• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Int J Psychol Relig. Author manuscript; available in PMC Mar 21, 2013.
Published in final edited form as:
Int J Psychol Relig. 2012; 22(2): 135–154.
Published online Mar 21, 2012. doi:  10.1080/10508619.2011.638589
PMCID: PMC3314245
NIHMSID: NIHMS364022

Parental Religious Socialization Practices, Connectedness With Others, and Depressive Symptoms in Late Life

Abstract

This purpose of this study is to examine two constructs that have been largely overlooked in the study of religious involvement among older people: parental religious socialization practices and feelings of connectedness with others. The data are from an ongoing nationwide survey of older people. Findings from a latent variable model that was designed to examine the two focal constructs provides support for the following relationships:(1) older people whose parents encouraged them to become more involved in religion are more likely to attend worship services; (2) older people whose parents promoted religious involvement and older individuals who attend church more often are more likely to report that they see a fundamental connection among all human beings; (3) older adults who feel more closely connected to others will be more likely to forgive people for the things they have done; and (4) older people who are more forgiving are likely to experience fewer symptoms of depression over time.

Keywords: parental socialization, connectedness, depression

Research indicates that some form of religion is found in virtually every culture on earth (Smith, 1991). Moreover, compelling evidence suggests that some type of religion has been practiced for over 100,000 years (Eliade, 1978). So if religion is so widespread and so impervious to the ravages of time, then it must perform some vitally important function or satisfy some basic human need. Although researchers have identified a range of functions and needs that are satisfied by religion, one of the more frequently discussed benefits involves the opportunity to form meaningful social relationships with others (Hood, Hill, & Spilka, 2009). But these investigators had more in mind than the mere establishment of social ties. In order to clarify their view on religion and social relationships, Hood, Hill, and Spilka (2009) evoke the term “sociality”, which has to do with feeling connected to other people as well as a sense of belonging to the wider religious group (e.g., congregation) in which the individual is embedded (p.19).

The notion that religion fosters a sense of connectedness with others is hardly new. Evidence of this may be found, for example, in the work of Edward Alsworth Ross, who was one of the first presidents of the American Sociological Association. Writing in 1896, he discussed the ways in which a sense of connectedness with others promotes some of the core values of religion: “The mere perception of likeness fosters sympathy, but the conviction of underlying oneness does more. It destroys the egocentric world which each unreflecting creature builds for itself … It fosters respect for others … It lessens our willingness to use them as means to our own ends” (Ross 1896: 441, emphasis added). A similar view was expressed by John Stewart Mill (1874/2008) in his widely cited volume, Three Essays on Religion. In this work, Mill (1874/2008) criticized religious traditions for falling short of what they could be and in his opinion, should be. More specifically, he argued that the primary goal for any religion is to instill a sense of unity with all mankind

As the arguments of these early intellectuals reveal, one of the basic functions of religion is to foster a sense of connectedness that is built on sympathy, duty, and selflessness. The purpose of the current study is to probe more deeply into this religiously-based notion of connectedness with others. This is accomplished by developing and testing a latent variable model that examines the social factors that promote these fundamental beliefs and explores the mental health consequences of developing a deep sense of connectedness with others. However, before turning to this conceptual scheme, it is important to briefly review research on social relationships in the church in order to more clearly delineate the nature of a religiously-motivated feeling of being closely connected with others.

Religion and Feelings of Connectedness With Others

A growing number of studies point to the importance of studying social relationships that arise in religious institutions (e.g., Krause, 2008). But rather than being a recent innovation, the study of social relationships in religious settings is at least a century old. For example, Georg Simmel (1898/1997), a classic social theorist, argued that the very basis and substance of religion arises from and reflects the relationships that exist among the faithful. Building on these classic insights, contemporary researchers have begun to tease out some of the finer nuances of church-based social relationships. For example, this work reveals that social relationships in the church may be even more consequential for health than social ties that are formed in wider secular world (Krause, 2006). But these findings hardly exhaust all that can be known about the social support process at church. One issue, which forms the focal point of the current study, has to do with isolating the underlying belief system that may help explain the potentially unique effect of church-based social ties.

Researchers have examined the religious beliefs that foster potentially unique social relationships in the church, but this work is in its infancy. In one of the earlier studies on this issue, Davidson (1972) made a distinction between the vertical and horizontal dimensions of religion. The vertical dimension refers to the relationship that a person establishes with God, whereas the horizontal dimension has to do with relationships that an individual develops with other people. Davidson (1972) developed a measure of the horizontal dimension that consists of two indicators. The first item deals with the need to love one’s neighbors and the second involves the importance of helping one’s fellow human beings. Although loving and helping others can clearly affect the quality of the social relationships that people develop, a deeper and more fundamental set of beliefs may lie behind them. More specifically, it may not be enough to believe that people should love and help each other – instead, these beliefs will be more likely to influence the development of social relationships if people have a clear sense of why loving and helping others is important. A core premise in the current study is that beliefs about loving and helping others arise from seeing the fundamental connectedness among all people.

There appears to be only one study in the literature that empirically evaluates the extent to which social relationships in the church are influenced by a deep sense of connectedness with others. This research, which was conducted by Krause and Bastida (2009), reveals that a strong sense of connectedness with others is associated with providing more emotional support and tangible assistance to fellow church members. The data from this study further reveal that older people who feel more closely connected with others are more likely to pray for other people, as well.

