Religious Retention

Much has been written about the decrease in religious affiliation in the United States during the most recent decades.⁴¹ For example, the PRLS found that 78% of US adults identified as Christians in 2007, but that the number dropped to 71% by 2014 and then continued to drop, reaching 63% in 2024. Across the same period, those who identified as religiously unaffiliated rose from 16% in 2007 to 28% in 2024.⁴² Religious scholars have identified 1991 as the crucial turning point, marking the beginning of a substantial drop in religious commitment.⁴³

Retention in 2024–2025 Across Generations

To estimate Latter-day Saint retention rates, we used the nationally representative PRLS and SSS data. We also explored retention of youth ages 12–18 using the longitudinal FFYD data. In table 3, we show retention rates in 2024 and 2025 from the PRLS, SSS, and FFYD. In the PRLS and SSS, approximately half of those who were raised Latter-day Saint continued to identify as Latter-day Saint as an adult. Looking specifically at youth in the FFYD, around three quarters of those who indicated they were a Latter-day Saint in 2016 (when they were 12–14 years old) also indicated they were a Latter-day Saint eight years later in 2024 (when they were 20–22 years old). It should be noted that retention is likely much lower in the PRLS and SSS than the FFYD because the PRLS and SSS contain those of all age groups and the FFYD contains only those in their late teens to early 20s. In other words, there was less time for religious change to occur for these younger Latter-day Saints.

We then examined how retention rates may differ across religions. As noted in the introduction, the comparison group for retention is important. We therefore made comparisons to the most specific level of analysis we were able, denominations or religious families (as defined by the PRLS). For the most part in the PRLS data, the smallest unit of analysis is religious family (for example, Baptist family, Lutheran family, and others). This level of analysis with the PRLS is not ideal given that it would still count as retention if a denomination within a religious family lost members to another denomination within the same family. Thus, the retention rates of these families are higher than retention rates for the individual denominations within the religious families. We exclude denominations with fewer than 200 respondents from our comparisons, since small sample sizes produce large margins of error—especially when comparing across generations.

In figure 16, we used the PRLS data and compared retention rates across religions in 2024. The highest retention rate is for Hindus and then Muslims, Jews, Orthodox Christians, Catholics, Latter-day Saints, and Baptists. However, the Latter-day Saint rate is not significantly different from Catholics. Latter-day Saints have a higher retention rate than other Christian religions (except Orthodox Christian and Catholics) and Buddhists. However, with the confidence intervals, Latter-day Saints are statistically the same as Catholics and Baptists. Given that nonparticipation in an organization is conceptually and practically different from participating in an organization, we did not include those who identify as “unaffiliated” (meaning there is no religious organization to which they belong) in figure 16. However, we did examine the stability of such religious nonparticipation, finding that 74% of those raised nonreligious remain nonreligious as adults, comparable to the retention rate of Muslims, Hindus, and Catholics.

Retention Across Time

We also explored how Latter-day Saint retention rates have changed over time. We did so using the GSS data. Although the GSS contains fewer Latter-day Saints than the other surveys, it is the only survey where retention can be estimated back to the 1980s. When estimating retention in the 2000s, the best data come from the PRLS and SSS data. However, one cannot use those data to examine changes in retention across several decades. As with the above, we define “retention” as someone saying they were a Latter-day Saint as a teen, and as an adult they continue identifying as a Latter-day Saint. Importantly, this is independent of how active they were as a Latter-day Saint during their teenage years.

As found in the GSS data, in the 1980s, we calculated Latter-day Saint retention at 82%. That is, 82% of those who were raised Latter-day Saint indicated that they still identified as Latter-day Saint as adults. This identification number dropped slightly to 76% by the 1990s and then more steeply to 58% by the 2000s. In line with religious identification trends broadly, by the 2010s and beyond, the retention rate was slightly less than half at 46%. The rate of 46% from the GSS is similar to what was reported above from the PRLS and SSS data, which put the retention rate at just over 50% in 2024 and 2025.

