Religious Attachment and the Sense of Life Purpose among Emerging Adults

The salubrious association between religious involvement and well-being is evident among the general population of religious individuals. In particular, the sense of attachment to a deity is linked to promoting healthy behavior and positive well-being. The link between religious attachment and well-being is particularly salient for emerging adults during a life stage where they are developing their own sense of self while also renegotiating religious commitments. The current study uses OLS regression and a lagged dependent variable model to analyze how perceived closeness to God is linked to the sense of life purpose among a diverse, national sample of emerging adults. We find that relative to those perceiving closeness to God, those who reported feeling neutral or did not believe in God experienced lower levels of life purpose. Respondents who reported feeling distant from God had the lowest scores on the Life Purpose Index. The findings are examined within the framework of religion and attachment theory literature. The study encourages researchers to consider attachment to a deity as an important link in explaining well-being outcomes, especially among religious individuals.


Introduction
Religion has the potential to enhance various aspects of well-being through integration and social support, and become a positive influence in the lives of individuals by promoting a deeper sense of moral order and bolstering psychological resources (Ellison 1991;Smith 2003). One such important resource that is linked to religion is life purpose. Life purpose is a multi-faceted and complex construct, operationalized and conceptualized in diverse ways. In the broadest definition, Life purpose refers to the value and meaning individuals ascribe to their lives that cultivates and facilitates an underlying sense of direction, morality, goal-striving, and the pursuit of maximizing potential (Frankl 1946;Crumbaugh and Maholick 1964;Crumbaugh 1968;Ryff 1989;Ryff and Keyes 1995;Steger et al. 2006). Researchers have found the sense of life purpose to be an important link in many physical and mental health outcomes across a variety of contexts. For instance, higher baseline measures of purpose have been found to be negatively associated with mortality, dementia, and psychiatric and somatic symptoms (Boyle et al. 2009(Boyle et al. , 2010. Life purpose has also been found to reduce the risk of anxiety, depressive symptoms, stroke, and other cardiovascular events, and it can ease responses to emotional stress (Ishida and Okada 2006;Krause 2007). The search for meaning and life direction has been shown to act as a regulatory function of health that predicts healthy eating and activity among adolescent and adult populations (Brassai et al. 2015;Piko and Brassai 2016). Given the links to physical and mental health outcomes, it is important to understand the factors that contribute to the development of a positive sense of life purpose. The following section examines religion as an important contributor to one's overall sense of purpose in life.

Religion and Life Purpose
One fundamental function of religion is to fulfill a yearning for a sense of purpose among individuals (Kirkpatrick and Hood 1990;Galek et al. 2015). Religion is uniquely associated with the sense of life purpose in that doctrine and texts tend to reinforce the idea of a divine purpose. Coupled with doctrine and texts, social networks exercised in religious communities may also reinforce a sense of purpose, so long as participants have an intrinsic religious orientation that compels positive perceptions of God or closeness to a respective deity (Allport and Ross 1967;Maltby and Day 2003). Empirical studies relying on qualitative interviews, as well as quantitative scales measuring Life Purpose, have found positive associations between religious engagement and the sense of life purpose. For example, significant positive correlations have been found between religious service attendance and the sense of purpose (Robbins and Francis 2000) and between frequent reading of religious texts and exploration of meaning and identity (Francis 2000), with attendance and reading leading to internalized religious beliefs and a subsequent reliance on religion as a daily moral compass (Tirri and Quinn 2010). We draw on attachment theory to help explain one potential mechanism by which religion might bolster the sense of life purpose.

Attachment Theory
John Bowlby (Bowlby 1969(Bowlby , 1973(Bowlby , 1980 originally proposed a motivational system to explain how infants seek comfort and safety, mainly from their mothers (i.e., primary caregiver). Infants emit social cues such as crying, smiling, and clinging to which a caregiver can be more or less attentive and vary in "warmness" (Bowlby 1969;Kirkpatrick 1992). Infants characterized as having a secure attachment are thought to explore their environment with confidence and seek comfort in their primary caregiver when a threat is present. Bowlby proposed additional attachment types-anxious/ambivalent and avoidant-with each type characterized by a sense of distance and less trust with their primary caregiver. Bowlby theorized the mother-infant relationship influences behavior "from the cradle to the grave" (Bowlby 1979, p. 129).