Although the study by Krause and Bastida (2009) helps bring the notion of connectedness to the foreground, a considerable amount of research with this fundamental construct remains to be done. The conceptual model that is presented in the next section was designed to advance research on feelings of connectedness with others in four potentially important ways.

First, if a sense of connectedness with others occupies a fundamental place in religious life, then it is important to know how these beliefs arise. The model that was developed for the current study examines this issue by seeing whether feelings of connectedness with others may be traced to the religious socialization practices of the study participants’ parents.

Second, if a sense of connectedness forms the foundation of social relationships in the church and church-based social ties are related to a range of health-related outcomes, then it follows that feelings of connectedness with others should be associated with health-related outcomes, as well. An effort is made in the current study to explore this possibility by assessing whether strong feelings of connectedness are associated with fewer symptoms of depression.

Third, rather than merely demonstrating that feelings of connectedness are associated with mental health, the model that was developed for this study was designed to evaluate a key intervening variable that may explain this relationship: forgiveness of others.

Fourth, the relationships in the study model are estimated with data that have been gathered at more than one point in time. This makes it possible to see if factors such as feelings of connectedness with others are associated with change in depressive symptoms over time.

Modeling the Genesis and Consequences of Feeling Connected with Others

Figure 1 contains the latent variable model that was designed to explore the relationships among parental religious socialization practices, feelings of connectedness with others, and change in depressive symptoms over time. Two steps were taken to simply the presentation of this complex conceptual scheme. First, the elements of the measurement model (i.e., the factor loadings and measurement error terms) are not depicted in Figure 1 even though a full measurement model was specified when this conceptual scheme was estimated. Second, although they are not shown in Figure 1, measures of age, sex, education, and race were included in the model when the relationships among the core study constructs were evaluated empirically.

Figure 1
A Conceptual Model of Parental Religious Socialization Practices and Depressive Symptoms

Even though a number of relationships are specified in Figure 1, the following core linkages capture the overall theoretical thrust of this conceptual scheme: (1) people whose parents encouraged them to become more involved in religion are more likely to attend worship services than individuals whose parents did not encourage them to become involved in religion; (2) people whose parents promoted greater involvement in religion and individuals who attend church more often are more likely see a fundamental connection among all human beings; (3) individuals who feel more closely connected to others will be more likely to forgive people for the things they have done; and (4) people who are more forgiving are likely to experience fewer symptoms of depression over time. The theoretical rationale for each of these core relationships is provided below.

Parental Religiousness, Church Attendance, and Feelings of Connectedness

According to the model in Figure 1, the genesis of feelings of connectedness with others may be found in the interplay between parental religious socialization practices and the frequency of church attendance. A vast literature suggests that the religious socialization practices of parents exert a profound influence on the religious beliefs and behaviors of their children. Spilka, Hood, Hunsberger, and Gorsuch (2003) point to the striking degree of consensus in this literature: “All the different approaches to studying parental influence in the religious socialization process converge on a single conclusion: Parents play an extremely important role in the developing religious attitudes and practices of their offspring. In fact, few researchers would quarrel with the conclusion that parents are the most important influence in this regard” (Spilka et al. 2003, p.115; emphasis in the original). So if parents exert the most important influence on religious beliefs, and feelings of connectedness with others is a core religious belief, then it follows that researchers who seek to understand how feelings of connectedness arise would be well advised to focus on parental religious socialization practices.

Although research on parental religious socialization practices provides a number of important insights into the factors that shape religiousness in their offspring the analyses that follow are designed to expand the scope of inquiry in a potentially important manner. Virtually all the studies that have been done so far focus on how parental religious socialization practices shape the religious involvement of children, adolescents, and young adults (e.g., Regnerus, Smith, and Smith 2004; Sherkat 1998). In contrast, very little is known about whether the influence of parents extends further into the life course, especially into old age. This is unfortunate because the conclusion reached by Spilka et al. (2003) would be even more compelling if it can be shown that the influence of parental religious socialization practices is still evident fifty or sixty years after a child has left the home. This issue can be examined in the current study because the analyses are based on data that have been provided by older people.

There appears to be only one study in the literature that explores the ways in which parental religious socialization practices influence religious involvement in late life. This research, which was conducted by Krause and Ellison (2007), reveals that individuals whose parents encouraged them to become more involved in religion are more committed to their faith in their later years, and older adults who are more committed to their faith tend to have a stronger sense of self-worth. The current study aims to move beyond the broad notion of religious commitment so that greater insight may be obtained into the specific ways in which parental religious socialization practices operate. When individuals are committed to their faith they are dedicated to the practice of a set of specific religious beliefs. But when these beliefs are examined collectively under the broad rubric of religious commitment, it is more difficult to determine what individuals have actually been taught by their parents and it is hard to identify particular beliefs that are consequential for health. Focusing on a sense of connectedness with others provides a clearer sense of the type of religious lessons that parents may have instilled in their offspring and in the process, it makes it possible to determine whether this belief is associated with mental health..

