Growing Up UU Survey: Preliminary Results

Thanks to everyone who contributed to my survey, “Growing Up UU,” in the fall of 2011.  Many of you have expressed interest in reading the survey results once completed, so I have created this website specifically to share those data.  However, in the amount of time I had to write the paper for the seminar which inspired this research, I could not make any claims regarding the relationship between growing up UU and educational or career attainment (my stated goal for this project).  I did make some interesting discoveries.

In the paper below (a research proposal for the type of analysis mentioned), I do provide an informative overview.  I encourage those of you who are curious about things like how many of us graduated from college, believe in God, or consider ourselves gay (among other things), to scroll on down to the What can Unitarians contribute to this conversation? for a little background on how the data was used or to Data Collection and Data to see our actual numbers.  Also, the paragraph immediately before Discussion contains probably the most interesting finding the survey yielded – a clear indicator that sexual minority youth seek out UU churches.

Anyone interested in doing something more thorough with the dataset which came out of the survey is welcome to contact me at

Thanks again for your contributions!

——————————————————————————————-*This paper was written by Mallary Allen and submitted for credit in SOC 551 (D. Sherkat) at Southern Illinois University on 12/15/11.

Unitarian Upbringing and Status Attainment: A Research Proposal

The impact of religion on work and prosperity has interested sociologists since the field’s inception. Durkheim (1915) emphasizes the importance of religion in building solidarity, structuring complimentary roles, and imbuing life and work with a sense of purpose.  Weber (1922) asserts that Protestant notions of salvation and Predestination contributed to a distinctive cycle of hard work and savings amongst these denominations in early American history.  Modern theorists, too, have elaborated upon these associations.

These contemporary scholars acknowledge that the relationship between religion, wealth, and educational outcomes is a complex one and strive to identify mechanisms which contribute to differential patterns in achievement, earnings, and saving among adherents of various faiths.  One religious orientation which has received little attention in work on this topic is Unitarian Universalism (hereafter UU or Unitarian), a liberal denomination with roots in Calvinism whose members represent about 0.3% of the U.S. adult population (Pew Forum 2010).

Reasons that little attention has been directed to the study of UUs are manifold, their small numbers being primary.  But reasons for examining this group more closely are also numerous.  Chief among them are UUs’ unique patterns of conversion and member loss coupled with high levels of educational attainment and one of the highest mean incomes of any religious group.  A thorough analysis of UUs and factors associated with their high levels of socioeconomic status (SES), like the one that I purpose in this paper, will usefully contribute to sociological knowledge of religious and other mechanisms which shape life outcomes in theU.S.

In the research prospectus that follows, I will examine literature on SES attainment as moderated by religion.  I will then discuss social mechanisms which may be pertinent to education and earning outcomes.  Next, I will discuss relevant demographic information about UUs gathered from a recent survey entitled Growing up UU (GUU),which I composed specifically for this study.  Finally, I will discuss my proposed research plan, multivariate regression analysis combining General Social Survey (GSS) and (GUU) data to explore the relationship between achievement and religion among Unitarians.

Religion and Socioeconomic Status

Sociologists have identified trends in religious affiliation among Americans and its relationship to differential levels of education and wealth (e.g. Keister 2011; Sherkat 2009; 2011) While it is impossible to isolate its specific contribution, it is clear that religion, in combination with variables such as inheritance, parental SES, schools, family size, and race and ethnicity, has a profound impact on educational and career attainment, with Jews and Liberal Protestants enjoying higher average levels of SES and Conservative Protestants and (especially Black and Hispanic) Catholics falling on the lower end of the continuum.  Below, I discuss specific ways in which religion influences human capital and rational choice processes (Keister 2011), leading to these observable differences.

Conservative Christian religions face many obstacles to obtaining human capital.  These groups tend to be rural or located in poorer regions where schools are under-resourced (Keister 2011).  Unlike wealthier denominations, these congregants are also less likely to go to church with professionals who may represent valuable loose ties (Granovetter 1973) and networking opportunities.  Additionally, there is a strong relationship between poor churches and strict churches (e.g. Iannaconne 1994), and membership in such congregations may mean interacting with non-members infrequently.

