• The Cardus Education Survey (CES), conducted in the United States in 2011, 2014, 2018, and again in 2023, examines a range of outcomes for a nationally representative sample of adults aged 24 to 39 who attended public schools, Protestant schools, Catholic schools, nonreligious independent schools, or were homeschooled.
• The CES examines respondents’ academic, spiritual, cultural, civic, and relational outcomes, and their life patterns, views, and choices. The 2023 iteration also included questions about mental health and personal values.
• The CES controls for a range of respondents’ demographic characteristics in order to estimate the specific effect of school type on graduate outcomes.
• Top-line findings:
  • High school experiences and quality: Graduates from Protestant, Catholic, and nonreligious independent schools are significantly more positive in their ratings than those from public schools. Those from Protestant and Catholic schools believe that their schools prepared them well to have a vibrant spiritual life.
  • Educational attainment and employment outcomes: Graduates from Catholic or nonreligious independent schools are more likely to attain a graduate degree and have higher income levels. Those from Protestant and Catholic schools are more likely to value having a job that is directly helpful to others.
  • Civic behaviors and values: Graduates from Protestant schools and homeschooling have higher rates of charitable giving, regardless of income, and are more likely to volunteer than are graduates from public schools. Graduates from nonreligious independent schools are more likely to cite community involvement and tolerance as very important.
  • Mental health and well-being: Graduates who were homeschooled report the lowest levels of depression and anxiety. Graduates from Protestant and Catholic schools have the highest levels of life satisfaction, though this difference disappears after controlling for respondents’ demographic characteristics.
  • Faith formation: The strongest sector-level effects are observed here. Graduates from Protestant schools or homeschooling are much more likely to report that they believe in God and life after death and regularly engage in religious practices.

In the United States, the history of education is a pluralistic one, with multiple sectors responsible for schooling. These sectors often have different goals for students, though all necessarily contribute to preparing students for shared life in a democracy. 1 1 M. Casagrande, R. Pennings, and D. Sikkink, “Rethinking Public Education: Including All Schools that Contribute to the Common Good,” Cardus, 2018, 3, https://cardus.ca/research/education/reports/rethinking-public-education. Given the central role that education plays in the lives of students, families, and society writ large—as well as the investment of both tangible and intangible resources in schooling—it is worth asking how the various educational sectors in the US are performing in preparing students. Research has consistently demonstrated differing patterns of outcomes by educational sector. 2 2 See D. Hamlin and A. Cheng, “Parental Empowerment, Involvement, and Satisfaction: A Comparison of Choosers of Charter, Catholic, Christian, and District-Run Public Schools,” Educational Administration Quarterly 56, no. 4 (2020): 641–70, https://doi.org/10.1177/0013161X19888013; Y. Chen et al., “School Types in Adolescence and Subsequent Health and Well-Being in Young Adulthood: An Outcome-Wide Analysis,” PLOS ONE 16, no. 11 (2021): e0258723, https://doi.org/10.1371/journal.pone.0258723; M.D. Shakeel et al., “The Public Purposes of Private Education: A Civic Outcomes Meta-Analysis,” Educational Psychology Review 36, no. 40 (2024): 1–41, https://doi.org/10.1007/s10648-024-09874-1; L.M. Foreman, “Educational Attainment Effects of Public and Private School Choice,” Journal of School Choice 11, no. 4 (2017): 642–54, https://doi.org/10.1080/15582159.2017.1395619.

The Cardus Education Survey (CES), conducted in the United States in 2011, 2014, 2018, and in 2023, examines a range of outcomes for a nationally representative sample of high school graduates aged 24 to 39 across five sectors: public schools, 3 3 The term “public schools” refers to traditional brick-and-mortar schools that are government-funded and operated. Results for respondents who spent most of their high school in a nontraditional public school (such as a charter school, whether brick-and-mortar or online) are not included in the CES analysis, given the small numbers of such adults. Protestant schools, Catholic schools, nonreligious independent schools, and homeschooling. One of the most comprehensive surveys of its kind, the CES seeks to understand school-sector influence on a range of academic, spiritual, cultural, civic, and relational outcomes for graduates, as well as to assess graduates’ life patterns, views, and choices and their contribution to a shared good. The 2023 iteration of the CES also includes a new, explicit focus on mental health as well as questions about respondents’ personal values.

By measuring a range of holistic outcomes, the CES advances a view of education as contributing to the flourishing of both students and society as a whole. Flourishing is far from a new idea, dating back to the Aristotelian concept of eudaimonia—of human flourishing that is derived from “doing and living well.” 4 4 K. Kristjánsson, Flourishing as the Aim of Education: A Neo-Aristotelian View (Routledge, 2020). In psychology, flourishing has been used conceptually to redefine well-being as a holistic state, one that is greater than just the absence of illness. 5 5 C.L.M. Keyes, “The Mental Health Continuum: From Languishing to Flourishing in Life,” Journal of Health and Social Behavior 43, no. 2 (2002): 207–22, https://doi.org/10.2307/3090197. Theologians, such as Aquinas and Augustine in the Christian tradition, have also explored deeply the question of human flourishing. Regardless of its specific framing, a view toward educational flourishing is a capacious one that prioritizes the holistic development of persons—in mind, heart, body, and spirit—while situating them within larger and reciprocal social and communal contexts. Such a view broadens the apertures of educational theory, research, and practice, thereby opening up greater possibilities not only of what can be seen but also of what might be imagined and enacted in schools.

To this end, the CES attempts to explore, describe, and ultimately spark discussion about the larger story of what is happening educationally across school sectors. The CES also provides readers who have an interest in a particular sector with insights on its unique strengths and potential areas for growth, which may inform future research as well as sector-specific improvement efforts. Finally, CES data may be helpful at a cross-sector level, by informing the work of those responsible for curating the diverse and vibrantly pluralistic educational system in America—ideally in the direction of greater individual and collective flourishing.

Many aspects of the CES have been preserved from prior administrations. For example, the 2023 iteration continues to focus on the same five school sectors—public schools, Protestant schools, Catholic schools, nonreligious independent schools, and homeschooling—which are the largest school sectors in the US. Similar to earlier administrations, the 2023 data presented in this report describe how graduates from each sector say they are faring in adulthood relative to a range of outcomes, as well as how much each sector may have contributed to those outcomes. New for 2023, as already mentioned, is a set of questions regarding respondents’ mental health, as well as items that gauge the importance that respondents place on various personal values.

It is important to note that the 2023 CES is the first administration after the COVID-19 pandemic of 2020. Respondents for this iteration graduated from high school prior to the pandemic and thus did not experience pandemic-related disruption as students. While COVID undoubtedly affected their lives as adults, the CES does not directly measure its impact. At the same time, the results still provide valuable data on how graduates of each sector say they are faring post-pandemic. Because future administrations of the CES will include adults who were in high school during the pandemic years, future insights into pandemic effects and the role of each sector (in amplifying or mitigating those effects) will likely be possible—especially given the new questions relating to mental health that have been added to the CES. It is also likely that sector-switching, which occurred in significant numbers both during and immediately following the pandemic—observable in the growth of homeschooling and enrollment in Protestant schools, for example 6 6 M.H. Lee and E.W. Price, “ACSI Schools Weather Pandemic Storm with Steady Student Achievement, In-Person Instruction, and Enrollment Growth,” ACSI Research Reports October 2022, https://www.acsi.org/docs/default-source/website-publishing/research/acsi-schools-weather-pandemic-storm.pdf; G. Smith and A.R. Watson, “New U.S. Census Bureau Data Confirm Growth in Homeschooling Amid Pandemic,” Education Next 24, no. 4 (2024), https://www.educationnext.org/new-u-s-census-bureau-data-confirm-growth-in-homeschooling-amid-pandemic/. —will play an important role in future analyses.