Although parental religious socialization practices are a vitally important source of religious beliefs for their children, parental encouragement alone may not be able to sustain religious beliefs for extended periods of time. The classic insights of Berger (1967) show why this may be so. He maintains that religious world views, “… are socially constructed and socially maintained. Their continuing reality, both objective … and subjective … depends upon specific social processes, namely those processes that ongoingly reconstruct and maintain the particular worlds in question” (Berger, 1967, p. 45). Berger’s (1967) emphasis on the continuous reconstruction and the ongoing maintenance of religious beliefs suggests that some sort of reinforcement is needed to sustain religious lessons that were originally learned from one’s parents. And as the model in Figure 1 suggests, attendance at worships services represents one setting in which this type of reinforcement is provided. Fundamental religious precepts are embedded in the hymns, group prayers, and sermons that typically are found in formal worship services. This is one reason why Stark and Finke (2000) maintain that the more a person participates in religious rituals, the more their faith in religion deepens. Taken together, the work of Berger (1967) as well as Stark and Finke (2000) show that more frequent attendance at worship services provides one way in which parental teachings about religion can be reinforced and maintained.

Connectedness with Others and Forgiveness

Verbally endorsing religious beliefs is important, but practicing religious beliefs in daily life is a better gauge of the extent to which an individual is committed to his or her faith. So in order to more fully understand the impact of parental religious teachings in late life, it is important to move beyond beliefs to the assessment of specific ways in which these religious beliefs are put into practice. There are likely to be a number of ways in which a sense of connectedness with others may be manifest behaviorally. For example, individuals with a strong sense of connectedness with others may be more likely to perform volunteer work at church or they may be more likely to provide informal support to fellow church members, as the research of Krause and Bastida (2009) suggests. An effort is made in the current study to strike out in a different direction by seeing whether a sense of connectedness with others influences one of the core practices of virtually every major faith tradition in the world: forgiveness (Rye et al., 2000). There are two reasons why it is important to focus specifically on this aspect of religion. First, a good deal of research has focused on the ways in which forgiving others influences physical and (especially) mental health (McCullough, Pargament, & Thoresen, 2000; Worthington, 1998). Second, a number of studies that assess the health-related effects of forgiveness have been conducted with older people (Lawler-Row, 2010), who are the focus of the current study.

Having a strong sense of connectedness with others is likely to promote forgiveness because individuals who believe they are closely connected with others are more likely to be compassionate, and people who are more compassionate are, in turn, more likely to forgive. Although researchers have yet to empirically examine the relationship between connectedness with others and compassion, evidence that the two may be related can be found in the literature on perceived self-other similarity. Perceived self-other similarity has to do with recognizing that one or more features or attributes that a person possesses are shared in common with another individual or group (Oveis, Hosberg, & Keltner, 2010). Perceived self-other similarities and feelings of connectedness with others differ primarily in terms of scope. Feelings of connectedness deal with perceived similarity between one’s self and all of humanity whereas perceived self-other similarity is limited to comparisons involving specific individuals or particular groups. But this issue aside, both constructs are based on a perceived bond with others. This similarity is important for the purposes of the current study because a number of investigators report that greater perceived self-other similarity promotes greater compassion (see Penner et al., 2005 for a review of this research).

Support for the notion that compassionate people are more forgiving is found in a study by Berry and his colleagues (Berry et al., 2005). These investigators studied crime victims and their significant others in an effort to identify the factors that are associated with their willingness to grant forgiveness to the offender. They found that individuals who value compassion were more likely to forgive than people who placed a higher value on other virtues, such as moderation, self-control, and justice.

Arguing that a close sense of connectedness promotes forgiveness belies the difficulty that is often encountered in this process. Earlier, helping others was identified as an example of how feelings of connectedness with others may be manifest. However, the personal costs of helping others are relatively low in many instances. But forgiving a transgressor for perpetrating a hurtful act may involve much greater costs and require considerably more effort. In order to forgive a transgressor, the victim must often overcome strong feelings of anger and resentment. Moving beyond the pain these emotions create may require considerable fortitude, especially when the hurtful act was unjustified. In addition, as Enright, Freedman, and Rique (1998) point out, forgiving others may make the victim feel vulnerable because the perpetrator may view forgiveness as a sign of weakness and an invitation to commit additional hurtful acts. Simply put, overcoming strong emotions and putting aside fears of additional transgressions is often not an easy task. To the extent that this is true, forgiveness may provide an important way of assessing the extent to which a person truly feels that he or she is connected with others.

Forgiving Others and Depressive Symptoms

A rapidly growing body of research suggests that individuals who forgive others are less likely to experience a range of mental health problems, including depressive symptoms (e.g., Toussaint et al., 2008). A number of intervening mechanisms have been identified to explain the relationship between forgiving others and depressive symptoms. The role that these mediators play is easier to grasp when they are viewed from the perspective of being unable or unwilling to forgive others. For example, several studies indicate that individuals who are not able to forgive transgressors are more likely to ruminate. This means that they repeatedly return to the hurtful act, reliving it over and over again in their minds (Burnette et al., 2009). This is important because research suggests that rumination that arises from the inability to forgive tends to promote greater depressive symptomatology (Ingersoll-Dayton, Torges, & Krause, 2010). Other investigators report that individuals who do not forgive a transgressor are more likely to experience a sense of hopelessness, and those who feel hopeless are, in turn, more likely feel depressed (Toussaint et al., 2008). Further intervening mechanisms are identified by Baumeister, Exline, and Sommer (1998). These investigators maintain that holding a grudge encourages an individual to assume the role of the victim, which has to do with embracing feelings of suffering, weakness, and distress. Moreover, Baumeister et al. (1998) argue that the role of the victim breeds passivity and failure in unrelated aspects of life, which in turn may create further mental health problems.