Using the GSS, we also explored how active participation has changed over time for those raised Latter-day Saint. In 1980, 65% of those raised Latter-day Saint attended religious services at least monthly, while 16% attended several times a year or less and 19% no longer identified as Latter-day Saint. Over the following decades, the percentage of those who attended a few times a year or less but continued to identify as Latter-day Saint decreased. From 2010 to 2024, only 7% of those raised Latter-day Saint continued to identify as Latter-day Saint when they attended church a few times a year or less. This suggests that today, Latter-day Saints who do not attend church are less likely to identify as Latter-day Saint than they were in the past. It may be that the increased negative perception of religion in general (or The Church of Jesus Christ of Latter-day Saints in particular)⁴⁴ makes it less desirable to identify as actively religious.⁴⁵ In the past, people may have identified with a particular religion even though they were not attending church, whereas today, continuing to identify may come with an increase in perceived social costs.⁴⁶

Retention for Younger Generations

Using the GSS, we also explored how Latter-day Saint Millennial+ retention rates compare to retention rates of previous generations. Retention rates across the generations decreased from the 1990s to the 2000s. Boomers retention held steady from the 2000s to the 2010s, while Millennial+ had the greatest drop in retention, from just over 60% in the 2000s to just above 40% in the 2010s. We should note that many Millennial+ were still young prior to the 2000s and therefore did not have as much time to leave their religion than those of other generations. These data also do not provide insights into individuals who may have returned to religious faith later in life.

Active Retention

We were not only interested in whether an adult continues to identify with their childhood religion but whether he or she is, to some extent, engaged in that religion as an adult.⁴⁷ In figure 17, we broke down retention by frequency of religious service attendance. We divided PRLS respondents who were raised Latter-day Saint into three groups: (1) those who continue to identify as Latter-day Saint as adults and attend religious services at least monthly, (2) those who continue to identify as Latter-day Saint as adults but attend religious services only a few times a year or less, and (3) those who no longer identify as Latter-day Saint. We did the same for those raised in other religions, examining continued identification and religious service attendance together.

At 42%, Latter-day Saints have the highest percentage of retention (i.e., continuing to identify as a Latter-day Saint as an adult) and monthly attendance of any religion. This is followed by Muslims at 38%, then Nondenominational at 28%. For Jews, who are the third highest in overall retention, only 18% of respondents are retained and attend monthly.

Essentially, it is a minority of Latter-day Saints who do not attend at least monthly and still identify as Latter-day Saint (12%). In contrast, Jews, Hindus, and Muslims all have substantial numbers (around half) of adherents who were raised in these religions and still identify as such but attend religious services a few times a year or never. For these religions, expectations for attending religious services may differ from Latter-day Saints, where frequent attendance is emphasized. These findings indicate the importance of attendance as a feature of identifying as Latter-day Saint.

We conducted the same analysis for Millennial+ (born in or after 1980) with similar trends observed. As indicated in figure 18, Latter-day Saint Millennial+ are statistically even with Muslims in the percentage retained who also attend services at least monthly (39%). The next closest is the Nondenominational family at 30%.

Retention by Demographics

We examined whether retention rates among Latter-day Saints differ across a host of demographic characteristics, such as gender, age, income, race, and location of residence.

Gender and Age

Using the GSS, we compared the retention rates of Latter-day Saint women with Latter-day Saint men over time. Latter-day Saint men’s retention rate remained steady at just over 80% from the 1980s to the 1990s and then dropped to just under 50% in the 2000s. It then declined only slightly through 2024 (combining those years together). In comparison, Latter-day Saint women’s retention rate remained stable at just under 70% from the 1990s to the 2000s and then declined to just under 50% from the 2000s through 2024.

We then explored whether retention rates differ by gender and age (fig. 19). Using the 2025 SSS data, we found that 18-year-old Latter-day Saint men have a retention rate of just below 60%, while older men have slightly higher rates, reaching just above 60% for the oldest men. In contrast, younger Latter-day Saint women have a much lower retention rate compared to older Latter-day Saint women: Eighteen-year-old Latter-day Saint women have a retention rate of just below 40%, while the oldest Latter-day Saint women in the data have a 60% retention rate.

This finding suggests a striking generational difference for women, with younger women much less likely to continue identifying as Latter-day Saint than women of older generations. In contrast, men are nearly just as likely to stay in the faith no matter what generation they were born into. However, these data do not account for those who may have left the Church and returned. It may be that women are more likely to leave at younger ages but are also more likely to return later.