Other barriers are more representative of self-selection, or rational choice.  For instance, conservative and sectarian Protestants are often opposed to liberal curricula and may be apposed to taking (or allowing their children to take) various science, literature, and humanities courses in school; these students are also less likely to attend college (Darnell & Sherkat 1997) and thus obtain prestigious and high-paying jobs. An orientation towards other- versus this-worldly rewards, too, may mold believers’ decisions to earn early and save often or to de-emphasize long-term wealth in favor of immediate spending (Burnstein 2007 in Keister 2011).

The vast majority of differential access to wealth and opportunity among various religious groups, however, is attributable to a combination of structural obstacles and religiously motivated decision-making.  Keister (2011) discusses the mutually reinforcing role of early marriage, as encouraged by religious abstinence-only campaigns, with lower levels of education, greater numbers of children (and thus fewer social and material resources to go around), higher divorce rates, and higher rates of single parenthood and blended families.

Compounding the cycle is the role-modeling of work and financial habits by parents whose levels of education and occupational prestige is lower than in other religions.  These parents, as well as the poorer schools that educate their children, also tend to favor child-rearing techniques thought to foster obedience rather than critical thinking skills.  This stricter orientation prepares children for lower-paying, low-autonomy work rather than creative and leadership positions (Laureau 2003).

Incidentally, the latter type of work in less abundant in the rural areas where conservative Protestants may choose to live.  High levels of religious intramarriage, too, correlate to shared social processes in family formation.  Marrying a spouse of the same faith, Keister (2011) finds, often correlates to shared parenting techniques, money habits, educational background, and so forth, perpetuating the education and financial habits of religious groups.

Keister (2011) observes that liberal Protestants and Jews, on the other hand, have parents with higher levels of education to begin with, who are also not opposed to liberal curricula throughout high school and beyond.  These prosperous parents role model prosperous work and financial habits and have fewer children, so they are more able to support their children’s educations materially and socially.  Additionally, Jews and liberal Protestants tend to marry later (in part because pre-marital sex, birth control, and co-habitation are more acceptable to them (Sherkat forthcoming) and to delay childbearing.  Having fewer children later in life generally allows more time for education and career, while later marriage typically means a combining of assets without the high risk of divorce.

Unitarian Universalists are lumped with liberal Protestants in most surveys, and mirror these general trends. However, UUs warrants an individual analysis for numerous reasons.  Below, I discuss the unique factors which set this group apart from other Protestants and explain this group’s potential contribution in furthering sociological understandings of the role that religion plays in shaping socioeconomic outcomes generally.

What can Unitarians contribute to this conversation?

The relationship between Unitarian Universalism and education in particular, is an old one.  Many early American thinkers and politicians, such as Thomas Jefferson, identified with Unitarianism, or the rejection of the Christian trinity in favor of the belief that God and Christ are distinct entities.  Early female scholars, a host of female clergy (prior to their ordination by most other groups), abolitionists, and later suffragists, make Unitarianism an early cite of American feminism (Green 2003).  In addition, prestigious American universities, such as Harvard and Antioch, claim Unitarian roots.

Since the 1961 merging of Unitarianism with Universalism – a similarly liberal branch of Protestantism which takes its name from its historical belief in universal salvation – the UU church has taken a decidedly less Christian turn. Members now mainly identify themselves with Humanism or, for some, Spiritualism (such as Wicca, Peganism, new age movements, etc.) (Lee 1995). Contemporary services in UU churches tend to emphasize humanitarian and political causes (Green 2003).  In one survey of several congregations in the Midwest, 13 percent of respondents described themselves as Christian, while 18 percent described themselves as atheist (Casebolt and Niekro 2005).

Today’s UUs are distinguished from adherents to other religions both educationally and economically.  In 1967, shortly after the unification of the two branches under the Unitarian Universalist Association (UUA), survey data indicates that UUs claimed the highest levels of education and income in the country, as well as the highest per capita representation of members employed in professional careers (UUA in Lee 1995).  Sources like Lee (1995) place UUs second to Jews in income and educational attainment, while Sherkat (forthcoming) places contemporary UUs again in first place for earnings.