But beyond the intrapsychic costs of un-forgiveness, the inability to forgive others may extract interpersonal costs, as well. For example, failure to forgive a transgressor may lead to the termination of a previously satisfying social relationship (Baumeister et al., 1998). This is noteworthy because the loss of close social relationships is linked to a wide range of health-related problems (Cohen, 2004). In addition, when victims are unable to forgive a transgressor, they are more likely to engage in acts of retaliation and they are more likely to require the transgressor to perform acts of contrition. These responses on the part of the victim may prompt the transgressor to commit further hurtful acts. This creates a vicious cycle of recrimination and revenge that has been shown to have exert an adverse effect on mental health (Krause & Ellison, 2003).

Although the discussion of how the inability to forgive others may lead to depression, a nagging issue remains that has to do with the direction of causality between these constructs. Instead of specifying that the inability to forgive others increases the odds of feeling depressed, it is just as easy to argue that people who are initially depressed are subsequently less likely to forgive others. There are two ways to address this issue. First, a number of studies indicate that levels of depression and anxiety can be reduced by participation in forgiveness therapy (e.g., Reed & Enright, 2006). This research is noteworthy because the effects of forgiveness therapy have been assessed with studies that employ experimental designs. As a result, the notion that forgiveness affects depression is difficult to dispute.

In addition to experimental assessments of forgiveness therapy, a second way to explore the causality issue is provided in the analyses that are presented below. This strategy is depicted in Figure 1. As Menard (1991) points out, there are several ways to analyze survey data that have been gathered at more than one point in time. In fact, he identifies four “pure” longitudinal models (Menard, 1991, p. 59). The model depicted in Figure 1 is among them. In this type of conceptual scheme, change in the dependent variable is expressed in terms of change in the independent variable. The logic of this specification is straightforward: if the level of one variable (e.g., depressive symptoms) depends upon the level of a second variable (e.g., forgiveness), then if the second variable changes, the first variable must also change. Menard (1991) goes on to point out that many researchers express their hypotheses in terms of this model but then unknowingly test their hypotheses with a different model. Although the analysis of longitudinal survey data does not conclusively resolve the issue of causality, bringing change to the foreground helps relax stringent assumptions that are inherent in cross-sectional analysis thereby making it possible to make statements about the temporal ordering between key constructs with greater confidence.

Methods

Sample

The data for this study come from an ongoing nationwide survey of religion and health among older people. The population for this study was defined as all household residents who were either African American or white, noninstitutionalized, English-speaking, and at least 66 years of age. Geographically, the study population was restricted to all eligible persons residing in the coterminous United States (i.e., residents of Alaska and Hawaii were excluded). Finally, the study population was restricted to currently practicing Christians, individuals who were Christian in the past but no longer practice any religion, and people who were not affiliated with any faith tradition at any point in their lifetime. This study was designed to explore a range of issues involving religion and health. As a result, individuals who practice a faith other than Christianity were excluded because it would be too difficult to devise a comprehensive battery of religion measures that would be suitable for individuals of all faiths.

The sampling frame for this study consisted of all eligible persons contained in the beneficiary list maintained by the Centers for Medicare and Medicaid Services (CMS). A five-step process was used to draw the sample from the CMS Files (see Krause, 2002a, for a detailed discussion of these steps).

The baseline survey took place in 2001. The data collection for all waves of interviews was conducted by Harris Interactive (New York). A total of 1,500 interviews were completed, face-to-face, in the homes of the study participants. Older African Americans were over-sampled so that sufficient statistical power would be available to assess race differences in religion. As a result, the Wave 1 sample consisted of 748 older whites and 752 older African Americans. The overall response rate for the baseline survey was 62%.

The Wave 2 survey was conducted in 2004. A total of 1,024 study participants were re-interviewed successfully, 75 refused to participate, 112 could not be located, 70 were too ill to participate, 11 had moved to a nursing home, and 208 were deceased. Not counting those who died or moved to a nursing home, the re-interview rate for the Wave 2 survey was 80%.

A third wave of interviews was completed in 2007. A total of 969 older study participants were re-interviewed successfully, 33 refused to participate, 118 could not be located, 17 were too sick to take part in the interview, and 155 older study participants had died. Not counting those who died, the re-interview rate was 75 percent.

Wave 4 was completed in 2008. A total of 718 older study participants were re-interviewed successfully, 61 refused to participate, 92 could not be located, 77 were too sick to take part in the interview, and 153 had died. Not counting study participants who died, the re-interview rate at Wave 4 was 90 percent.

The analyses presented below are based on data from the Wave 3 and Wave 4 surveys. These data collection points are used because question about parental religious socialization practices were administered for the first time at Wave 3.

The full information maximum likelihood estimation (FIML) procedure was used to impute missing values in the data. As Graham, Olchowski, and Gilreath (2007) report, FIML produces results that are equivalent to more time consuming procedures for dealing with item non-response, such as multiple imputation.

Preliminary analysis revealed that the average age of the participants in this sample was 78.1 years (SD = 5.1 years), approximately 35 percent were older men, 54 percent were white, and the average number of years of schooling was 11.8 (SD = 3.2).