Using the PRLS, we also explored this in other religions (see appendix C). The findings were somewhat similar, with men’s retention rate 10% higher than women’s at ages 18–24 (fig. C1). However, this difference is not statistically significant, and at no age are men and women significantly different from each other. For women, the retention rate does significantly increase as women age. In other words, women at older ages are more likely to be retained than women at younger ages, there being an approximate 30% difference.

We also explored whether this age difference in retention is the same across various religious families (fig. C2). Latter-day Saint women follow similar retention trends with Catholics and Baptists (both in absolute level and in the retention rate increasing with age).⁴⁸ Female Latter-day Saints ages 18–24 are relatively average in their retention rate, though older Latter-day Saint women have the highest retention rate of any religious family. Given that their retention at 18–24 is average, Latter-day Saint women appear to have one of the largest generational differences, though other religious families follow a similar pattern. At the same time, we again note that when referring to a religious family’s retention rate, each denomination within that family will necessarily have a lower retention rate than the overall family due to individual members switching denominations within the family.

Income and Education

Using the SSS data, we specified logistic regressions to see whether income and education levels are related to Latter-day Saint retention rates. We controlled for age because younger individuals will likely have lower incomes and less education than older individuals. Figure 20 shows retention rates by categories of income in the SSS. Those making less than $50,000 a year are 50% or lower in their retention rates, while those making above that are almost all close to 60% retention. We note that the two highest income categories have particularly large 95% confidence intervals making us less certain of their retention rates. Other research finds a similar relationship between religiousness and income, with religiousness increasing across income categories until the highest categories of income, where religiousness levels off or even decreases.⁴⁹

Our analyses also showed that those with higher education are more likely to remain Latter-day Saint (fig. 20): Those who have less than a college degree have nearly a 50% retention rate, while those with a bachelor’s degree or some graduate school education have about a 60% retention rate. This trend fits with research that finds a positive correlation between religiousness and education level.⁵⁰

Race

In the SSS data, of those raised Latter-day Saint, 826 identified as White, 40 as Black, 180 as Hispanic, and 31 as Asian. Retention rates for these groups are as follows:

White, 54%

Black, 49%

Hispanic, 50%

Asian, 59%

There are no statistically significant differences in these percentages. However, given that the sample sizes of Blacks and Asians are rather low, comparisons with these demographics should be interpreted with caution.

Location of Residence

Using the SSS, we divided the sampled individuals into the nine census regions:

• the New England region: Maine, New Hampshire, Vermont, Massachusetts, Rhode Island, and Connecticut
• the Middle Atlantic region: New York, New Jersey, and Pennsylvania
• the East North Central region: Ohio, Indiana, Illinois, Michigan, and Wisconsin
• the West North Central region: Minnesota, Iowa, Missouri, North Dakota, South Dakota, Nebraska, and Kansas
• the South Atlantic region: Delaware, Maryland, Virginia, West Virginia, North Carolina, South Carolina, Georgia, Florida, and the District of Columbia
• the East South Central region: Kentucky, Tennessee, Alabama, and Mississippi
• the West South Central region: Arkansas, Louisiana, Oklahoma, and Texas
• the Mountain region: Montana, Idaho, Wyoming, Colorado, New Mexico, Arizona, Utah, and Nevada.
• the Pacific region: Washington, Oregon, California, Alaska, and Hawaii

Because the New England region had only fifteen participants, we combined it with the Middle Atlantic region.

Figures 21–22 have retention rates for the various US regions. The Mountain and West North Central regions have the highest rates, while the lowest rates are in the New England and Middle Atlantic, South Atlantic, and West South Central regions.

Predictors of Deidentification and Groups of Deidentifiers

We used the FFYD data to examine predictors of deidentifying as Latter-day Saint because, with these data, we could explore how earlier characteristics predict whether someone would deidentify later in life. Specifically, we looked at predictors of whether youth would deidentify as they transitioned into young adulthood between 2020 and 2024.

