant or shrinking population growth, it may have the most use in the United States for understanding social and cultural conflict.

Conflict between modernist secular perspectives and traditional, often Christian, values has been described as a culture war.42 Our work indicates that this culture war is not merely reinforced by differences in political and cultural ideologies but also by distinctive ideals about family and community. Individuals existing in social networks with others with similar ideals of family and community can deepen the divide in our culture war. The local-focused and child-centered family can support a perspective whereby deference to a traditional authority structure can provide order to that community. In this way, family and community values of conservative Christians can reinforce their conservative social viewpoint. On the other hand, the individualist, global ideals promoted by the nonreligious may reinforce a cultural progressive agenda centered in a critique of our larger society. This can help explain how the more cosmopolitan family perspectives of the nonreligious can lend support to more progressive values within ideological constructions such as feminism, environmentalism, and critical theories.

For example, some have observed a growing secular ideology emerging in the United States.43 This suggests that the secular community is gaining in strength. Perhaps the possibility of a European-style society may emerge over time. Yet as we document within that secular community, quality and not quantity of children is the presiding value. Thus one may suspect that the secular communities will not maintain themselves. However, given that about sixty percent of Christians decrease frequency of church attendance after going to college, one can see that the secular community can gain enough adherents to maintain a trend toward secularization.44

Our work suggests that conservative Christian communities have an inherited advantage over progressive communities with a child-centered notion of families that allows them to procreate at a higher rate. But in the quantity versus quality calculation, there is a potential problem of having many children who are not properly prepared for the intellectual and cultural challenges of modern society. This possibility may explain the inability of conservative Christians to retain their children as believers. It is beyond the scope of this paper to explore Christian socialization; however, the tendency of conservative Christians to add children to one’s family due to the pure joy of having children does suggest that care may not be taken to make sure these children are sufficiently socialized into the religious values of the parents. Long-term changes in both secularization and the current culture war can emerge if conservative Christians are able to find ways not only to become enthusiastic about having a high number of children but also to develop the concern of the nonreligious in raising these children to possess the desired values of the parents. Perhaps a revitalization of the local community, which conservative Christians tend to value, as a socialization agent can provide a partial answer for such Christians.

Cite this article
George Yancey and Michael O. Emerson, “Having Kids: Assessing Differences in Fertility Desires between Religious and Nonreligious Individuals”, Christian Scholar’s Review, 47:3 , 263-280