Measures

Measures of the core study constructs are presented in Table 1. The procedures that were used to score these indicators are provided in the footnotes of this table.

Table 1
Core Study Measures

Depressive Symptoms

Three items were taken from the Center for Epidemiologic Studies Depression Scale (CES-D) to assess depressive symptoms (Radloff, 1977). Identical indicators were used to measure depressive symptoms in the Wave 3 and Wave 4 interviews. These items reflect the cognitive-affective aspects of depressive symptoms, including feeling sad, blue, and depressed. A high score on these measures denotes greater depressive symptomatology. The mean depressive symptom score at the Wave 3 survey is 4.3 (SD = 2.0) and the mean at Wave 4 is 4.2 (SD = 1.9).

Parental Religious Socialization Practices

Four items were used to see if older study participants were encouraged by their parents to become involved in religion. These indicators ask how often their parents encouraged them to attend church, pray, read the Bible, and how often their parents talked to them about religious principles and teachings. These questions were administered twice during the Wave 3 survey. The first time the questions asked about the influence of mothers (or mother surrogates) while questions assessing the religious influence of fathers (or father surrogates) were administered the second time. A high score on the items represents greater encouragement by parents to adapt religious teachings and practices.

The measures of parental religious socialization practices present a data analytic challenge because some study participants were not raised by either their natural fathers or a father surrogate (N = 105), while others were not raised by either their natural mother or a mother surrogate (N = 19). In order to insure that these participants were not excluded from the analyses, a strategy for assessing parental religious socialization practices was devised that was based on the following decision rules. First, if an older study participant was not raised by either their natural father or a father surrogate, then their parental religious socialization score was based on their report of their mother’s religious influence. Second, if an older respondent was not raised by either their natural mother or a surrogate mother, then their parental religious socialization score was based on their report of their father’s religious influence. Third, if an older adult was raised by both their mother and their father, then the higher of the two scores was selected for use as the parental religious socialization measure. Fourth, if an older study participant was raised by both parents and the religious socialization score of both parents was the same, then their religious socialization score was based on their report of their mother’s influence. The mean of the parental religious socialization measure is 12.8 (SD = 3.5).

Church Attendance

A single widely-used indicator was included in the Wave 3 survey to assess how often the older people in this study attend worship services. A high score denotes more frequent church attendance. The average level of church attendance is 5.9 (SD = 2.8).

Connectedness With Others

Three items were used to measure a sense of connectedness with others. For example, one question asked the older study participants if they believe their faith helps them see a common bond among all people. An extensive program of qualitative research was conducted in order to develop these (and other) measures of religion (Krause 2002b). These items were written so that they refer to a sense of connectedness that arises specifically from one’s own faith. This strategy was implemented in order to ground feelings of connectedness with others in an explicitly religious context. The items in Table 1 are coded so that a high score denotes a greater sense of connectedness with others. The mean value of this brief composite at Wave 3 is 10.5 (SD = 1.8).

Forgiveness of Others

Forgiveness of others was measured with three indicators. These items assess whether older study participants feel resentful toward others, hold a grudge, and whether they are able to forgive other people for things they have done. The measures of forgiveness are coded so that a high score stands for older people who are more forgiving. Identical measures of forgiveness of others were included in the Wave 3 and Wave 4 interviews. The mean at Wave 3 is 10.2 (SD = 1.7) and the mean at Wave 4 is 10.6 (SD = 1.6).

Demographic Control Variables

Recall that the relationships among the constructs in Figure 1 were estimated after the effects of age, sex, education, and race were included in this conceptual scheme. Age and education are continuous variables that are scored in years. In contrast, sex (1 = men; 0 = women) and race (1 = white; 0 = black) are coded in a binary format.

Results

The findings from this study are presented in three sections. Several issues about the way in which the study model was estimated are discussed in the first section. Data on the reliability of the multiple item constructs is presented in section two. Section three contains the substantive findings this study.

Model Development Issues

The model depicted in Figure 1 was evaluated with the maximum likelihood estimator in Version 8.80 of the LISREL statistical software program (du Toit and du Toit, 2001). Researchers who use this estimator must assume that the observed indicators in their models have a multivariate normal distribution. Preliminary tests (not shown here) revealed that this assumption had been violated in the current study. Although there are a number of ways to deal with departures from multivariate normality, the straightforward approach that is discussed by du Toit and du Toit (2001) was followed here. These investigators report that departures from multivariate normality can be handled by converting the raw scores of the observed indicators to normal scores prior to estimating the model (du Toit and du Toit, 2001, p.143). Based on these insights, the analyses presented below are based on observed indicators that have been normalized.

Forgiveness and depressive symptoms were measured at two points in time. Consequently, two issues involving the measurement of these constructs must be addressed so that the model with the best fit to the data can be identified. The first has to do with seeing whether the measurement error terms for identical indicators of forgiveness, as well as the measurement errors associated with identical indicators of depressive symptoms, are correlated over time. Preliminary tests (not shown here) reveal that the measurement error terms are significantly correlated over time (details of these analyses are available from the author). The second issue involves testing factorial invariance over time (Bollen, 1989). These tests are performed in order to see if the elements of the measurement model (i.e., the factor loadings and measurement error terms) are the same over time for the forgiveness and depressive symptom items, respectively. Preliminary tests (not shown here) indicate that the factor loadings, but not the measurement error terms, are invariant over time. Although it would have been preferable to find that the measurement error terms are also invariant over time, Reise, Widaman, and Pugh (1993) argue that achieving even partial invariance is acceptable, and that meaningful conclusions can be drawn about the nature of the relationships among the constructs in a model.