Since Bowlby's initial observations and hypotheses of the mother-infant relationship, a multitude of empirical research has extended the impact of attachment beyond personality traits to also include romantic relationship quality (Kirkpatrick and Davis 1994;Kirkpatrick and Hazan 1994) and psychological functioning (Riggs et al. 2007;Ellison et al. 2012). Researchers have theorized that secure attachment during infancy and early adolescence bodes well for the positive development of internal working models which provide youth with a cognitive framework that facilitates confidence, exploration, and independent functioning (Bretherton and Munholland 1999;Thompson 2000). This cognitive framework may aid in building the capacity to form meaningful and healthy social and romantic relationships, as behavior in social interactions mirrors that of the secure attachment relationship. Healthy relationships can then enhance mental health, reduce the impact of stress, and cultivate a sense of purpose (Thoits 1995;Cohen 2004;Umberson and Montez 2010). In contrast, the consequences of avoidant and anxious/ambivalent attachment may be the formation of less healthy relationships and increased stress that operates as a barrier to exploring one's purpose.

Religion and Attachment Theory
Researchers in both the psychology and sociology of religion have integrated perspectives and concepts from attachment theory, examining the relationship with the divine as a source of attachment that impacts the life outcomes of religious adherents (Kirkpatrick 1992;Kirkpatrick 1997). Those adhering to a monotheistic religion may perceive God as being warm, responsive, and a secure base during times of increased stress and perceived threat. For these individuals, God may be the perfect attachment figure that supersedes even the influence of the mother-infant relationship or compensates for the absence or lack of attachment to a parent. Particularly among Protestant religious traditions, the idea of God as a source of security mirrors secure attachment as characterized by Bowlby, which becomes the catalyst for further religious exploration and confidence (Kirkpatrick 1992). Given the compatibility of attachment theory with religious beliefs, practices, and imagery, it is plausible to expect that in the lives of religious individuals, attachment to God may operate similarly to the mother-child attachment in Bowlby's work (Kirkpatrick 1992(Kirkpatrick , 1997(Kirkpatrick , 2005. Religious individuals perceiving God as the ultimate attachment figure may find attachment to this deity fundamentally involved in many (if not all) aspects of their lives. Therefore, holding warm beliefs about God as a source of security or haven of safety may bode well for the sense of life purpose. Our expectations regarding the outcomes for the current study are based on the compatibility between attachment to God literature and Bowlby's attachment theory. Attachment security is associated with various kinds of positive affect such as feeling valued by others, and is also likely to be associated with a sense of purpose (King 2012;Mikulincer and Shaver 2013). In contrast, insecure attachment can erode positive affect which may foster feelings of vulnerability and doubts that are barriers to the sense of purpose. Therefore, just as general attachment literature asserts secure attachment bodes well for individuals' cognitive framework, so might attachment to God provide a religious framework that enhances the sense of direction and meaning, so long as individuals perceive God as benevolent and actively involved in their lives (Kirkpatrick 1992;Kelley and Chan 2012;Ellison et al. 2014;Galek et al. 2015).
Studies have found secure attachment to and intimacy with God to be inversely associated with psychological distress and psychopathology (Bradshaw et al. 2008;Bradshaw et al. 2010), and positively associated with greater resilience after perceived negative events (Ellison et al. 2014). Anxious attachment, in contrast, was positively correlated with psychiatric symptoms such as general and social anxiety, obsession, and compulsion (Ellison et al. 2014). Interestingly, Ellison et al. (2014) also found that individuals that pray frequently while perceiving a close relationship with God experienced certain health benefits, while those that pray yet perceive God as distant do not experience the same benefits in return. Bradshaw and Kent (2017) find similar results, suggesting prayer does not have a main effect on psychological well-being in later life, but rather the association is moderated by attachment to God.