Certainly there is nothing novel about a liberal faith having high levels of education and earning; this is consistent with findings about liberal Protestants in general, as discussed above.  What is unique about Unitarians is their high rates of conversion and intergenerational attrition. 1967 survey data from the UUA (in Lee 1995) shows that 89 percent of adherents to the newly merged faith were converts.  Sources estimate that membership is similarly comprised of converts today.  What is more, an estimated 90-95 percent of children raised in UU churches disaffiliate at some time in their lives (Lee 1995; Shister 2002).

Because it is likely that a significant number of UUs (likely half or more, as conversion likely takes place in adulthood following family formation (Sherkat 1991)) obtain their degrees prior to joining the church, it is difficult to ascertain the role that religion plays in shaping SES trajectories for Unitarians, net of other factors.  And because the children of UUs overwhelmingly leave the church in young adulthood, it is difficult to know how being raised in this liberal, affluent faith influences educational trajectories beyond high school.

Examination of this group, however, would help in explaining causal mechanisms between religion and life outcomes.  In short, if the generation brought up within the church has markedly different educational and career outcomes as compared with the converts who account for these high levels of education and income, we must consider that 1.) the religious mechanism is not working the way we think it works in relationship to this group (e.g. perhaps growing up UU makes one poorer/richer in comparison with her parents in the convert generation than if she had been raised in another faith) or 2.) that the role of religion in shaping outcome trajectories must be reconsidered.

In the remainder of this paper, I describe cursory (and mostly untreated) survey data from a convenience sample of adults who grew up in the UU denomination in the United States.  I go on to describe ways in which this data could be usefully combined with GSS data and analyzed in multivariate regression to better understand the role of religion in education and career attainment trajectories.


In the fall of 2011, I created a web-based survey, Growing up UU survey (or GUU) and shared it via a social networking website with 84 members of an online group in which I am enrolled.  The members and I have in common that we grew up Unitarian; the purpose of our group is to keep in touch with others like ourselves who attended a number of Unitarian youth conferences (or “cons”) throughout the Midwest during the mid- to late-1990s and early 2000s. In addition, I contacted 10 or so peers with whom I attended church growing up but who were not involved in the “con” scene. I also emailed the survey to a few personal acquaintances and contacts within Midwestern Unitarian churches, asking them to pass it along to qualified potential subjects (as described below).

In an online introduction to the survey, I informed participants that I created the survey our of general interest in this little-studied population (of which I am also a part) and disclosed any anticipated risks associated with completing the survey (few to none). Criteria for inclusion were simply that participants have attended UU services prior to adulthood, and that they are over the age of 18.  In a one month period, the survey was shared with and completed by 329 respondents, two of whom were disqualified due to indicating on their surveys that they had not attended services as a minor. Ns are also smaller for most items (315 at least). I received emails from two respondents asking permission to share the survey online with populations in the East Coast, which I welcomed.

The GUU survey consists of 42 items divided into four sections, each of which pertains to a specific period of life course development.  The first section, labeled “Your UU Upbringing” solicits demographic information such as age and gender along with information regarding age when respondents began attending the UU church and whether they attended with family members or by themselves.

The second portion, titled “Your Parents” asks about parents’ marital status, parents’ religion/s of origin, their educational attainment, and careers.  The third portion, labeled “Your Intimate Relationships” asks similar information in regard participants’ romantic partners as well as religious training of any children.  The final portion, “You Today” solicits information about applicants’ current religious affiliation as well as educational and career attainment.  This portion also contains the only qualitative item included, the question, “If you have stopped attending UU services and/or have stopped identifying as a UU, please describe briefly why you stopped.”  (Responses to this item are cataloged on this site under the heading Growing Up UU Survey: Why We Stopped Attending) Below, I discuss preliminary descriptive statistics yielded from this survey.