This summary report of CES findings, across all sectors and outcomes measured, is structured as follows. First, an overview of survey methods is provided. Next, results related to graduates’ perceptions of their high school experiences are shared. This is followed by a discussion of results for five groupings of outcomes: educational attainment and employment outcomes, civic formation, mental health and well-being, faith formation, and family formation. A final section presents sector-specific highlights for three of the five sectors (Protestant, Catholic, and nonreligious independent schools). This is provided not for comparative purposes but rather to give readers who have an interest in one or more of those particular sectors a narrative description of CES findings for that sector.

This administration of the CES was conducted in 2023 from October 10 through October 22 by the polling firm Ipsos Public Affairs, using its Knowledge Panel. The Knowledge Panel is the largest online panel in the United States and uses probability-based sampling techniques to obtain a representative sample of adults in the United States. The target population comprised US adults aged 24 through 39 who have completed their education through high school. Adults who predominantly attended at least three years of high school (grades 9–12) in Protestant schools, Catholic schools, nonreligious independent schools, or were homeschooled were oversampled. Because adults from these educational sectors make up a small minority of the US population, oversampling was needed to obtain a sample size that was sufficiently large enough to make inferences about them. Thus, the analytic sample includes more responses from those adults, making those adults overrepresented in the data. Following standard practice, survey weights are incorporated into all the analyses to account for this oversampling so that the results remain nationally representative of 24- to 39-year-olds who completed high school. 7 7 Some respondents also had postsecondary education; see the later section on postsecondary degree attainment.

The sample size for this iteration of the CES was 2,350 respondents. Respondents indicated the sector in which they attended each year of their primary and secondary schooling. For each year, respondents were able to indicate whether they attended a public school, a charter school, a Catholic school, a Protestant school, another type of religious independent school, a nonreligious independent school, or another type of school. Respondents were also able to indicate whether they were homeschooled for any of their primary and secondary schooling years.

The first column of table 1 lists the number of respondents who spent at least one year of their primary and secondary schooling in each sector. Of the 2,350 respondents, close to 90 percent (2,095) spent at least one year of their primary and secondary schooling in a public school. Four hundred and six respondents spent at least a year in a Catholic school, and 291 spent at least one year in a Protestant school. Finally, 181 respondents were homeschooled at some point for their primary and secondary education.

The second column of table 1 lists the average number of years that respondents spent in each educational sector, conditional on having spent at least one year in that sector. For instance, respondents who attended public schools for at least one year of their primary and secondary schooling spent an average of 11.6 years in that sector. In other words, respondents who had ever attended a public school spent, on average, almost all thirteen years of their primary and secondary schooling years in that sector. Respondents who spent some time in Catholic and Protestant schools, on average, spent 6.8 and 5.7 years of their K–12 education in those sectors, respectively. Put differently, adults who had ever attended a Catholic or Protestant school spent approximately half of their primary and secondary schooling years in those sectors. Meanwhile, the 181 adults who were ever homeschooled during their primary and secondary education spent about 6.0 years in that setting.

The last column of table 1 lists the number of respondents who spent at least three of their high school years (grades 9–12) in each sector. The data in table 1 underscore that although a majority of primary and secondary education occurs in public schools, a sizeable proportion of 24- to 39-year-olds in the US attended schools in other sectors, even if for a brief period.

For all figures presented in this report, respondents who spent a majority of their high school years (that is, at least three years) in public school are compared to respondents who spent the majority of their high school years in Catholic school, Protestant school, nonreligious independent school, or homeschool. 8 8 For the questions that specifically asked respondents about their experiences in high school, the following instructions were provided: “If you attended more than one high school, please consider the high school that you attended longest.” Results for respondents who spent most of their high school in some other type of school are not presented, given the small numbers of such adults. Throughout this paper, respondents are referred to as “graduates” of the school sector in which they spent a majority of their high school years.

Two sets of results are presented. First, raw scores reflect outcomes across the school sectors without adjusting for demographic characteristics. These scores do not isolate the effect that each school sector had on its graduates because they also reflect the effect of other variables. Second, sector-effect scores are presented. These isolate the effect that each school sector had on its graduates by controlling for other variables. These results are obtained by using linear regression techniques to account for the following demographic differences: respondent’s age, gender, race, whether they live in a metropolitan area, region of the US in which they currently live, and whether they grew up in poverty, 9 9 Respondents were asked, “Growing up, what was the money situation at your home like?” Respondents who selected the response options “We didn’t have enough money for bills and food,” “We could pay some of our bills and food,” or “We could pay most of our bills and food” were considered as living in poverty. Respondents who selected “We paid all of our bills, but didn’t have enough for much more” or “We had plenty of money for our bills and other things we needed” were not considered as living in poverty. grew up with both biological parents, were raised in a nonreligious household, 10 10 Respondents were asked, “What was the primary religious identity that you were raised in, if any?” Respondents who selected the response option “atheist” or “no religion” were considered raised in a nonreligious household. and had a college-educated mother.

These regression models essentially adjust raw scores for the influence that the home and other formative life experiences potentially have on respondents, so that any observed differences between respondents across the school sectors are more reflective of the school’s contribution to those outcomes. This set of results assumes that influences outside of school sector have been adequately accounted for in the demographic variables included in the models. While this cannot be guaranteed, the CES follows industry standards in accounting for these variables as much as is methodologically possible.

Finally, while this report notes all differences between sectors, not all findings are statistically significant (meaning that the observed result is not likely to have occurred by chance alone). The statistical significance of findings is noted throughout the report and in all figures.

For all figures presented, the public school average response is the baseline (at zero). This decision, which is consistent with the way results were reported for previous administrations of the CES in the US, does not just reflect the reality that public schools are the largest educational sector. It also underscores the fact that public schools are the default schooling option for US families, as these schools are fully funded by the government and, in most cases, students are automatically assigned to a specific school based on their place of residence.

Finally, many of the outcomes are measures obtained from Likert-type items, with some items having up to seven response options. Presenting the percentage of respondents from each school sector who selected each response option would result in an excessive amount of information and obscure the general picture that the data are revealing. To maximize clarity in the communication of results, response options are grouped together and the percentage of respondents in each school sector who selected a response option from that group of response options is presented. 11 11 Three guiding principles were used in grouping responses. First, where response options flip meaning (for example, in the case of Likert items that have the response options “strongly disagree,” “disagree,” “slightly disagree,” “slightly agree,” “agree,” and “strongly agree"), the responses that expressed agreement were grouped and separated from the ones that expressed disagreement. Second, where items were adopted from other research studies, results are presented in the same way as in those studies (unless otherwise noted). And third (and more technically), in some cases, very few respondents selected particular response options, so a grouping decision was made in order to limit the amount of measurement error in the results and maximize the validity and precision of the results. The way response options are grouped for a particular outcome is noted in each figure. For readers who are interested in a finer-grained examination of the results, the appendix lists the percentage of respondents from each school sector who selected each respective response option for each question.