The fit of the model that emerged after tests for these measurement issues were evaluated is good. Because the FIML procedure was used to deal with item non-response, the LISREL software program provides only two goodness-of-fit measures. The first is the full information maximum likelihood chi-square (394.888 with 201 degrees of freedom, p < .001). Unfortunately, this statistic is often not useful because it tends to underestimate the fit of the model to the data when samples are large, like the sample in the current study. A better grasp of the fit of the model to the data is provided by the second goodness-of-fit measure - the root mean square error of approximation (RMSEA). The RMSEA value for the model in Figure 1 is .037 (90 % C.I. = .031; .042). As Kelloway (1998) and others suggest, RMSEA values below .05 indicate a very good fit of the model to the data. Because the lower bound of the confidence interval is below .05 (i.e., .031), the null hypothesis that the model has a close approximate fit in the population cannot be rejected. Moreover, because the upper bound of the confidence interval is below .10 (i.e., .042), the null hypothesis that the fit of the model in the population is as bad, or worse, than the observed fit can be rejected.

Reliability of the Observed Indicators

Table 2 contains the factor loadings and measurement error terms that were derived from estimating the study model. These coefficients are important because they provide information about the reliability of the multiple item study measures. Kline (2005) points out that items with standardized factor loadings in excess of .600 tend to have reasonably good reliability. As the data in Table 3 indicate, the standardized factor loadings range from .450 to .921. Although it would have been desirable to find that all the factor loadings exceed .600, it is important to keep two issues in mind. First, only two of the nineteen standardized factor loadings fall below the recommended value. Second, the .600 guideline is not endorsed by all investigators. Instead, some researchers, such as Brown (2006), suggest that standardized factor loadings above .400 are acceptable (see also Ximenez, 2006).

Table 2
Measurement error parameter estimates for multiple item study measures (N = 718)
Table 3
Connectedness With Others and Change in Depressive Symptoms (N = 718)

Although the factor loadings and measurement error terms associated with the observed indicators provide useful information about the reliability of each item, it would be helpful to know something about the reliability of the scales as a whole. Fortunately, it is possible to compute these reliability estimates with a formula provided by DeShon (1998). This procedure is based on the factor loadings and measurement error terms in Table 2. Applying the procedures described by DeShon to these data yield the following reliability estimates for the multiple item constructs in Figure 1: parental religious socialization practices (.919), feelings of connectedness with others (.937), forgiveness of others - Wave 3 (.704), forgiveness of others - Wave 4 (.729), depressive symptoms - Wave 3 (.870), depressive symptoms - Wave 4 (.866). Taken as a whole, these estimates indicate that the reliability of the multiple item constructs is adequate.

Substantive Findings

Table 3 contains estimates of the relationships among the measures in the study model as well as the relationships between the demographic control variables and these core constructs. Taken as a whole, these data provide support for the theoretical rationale that was developed for the current study. More specifically, the findings suggest that older people who were encouraged by their parents to become more involved in religion report that they attend worship services more often than older adults whose parents offered less encouragement to become involved in religion (Beta = .211; p < .001). The results further reveal that older study participants who attend church more often are more likely to believe there is a deep connection among all people (Beta = .360; p < .001). The fact that both relationships are statistically significant is consistent with the notion that core religious beliefs are initially instilled by one’s parents and subsequently reinforced through continued religious practice (i.e., church attendance).

The findings in Table 3 also suggest that older adults who feel closely connected with others are more likely to forgive transgressors for the hurtful acts they have committed (Beta = .142; p < .01). The fact that connectedness with others is associated with forgiveness at Wave 3 but not forgiveness at Wave 4 (Beta = .053; n.s.) provides some preliminary insight into the temporal nature of the relationship between these constructs. More specifically, the data suggest that the influence of connectedness with others on forgiveness tends to become manifest rather quickly (i.e., at Wave 3) and that a longer period of time (i.e., at Wave 4) is not required for the influence of feelings of connectedness on forgiveness to become evident.

The data in Table 3 indicate that older study participants who forgive others for the things they have done tend to report that they have experienced fewer symptoms of depression at Wave 3 (Beta = −.213; p < .001). However, the relationship between the Wave 4 measures of forgiveness and depressive symptoms is of greater interest because it provides a more stringent test of the association between these two constructs. The estimate of this relationship suggests that an increase in forgiveness over time is associated with a decrease in the frequency of depressive symptoms over time (Beta = −.283; p < .001).

One of the advantages of working with latent variable models arises from the fact that it is possible to assess the indirect and total effects of a variable that operate through a model. The meaning of these terms can be clarified with an example. The model in Figure 1 specifies that older people who were encouraged by their parents to become involved in religion are more likely to attend worship services in late life, and more frequent church attendance is, in turn, associated with developing a stronger sense of connectedness with others. This means that parental religious socialization practices affect feelings of connectedness with others indirectly through the frequency of church attendance. When the direct effect of parental religious socialization practices is added to the indirect effect that operates through church attendance, the resulting total effect provides a broader vantage point for assessing the overall influence of parental religious socialization practices. Breaking down the total effects of a relationship into direct and indirect effects is known in the literature as the decomposition of effects (Alwin 1988).