These findings indicate that secure attachment to God promotes favorable health and well-being outcomes compared to the anxious and distantly attached, and may be an important link in understanding well-being outcomes among religious individuals. Holding positive beliefs about God and perceiving God as a close and secure base provides unique physical, mental, well-being and psychological resource benefits for religious individuals. Research has established links between secure attachment and healthy social and romantic relationships that are positively associated with well-being (Thompson 2000;Mikulincer and Shaver 2013). The current study extends this research by examining the specific link between attachment to God and the sense of purpose among emerging adults.
The aforementioned patterns regarding the positive association between perceived attachment to God and the sense of life purpose and overall well-being are evident among the general population of adults as well as some studies focusing on adolescents. However, less attention has been given to how these relationships manifest among national samples of emerging adults with diverse lived-experiences. Emerging adulthood is a critical time for exploration and the formation of personal identity and establishing trajectories of adulthood. Yet this is also a period in the life course that is marked by transitions and instability as young people navigate through the "capstone" experiences of adulthood such as marriage, post-secondary education, and family formation (Arnett 2000;Arnett and Jensen 2002). Religious commitment and engagement typically decline during this time period as well, as young people establish independence from the religion of their parents and learn to balance many competing demands on their time (Smith and Snell 2009). Emerging adulthood, then, rests at the intersection of religious decline and the search for identity and life purpose. Given the established importance of religion in shaping the sense of life purpose and subsequent health and well-being, it is especially important to examine how this operates among emerging adults and understand how lived-experiences during this unique life stage can impact one's attachment to God and sense of purpose.
The current study draws on a diverse, national sample of emerging adults to enhance our understanding of religiosity, specifically variations in attachment to God, and its effects among this unique population's sense of purpose-an important psychological resource correlated with health and well-being.
Despite the compelling empirical evidence that indicates parental attachment matters for child development, there is still a debate on how fixed attachment patterns are over time and how closely they are linked to well-being outcomes in adulthood (Thompson 2000). Some research highlights stronger contemporaneous associations between attachment and socioemotional adaptation than longitudinal analyses (Thompson 1999). Other studies have found supporting evidence for Bowlby's hypotheses that suggest individual differences in attachment security are stable for long periods of time across the lifespan (Waters et al. 2000). The current study speaks to this debate by conducting a cross-sectional analysis and further assessing how attachment, specifically attachment to God, is related to changes in the sense of life purpose over time using a lagged dependent variable model with panel data. We suspect attachment to God will operate similarly to general attachment theory in predicting long-term well-being. Secure attachment (whether divine or human) in the early stages of life is associated with adaptive coping mechanisms and healthy social and romantic relationships in adulthood, both of which provide individuals with the capacity to feel confident in pursuing their life purpose. By conducting cross-sectional and longitudinal analyses, we seek to expand attachment literature while answering the research question: "How does religion-particularly religious attachment to the divine-influence the sense of life purpose among emerging adults?" Building upon previous research we hypothesize a cross-sectional relationship between one's current attachment to the divine and their reported sense of life purpose.
Hypothesis 1. Emerging adults who report feeling very or extremely close to God at Wave 3 will be more likely than their peers to concurrently report a greater sense of life purpose.
In addition, we hypothesize that divine attachment will continue to impact future feelings of life purpose.
Hypothesis 2. Emerging adults who report feeling very or extremely close to God at Wave 3 will be more likely than their peers to report a positive sense of life purpose measured five years later at Wave 4.
Our study contributes to the literature in several meaningful ways. We bring together two existing strands of attachment literature: Research that links general attachment with positive affect and positive sense of life purpose, and research arguing that attachment to God operates as one source of attachment that shapes individual life outcomes. Relying on assertions from these two bodies of literature provides us a framework to help explain the potential association between attachment to God and life purpose which has not been sufficiently addressed by existing research. In addition, the longitudinal analysis increases our confidence in arguing that attachment to God predicts changes in life purpose over time. In doing so, we encourage researchers to account for attachment to a deity as an important link in explaining health and well-being outcomes, especially among religious individuals.