The CES measures graduates’ perceptions of their high school experiences across a range of dimensions: the overall quality of the education they received, how well their schools prepared them for life after graduation, the quality of school relationships, and how their schools handled challenging topics.

When it comes to key findings in these areas, Protestant, Catholic, and nonreligious independent school graduates are significantly more positive about many aspects of their schools than are public school graduates. Protestant and Catholic school respondents also believe that their schools prepared them well to have vibrant spiritual lives. This section of the report unpacks the CES findings relative to graduates’ perceptions of high school experiences by sector. 12 12 Because these items were designed for respondents to rate their high schools, respondents who were primarily homeschooled are omitted from these results. Also, as noted earlier, for all figures presented in this report, the public school average response is the baseline, at zero.

Two items assessed respondents’ perceptions of the overall experience and quality of their high school education. The first item asked them to indicate how favorable they felt about their overall high school experience. Both raw scores and sector-effect estimates reveal that Protestant school graduates were as likely as public school graduates to rate their high school experience favorably. Catholic or nonreligious independent school graduates were about 20 percentage points more likely to offer a favorable rating compared to adults from public schools. When asked to rate the overall quality of their high school education, responses by sector were similar.

The CES also asked respondents to rate the quality of the extracurricular opportunities at their school. While 50 percent of public school graduates rated their extracurricular opportunities as favorable, Protestant school graduates were between 10 and 20 percentage points less likely to do so. Graduates of nonreligious independent schools rated the quality of their schools’ extracurricular opportunities about the same as public school graduates. Only Catholic school graduates were more likely to provide a favorable rating, by a statistically significant 8 percentage points.

The CES asked respondents to rate how well their high schools prepared them for life. Graduates from the independent sectors (Protestant, Catholic, nonreligious) were more likely than public school graduates to say that their school prepared them very well in the areas of academic success at the postsecondary level, success in a job or a career, and strong personal relationships. These differences remain large and statistically significant even after adjusting for demographic characteristics. Additionally, respondents who said that their schools prepared them well on these aspects of life were more likely to provide favorable ratings of their overall high school experience and the quality of their education.

When it comes to preparation for a vibrant spiritual or religious life, only graduates of Protestant and Catholic schools reported feeling more prepared than their public school counterparts; those from nonreligious independent schools felt as prepared as graduates from public schools. This is perhaps unsurprising, given the focus on spiritual formation in religious independent schools.

As for how well schools prepared their students for interacting with the range of people in society, graduates of Protestant schools responded similarly to those of public schools; around one out of every three respondents from those two sectors reported that their school prepared them very well for that goal. In contrast, graduates of Catholic and nonreligious independent schools were more likely to say that their schools prepared them very well for that goal, even after controlling for demographic characteristics.

The CES asked respondents to characterize the level of social connectedness found in their high schools. Importantly, graduates’ responses to questions in this area were strongly correlated with their evaluation of their schools’ overall educational quality and experience. After adjusting for demographic characteristics, sizeable differences between public school graduates and those who attended independent schools were evident in terms of agreeing that their school was a close-knit community and that their teachers really cared about students.

Given these findings about the closeness of the school community and the care of teachers, it is interesting to note the differing sector-level responses when it comes to feeling a sense of belonging at school. Only graduates of nonreligious independent schools were more likely than their public school counterparts to agree that they felt they belonged at their schools, while Catholic school graduates were about as likely to agree, and Protestant school graduates were slightly less likely to agree. Differences were also observed for the Protestant school sector when it comes to agreeing with the statement “I enjoyed being a student at my school.” Protestant school graduates were as likely as adults from public schools to agree with the statement, while Catholic and nonreligious independent school graduates were more likely to agree with it.

When it comes to relationships with their teachers, respondents from all independent school sectors were more likely than public school graduates to rate those relationships favorably, though the greatest effect size was observed for Catholic school graduates. In terms of peer relationships, Catholic and nonreligious independent school graduates were still more likely than public school graduates to rate those relationships favorably; however, Protestant school graduates showed no statistically significant difference from their public school counterparts.

Respondents also rated how well they thought their school handled challenging topics such as religious and spiritual matters or national politics. Graduates from all independent school sectors were more likely than their public school counterparts to provide a favorable rating of their school when it comes to religious and spiritual matters, though the difference was greater for Catholic and nonreligious independent school graduates. In terms of their schools’ handling issues of national politics, however, nonreligious independent school graduates were the only ones more likely to provide favorable ratings as compared to their public school counterparts, after adjusting for demographic differences.

Postsecondary degree attainment and improved economic outcomes are typical expectations of K–12 education in the US. The CES measures educational attainment, income, and employment outcomes of graduates. In terms of key findings in this domain, attending a Catholic or nonreligious independent school is correlated with a greater likelihood of attaining a graduate degree and with higher income levels. When it comes to careers, Protestant and Catholic school graduates are more likely to value having a job that is directly helpful to others. The remainder of this section explores the CES findings relative to educational attainment and employment outcomes by sector.

Similar to the results of prior administrations of the CES, Catholic school graduates have the highest levels of degree attainment compared to other sectors. This is true both for attaining a bachelor’s degree and for attaining a graduate degree. Respondents from Protestant and nonreligious independent schools are also more likely than those from public schools to have at least a bachelor’s degree,yet those differences are likely due to the influence of other factors rather than the school (as controlling for demographic characteristics reduces those differences to only a few percentage points and statistical insignificance). Nonreligious independent school graduates are more likely to obtain a graduate degree, albeit not at the same rates as Catholic school graduates. Finally, adults who were homeschooled for a majority of their high school years are less likely to have any postsecondary degree when compared to public school graduates, a finding observed even after controlling for demographic characteristics.

The connection between educational attainment and income has been repeatedly demonstrated in economic research, with earnings generally higher for workers with more education. 13 13 D.J. Deming, Why Do Wages Grow Faster for Educated Workers? Working Paper 31373, National Bureau of Economic Research, 2023, https://www.nber.org/papers/w31373. The results from the CES are consistent with this economic reality, as evidenced in the income distribution by quartiles for graduates from each school sector. According to the analysis based on raw scores, graduates of Catholic and nonreligious independent schools are overrepresented in the two highest income quartiles. However, when adjusting for background characteristics to estimate school effects, Catholic school graduates are no longer overrepresented, suggesting that differences in income between Catholic and public school graduates are not attributable to school sector. Protestant school graduates most closely resemble their public school counterparts, while adults who were homeschooled for a majority of their high school education are overrepresented in the two lowest income quartiles.

The findings regarding educational attainment and income are also generally, though not always, consistent with findings regarding employment. In terms of the likelihood of not being employed, Protestant school graduates are again no different from graduates from public schools, and Catholic school graduates are less likely not to be employed, based on sector effect estimates. Homeschooled adults, on the other hand, are much more likely than public school graduates not to be employed. In one of the more intriguing findings, nonreligious independent school graduates are more likely not to be employed compared to those from public schools. This result seems to run counter to these graduates’ higher incomes and levels of educational attainment. For reasons that cannot be ascertained from the CES data, the income penalty for not being employed is not as drastic for nonreligious independent school graduates, as compared to graduates from other sectors.