  1. Tineke Fokkema and Ingrid Esveldt, “Motivation to have Children in Europe,” in People, Population Change and Policies, ed. Charlotte Hohn, Dragana Avramov and Irena E. Kotowska (Springer Netherlands, 2008); Catherine Krull and Frank Trovato, “Where Have All the Children Gone?: Quebec’s Fertility Decline: 1941-1991,” Canadian Studies in Population 30.1 (2003): 193-220; Meilin Lee, “Transition to Below Replacement Fertility and Policy Response in Taiwan,” The Japanese Journal of Population 7.1 (2009): 71-86; Philip S. Morgan, “Is Low Fertility a Twenty-First-Century Demographic Crisis?,” Demography 40.4 (2003): 589-603.
  2. R. Lesthaeghe and K. Neels, “From the First to the Second Demographic Transition: An Interpretation of the Spatial Continuity of Demographic Innovation in France, Belgium and Switzerland,” European Journal of Population 18 (2002): 325-60; Dirk J. van de Kaa, “Europe’s Second Demographic Transition,” Population Bulletin 42.1 (1987): 1-59.
  3. Hans-Peter Kohler, Francesco C. Billari, and Jose Antonio Ortega, “The Emergence of Lowest-Low Fertility in Europe During the 1990s,” Population and Development Review 28.4 (2002): 641-80; Lareen A. Newman and Graeme J. Hugo, “Women’s Fertility, Religion and Education in a Low-Fertility Population: Evidence from South Australia,” Journal of Population Research 23.1 (2006): 41-66; Dirk J. van de Kaa, “Europe’s Second Demographic Transition,” Population Bulletin 42.1 (1987): 1-59.
  4. Gary Becker, A Treatise on The Family (Cambridge, MA: Harvard University Press, 1981); Timothy T. Schwartz, “’Children are the Wealth of the Poor.’ Pronatalism and the Economic Utility of Children in Jean Rabel, Haiti,” Research in Economic Anthropology 22 (2003): 61-105.
  5. Alaka M. Basu, “Why Does Education Lead to Lower Fertility?: A Critical Review of Some of the Possibilities,” World Development 30.10 (2002): 1779-1790; Jean Dreze and Murthi Murthi, “Fertility, Education and Development: Evidence from India,” Population and Development Review 27.1 (2001): 33-63; Trude Lappegard and Marit Ronsen, “The Multifaceted Impact of Education on Entry into Motherhood,” European Journal of Population 21.1 (2005): 31-49; Teresa Castro Martin, “Women’s Education and Fertility: Results from 26 Demographic and Health Surveys,” Studies in Family Planning 26.4 (1995): 187-202; Samuel H. Preston and Caroline S. Hartnet, “The Future of American Fertility,” in National Bureau of Economic Research Conference on Demography (Napa Valley, CA, 2008).
  6. Francesco C. Billari, “Synthetic Fertility Measures and the Search for Commonalities,” Population Studies 58.1 (2004): 84-85; Dirk J. van de Kaa, “The True Commonality: In Reflexive Modern Societies Fertility is a Derivative,” Population Studies 58.1 (2004): 77-81.
  7. Alicia Adsera, “Marital Fertility and Religion in Spain,” Population Studies 60.2 (2006): 205-21; F. Althaus, “Differences in Fertility of Catholics and Protestants are Related to Timing and Prevalence of Marriage,” Family Planning Perspectives 24.5 (1992): 234-35; Kevin J. Christiano, “Religion and the Family in Modern American Culture,” in Family, Religion, and Social Change in Diverse Societies, eds. Sharon K. Houseknecht and Jerry G. Pankhurst (Oxford, UK: Oxford University Press, 2000), 43-78; Tomas Frejka and Charles F. Westoff, “Religion, Religiousness and Fertility in the US and in Europe,” European Journal of Population 24 (2008): 5-31; Tim B. Heaton, “Religious Influences on Mormon Fertility: Cross-National Comparisons,” Review of Religious Research 30.4 (1989): 401-411; Tim B. Heaton and Kristen L. Goodman, “Religion and Family Formation,” Review of Religious Research 26.4 (1985): 343-59; Lareen A. Newman and Graeme J. Hugo, “Women’s Fertility, Religion and Education in a Low-Fertility Population: Evidence from South Australia,” Journal of Population Research 23.1 (2006): 41-66; Iqbal M. Zafar, Fawad Asif, and Sutan Adil, “Religiosity as a Factor of Fertility and Contraceptive Behavior in Pakistan,” Journal of Applied Science 3.1 (2003): 158-66.
  8. Kathryn Joyce, Quiverfull: Inside the Christian Patriarchy Movement (Boston, MA: Beacon Press, 2009).
  9. Sarah R. Hayford and S. Philip Morgan, “Religiosity and Fertility in the United States: The Role of Fertility Intentions,” Social Forces 86.3 (2008): 1163-88; Tim B. Heaton, “How Does Religion Influence Fertility?: The Case of Mormons,” Journal for the Scientific Study of Religion 25.2 (1986): 248-58; Kevin McQuillan, “When Does Religion Influence Fertility?,” Population and Development Review 30.1 (2004): 25-56.
  10. In our total sample 70% of the respondents are Christian. Among those who attend religious service at least once a week, 73% of the respondents are Christian. In our findings we will note any times the Christians sample differs from the non-Christian sample among the religious.
  11. Conrad Peter Hackett, Religion and Fertility in the United States: The Influence of Affiliation, Region, and Congregation (Princeton, NJ: Princeton University Press, 2008).