Two decompositions will be reviewed below in order to clarify and extend the findings that have been presented up to this point. A major emphasis is placed in the current study on parental religious socialization practices. Exploring two of the decompositions sheds additional light on the lifelong influence that parents can exert on the religious beliefs and practices of their children. First, the data in Table 3 indicate that older people who were encouraged by their parents to become involved in religion are more likely to feel a deeper sense of connectedness with others (Beta = .210; p < .001). However, when the indirect effect of these socialization practices that operates through church attendance (Beta = .076; p < .001; not shown in Table 3) is taken into account, the way in which parental religious socialization practices operate comes into sharper focus. The total effect of parental religious socialization practices on feelings of connectedness with others is: Beta = .286; p < .001 (not shown in Table 3). When the indirect effect is divided by the total effect (.076/.286 = .266) it can be seen that 26.6 percent of the effect of parental religious socialization practices on feelings of connectedness is due to reinforcement of this core religious belief in worship services. The fact that the bulk of the influence of parental religious socialization practices on feelings of connectedness is captured by the direct effect (.210/.286 = 73.4 percent) attests to the incredible endurance of these early teachings that operate above and beyond current church attendance.

The second decomposition has to do with the relationship between feelings of connectedness with others and depressive symptoms. Initially, the data in Table 3 may create the impression that feelings of connectedness with others is not significantly associated with depressive symptoms at Wave 3 (Beta = -.068; n.s.) or Wave 4 (Beta = .059; n.s.). However, further analysis (not shown in Table 3) suggests that feelings of connectedness with others exerts a statistically significant indirect effect on depressive symptoms at Wave 3 (Beta = −.030; p < .05) and Wave 4 (Beta = −.070; p < .01). The indirect effect on depressive symptoms at Wave 4 is the more important of the two. This relationship provides two important insights into the nature of the relationship between these constructs. First, this indirect effect suggests that feelings of connectedness with others are associated with change in depressive symptoms over time (i.e., a decline in depressive symptoms). Second, the fact that the indirect effect, but not the direct effect, is statistically significant suggests that the relationship between connectedness with others and change in depressive symptoms is due entirely to the mediating role that is played by forgiveness of others and depressive symptoms at Wave 3.

Conclusions

Viewed broadly, the purpose of this study was to bring together two fundamental issues that have yet to be fully integrated into the literature on religion even though each has a long history. The first has to do with seeing the connection that exists among all people whereas the second involves parental religious socialization practices. Writing in 1799, Schleiermacher, an eminent Protestant theologian, captured the essence of feeling connected with others. In the process of discussing the social basis of religion, he observed that, “The highest and most cultured always see a universal union, and, in seeing it, establish it. Every man is only in contact with his neighbour; but on every side and in every direction he has neighbours and is inseparably bound up with this whole” (Schleiermacher, 1799/1994). But Schleiermacher (1799/1994) did not fully illuminate how this sense of connectivity arises. The data in the current study suggest that these beliefs are shaped, in part, by parental religious socialization practices. Linking the two provides one way of showing that the experiences and lessons learned in childhood play a significant role in shaping the way people turn out in late life. However, the notion that childhood plays such a pivotal role in adult development is hardly new. Writing in 1807, the poet Wordsworth coined the frequently-cited phrase: “The child is father to the man” (as quoted in Kaplan, 1992, p. 374). One of the unique contributions of the current study arises from documenting that the influence of religious teachings during childhood may be traced across the entire life course. Recall that the people in the current study were, on average, 78 years old. This means that the effects of parental religious socialization practices can be observed over a period of 60 years and longer. The fact that this relationship was observed even after the effects of church attendance in late life were taken into account provides further evidence of the long reach of religious lessons that were learned in childhood.

In addition to exploring the genesis of beliefs about connectedness with others, another goal of the current study was to examine the health-related consequences of these beliefs by seeing if they are associated with depressive symptoms. But rather than merely showing that the two are related, an effort was made to explore one way in which this relationship may arise. The data indicate that older people who believe a connection exists among all individuals are more likely to forgive those who have hurt them, and people who forgive others tend to experience fewer symptoms of depression. In the process, an effort was made to instill greater confidence in the findings by using longitudinal data to show that change in forgiveness is associated with change in depressive symptoms over time.

Although the findings from the current study may have contributed to the literature, a considerable amount of work remains to be done. As discussed earlier, the items that assess feelings of connectedness were written so that they were specifically grounded within the context of religion. However, there are other reasons why people may believe there is a connection among all individuals. For example, Kasprow and Scotton (1999) discuss a type of psychotherapy that is based on transpersonal theory, which focuses specifically on deepening and person’s sense of interpersonal connectedness. Because factors other than religion may influence a sense of connectedness, researchers need to know if the influence of religiously-based feelings of connectedness are unique, or whether they have the same affect on mental health as feelings of connectedness that are fostered by factors in the wider secular world.