In addition to employment outcomes, the CES presented respondents with various aspects of their work and asked them to rate how important each aspect was to them on a six-point Likert scale. All adults from Protestant, Catholic, and public school settings are as equally likely as their public school counterparts to say that it is “very important” for them to have a job that pays well, while nonreligious independent school graduates are numerically more likely to say so (though this difference is not statistically significant). However, there is statistically significant evidence that adults from homeschool settings are less likely to prioritize a job that pays well compared to public school graduates.

When it comes to the type of job they prefer, graduates of religious independent schools (both Protestant and Catholic) are statistically more likely than public school graduates to say that it is “very important” to have a job that provides an opportunity to be directly helpful to others. Nonreligious independent school graduates were the only group that was statistically more likely to say having a job that fulfills their calling is very important, as compared to their public school counterparts. The CES also queried respondents about how important they felt it is to have a job that provides a chance to be creative, allows for establishing roots in a community instead of moving from place to place, or allows one to be near family. Regarding these aspects of a job, there is no evidence of any differences by educational sector.

Two additional data points potentially illuminate respondents’ perspectives towards education, employment, and money. Following the approach of a Wall Street Journal/NORC poll conducted in 2023, 14 14 “WSJ/NORC Poll March 2023,” Wall Street Journal/NORC at the University of Chicago, https://s.wsj.net/public/resources/documents/WSJ_NORC_ToplineMarc_2023.pdf. the CES presented respondents with a list of values to rate in terms of personal importance. Homeschooled adults and Catholic school graduates are statistically less likely than their public school counterparts to say that money was a very important personal value, though the difference for homeschooled adults is the greatest. This difference was also observed for Protestant school graduates, but not at a level of statistical significance. These perspectives toward money are consistent with results discussed earlier regarding the level of importance for respondents to have a job that pays well, along with findings regarding lower levels of income among adults from Protestant and homeschooled settings. Finally, adults from Catholic schools, nonreligious independent schools, and homeschool settings are more likely than public school graduates to say that hard work is a very important personal value. Adults from Protestant schools are also less likely than adults from public schools to say so. However, many of those differences are not statistically significant.

A key role of schooling in a democracy is to instill, nurture, or otherwise inculcate the dispositions and skills needed for living well together in a pluralistic society. In terms of outcomes, this is often referred to as civic formation. The CES measures civic values as well as two civic behaviors—charitable giving and volunteerism—as indicators of civic formation for graduates. 15 15 It should be noted that there are other forms of civic engagement, as well as cultivating civic knowledge and civic virtue, which are outside the scope of the available data. A major finding here is that Protestant school graduates have one of the highest rates of charitable giving, regardless of income. Another is that they are more likely to volunteer than are public school graduates, as are homeschooled adults. Additionally, compared to other sectors, graduates of nonreligious independent schools are more likely to cite community involvement and tolerance as “very important.” The remainder of this section examines CES findings relative to civic formation of graduates by sector.

The CES first considers civic formation by asking for self-reports of how important certain civic values are to respondents. Following the Wall Street Journal/NORC survey items as described earlier, respondents indicated how important tolerance and community involvement are to them as personal values. These two values are conventionally thought of as important civic values that schools should instill. In terms of rating tolerance as a very important value, there is a strong sector effect among nonreligious independent school graduates, who are about 26 percentage points more likely than public school graduates to respond thus. While Catholic school graduates numerically rate tolerance as very important, this difference disappears after controlling for demographic characteristics. When it comes to the importance of community involvement, nonreligious independent school graduates are twice as likely as adults from public schools to say this is a very important personal value. Protestant and Catholic school graduates are less likely than public school graduates to consider community involvement as a very important personal value. 16 16 While the CES data cannot provide an explanation for this finding, it is nonetheless interesting, given these schools’ religious missions as well as the findings of previous research, which correlates community engagement and flourishing in Christian schools; see L.E. Swaner et al., “Flourishing Schools: Research on Christian School Culture and Community,” Association of Christian Schools International, 2019, https://www.acsi.org/docs/default-source/website-publishing/research/fsci-research-report-2019.pdf.

However, self-reports about one’s beliefs do not always perfectly align with one’s actions. To address this issue, the CES inquires about civic behavior, specifically respondents’ engagement with two important civic practices: charitable giving and volunteering.

Graduates of all independent school sectors are more likely to report giving charitably than their public school counterparts, while no difference is observed for the homeschooling sector. Interestingly, Protestant school graduates report the highest rates of charitable giving, despite having lower average incomes (see section 3). This finding is consistent with existing evidence suggesting that charitable-giving rates are higher among religious adults than nonreligious adults. 17 17 K. Zinsmeister, “Less God, Less Giving? Religion and Generosity Feed Each Other in Fascinating Ways,” Philanthropy Magazine (Winter 2019), https://www.philanthropyroundtable.org/magazine/less-god-less-giving/. Additionally, based on a separate analysis of the CES data, adults who attend religious services at least weekly (a standard measure of religiosity) are about 27 percentage points more likely to report giving charitably than adults who do not attend religious services at least weekly, even after accounting for household income and other demographic characteristics. Section 6 of this report discusses the religious formation of respondents, but readers should note here the suggestive evidence from the CES data—that the religious formation that occurred among Protestant school graduates appears to have translated into more frequent charitable giving.

According to data about the full CES sample, 28 percent of 24- to 39-year-olds report having done unpaid, volunteer work in the last twelve months. Protestant school graduates and homeschooled adults are more likely to report volunteering than are adults from public schools. The same is true for Catholic school graduates, but this is largely explained by demographic factors rather than a sector effect. The 2023 CES included follow-up questions asking respondents who had volunteered how frequently they participated in their volunteer work; no differences between sectors were observed in terms of the likelihood of respondents’ reporting that they volunteer on at least a monthly basis. The CES also included an item that queried respondents who volunteered about whether they held a leadership position in their volunteer work, for which no significant sector-level differences were observed.

Lastly, the CES included a list of seven types of service organizations and asked respondents if they had ever been involved with them. When looking at the full CES sample, among those who reported volunteering, the type of organizations most frequently selected were (in descending order): nonreligious social service (36 percent of all who volunteered); arts, sports, or cultural (24 percent); religious, other than one’s church or congregation (21 percent); neighborhood (16 percent); environmental (14 percent); and political (9 percent). As found in prior administrations of the CES, school sectors seem to channel individuals into different types of service work. 18 18 A. Cheng and D. Sikkink, “A Longitudinal Analysis of Volunteerism Activities for Individuals Educated in Public and Private Schools,” Youth & Society 52, no. 7 (2020): 1193–219, https://doi.org/10.1177/0044118X19861979. For instance, nonreligious independent school graduates are almost twice as likely as public school graduates to report volunteering in a nonreligious social-service organization, while Protestant school graduates and adults who were homeschooled are less likely to report the same. Adults who were primarily homeschooled during high school are more likely to report serving in religious organizations besides their own church or congregation than their public school counterparts, but the same cannot be said for Protestant school graduates (who report similarly to their public school counterparts) or for Catholic school graduates (who are less likely to report volunteering with a religious organization).