In these analyses, 709 youth identified as Latter-day Saint in 2020, with 92 (13%) no longer identifying as such by 2024. We examined whether the following characteristics in 2020 predicted deidentifying by 2024: parent education, parental divorce status, family income, parent and adolescent political views, adolescent depression, adolescent anxiety, and adolescent personality. We also included numerous adolescent religious behaviors: attending church, praying, reading scriptures, keeping the Sabbath day holy, attending seminary, attending the temple, researching ancestors, giving a talk or lesson in church, sharing religious beliefs with someone not of their faith, and spending time performing Church duties.⁵¹

We also examined more internal aspects of religion, such as whether adolescents felt a secure attachment to God, how often they felt God’s presence, and whether they had a testimony that The Church of Jesus Christ of Latter-day Saints is the true Church. Finally, we examined how support for somewhat controversial issues related to deidentification: whether adolescents felt women should be ordained to the priesthood, whether they thought the Church should fully accept same-sex marriage, the degree to which issues of Church history had troubled them, and the degree to which the Church’s views on social and political issues had bothered them.

When we included all of the above in a logistic regression model⁵² that predicted retention after four years (between 2020 and 2024), the following significantly predicted greater likelihood of deidentifying: not feeling God’s presence in one’s daily life, agreeing that the Church should accept same-sex marriage, and feeling troubled by the Church’s stance on political issues (see table 4). For those who almost never felt God in their daily lives, 24% deidentified.⁵³ However, for those who felt God’s presence every day, only 4% deidentified. For those who strongly agreed that the Church should accept same-sex marriage, 28% deidentified, while only 5% of those who strongly disagreed deidentified.⁵⁴ Finally, for those who felt very concerned or troubled about the Church’s stance on social and political issues, 22% deidentified, while fewer than 10% of those who felt little or no concern deidentified.

The relationship between political issues and religion is not unique to The Church of Jesus Christ of Latter-day Saints. Indeed, nationally, while only 17% of religiously affiliated individuals are liberal or very liberal, over 55% of Atheists/Agnostics are.⁵⁵ And this divide is relatively recent. Before the 1980s, unaffiliated individuals were just as likely to be Republican as Democrat. However, since then, the divide has widened such that Republicans are far more likely to be religious than Democrats.⁵⁶

Many of the characteristics we examined did not significantly predict deidentifying. However, this does not necessarily mean they are unrelated to religious deidentification. What these analyses suggest is that if a characteristic does play a significant role in deidentification, it would be through its effect on experiencing God’s presence and on the individual’s social/political views (see table 4).⁵⁷ For example, although scripture reading in 2020 did not have a direct relationship with deidentification, scripture reading may impact deidentification by increasing the feeling of God’s presence in an adolescent’s daily life or by decreasing the trouble felt from his or her religion’s views on social and political issues. The statistically significant predictors we identified in table 4 are likely the most proximal influences on deidentification.

Digital Media

While digital media can enable religious leaders to expand their reach, digital technology also encourages individuals to explore a plurality of religious ideas and feel less bound to or trusting in specific religious institutions or belief systems.⁵⁸ For young adults, more social media use was linked to a “pick and choose” approach to faith, an increase in being religiously unaffiliated, and less certainty in the exclusive truth of one’s religion.⁵⁹ For adolescents with religious parents, more screen time was linked with diminished religious commitment and decreased scripture reading.⁶⁰

As with the analyses in table 4, we used the FFYD data in a logistic regression to examine whether the following were related to deidentification: smartphone ownership (11% of the sample did not own a smartphone), hours spent watching video clips (like on YouTube), and hours spent on social media platforms. We controlled for gender and parental income (as families with higher incomes may have had more access to devices and media). We included all variables in the same model to control for each other. We used the 2020 levels of these variables to predict whether an adolescent would deidentify as Latter-day Saint by 2024. (In 2020, all youth identified as Latter-day Saint.)

From the FFYD data, we found that owning a smartphone was related to a statistically significant greater likelihood of deidentification by 2024. Youth with a smartphone were nearly three times more likely to deidentify (13.0%) compared to those without a smartphone (4.9%). The amount of time spent watching video clips was also statistically related to deidentification: While only 7.6% of those who watched no video clips deidentified, 12.1% of those who watched one hour a day deidentified, 16.3% of those who watched two hours a day deidentified, and 21.6% of those who watched three hours a day deidentified.

It is worth noting that when we put these digital media variables in a logistic regression with daily spiritual experiences and social and political concerns (see table 4), they were no longer statistically significant. This may indicate that digital media has an influence on retention through its impact on spiritual experiences. Indeed, as mentioned above, research has found that digital media use relates to lower religious commitment and less engagement in religious activities.⁶¹ Digital media may also increase dissonance between youth and their religion’s positions on social issues by negatively portraying such positions.