Data
The National Study of Youth and Religion (NSYR)'s longitudinal telephone survey began as a nationally representative telephone survey of 3290 English and Spanish speaking teenagers between the ages of 13 and 17. The baseline survey was conducted with the teen respondents and one of their parents between July 2002 and April 2003 by researchers at the University of North Carolina at Chapel Hill. A random-digit dial (RDD) telephone method was employed to generate numbers representative of all household telephones in the 50 United States. The total number of completed cases in the first wave of NSYR was 3370. 1 Subsequent waves of the NSYR are re-surveys of the Wave 1 English-speaking teen respondents. Wave 2 of the survey was conducted from June 2005 through November 2005 when the respondents were between the ages of 16 and 21. Wave 3 of the survey was fielded from September 2007 through April 2008 when the respondents were between 18 and 24 years old. Wave 4 was conducted 2012 through 2013, when respondents were ages 23-28. In Wave 4, 8% of respondents responded to the survey online, while the balance completed their survey on the telephone. On average, surveys in the second, third and fourth waves took about 45 min. Every effort was made to contact and survey all original NSYR respondents at each wave, including those out of the country and in the military. The current study draws on data from Wave 3 and Wave 4 of the survey.
Diagnostic analyses comparing NSYR data with U.S. Census data on comparable households and with comparable adolescent surveys-such as Monitoring the Future, the National Household Education Survey, and the National Longitudinal Study of Adolescent Health-confirm that Wave 1 NSYR provides a nationally representative sample without identifiable sampling and non-response biases of U.S. teenagers ages 13-17 and their parents living in households (for details, see (Smith and Denton 2005).

Dependent Variable: Life Purpose Index
The dependent variable for the current study is the sense of life purpose reported by respondents. The Life Purpose Index is constructed from a series of three survey items that were asked at both Wave 3 and Wave 4. These three items included the statements: (1) "Your life often seems to lack any clear goals or sense of direction;" (2) "You do not have a good sense of what it is you are trying to accomplish in life;" and (3) "Some people wander aimlessly through life, but you are not one of them." Responses for these statements include: Strongly agree, Agree, Do not know/Unsure, Disagree, and Strongly disagree. Responses were coded such that a higher value represents a stronger sense of purpose and a lower value represents a lack of purpose or direction in life. An index of all available responses was created in which the value of the index represents an individual's average response across all available variables. The minimum response for the Life Purpose index is "1" which suggests the respondent has a low sense of purpose in life, while the maximum "5" suggests a high sense of purpose in life. The Wave 3 Life Purpose Index has a mean of 3.82 (α = 0.65) and the Wave 4 index has a mean of 3.76 (α = 0.74).

Independent Variable: Closeness to God
Attachment to the divine is operationalized as reported Closeness to God. At each wave of the NSYR survey, respondents were asked whether or not they believe in God, and those who said they believe in God were then asked how close they feel to God. The response options for the closeness to God question were extremely distant, very distant, somewhat distant, somewhat close, very close, and extremely close. We collapsed these responses, along with the original belief question, into four indicator variables. Respondents that indicated being either very close or extremely close to God were combined into one variable as Close to God (close to God = 1, other = 0), and this is used as the reference category in the regression models. Respondents that indicated being somewhat distant or somewhat close were combined into one variable as Neutral (neutral = 1, other = 0). Respondents that indicated being very distant or extremely distant were combined into one variable as Distant from God (distant from God = 1, other = 0). The final indicator variable is Does not believe in God (does not believe = 1, other = 0). This variable includes respondents that indicated not believing God and therefore did not get asked about how close they feel to God. Respondents that reported being unsure about their belief in God were asked the Closeness to God follow-up question. Of these respondents, some answered that the 1 N = 3370 includes 80 oversampled Jewish households that were not part of the nationally representative sample. These 80 cases are excluded from this analysis. The original nationally representative sample includes 3290 cases.
Closeness to God survey question did not apply to them. These responses were also included in the Does not believe in God indicator variable.