For the other types of service organizations, the following sector-level differences were observed. Adults from the homeschooling sector were much less likely (by 15 percentage points) than public school graduates to have volunteered in a sports, arts, or cultural organization. Volunteering in an environmental organization was also less likely among adults from the homeschooling sector as well as Protestant school graduates (at a difference of about 5 percentage points, as compared with public school graduates). Protestant and Catholic school graduates are relatively less likely to volunteer in a neighborhood association, by 10 percentage points. Finally, there is no evidence of differences by educational sector in the likelihood of volunteering in a political organization, a kind of organization in which very few respondents participated.

Growing attention is being paid to the role of schooling in student mental health and well-being, given the rise of youth mental health issues both prior and subsequent to the COVID-19 pandemic. 19 19 US Centers for Disease Control and Prevention, Youth Mental Health: The Numbers (2024), https://www.cdc.gov/healthyyouth/mental-health/mental-health-numbers.html. For the first time in its administration, the CES included several scales designed to measure aspects of mental health, including life satisfaction, depression and anxiety, and closeness of social relationships. The CES also measured subjective dimensions of respondents’ well-being, such as gratefulness, sense of purpose, optimism, and self-efficacy. The key finding for this domain concerns adults who were homeschooled, who are the only sector graduates to report the lowest levels of depression and anxiety; all other sector graduates report about the same (higher) levels. Protestant and Catholic schools can claim more graduates with the highest levels of life satisfaction, though this is largely attributable to other demographic factors rather than to school influence. Interestingly, about three-quarters of adults from all the educational sectors agreed with the statement “When I think about the future, I am positive.” This finding is encouraging, given prevalent accounts about post-pandemic hopelessness, disillusion, and disengagement in society. The remainder of this section unpacks the CES findings relative to a range of mental health and well-being outcomes by sector.

In the 2023 administration, CES respondents completed the Satisfaction with Life Scale. 20 20 E. Diener et al., “The Satisfaction with Life Scale,” Journal of Personality Assessment 49 (2010): 71–75, https://doi.org/10.1207/s15327752jpa4901_13. Respondents were asked to “State your agreement or disagreement with the following statements”: (1) “In most ways, my life is close to my ideal,” (2) “The conditions of my life are excellent,” (3) “I am satisfied with my life,” (4) “So far, I have gotten the important things I want in life,” and (5) “If I could live my life over, I would change almost nothing.” Possible response options were: strongly disagree (value=1), moderately disagree (value=2), slightly disagree (value=3), slightly agree (value=4), moderately agree (value=5), and strongly agree (value=6). Answers for all five items were then averaged to generate a composite score for each respondent. According to this measure, the average satisfaction-with-life score for adults from public schools was 3.99 out of a possible 6 points. Relative to this finding, adults who attended Protestant and Catholic schools report the highest levels of life satisfaction. However, there is no evidence of a sector effect, as adults from all educational sectors exhibited statistically indistinguishable levels of life satisfaction after accounting for demographic characteristics.

The CES also included selected items from two scales designed to screen for depression and anxiety. 21 21 Kroenke et al., “The PHQ-9: Validity of a Brief Depression Severity Measure,” Journal of General Internal Medicine 16, no. 9 (2001): 606–13, https://doi.org/10.1046/j.1525-1497.2001.016009606.x; R.L. Spitzer et al., “A Brief Measure for Assessing Generalized Anxiety Disorder: The GAD-7,” Archives of Internal Medicine 166, no. 10 (2006): 1092–97, https://doi.org/10.1001/archinte.166.10.1092. Respondents were asked, “Over the last 7 days, how often have you been bothered by: (1) “Having little interest or pleasure in doing things?,” (2) “Feeling down, depressed, or hopeless?,” (3) “Feeling nervous, anxious, or on edge?,” and (4) “Not being able to stop or control worrying?” Possible response options for each of these items were: not at all (value=1), one day (value=2), two to three days (value=3), more than half the days (value=4), and nearly every day (value=5). Answers for all four items were then averaged to generate a composite score for each respondent. 22 22 Importantly, while these items were drawn from the aforementioned two clinical scales, the CES does not use all of the items on these scales, nor does it follow the scoring methods required for clinical use. This means that the CES does not provide a diagnostically accurate picture of clinical depression or anxiety. Instead, data from these questions were used to generate a snapshot of the depression and anxiety levels self-reported by graduates of each sector. These results cannot, and should not, be construed as a clinically accurate measurement of depression and anxiety for an individual respondent, a group of respondents, the overall CES sample, or for the larger population of 24- to 39-year-olds in the US. The average score for adults from public schools was 2.05 out of a possible 5 points. Respondents across all educational sectors reported similar levels of depression and anxiety, with the notable exception of adults who were homeschooled, who reported the lowest levels. Considering sector-effect estimates, their depression and anxiety scores are, on average, 0.28 scale-points lower than the scores for adults from public schools—a large difference, considering that the standard deviation of depression and anxiety scores is about 1.13.

Respondents were asked questions from the Close Social Relationships domain of the Flourishing Measure, developed by the Human Flourishing Program at Harvard University, 23 23 T.J. VanderWeele, “On the Promotion of Human Flourishing,” Proceedings of the National Academy of Sciences 31, no. 114 (2017): 8148–56, https://doi.org/10.1073/pnas.1702996114. along with other items designed to measure sense of belonging. Specifically, the CES asked respondents to “State your agreement or disagreement with the following statements”: (1) “I am content with my friendships and relationships,” (2) “My relationships are as satisfying as I would want them to be,” (3) “I feel lonely,” (4) “I feel a sense of belonging to my local community,” and (5) “If I were in trouble, I have relatives or friends whom I can count on to help me when I need them.” Possible response options were: strongly disagree (value=1), moderately disagree (value=2), slightly disagree (value=3), slightly agree (value=4), moderately agree (value=5), and strongly agree (value=6). Answers for all five items were then summed and averaged to generate a composite score for each respondent.

The average Close Social Relationships score for adults from public schools was 4.19 points out of a possible 6 points. Catholic school graduates are the only group to report higher levels of closeness in their social relationships as compared to their public school counterparts. Scores on the measure for the closeness of social relationships are, on average, 0.25 scale-points higher for Catholic school graduates. However, the difference is not conclusively due to a sector effect, as the statistical difference in the measure of the closeness of social relationships between adults from Catholic schools and adults from public schools disappears after accounting for demographic characteristics. Nor is there any evidence that scores on the measure of the closeness of social relationships is different among adults from the sectors other than Catholic schools.