Religious Patterns After Deidentifying

To better understand Latter-day Saints who chose to deidentify, we explored their religious beliefs and behaviors after they deidentified. Using the SSS and PRLS data, we could see how former Latter-day Saints currently identified religiously. In the SSS, we found that 30% identified with another Christian religion (such as Protestant or Catholic), 15% identified with some other religion (such as Buddhism or Islam), 26% identified as Atheist/Agnostic, and 29% identified as “nothing in particular.” In other words, almost half switched to another (primarily Christian) religion, 1 in 4 became Atheist/Agnostic, and about 1 in 3 were ambivalent. When compared to other religions, those raised Latter-day Saint become religiously unaffiliated at a similar rate as those raised in other Christian religions.

In the PRLS data, 28% of those raised Latter-day Saint were unaffiliated with religion later in life. (Given the different data and somewhat different definition of being religiously unaffiliated, this will differ from the SSS numbers.) For other Christian religions, of those raised in their religious tradition, between 15% and 31% become unaffiliated with any religion later in life.⁶²

Groups Among Latter-day Saints Who Deidentified

Deidentification findings reported thus far have provided a picture of averages, such as determining, on average, whether those who deidentify are more likely to live in the Mountain region or the Pacific region. Yet, as Jan Stets and James Heft state, it is important to avoid overly simplistic views of those who deidentify from their religion.⁶³ To better understand the diversity within deidentified Latter-day Saints, we conducted sophisticated latent class analyses based on a variety of traits to identify groups of those who deidentified.⁶⁴ For this, we used the SSS data, which contains 527 individuals ranging in age from 18 to 95 who were raised Latter-day Saint but did not identify as such as adults (at least, not when they took the survey).

Distinct groups of those who deidentified as Latter-day Saint emerged from the data based on the following: how important religion and spirituality was to them, how warm they felt toward Latter-day Saints, whether they believed that death is the end of existence, whether they believed Jesus is the Son of God, and whether they felt religion has a place in society. Participants were also able to select reasons why they no longer identified with the Church. They could select any or all of the following reasons: getting out of the habit of attending church due to COVID-19 lockdowns, moving, no longer believing the tenets of the religion, finding religious affiliation inconvenient, prioritizing other activities, and not feeling love from the members.

Our analyses identified four groups,⁶⁵ two of which felt religion was still an important part of their lives and two that did not (fig. 23). We refer to the first two groups as “Faith Staying” (22% of the sample) and “Faith Bordering” (32% of the sample). For these groups, religion and spirituality were still important. For the Faith Staying, the most common reasons for deidentifying were that they had moved or that Church affiliation had become inconvenient (fig. 24). They felt relatively warm toward Latter-day Saints, believed that Jesus is the Son of God, and felt that religion should play some part in society. Of these, only 3% indicated they were Atheist/Agnostic (fig. 25). Most of them identified with another religion, primarily Christian religions (70%). Given that they felt positively toward religion and Latter-day Saints and still held central beliefs, Faith Staying may have simply “drifted away” from the Church due to various life circumstances that made it difficult to stay connected. Most began identifying with another denomination, perhaps one more convenient to the location they moved to.

Compared to Faith Staying, Faith Bordering felt religion and spirituality were somewhat less important and were more likely to have left the Church because they did not believe or they felt religion was not as high a priority to them. While they still felt somewhat warm toward Latter-day Saints, they were less likely to believe that Jesus is the Son of God and seemed more skeptical that religion should have a place in society (fig. 23). They were more likely to identify as Atheist/Agnostic (9%), and 30% of them identified as “nothing in particular.” However, 57% identified with another religion. (For most of these, it was another Christian religion. See fig. 25.) These individuals seem to have lost connection with religious beliefs, and with other priorities emerging, they chose to no longer identify as Latter-day Saint. At the same time, they appear to be leaning toward faith, though they were less connected to it than the Faith Staying.