Attachment to God measures are conceptually diverse with many researchers emphasizing those aspects of attachment most salient to a specific outcome (Rowatt and Kirkpatrick 2002;Calvert 2010). Some researchers treat attachment to God as discrete categories (e.g., avoidant, anxious, secure) (Kirkpatrick and Shaver 1992), while others measure attachment security as a single dimension (Belavich and Pargament 2002;Calvert 2010). Each conceptualization comes with its own set of limitations, despite many improvements that attempt to capture a broad spectrum of characteristics hypothesized to be related to attachment to God (Beck and McDonald 2004). Given the data available in the NSYR, the current study operationalizes attachment to the divine as reported closeness to God. Existing literature posits that the sense of closeness to God is linked to perceiving that one's attachment to God is secure (Kirkpatrick 1992;Kirkpatrick and Shaver 1992), and is therefore an essential component of attachment to God. We expect the Closeness to God measure to operate similarly to other measures of attachment to God in predicting well-being outcomes. However, data limitations do not allow us to measure the full spectrum of characteristics related to attachment to God, and dimensions of higher order, such as avoidance and anxiety, are not fully captured in this analysis.

Control Variables
Our analysis seeks to examine how Closeness to God impacts the life purpose of emerging adults. One important question is whether the mechanism at work is a general religiosity or if the closeness to God dimension of religiosity has a unique impact on one's sense of life purpose. In order to assess this, we add a control for religious attendance to our models. In this way, we can determine whether Closeness to God is simply operating as a proxy for overall religiosity or if it is a specific dimension of religion that offers its own contribution to the well-being of young people. Religious service attendance is measured on a 7-point scale, where 0 = Never attends and 6 = Attends more than once per week.
Each of the models includes a set of demographic control variables thought to be related to the variables of interest. Variables that control for the ascribed characteristics of the respondents include sex (1 = female), age measured as a continuous variable from youngest to oldest, and a series of indicators for the race of the respondent (black, Latino/a, and other, with white as the reference category). One characteristic of emerging adulthood is that there is more variation and ambiguity surrounding life events that have traditionally served as markers of adulthood. Achieving these traditional markers of adulthood may contribute to reduced ambiguity and a more certain sense of life purpose for emerging adults. We measured several achieved status characteristics associated with the transition to adulthood that we expect to be related to both a sense of life purpose and one's closeness to God. Education is one such characteristic included in our models. All respondents who have at least a bachelor's degree or are currently enrolled in a four-year degree program are coded as 1, and all other respondents are coded as 0. Marriage has been shown to be correlated with higher levels of religiosity (Mahoney et al. 2003;Waite and Lehrer 2003). In addition, the commitment of marriage can serve to imbue individuals with a heightened sense of life purpose (Umberson et al. 2006;Manzoli et al. 2007;Uecker 2012). In order to capture these relationships, we include marital status of respondents at the time of the survey (1 = married, 0 = not currently married). Young people's faith is often shaped by family relationships, such that living apart from parents may usher in changes to one's faith practices. Living independently is also a marker of adulthood that might signal life success for emerging adults. The models include an indicator variable for those respondents who are living on their own (1 = living on their own, 0 = all other living arrangements). Finally, given that physical health has been correlated with both religiosity (Powell et al. 2003) and mental and emotional well-being (Repetti et al. 2002;Umberson and Montez 2010), we include a measure of self-reported health ranging from poor health to excellent health.