  12. Annette Mahoney, “Religion in Families, 1999–2009: A Relational Spirituality Framework.” Journal of Marriage and Family 72.4 (2010): 805-27.
  13. Bradford W. Wilcox, Soft Patriarchs, New Men: How Christianity Shapes Fathers and Husbands (Chicago, IL: University of Chicago Press, 2004).
  14. We collected the majority of the first survey before discovering that it was more economical to purchase responses from Survey Monkey. Given the qualitative nature of this project, multiple sources of respondents enhances the diversity of our sample.
  15. Those in the Survey Monkey sample are more female, religious, older, wealthier, and more likely to be married than those in the Amazon Mechanical Turk. These characteristics suggests a more traditional outlook on cultural and family issues than Amazon Mechanical Turk respondents. Religiosity/non-religiosity differences found in both samples supplies us with more confidence that these findings are not merely tied to the demographic makeup of a particular sample.
  16. This is an estimate since our highest category the respondents were given was “more than five children”. We assigned six children for those who checked this category. This may create a slightly lower estimated TFR than what the respondents actually want but since only 9 respondents, all religious individuals, used this category possible deflation of the desired TFR is limited to religious individuals and not likely to be very high.
  17. There was no significant difference between the religious individuals in the Amazon Mechanical Turk sample and the Survey Money samples (2.387 v. 2.145: ns). There was a difference between the nonreligious individuals in these samples as the nonreligious individuals in the Amazon Mechanical Turk sample desired more children that those in the Survey Monkey sample (1.336 v. .882: p < .01). However, even in the Amazon Mechanical Turk sample religious respondents have a significantly higher desire for children (2.387 v. 1.336: p < .001)
  18. Yet, the level of loving statements in the Amazon Turk sample was quite similar between religious and nonreligious samples in large part due to the lower percentage of religious individuals (30.8%) who made such statements. Given that there were only 26 religious individuals left in that sample when we make this comparison we are cautious about drawing conclusions.
  19. However, this was a stronger tendency among non-Christian religious individuals as they were significantly more likely state that they wanted a larger family than Christian religious individuals (21.7% v. 6.5%: p < .05).
  20. These last two differences were found in both the Amazon Mechanical Turk and the Survey Monkey samples.
  21. Among respondents who wanted at least one child the religious still wanted more children than the nonreligious (2.846 v. 1.884; P < .001). This difference was significant in both the Amazon Turk and the Survey Monkey subsamples.
  22. However, a contradictory finding to this notion of altruism is that a small percentage (4.9%) of nonreligious respondents discussed the desire to pass their genes to the next generation while none of the religious respondents exhibited such a desire (p < .05).
  23. Five individuals, four of them religious, indicated that they wanted more than five children indicating a slight potential deflation effect on estimated desired TFR.
  24. With all three reasons religious individuals were more likely to state these factors over nonreligious individuals at least p < .05.
  25. Liselotte Frisk and Peter Nynäs, “Characteristics of Contemporary Religious Change: Globalization, Neoliberalism, and Interpretative Tendencies,” in Post-Secular Society (New Brunswick, NJ: Transaction Publishers, 2012), 47-70; Daniel A. Helminiak, Spirituality for our Global Community: Beyond Traditional Religion to a World at Peace (Lanham, MD: Rowman & Littlefield Publishers, 2008); Nobutaka Inque, “Globalization and Religion: The Cases of Japan and Korea,” in Religion, Globalization and Culture, eds. Peter Beyer and Lori Bearman (Danvers, MA: Brill, 2007), 431.
  26. Greg G. Armfield and R. Lance Holbert, “The Relationship Between Religiosity and Internet Use,” Journal of Media and Religion 2.3 (2003): 129-44.
  27. Richard Cimino and Christopher Smith, “Secular Humanism and Atheism beyond Progressive Secularism,” Sociology of Religion 68.4 (2007): 407-24.
  28. Sky L. Ammann, “Is There an Attendance Effect? Examining the Causal Link Between Religious Attendance and Political Participation,” American Politics Research (2014): 1532673X14533720; Kenneth D. Wald, Lyman A. Kellstedt, and David C. Leege, “Church Involvement and Political Behavior,” in Rediscovering the Religious Factor in American Politics, ed. David C. Leege (Abingdom, UK: Routledge, 1993), 121-38.