The other two groups, “Faith Distancing” (36% of the sample) and “Faith Leaving” (10% of the sample), both indicated religion was not at all important to them. The most prominent reason for these two groups no longer identifying as Latter-day Saint was nonbelief. Faith Distancing still placed at least some importance on spirituality and maintained some, though seemingly tepid, religious beliefs (such as death not being the end and Jesus being the son of God); they were also tepid about religion having a place in society. Forty-one percent of them identified as Atheist/Agnostic, and 40% identified as “nothing in particular” (fig. 25). These individuals seem to have lost any interest in and are somewhat skeptical of organized religion. However, they still place some value in “spirituality,” somewhat believing in life after death. But their primary reason for deidentifying was not believing. It is notable they also were the highest in selecting “did not feel love” from their religion as a reason for deidentifying (fig. 24). This may have been part of the reason for their soured relationship with religion.

The Faith Leaving placed no importance on either religion or spirituality. They had the lowest warmth toward Latter-day Saints, felt that death is the end, and had no belief that Jesus is the Son of God. They were also the most likely to believe that religion has no place in society (fig. 23). Of this group, 73% identified as Atheist/Agnostic, while 22% identified as “nothing in particular.” None of them identified as part of a Christian religion. These individuals seem to be entirely disconnected from a religious or spiritual view of the world and left the Church primarily because of their rejection of religious beliefs.

We also explored other characteristics of these groups, including age, gender, sexual orientation, education level, income level, relationship status, church attendance frequency in childhood, current region of residence, the percentage of religious adherents in their counties, political ideology, and experience of abuse in a religious context (physical, psychological, or sexual). We also examined whether they had any adverse childhood experiences, which included experiencing things such as parental divorce, abuse by someone living in the home, residence with someone who had a drug problem, residence with anyone who had been incarcerated, and other experiences. When included in the same statistical model, the following significantly differentiated the four groups (Faith Staying, Faith Bordering, Faith Distancing, and Faith Leaving): political ideology, relationship status, region of residence, adverse childhood experiences, and abuse experienced in a religious context.

In terms of political ideology (fig. 26), Faith Staying were the most conservative group, with 47% either conservative or very conservative and only 14% liberal or very liberal. Every other group had less than half that percentage of conservatives, and 51% of Faith Leaving were liberal or very liberal.

In terms of relationship status (fig. 27), Faith Staying were the most likely to be married or partnered (50%). In contrast, Faith Leaving were the least likely to be married or partnered (35%). This follows other research finding that religious individuals are more likely to be married than nonreligious individuals.⁶⁶

In terms of region of residence (fig. 28), Faith Staying were the least likely to live inside Utah or Idaho (16% live in Utah or Idaho). Faith Distancing and Faith Leaving were the most likely to live in those locations (30% and 25%, respectively).

Faith Staying were the most likely to say they left their faith due to an inconvenience or a move (fig. 24). It may be that these individuals were more likely to have moved from a location with a high density of Latter-day Saints (Utah and Idaho) to a location with a lower density, possibly making it more difficult to attend church.

In terms of adverse childhood experiences (fig. 29), Faith Leaving were the most likely to have had an adverse childhood experience at 27%, while other groups ranged between 15% and 18%.

Participants were asked questions about whether they experienced emotional, sexual, or physical abuse in a religious context (fig. 30).⁶⁷ Faith Staying had the lowest percentage of those who experienced some kind of abuse in a religious context at 13%, with Faith Bordering at 17%, Faith Distancing at 36%, and Faith Leaving at 31%. Other research has found that those who deidentify were more likely to have experienced abuse in a religious context than those who do not deidentify.⁶⁸ While this analysis is about the profiles of those who deidentified as Latter-day Saint, it would not be surprising if deidentification from any religion is strongly associated with experiencing abuse within that religion’s context.⁶⁹

When included as predictor without other controls, sexual orientation was statistically significant in predicting group membership (fig. 31). However, when political ideology was also included as a predictor in the regression, sexual orientation dropped from significance. This suggests that sexual orientation itself is not predictive of group membership per se, but rather those who were nonheterosexual were more likely to be politically liberal and therefore more likely to be Faith Leaving. For example, conservative heterosexuals and conservative bisexuals were just as likely to be Faith Staying. However, compared to heterosexual individuals, bisexual individuals were more likely to be liberal than conservative and thus more likely to be in the Faith Leaving group. Figure 31 shows the percentage of each group that identified as homosexual/gay, bisexual, and “other.” For Faith Staying, 13% identified as one of these, while a full 33% of Faith Distancing and 33% of Faith Leaving identified as such. It is important to note that of those who continued to identify with the Church, 6% were nonheterosexual, and of those who deidentified from the Church, 26% were nonheterosexual.