Respondents answered a series of items designed to measure aspects of subjective well-being such as gratefulness, sense of purpose, optimism, and self-efficacy. 24 24 M.E. Lachman and S.L. Weaver, “Sociodemographic Variations in the Sense of Control by Domain: Findings from the MacArthur Studies of Midlife,” Psychology and Aging 13 (1998): 553–62, https://doi.org/10.1037/0882-7974.13.4.553; M.E. McCullough et al., “The Grateful Disposition: A Conceptual and Empirical Topography,” Journal of Personality and Social Psychology 82 (2002): 112–27, https://doi.org/10.1037//0022-3514.82.1.112; J.M. Salsman et al., “Assessing Meaning & Purpose in Life: Development and Validation of an Item Bank and Short Forms for the NIH PROMIS,” Quality of Life Research 29, no. 8 (2020): 2299–310, https://doi.org/10.1007/s11136-020-02489-3; A. Zuffianò et al., “The Positivity Scale: Concurrent and Factorial Validity Across Late Childhood and Early Adolescence,” Frontiers in Psychology 24, no 10. (2019): 831, https://doi.org/10.3389/fpsyg.2019.00831. These items have been asked on prior administrations of the CES. With all of these items, differences across the educational sectors are not pronounced. For instance, about 78 percent of respondents from all the educational sectors agreed with the statement “When I think about the future, I am positive.” There is no evidence of sector differences relative to graduates of traditional public school except that homeschoolers were a statistically significant 9 percentage points more likely to agree with that statement. Likewise, 34 percent of respondents from public schools and all independent school sectors agreed with the statement “I often feel helpless dealing with the problems of life”; however, homeschoolers were a statistically significant 11 percentage points less likely to agree with the statement after accounting for background demographic characteristics. Levels of sense of purpose also seem to be similar across all the sectors. In this case, however, respondents who were homeschooled were most likely to agree with the statement “My life seems to lack any clear goals or sense of direction.” Specifically, those from the homeschooling sector were 14 percentage points more likely than public school graduates to agree with this statement. In sum, compared to graduates of traditional public schools, graduates from homeschool settings feel more optimistic about the future, more able to deal with life’s problems, but less of a sense of life aim.

Finally, respondents’ level of gratefulness was measured by soliciting their agreement or disagreement with the statement “I have so much in life to be thankful for.” Catholic school graduates appear most likely to agree with this statement, at 6 percentage points more likely than public school graduates to express agreement. Those from all the other sectors are a few percentage points more likely to express agreement as well, but only the Catholic school difference is statistically significant. Nonetheless, there is no evidence of a sector effect, as all differences shrink and are not statistically significant after controlling for demographic variables.

The CES measures religious belief, practice, and values as indicators of faith formation. Questions for this domain were adapted from the Spectrum of Spirituality Index developed by Cardus and the Angus Reid Institute. Consistent with previous administrations of the CES, faith formation is the domain in which the strongest evidence for sector-level effects can be observed, with a clear benefit to respondents who graduated from Protestant schools or were homeschooled. These graduates are much more likely to report believing in God and life after death, as well as engaging in religious practices (regular prayer, Bible reading, attending weekly religious services), feeling God’s presence regularly, and agreeing that it is important for parents to teach their children about religious beliefs. The remainder of this section explores the CES findings relative to faith formation by sector.

When asked whether they believe “that God or a higher power exists,” about 40 percent of respondents from public schools answered, “Yes, I definitely do believe,” as opposed to “Yes, I think so,” “No, I don’t think so,” or “No, I definitely do not believe.” Respondents from Protestant schools and homeschooling settings are, respectively, 29 and 27 percentage points more likely than public school graduates to believe in the existence of God or a higher power. There is evidence for a sector effect as well: after controlling for demographic characteristics, the difference in likelihood of belief in God or a higher power between adults from Protestant schools and adults from public schools decreases but remains a sizable and statistically significant 20 percentage points. The same can be said for adults who were mostly homeschooled in their high school years.

About one third of respondents from public schools reported, “Yes, I definitely believe” in life after death. Protestant school graduates are 32 percentage points more likely to believe in life after death—double the rate of public school graduates. Homeschooled adults were 26 percentage points more likely to believe in life after death. As with belief in God, there appears to be a sector effect: after controlling for demographic characteristics, the difference is still a large and statistically significant 24 and 21 percentage points for Protestant school graduates and adults from homeschool settings, respectively.

Like sector differences regarding religious belief, sector differences regarding religious practice are quite apparent. In general, Protestant school graduates and adults who were homeschooled are more likely to report engaging in religious practices. While 37 percent of adults from public schools report they pray at least weekly, those from Protestant schools or homeschool settings are about 25 percentage points more likely to say they pray at least weekly. Even after controlling for demographic variables, adults from Protestant schools are 14 percentage points more likely to report praying weekly, suggesting a sector-level effect. Similarly, demographic variables do not explain the higher report of weekly prayer among homeschooled adults. Catholic school graduates are also more likely to say they pray at least weekly compared to adults from public schools, by about 8 percentage points, but there is no evidence of a Catholic school-sector effect.

When asked how often they “read the Bible or other sacred text,” only 18 percent of adults from public school reported reading a sacred text on at least a weekly basis. Comparatively, adults from Protestant schools and homeschool settings were more likely to say they did so on at least a weekly basis (24 and 36 percentage points, respectively). After accounting for demographic characteristics, adults from Protestant schools are still 16 percentage points more likely to report reading a sacred text on a weekly basis, while the difference for homeschooled adults does not change. Results concerning attendance at religious services are almost the same as results for reading a sacred text.

Given the findings for religious practice, it is unsurprising to also find that adults from Protestant schools and homeschool settings were more likely than other adults to feel God’s presence on at least a weekly basis. After adjusting for demographic differences, Protestant school graduates were 9 percentage points more likely to feel God’s presence on at least a weekly basis, relative to adults from public schools (although this is a large gap, considering that only 30 percent of adults from public school feel God’s presence at least weekly, this finding falls short of conventional thresholds to be considered statistically significant). The difference in likelihood between homeschooled adults and public school graduates is even larger, at 29 percentage points, and persists even after controlling for demographic characteristics.

These results regarding faith formation are consistent with respondents’ self-reports about how important religion is to them. Again following the approach of the Wall Street Journal/NORC poll, 25 25 “WSJ/NORC Poll March 2023.” respondents were asked to rate how important religion was as a personal value. About one-fifth of adults from public schools view religion as a very important value. Adults from Protestant schools and homeschool settings were more likely than public school graduates to consider religion a very important personal value; after controlling for demographic characteristics, however, this finding was statistically significant only for homeschooled adults.

Finally, respondents were asked, using a six-point Likert scale, how strongly they agreed or disagreed that it was important for parents to teach their children about religious beliefs. Public school graduates were much more likely to agree with this statement than to consider religion a very important personal value. These findings, perhaps, reveal that many public school graduates think it is important that children learn about religion but place less emphasis on practicing or being influenced by a particular religion. Catholic and nonreligious independent school graduates were no different from adults from public schools when it comes to religious practice but were slightly more likely to agree that parents should teach their children about religion, compared to adults from public schools. It is possible that adults from these two sectors might likewise emphasize the importance of learning about religion while deemphasizing its practice or how it should influence one’s life. The closest alignment between values was evident among adults from Protestant schools and homeschool settings, who not only more frequently reported engaging in religious practices and holding religion as a very important personal value but were also much more likely to believe that parents should teach their children about religious beliefs.

Nevertheless, it is not as if adults from the nonreligious independent school or public-school sectors are not concerned with developing a sense of purpose or experiencing transcendence. Based on the CES data, it appears that those adults are looking for self-fulfillment but not by engaging in religion as it is traditionally understood. When asked how important the value of self-fulfillment was to them personally, graduates of these two sectors were much more likely than adults from Protestant schools and homeschool settings to consider it a very important personal value. The contemporary search for self-fulfillment may be, as some scholars have observed, the modern alternative for religion. 26 26 C. Taylor, A Secular Age (Harvard University Press, 2007).