Limitations to this data are numerous, and, in some cases apparent.  While I was fortunate to have access to a network of adults who have in common only their Unitarian youth (and not necessarily current membership status), members of the web-based group I described, who made up my recruitment base, were more involved than typical UU teenagers.  These contacts are also relatively young, having been in high school in or around the late 1990s. Additionally, the survey was shared at greater rates than I anticipated, and I am unaware of how typical or atypical the majority of research subjects are.  With these limitations in mind, data gathered are nonetheless instructive.  I discuss some basic numbers here.

Respondents’ Demographic and Religious Backgrounds

The average age of GUU participants was approximately 33 years. Median age, however, was 29, and modal age was 27. Respondents were 90% white, with the next largest group, 7%, being bi-racial (white and another race/ethnicity). Approximately 68% were female (n 220), 31% male (n 101), and 2% transgendered (n 7).

The approximate mean age when respondents began attending UU services was 5 years.  91% of respondents reported that they attended church with their families growing up, while about 9% said that they attended the UU church on their own and that their parents were not UU.  Additionally, 17% of respondents reported that their mothers were raised UU, and 7% reported that their fathers were raised UU, indicating that the majority of respondents grew up with parents who converted to the religion later in life (presumably, when respondents were, on average, 5 years old).  The majority of parents (27% of mothers and 35% of fathers) were raised mainline/moderate Protestant, followed by Catholic (about 21% across both groups).

Parents and Cultural Capital

Table 1.1. Parent Educational Attainment by percentage

                          H.S.    Some college    AssocDeg.       4 yr degree      Some grad    Grad    Total   

Mother               4.4                4.7                 3                   23.8                9.4              53.4    n 317

Father                 4.7                 6                  3.7                 25                    4                51.7     n 306

Source:  Author, 2011

Educational attainment of respondents’ parents is quite high, which is consistent with other studies, for instance a 2005 survey which found that 52% of subscribers to the UUA publication UUWorld, hold graduate degrees (UUA 2005).  Respondents also reported that 65% of their mothers and 72% of their fathers were employed in the three combined prestigious categories of professional work (medicine, law, teaching, etc.), technological fields, and business/managerial.  Behind this lumped category for mothers was stay-at-home parent (22%), and, for fathers, skilled labor (8%). (N similar to table 1.1.)

Not surprisingly, 84% of respondents reported that their mothers were Democrat (6% Republican) while 70% of fathers were Democrat (9% Republican).  89% described their mothers as liberal or very liberal; 73% said the same about their fathers. (N similar to table 1.1.)  This population is clearly more liberal than the general population and unlikely to raise objections to public school curricula, such as the teaching Evolution or the reading of certain literature.

Three percent of respondents said that their parents were never married, and 28% had parents who divorced at some point.  Of all respondents, 7.5% were raised by a single mother; 9.3% had one or more heterosexual step-parents, and 2.8% had one or more gay or lesbian step-parents. At 28%, parent divorce rates for this sample are lower than divorce rates for the general population, wherein 50% of couples divorce within seven years of marriage (Bureau of the Census 1997 in Renzetti and Curran 2003).

Respondent Cultural Capital

Table 1.2.  Educational Attainment of Adults Raised UU (mean age 33)

H.S.       Some college     AssocDeg.      4 yr degree      Some grad      Grad     Total

2.3           11.8                   2.6                     28.4                  20.8           35.1     n 313

Source:  Author, 2011

While the educational attainment of adults raised in the UU faith is quite different from that of their parents’, the number of respondents who reported that they are currently students likely accounts for this difference.  Furthermore, 11 percent of respondents are not old enough to have completed a master’s degree (assuming they did so within a traditional time table).

Evidence of prolonged adolescence, however, should also be examined in the research I propose.  While the phenomenon of young adults returning home after college graduation and/or spending several years in graduate school and/or returning to school to obtain advanced degrees later in life has been observed across American adults in general (Mitchell 2006), differences in parenting strategies, particularly strictness (Sherkat, forthcoming), or, alternately, parent emphasis on education and accomplishment in this high-achieving group, may make this phenomenon unique among UUs.