Analytical Strategy
Hypothesis 1 will be tested in a multivariate context with a cross-sectional Ordinary Least Squares regression in which we use the Wave 3 Closeness to God measure to predict the Wave 3 Life Purpose score. Hypothesis 2 will be tested with panel data using a Lagged Dependent Variable model which predicts Wave 4 Life Purpose while controlling for the baseline Life Purpose score at Wave 3. This model will test the long-term impact of closeness to God at Wave 3 by measuring the related change in Life Purpose between Wave 3 and Wave 4. In other words, we ask whether closeness to God continues to have a longitudinal effect on future Life Purpose scores even after controlling for the cross-sectional relationship between Wave 3 Closeness to God and Wave 3 Life Purpose. Table 1 shows the descriptive statistics for the Wave 3 and Wave 4 variables included in the models. According to the indicators of Closeness to God at Wave 3, the largest group, at 54%, is the Neutral category. Twenty-eight percent of the young people report feeling very or extremely close to God and 12% feel very or extremely distant from God. The final 6% do not believe in God. The overall mean score for the Life Purpose Index was 3.82 in Wave 3 and 3.76 in Wave 4.  Table 2 reports the cross-sectional OLS regression analysis in which we regress Wave 3 Life Purpose on Wave 3 Closeness to God and control variables. In Model 1 we see that among the control variables, an increased sense of life purpose is positively associated with education, self-reported good health, and living on one's own. Even after controlling for the influence of these factors, closeness to God continues to be related to one's sense of life purpose. Feeling close to God serves as the reference category in this model and all of the other three groups report lower life purpose scores at statistically significant levels. The category of believing in God while feeling distant from God has the largest coefficient (b = −0.34, p < 0.001), demonstrating the largest difference in life purpose compared to those who are close to God. Those who do not believe in God and those who are in the neutral category also report lower life purpose scores, though the coefficients are smaller (b = −0.27, p < 0.05 and b = −0.23, p < 0.001, respectively). These findings offer support for Hypothesis 1. Model 2 adds the control for religious service attendance. Attendance itself is not a statistically significant predictor of life purpose. Including a control for how often individuals attend religious services, however, offers interesting insight about the impact of closeness to God. The coefficient for those who do not believe in God becomes statistically insignificant in Model 2 when the control for attendance is included in the model, suggesting that the results for this coefficient in Model 1 might be capturing the role of general lack of religiosity. However, the coefficients for Distant from God and Neutral remain statistically significant even after controlling for religious service attendance. Net of religious practice, feeling distant from God or neutral in one's attachment to God is associated with a lower reported sense of life purpose compared to those who report feeling close to God. In Model 2, being distant from God continues to have the largest impact on the sense of life purpose.  In addition to examining the cross-sectional relationship between closeness to God and Life Purpose, we are also interested in examining the potential effects over time. Using a lagged dependent variable model, we examine whether how close one feels to God at Wave 3 is related to changes in one's sense of life purpose between Waves 3 and 4. Table 3 reports the results of the LDV model in which the Life Purpose Index score from Wave 4 is regressed on the Closeness to God measure from Wave 3. The models also include a control for the Life Purpose score measured at Wave 3. The LDV model measures the extent to which Wave 3 Closeness to God predicts the change in Life Purpose from Wave 3 to Wave 4. The results in Table 3 illustrate a statistically significant relationship between closeness to God at one point in time and one's reported sense of life purpose at a future point in time. Model 1 presents the results without controlling for religious service attendance, while Model 2 includes the attendance control. Similar to the cross-sectional models, we see in the LDV models that how close one feels to God is a significant predictor of one's reported life purpose, even several years later. In Model 1, the coefficients for all three indicators of closeness to God are negative and statistically significant. Compared to respondents who feel close to God, all other respondents report lower levels of Life Purpose five years later, even after controlling for their initial Wave 3 Life Purpose score. And similar to the cross-sectional model, we see again that the largest effect is for those who report feeling distant from God (b = −0.28, p < 0.001). When the measure for religious service attendance is added to Model 2, the coefficient for Does not believe in God drops somewhat and loses statistical significance. This is not surprising given the substantial overlap between those who say they do not believe in God and those who say they never attend religious services. 2 However, controlling for religious service attendance does not eliminate the significant association with Life Purpose found for the other two indicators of Closeness to God. Young people who said they are neutral or distant from God continue to report significantly lower Life Purpose scores than their counterparts who at Wave 3 said they feel close to God (b = −0.11, p < 0.05 and b = −0.25, p < 0.01 respectively). In analysis not shown, among those who said they do not believe in God at Wave 3, 93% report never attending services at Wave 4.