The family continues to be a foundational institution that supports the stability and flourishing of society. The CES measures marriage status and number of children, including adoption and the value of having children. Across all sectors, there were no significant differences in graduates’ responses in most areas related to marriage, divorce, number of children, or the importance of having children. Only nonreligious independent schools appear to have an effect on graduate outcomes in this domain, specifically related to lower marriage rates. The remainder of this section explores the CES findings relative to family formation by sector.

After adjusting for demographic characteristics, the data show a sector-level effect only for nonreligious independent school graduates in terms of marriage rates. First, there is some evidence that nonreligious independent schools reduce the likelihood that their graduates report ever having been married, by nearly 15 percentage points compared to those from public schools. No statistically significant evidence was found in terms of differences in the likelihood of never marrying, as well as currently being married, for any other sector.

Despite this difference in marriage rates for nonreligious independent school graduates, there are no significant differences in the likelihood of being divorced or being separated for this sector or any other. Of respondents who have ever been married, only 4 percent of those from public schools are currently divorced or separated. Almost no adults from nonreligious independent schools and homeschool settings are divorced or separated. Meanwhile, adults from Protestant schools are 3 percentage points more likely to be divorced compared to adults from public schools, but that difference is not statistically significant. After controlling for demographic characteristics to estimate sector effects, the difference between adults from Protestant schools and adults from public schools still remains statistically insignificant, even though the gap increases to about 6 percentage points.

When asked how important marriage was as a personal value to respondents, 35 percent of adults from public schools said that it was very important. Protestant school graduates were 14 percentage points more likely to share that perspective, which seems borne out in their relatively higher rates of marriage. Homeschooled adults are also more likely than public school graduates to endorse marriage as a very important personal value. And homeschooled adults are more likely to marry and less likely to divorce compared to adults from public schools, though those differences are not statistically significant.

Another central aspect of family formation is childrearing. The CES asked whether respondents had children and if they adopted any of their children. According to the data, 42 percent of adults from public schools reported having children. The proportion of graduates from the other school sectors who reported having children was approximately equal to or less than the proportion of adults from public schools. In sector-effect estimates based on this data, lower proportions of Protestant school and nonreligious independent school graduates have children, though the differences are not statistically significant. Adults from homeschool settings are also less likely to have children compared to public school graduates, but once again, that difference is not statistically significant. These results are also consistent with data from the 2011 iteration of the CES, which found no differences in the likelihood of having children between adults who were homeschooled or public school graduates. When it comes to adoption, 4 percent of public school graduates said that they have adopted a child. Similar proportions of adults from Protestant school, nonreligious independent school, and homeschool settings have adopted a child. Catholic school graduates are least likely to have adopted, with only about 2 percent of them having done so—a rate that is statistically indistinguishable from the rate of adults from public schools.

Consistent with the few differences in the likelihood of adopting or having children, there is no evidence of association between school sector and the importance that respondents placed on having children. Although adults from all sectors were more likely to say that having children is a very important personal value, when compared with public school graduates, none of the differences are statistically significant except for an 8 percentage-point gap between Catholic and public school graduates. However, this difference is smaller and no longer statistically significant after controlling for demographic characteristics.

Results from the 2023 CES have thus far been presented by specific outcomes (perceptions of high school, educational attainment and employment outcomes, civic formation, and so forth). In this section of the report, findings are grouped by sector in order to show the unique profiles of each sector’s graduates. Three of the five school sectors are discussed in this section: Protestant, Catholic, and nonreligious independent. 27 27 Since public school data were used as a baseline for sector comparisons, outcomes for this sector are not reported separately. A forthcoming report will share detailed findings related to homeschooling, since homeschoolers often differ in outcomes based on the number of years spent in homeschooling versus other school settings.

Importantly, the purpose of sharing CES findings in this way is not to present a comparative view of sectors. Rather, it is to provide readers who have an interest in a particular sector with a narrative description of CES findings for that sector. It is hoped this description will be helpful to readers who are seeking to identify unique strengths and potential areas for growth and improvement for a specific sector.

Overall, graduates of Protestant schools rate their schools’ quality about the same as public school graduates do. When it comes to the specifics of their high school experience, however, graduates of Protestant schools were significantly more positive than public school graduates about their schools as close-knit communities, their relationships with teachers, the preparation their schools provided them to have strong personal relationships as adults, and their preparation for academic success in a postsecondary setting as well as for success in a job or career. They also were more likely to believe they are very well prepared to have vibrant spiritual lives, and rated their schools more highly in terms of how they handled religious and spiritual matters.

In terms of areas for improvement, graduates of Protestant schools were more likely to provide unfavorable ratings of the extracurricular activities at their school. This is perhaps unsurprising, given the funding realities for independent religious schools in the US, along with the overall lower tuition rates of this sector as compared to other types of independent schools. 28 28 In 2019, Christian school tuition at the median was $6,561, as compared with $26,866 for NAIS schools. See Association of Christian Schools International, “2018–2019 Tuition & Salary Survey: Member Report,” ACSI, 2019; National Association of Independent Schools, “Stats of Interest: 2019–20 NAIS National Tables and 2019–20 Facts at a Glance,” NAIS, 2020,  https://www.nais.org/parents/learn/stats-of-interest/. Of perhaps greater importance, however, is the finding that Protestant school graduates are slightly less likely than public school graduates to agree with the statement “I felt like I belonged at my school,” as well as slightly less favorable when it came to relationships with other students at their school. This was an interesting finding, given Protestant school graduates’ higher level of affirming their schools as close-knit communities. As mentioned earlier in this report, sense of belonging and relationship quality with other members of the school community are strongly correlated with graduates’ evaluations of their school’s educational quality and experience. For those seeking to improve the experiences of students in Protestant schools, overall sense of belonging and peer relationships may be important areas to consider.

In terms of academic and vocational outcomes, graduates of Protestant schools were no more likely to attain a bachelor’s degree or graduate degree than were graduates of public schools. They were also about as likely to be employed as were their public school counterparts. Differences between Protestant school and public school graduates were identified in terms of vocational values, however. Along with Catholic school graduates, Protestant school graduates were more likely than public and nonreligious independent school graduates to affirm that it is very important to have a job that provides an opportunity to be directly helpful to others. They were also less likely to consider money a very important personal value.

Moving from vocational to civic outcomes, Protestant school graduates were less likely to say that community involvement is a very important personal value. 29 29 While the CES data cannot provide an explanation for this finding, it is nonetheless interesting, given these schools’ religious missions as well as the findings of previous research, which correlates community engagement and flourishing in Christian schools; see Swaner et al., Flourishing Schools, 2019. Despite this, they have one of the highest rates of charitable giving, even with lower average incomes than those in at least two other sectors. They are also significantly more likely to volunteer than all other sectors, though not necessarily with any more frequency. Finally, compared to their public school counterparts, graduates of Protestant schools are no more or less likely to serve in a political organization or a religious organization besides their own church or congregation. They are, however, less likely to volunteer in an environmental organization or neighborhood association.

Numerically speaking, Protestant schools can claim more graduates with the highest levels of life satisfaction. This difference disappears after controlling for graduates’ demographic characteristics, however, which suggests that it is attributable to variables other than school type. Protestant school graduates reported similar levels of depression and anxiety compared with all other school types, except for homeschooling (for which graduates evidenced lower rates of both).