Asked their current career, 61.6 percent of reported that they were currently employed in professional jobs, technological fields, or business/managerial fields, while the next largest group, students, were 21 percent of the sample. (N similar to table 1.2).  A limitation of this data should be mentioned here, and that is that the category of professional work (which, alone, was selected in reference to 49% of mothers, 45% of fathers, and 47% of respondents) included the descriptor “teacher positions” which is likely to have yielded a wide array of prestige categories.

Respondents also provided demographic information regarding their partners.  While I will not describe educational and career attainment of respondents’ partners here, as more sophisticated data analysis is needed to separate characteristics of those to whom respondents are “married/partnered”  from those whom they have “divorced” and those with whom they are “in a relationship” (total N for these 3 groups= 253), religious partnering patterns here are interesting.  30% stated that their partner or recent ex-partner was Catholic, 22% had no religious affiliation, 13% were mainline Protestant, and 10% were UU.  Considering the small number of UUs and non-religious in the United States overall (around 10% for the latter (Sherkat 2010)), it appears that UUs are relatively successful in dating and marrying people similar to them in belief (or lack thereof).

On a related note, 40 percent of respondents reported that they believe in God or a higher power while 33.8 percent did not, and 26.4 percent said that they were not sure.  Interestingly, 80 percent of respondents (N 314) reported that they identify as Unitarian Universalist today, although 50 percent of all respondents (including the 20 percent who do not think of themselves as UU) indicate that they have stopped attending services.

Sherkat (1991) and others (Sherkat and Ellison 1999) observe that life course factors greatly influence religious participation.  Having children is regarded as the most compelling life stage in bringing congregants back to church.  Given the young age of GUU respondents, it will be difficult to glean useful information on this topic in reference to UUs.  It is worth mentioning that only about 22 percent of respondents report having children.  Of that number, 75 percent report that they were raising their children UU.  Approximately 20 percent report that their children have no religion, and 15 percent are raising their children in numerous other faiths.


The research project that I propose would merge the GUU dataset with GSS 2010 data.  The General Social Survey is a nationwide survey of Americans administered every one to two years since 1972 and contains demographic and opinion data similar to the variables I have described in the GUU.  The GSS has a much larger N, however, and addresses information about a variety of religious groups in theU.S.

Useful data analysis would regress the independent variable of religion and control variables such as parents’ SES and respondent age, with the dependent variable of respondentSES.  Models would examine both GUU data alone as well as combine GUU data with the GSS to generate a regression model where adults raised Unitarian are over-sampled, so that the effects of UU upbringing are not lost within a very small N of individuals who mostly converted to the faith, as is the barrier to studying Unitarians using GSS data.

Alternate models would compare the effect of being raised UU with the effect of being raised in other religions.  Evidence suggests that UUs are among the least active church participants (Creswell 2008) even prior to the exodus that many congregations report that they experience amongst graduated high schoolers (Shister 2002).  Strict, conservative religions, on the other hand, retain members and are successful in maintaining high levels of participation (Iannaconne 1994).  Alternate models which allow for examination of variation in strength of effect across groups are therefore needed.

It is necessary to mention that researchers have previously compared the strength of religious effects onSESacross denominations (e.g. Sherkat forthcoming; Keister 2011).  The addition of more UU cases to a regression model (one merging GSS and GUU data), however, will add to what can be known about the strength of religious effect in a faith where participation is thought to be very low (as indicated by arguably the lowest relative rates of tithing (Creswell 2008) and high attrition).

The effect of being UU on SES attainment may also function differently in light of other considerations, some of them touched upon in discussion of data above.  These features include prolonged adolescence possibilities, becoming UU through family or on one’s own, and the religious affiliation of respondents’ partners.

Another variable with a sometimes positive relationship to SES (citation needed), and likely a unique interaction with Unitarianism, is homosexuality.  As Unitarians were among the first religions to openly welcome gays and lesbians and to perform ceremonies honoring their commitments, the denomination enjoys more GLTB membership and participation than other churches (Lee 1995).  In the GUU, 10 percent of respondents identified as homosexual, and 13 percent identified as bisexual.