Recommitting to a Congregation

Participants were also asked, “What could possibly entice you to recommit to a congregation?”⁷⁰ This was asked only if an individual reported attending church “never” or “seldom.” Participants were given several reasons to choose from for potentially recommitting, and participants could select more than one reason (see table 5). Among all those who deidentified as Latter-day Saint, 36% selected the option that recommitting was “never going to happen,” indicating that about 2 in 3 of those who deidentified left open some possibility of recommitting. For the Faith Staying and Faith Bordering, a minority indicated that recommitting was “never going to happen” (19% and 25%, respectively). Nineteen percent of Faith Staying and 22% of Faith Bordering said they would recommit if a friend invited them or if they made new friends at church. For these two groups, other prominent reasons for recommitting were finding a church they liked and God telling them “in some significant way” to return.

By comparison, 56% of the Faith Distancing and 79% of the Faith Leaving indicated they would not recommit for any reason. Only 6% of Faith Distancing said they would recommit based on friendships, and no Faith Leaving said they would recommit based on friendships. For the Faith Distancing, the most cited reasons for recommitting were finding a church they liked and God telling them in some significant way to return. For Faith Leaving, the most cited reasons for recommitting were their children wanting to attend church and God telling them to go back in some significant way.

Summary of Deidentification Groups

In summary, over 50% of deidentified Latter-day Saints belong to groups that, on average, still consider religion and spirituality to be of some importance. Only about 10% of deidentified individuals appear to belong to a group wholly rejecting the importance of religion and spirituality in their lives (Faith Leaving), although another 36% are in a group that places no importance on religion and very little importance on spirituality (Faith Distancing). Groups who have deidentified from the Church and still consider religion and spirituality to be important are more likely to hold conservative political views, be married, and live outside of Utah and Idaho. They are less likely to have had adverse childhood experiences or to have experienced abuse in a religious context. Faith Leaving are far more likely to report having experienced difficulties in childhood and abuse in a religious context. Finally, a substantial number of those who no longer identify as Latter-day Saints indicate there is some chance they will recommit to a congregation one day. One in five of those who still feel religion is important say they will return if a friend invites them or if they make new friends.

Changes in the Percentage of the United States Who Identify as Latter-day Saint

PRLS data have put the percentage of the United States who are Latter-day Saint around 2%. A measure that has been used as an indicator of retention is the change over time in the percentage of the United States identifying as Latter-day Saint.⁷¹ While retention is related to this percentage, it is also related to overall US growth through birth and immigration. This means a religion can actually lose its share of the US population even if it has growing numbers. Indeed, population percentages are a zero-sum game—when one group grows, some others must shrink. If a single Catholic moves to the United States, the percentage of Americans who are Latter-day Saints automatically shrinks, even if not a single Latter-day Saint left the Church.

To examine change over time in the percentage of the United States who are Latter-day Saint, we examine three surveys: the CES, the PRLS, and The Public Religion Research Institute’s Census of American Religion (PRRI).⁷² Data from the CES finds a 45% drop in the percentage of the United States who identify as Latter-day Saint from 2014 to 2015.⁷³ However, during nearly that same time period, PRLS shows an 8% drop (between 2014 and 2024) and PRRI shows a 16% drop (between 2013 and 2025)—with neither drop being statistically significant. Given the substantial difference between CES data and the PRLS and PRRI, which survey may be more trustworthy? CES data rely on an “opt-in” panel, meaning participants selected to be part of a panel of individuals to take surveys. PRLS and PRRI, on the other hand, use random recruitment to build their samples. In other words, in contrast to the CES sample, an individual could not invite themselves to be part of the PRLS or PRRI. PRLS and PRRI also had their surveys available in Spanish, extending their reach and diversity. These surveys were also intentionally designed to capture the religiosity in the United States, whereas the CES was not specifically designed for that purpose. Extending its reach further, PRLS also allowed participants to take the survey over the phone, on paper, or online. The CES and PRRI could only be taken online. Given this, when these surveys disagree, the PRLS and PRRI are the better evidence, with PRLS likely the most reliable.