Consistent with previous administrations of the CES, faith formation is the domain in which the strongest evidence for sector-level effects can be observed, with a strong benefit for graduates of Protestant schools. For example, after controlling for demographic characteristics, the CES found that adults who graduated from Protestant schools were much more likely to report believing in God and in life after death, to engage in religious practices (including regular prayer, Bible reading, attending weekly religious services), and to report regularly feeling God’s presence. Graduates of Protestant schools are also much more likely to agree that it is important for parents to teach their children about religious beliefs. These observations persist after controlling for demographic characteristics, which suggests that these benefits in faith formation may be attributed to attending a Protestant school.

Finally, when it comes to marriage and children, more Protestant school graduates report having ever been married than their public school counterparts. They are also more likely to say that marriage is a very important personal value. However, neither finding appears to be attributable to school sector but rather may be due to other factors (such as the influence of family of origin). While not statistically significant, graduates of Protestant schools are slightly more likely to be divorced and less likely to adopt children than public school graduates.

For graduates of Catholic schools, their perceptions of high school were significantly more positive overall than public school graduates. Catholic school graduates were more positive about their schools as close-knit communities, their relationships with teachers, the preparation their schools provided them to have strong personal relationships as adults, and their preparation for academic success in a postsecondary setting as well as for success in a job or career. Catholic school graduates believe their schools prepared them to have vibrant spiritual lives at higher rates than public school graduates, and had a more favorable rating of how their school handled religious and spiritual matters. They were more likely than public school graduates to provide a favorable rating of their school’s extracurricular opportunities. Finally, Catholic school graduates rated their schools more highly than their counterparts when it comes to rating how well their school prepared them for interacting with the range of people in society, though they are not statistically different from public school graduates when it comes to how well their school handled politics.

In terms of educational attainment and employment outcomes, Catholic school graduates reported the highest levels on multiple metrics, including completing a bachelor’s degree. They are also twice as likely to have obtained a graduate degree as compared to graduates of all other sectors. Their self-reported income levels are also higher than graduates of public and Protestant schools, and they are less likely to be unemployed than graduates of all other sectors. For these reasons, it appears that attending a Catholic school provides a boost to educational attainment and employment outcomes, regardless of students’ background characteristics.

Responses from Catholic school graduates painted more of a mixed picture when it comes to values and civic behaviors. Along with Protestant school graduates, Catholic school graduates were more likely to affirm that it is very important to have a job that provides an opportunity to be directly helpful to others, and were less likely to consider money a very important personal value. However, they were less likely than their public school counterparts to say that community involvement is a very important personal value, less likely to serve in a religious organization or neighborhood association, and no more likely to volunteer in a political organization.

In terms of sheer numbers, Catholic schools can claim more graduates with the highest levels of life satisfaction, close social relationships, and levels of gratefulness. However, this difference disappears after controlling for graduates’ demographic characteristics, which suggests that they are attributable to variables other than school type. Catholic school graduates reported similar levels of depression and anxiety compared with all other school types except for homeschooling (for which graduates reported lower levels of both). Finally, graduates of Catholic schools had similar responses to public school graduates in terms of marriage and divorce rates, number of children, and adoption rates.

For graduates of nonreligious independent schools, their perceptions of high school were significantly more positive overall than were the perceptions of public school graduates. This also held true when they were asked about their schools as close-knit communities, their relationships with teachers, the preparation their schools provided them to have strong personal relationships as adults, and their preparation for academic success in a postsecondary setting as well as for success in a job or career. Graduates of nonreligious independent schools were more likely to report feeling a sense of belonging at their school and more likely to agree with the statement “I enjoyed being a student at my school.” As mentioned earlier, this may be linked or otherwise contribute to their more positive rating of their overall high school experience.

Interestingly, although nonreligious independent schools are not explicitly religious in nature, graduates from these schools were likely to have a favorable rating of how their school handled religious and spiritual matters. They also rated their schools more highly than public school graduates did when it came to how well school prepared them for interacting with the range of people in society. Nonreligious independent schools are also the only sector more likely to receive favorable ratings regarding how they handled issues of national politics.

In terms of educational attainment, nonreligious independent school graduates have the same likelihood of completing a bachelor’s degree as public school graduates. However, along with their Catholic school counterparts, they are more likely to obtain a graduate degree than other sectors. The self-reported income levels of nonreligious independent school graduates were also higher than those of public and Protestant school graduates. Intriguingly, nonreligious independent school graduates are more likely not to be employed compared to public school graduates. This result seems to run counter to the higher incomes and levels of educational attainment among nonreligious independent school graduates. For reasons that cannot be ascertained from the CES data, the income penalty for not being employed is not as strong among nonreligious independent school graduates compared to adults from other sectors.

In terms of vocational values, nonreligious independent school graduates were more likely than their public school counterparts to say it is very important to have a job that fulfills their calling. However, like their public school counterparts, nonreligious independent school graduates were much more likely to rate self-fulfillment as a very important value, compared with graduates of religious schools. They were also more likely than graduates of other sectors to say that community engagement and tolerance are very important values. This suggests that while nonreligious independent school graduates may not experience a boost in faith formation outcomes, their personal values are nonetheless shaped by their schools in significant ways.

A major finding for nonreligious independent schools is in the area of family formation. Compared to adults from public schools, graduates from this sector are the only ones who are significantly more likely to never marry. On all other metrics—such as divorce rates, number of children, and adoption rates—graduates of nonreligious independent schools are similar to adults who graduated from public schools.

The Cardus Education Survey examines a range of holistic outcomes for a nationally representative sample of high school graduates, aged 24 to 39, from five major school sectors (public schools, Protestant schools, Catholic schools, nonreligious independent schools, and homeschooling). This report shared the responses of graduates from these sectors regarding their perception of their high school quality and experiences, as well as how they are faring in adulthood on a range of holistic outcomes, including educational attainment and employment outcomes, civic formation, mental health and well-being, faith formation, and family formation. By controlling for a range of respondents’ demographic characteristics, the CES enables estimating sector-level effects—meaning, the difference that attending a specific kind of school makes.

Looking broadly across the CES results, sector-level differences are observable in many areas of attainment, perceptions, beliefs, values, and behaviors reported by graduates. Two important conclusions can be drawn. First, and most obviously, attending a specific type of school can contribute differentially to graduate outcomes. Because these differences often appear to align with the missions and distinctives of the school type (for example, faith-formation benefits accrue most strongly to Protestant school graduates and adults who were homeschooled), these differences should be considered by families as they make decisions about schooling for their children. Moreover, in a pluralistic society—where families hold diverse values, beliefs, and priorities—education can never be one-size-fits-all. It follows that regardless of economic background, families should be afforded the maximum opportunity to choose educational options they deem most suited for their children.

Finally, the CES provides evidence that the common-good outcomes essential to maintaining democratic vibrancy are being produced by schools of varying types. Public schools, attended by a majority of US students, remain important and essential to the common good. But CES results have consistently shown that for many common-good outcomes, other school sectors match—and in not a few cases, outperform—public schools. Beyond confirming that a single education-delivery model is unmerited, these findings provide strong evidence that educational pluralism holds promise for the flourishing of individuals and for society as a whole.