These figures, larger than national GLTB estimates, are interesting in themselves because this survey is comprised of individuals who attended UU churches as youth. Much of this finding is attributable to the fact that only 61 percent of those who reported attending services without their parents as minors said that they are heterosexual, compared with 74% of respondents whose parents were also UU.  This indicates that at least some youth who attend UU churches on their own do so because they are gay and the religion offers an environment supportive of their sexual identities.  In terms of SES, these high numbers are important because a non-heterosexual life course is less likely to be interrupted by early childbearing, a deterrent to educational and wealth attainment (Keister 2011).


Limitations to my proposed study are evident and have been discussed throughout this prospectus.  They mainly consist of unavoidable sampling bias (given my recruitment resources), the young mean, median, and modal ages of survey respondents, and some weaknesses to survey items (such as the lumping of all teaching careers with professional careers such as law and medicine).

The data collected, however, and the project proposed, will nonetheless yield valuable information.  Adults who grew up attending a uniquely liberal and decidedly transient church provide a novel, and hopefully instructive, layer to the study of religion’s influence onSES.  An analysis of Unitarian Universalism’s role in promoting (or deterring) educational attainment in adulthood, net of factors which largely-convert parents bring with them to the religious environment, will further our knowledge of religion’s role in status attainment.



Casebolt, James and Nieko, T. (2005). “Some UUs Are More U than U: Theological Self-Descriptors Chosen by Unitarian Universalists.” Review of Religious Research, 46(3): 235-242.

Creswell, Dick. (2008). “UUs and Money”. 12, 2008.  Accessed12/12/11.

Darnell, Alfred and Darren E. Sherkat. (1997). “The Impact of Protestant Fundamentalism on Educational Attainment.” American Sociological Review. 62:306-316.

Durkheim, Emile. (1915) 1965. The Elementary Forms of the Religious Life. New

York: Free Press.

Grannovetter, Mark S. (1973).  “The Strength of Weak Ties.” American Journal of Sociology, 6(73): 1360-1380.

Green, John C. (2003). “A Liberal Dynamo: The Political Activism of the Unitarian-Universalist Clergy.” Journal for the Scientific Study of Religion, 42(4): 577-590.

Iannaccone, Laurence R. (1994). “Why Strict Churches are Strong.” American Journal of Sociology.

Keister, Lisa. (2011) Faith and Money: How Religion Contributes to Wealth and Poverty.


Lareau, Annette. (2003). Unequal Childhoods: Class, Race, and Family Life.Berkeley:University ofCalifornia Press.

Lee, Richard W. (1995). “Strained Bedfellows: Pagans, New Agers, and ‘Starchy Humanists’ in Unitarian Universalism.”  Sociology of Religion, 56(4): 379-396.

Mitchell, Barbara A. (2006). “The Boomerang Age from Childhood to Adulthood:

Emergent Trends and Issues for Aging Families.” Canadian Studies in Population, (33)2:155-178.

Pew Forum on Religious Life. (2010). “US Religious Landscape Survey.” affiliations.  Accessed11/12/11.

Renzetti, Claire M. and Curran, D.J. (2003) Women, Men, and Society, 5th ed.Boston: Allyn and Bacon.

Sherkat, Darren. E. (1991). “Leaving the Faith: Testing Theories of Religious Switching using Survival Models.” Social Science Research, 20(2):171-187.

______. (2010). “The Religious Demography of the United States” in Robert Hummer and Christopher G. Ellison (eds) Religion, Families, and Health: Population-based Research in the United States. RutgersUniversity Press.

______.  (2011) Religion and Scientific Literacy in theUnited States. Social Science Quarterly, 92(5): 1134-1150.

______. (Forthcoming) “Chapter 5: Faith, Family, and Fortune.”

Shister, Neil (2002). Liberal Evangelists on Campus.  UUWorld, September/October 2002,

UUA. (2005) “About our Readers.” UUWorld. aboutourreaders. shtml. Accessed12/14/11.

Weber, Max. (1922). The Sociology of Religion.Boston. Beacon.

This entry was posted in Uncategorized. Bookmark the permalink.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s