  29. Alfred Darnell and Darren E. Sherkat, “The Impact of Protestant Fundamentalism on Educational Attainment,” American Sociological Review (1997): 306-15; Barry A. Kosmin and Ariela Keysar, American Religious Identification Survey (Hartford, CT: Institute for the Study of Secularism in Society & Culture, 2008).
  30. Steven E. Barkan, “Explaining Public Support for the Environmental Movement: A Civic Voluntarism Model,” Social Science Quarterly 85.4 (2004): 913-37; James L. Guth, John C. Green, Lyman A. Kellstedt, and Corwin E. Smidt, “Faith and the Environment: Religious Beliefs and Attitudes on Environmental Policy,” American Journal of Political Science (1995): 364-82; Clyde Wilcox, “Feminism and Anti-feminism Among Evangelical Women,” The Western Political Quarterly (1989): 147-60.
  31. John C. Caldwell, “Toward a Restatement of Demographic Transition Theory,” Population and Development Review (1976): 321-66; Ron Lesthaeghe and Johan Surkyn, “Cultural Dynamics and Economic Theories of Fertility Change,” Population and Development Review (1988): 1-45.
  32. Dick Houtman and Peter Mascini, “Why do Churches Become Empty, While New Age Grows? Secularization and Religious Change in the Netherlands,” Journal for the Scientific Study of Religion 41.3 (2002): 455-73; Annette Mahoney, “Religion in Families, 1999–2009: A Relational Spirituality Framework,” Journal of Marriage and Family 72.4 (2010): 805-27; Vassilis Saroglou, Vanessa Delpierre, and Rebecca Dernelle, “Values and religiosity: A Meta-analysis of Studies using Schwartz’s Model,” Personality and Individual Differences 37.4 (2004): 721-34.
  33. Conrad Peter Hackett, Religion and Fertility in the United States: The Influence of Affiliation, Region, and Congregation (Princeton, NJ: Princeton University Press, 2008).
  34. David R. Hall, “Risk Society and the Second Demographic Transition,” Canadian Studies in Population 29.2 (2002): 173-93.
  35. Anthony Giddens, The Consequences of Modernity, (Hoboken, NJ: John Wiley & Sons, 2013).
  36. Iain Wilkinson, Anxiety in a ‘Risk’ Society (Abingdon, UK: Routledge, 2002).
  37. Andrew Gelman, Red State, Blue State, Rich State, Poor State: Why Americans Vote the Way they do (Princeton, NJ: Princeton University Press, 2009); Alan I. Abramowitz and Kyle L. Saunders, “Is Polarization a Myth?,” The Journal of Politics 70.2 (2008): 542-55; James Davison Hunter, Culture Wars: The Struggle to Control the Family, Art, Education, Law, and Politics in America, (New York, NY: Basic Books, 1992); Shanto Iyengar and Kyu S. Hahn, “Red Media, Blue Media: Evidence of Ideological Selectivity in Media use,” Journal of Communication 59.1 (2009): 19-39; Geoffrey Layman, The Great Divide: Religious and Cultural Conflict in American Party Politics (New York, NY: Columbia University Press, 2001).
  38. Carol J. C. Maxwell, Pro-Life Activists in America: Meaning, Motivation, and Direct Action (Cambridge, UK: Cambridge University Press, 2002); Clyde Wilcox and Leopoldo Gomez, “The Christian Right and the Pro-life Movement: An Analysis of the Sources of Political Support,” Review of Religious Research (1990): 380-89.
  39. Jonathan Haidt, The Righteous Mind: Why Good People are Divided by Politics and Religion (New York: Vintage Books, 2013).
  40. Penny Edgell, Religion and Family in a Changing Society (Princeton, NJ: Princeton University Press, 2006); Neil Smelser, Toward a Theory of Modernization (Englewood Cliffs, NJ: Prentice- Hall, 1964); Judith Stacey, In the Name of the Family: Rethinking Family Values in the Postmodern Age (Boston, MA: Beacon Press, 1996).
  41. Jerome L. Himmelstein, “The Social Basis of Antifeminism: Religious Networks and Culture,” Journal for the Scientific Study of Religion (1986): 1-15; Peer Scheepers, Manfred Te Grotenhuis, and Frans Van Der Slik, “Education, Religiosity and Moral Attitudes: Explaining Cross-national Effect Differences,” Sociology of Religion 63.2 (2002): 157-76.
  42. James Davison Hunter, Culture Wars: The Struggle to Control the Family, Art, Education, Law, and Politics in America (New York, NY: Basic Books, 1992).
  43. Emily Dickinson, “The None-ing of American Religion & Spirituality,” in Choosing Our Religion: The Spiritual Lives of America’s Nones (Oxford, UK: Oxford University Press, 2016), 246; Ariela Keysar, “Shifts Along the American Religious-secular Spectrum,” Secularism and Nonreligion 3 (2014); Vegard Skirbekk, Eric Kaufmann, and Anne Goujon, “Secularism, Fundamentalism, or Catholicism? The Religious Composition of the United States to 2043,” Journal for the Scientific Study of Religion 49.2 (2010): 293-310.
  44. Jeremy E. Uecker, Mark D. Regnerus, and Margaret L. Vaaler, “Losing my Religion: The Social Sources of Religious Decline in Early Adulthood,” Social Forces 85.4 (2007): 1667-92.

George Yancey

Baylor University
George Yancey, Ph.D., is professor of sociology at Baylor University. He’s author of Beyond Racial Gridlock: Embracing Mutual Responsibility (IVP, 2006).

Michael O. Emerson

North Park University
Michael O. Emerson is provost of North Park University in Chicago.