Discussion
This study builds on the growing literature examining the relationship between religion, psychological resources, and overall well-being. We began by establishing a connection between attachment to God and a sense of life purpose. When young people feel an attachment to the divine, they are more likely to feel positive about the direction and purpose of their lives. This is consistent with our understanding of attachment theory and the salubrious effects of having a secure attachment in a significant relationship (Bretherton and Munholland 1999;Thompson 1999;Chopik et al. 2013). Interestingly, we found that those who believe in God but do not feel close to God report a lower Life Purpose score than those who say they do not believe in God at all. Given the hypothesized positive correlation between closeness to God and Life Purpose, this might initially seem surprising. However, this finding is actually consistent with previous findings of an ambivalent or avoidant attachment having a negative relationship with well-being (Bowlby 1973;Bretherton and Munholland 1999;Brumariu and Kerns 2010). If an individual believes that there is a God and yet feels distant from God, this would represent an ambivalent attachment to an acknowledged divine figure. This ambivalent attachment would then be a factor that influences their perception of life purpose. On the other hand, for individuals who do not believe in God, a relationship with God will not be likely to serve as factor in their psychological resources, either positive or negative. Thus, while God does not serve for them as a positive source of attachment, neither does God represent an ambivalent or avoidant attachment which might negatively influence their overall well-being. Consistent with our understanding of attachment theory, those who have the most secure attachment to God report the greatest sense of life purpose and those who report an ambivalent or distant attachment to God report the least confidence about their purpose in life.
Building on the connection between attachment and life purpose, we further establish the longitudinal impact of perceived closeness to God. Young adults with a closer attachment to God benefit from an increased sense of purpose about their lives, and this attachment continues to bolster their sense of life purpose when measured again five years later. Given the relatively long time frame between the two points in time, it is noteworthy that closeness to God has a persistent impact on Life Purpose scores of emerging adults and continues to contribute to positive psychological resources as they move into adulthood. In both the cross-sectional and the longitudinal analysis, the impact of closeness to God also persists independent of a measure of religious service attendance. Even after controlling for this measure of public religious practice, closeness to God is significantly related to a stronger sense of life purpose. This finding confirms that attachment to God is not simply a proxy for any measure of religiosity. Rather, there is something unique about attachment to God that bolsters one's sense of life purpose net of other facets of one's religiosity.
These findings are important for several reasons. Life purpose is linked to many other positive life outcomes and is a psychological resource that is particularly salient during the transitions and identify formation taking place during emerging adulthood (Arnett 2000;Thompson 2000;Umberson and Montez 2010;Uecker 2012). Understanding the mechanisms by which individuals develop their sense of life purpose is important for better facilitating positive health and well-being among emerging adults. And while religion has been previously shown to be positively related to psychological resources, this study explicates a specific mechanism-attachment to God-that shapes the life purpose of young people. The importance of attachment to God is noteworthy given that young adulthood is also a time of life during which many young adults are moving further away from religious engagement (Smith and Snell 2009). Among the young people in this study, for example, belief in God decreased by 14 percentage points between the two waves of data collection. Understanding the link between attachment to God and one's sense of life purpose sheds light on how patterns of religious decline may exacerbate this already difficult life stage for emerging adults, and helps us understand yet another layer of the challenges facing young people during their transition to adulthood.
The current study examines one facet of religion and its relationship to developing a strong sense of life purpose. We acknowledge that there are limits to the claims that can be made with data from only two points in time. However, the longitudinal design does increase confidence in the temporal order of this relationship, suggesting that closeness to God is a significant predictor of subsequent perceptions of Life Purpose. Future research can build on this study by examining additional facets of religion from a longitudinal perspective. Aspects of religion that foster social integration and embed individuals within a religious community might also have positive effects on the development of a sense of life purpose. While this study establishes a statistical relationship between Closeness to God and Life Purpose, we are unable to measure the extent to which emerging adults themselves understand these processes and attribute their own sense of life purpose to their religious faith or attachment to God. Qualitative research could expand on these findings by explicating the perspectives of emerging adults and the ways in which they draw on their faith as they develop their adult identities and come to understand their own purpose in life. This study contributes to the literature in the areas of attachment, religion, and well-being. Given the ways in which one's sense of life purpose can contribute to many health and well-being outcomes, expanding our understanding of the factors that contribute to this sense of life purpose is of particular importance. The results presented here highlight the value of accounting for attachment to a deity as an important link in explaining health and well-being outcomes, especially among religious emerging adults.