In the process of investigating this as well as other issues, research is also needed to address the limitations in the current study. In addition to the issues involving the direction of causality that were discussed earlier, two additional shortcomings should be mentioned here. First, data on parental religious socialization practices were obtained when the study participants reached old age. Because a significant amount of time has elapsed since these socialization practices were implemented, it is possible that reports of these practices may be biased. This issue represents a specific instance of the widely discussed problems with the validity of autobiographical memory (e.g., Piolino et al., 2010). The only conclusive way to resolve this potential problem is to literally interview people numerous times from childhood through late life. Because this ideal solution is not feasible, the only course of action that is available at the present time is to rely on retrospective recall while keeping the potential problems associated with autobiographical memory in mind. Second, the measure of feelings of connectedness with others was developed especially for this survey. As a result, additional psychometric work must be conducted to determine the convergent, discriminant, and predictive validity of the measures that assess feelings of connectedness with others (Campbell & Fiske, 1959).1

Charles Horton Cooley (1927) was one of the founders of social psychology. In the last book he wrote, Cooley (1927) emphasized the importance of seeing the fundamental connectedness and interdependence among all people. It is especially important to note that Cooley (1927) linked this sense of connectedness directly with religion: “The best religious education would be one that accustomed us from childhood to cooperation … in service of human wholes. Through the family, the play-group, the school, the community, the nation, humanity, we might acquire an enlarging sense of God” (Cooley 1927: 265). However, before implementing this recommendation, a necessary first step involves documenting the importance of seeing the connectedness among all people. The current study was conducted with this goal in mind. But considerably more research is needed before firm recommendations can be provided. Hopefully, the findings that were presented and the issues that were identified in the current study will motivate other investigators to explore this important aspect of religious life.

Acknowledgments

This research was supported by grants from the National Institute on Aging (RO1 AG014749; RO1 AG026259) and a grant from the John Templeton Foundation

Footnotes

1Two points should be made about the psychometric properties of the measure that assesses feelings of connectedness with others. First, the current study provides preliminary evidence about the construct validity of the measure that assesses feelings of connectedness with others. Construct validity refers to the extent to which a new measure relates to other constructs in a theoretically meaningful way (Zeller & Carmines, 1980). As the data from the current study show, older people who go to church more often tend to feel more closely connected with others. In addition, the results reveal that older adults who feel more closely connected with others are more likely to forgive people for what they have done. Both of these findings make sense conceptually. However, it should be emphasized that construct validity is never established in a single study. Instead, it is built up slowly in a literature as theoretically meaningful relationships between a new construct and a range of other variables slowly accumulates. The second psychometric issue has to do with the factor structure of the items that assess feelings of connectedness with others. An exploratory factor analysis (not shown here) was conducted with these indicators. This analysis revealed that the sense of connectedness items reflect a single underlying construct (i.e., one factor provided the best fit of the model to the data).

Bibliography

  • Alwin Duane F. Structural equation modeling in research on human development and aging. In: Schaie KW, Campbell RT, Meredith W, Rawlings SC, editors. Methodological issues in aging research. New York: Springer; 1988. pp. 71–170.
  • Baumeister RF, Exline JJ, Sommer KL. The victim role, grudge theory, and two dimensions of forgiveness. In: Worthington EL, editor. Dimensions of forgiveness: psychological research and theoretical perspectives. Philadelphia: Templeton Foundation Press; 1998. pp. 70–104.
  • Berger PL. The sacred canopy: Elements of a sociological theory. New York: Doubleday; 1967.
  • Berry JW, Worthington EL, Wade NG, Witvliet CV, Kiefer RP. Forgiveness, moral identity, and perceived justice in crime victims and their supporters. Humboldt Journal of Social Relations. 2005;29:136–162.
  • Bollen KA. Structural equations with latent variables. New York: Wiley; 1989.
  • Brown TA. Confirmatory factor analysis for applied research. New York: Guilford; 2006.
  • Burnette JL, Davis DE, Green JD, Worthington EL, Bradfield E. Insecure attachment and depressive symptoms: The mediating role of rumination, empathy, and forgiveness. Personality and Individual Differences. 2009;46:276–280.
  • Campbell DT, Fiske DW. Convergent and discriminant validation by the multi-trait-multimethod matrix. Psychological Bulletin. 1959;56:81–105. [PubMed]
  • Cohen S. Social relationships and health. American Psychologist. 2004;59:676–684. [PubMed]
  • Cooley CH. Life and the student. New York: Alfred A. Knopf; 1927.
  • Davidson JD. Patterns of belief at the denominational and congregational levels. Review of Religious Research. 1972;13:197–205.
  • DeShon RP. A cautionary note on measurement error correlations in structural equation models. Psychological Methods. 1998;3:412–423.
  • du Toit M, du Toit S. Interactive LISREL: User’s guide. Lincolnwood, IL: Scientific Software International; 2001.
  • Eliade M. A history of religious ideas, volume 1, from the Stone Age to the Eleusinian mysteries. Chicago: University of Chicago Press; 1978.
  • Enright RD, Freedman S, Rique J. The psychology of interpersonal forgiveness. In: Enright RD, North J, editors. Exploring forgiveness. Madison, WI: University of Wisconsin Press; 1998. pp. 46–62.
  • Graham JW, Olchowski AE, Gilreath TD. How many imputations are really needed? Some practical clarifications of multiple imputation theory. Prevention Science. 2007;8:206–213. [PubMed]
  • Hood RW, Hill PC, Spilka B. The psychology of religion: An empirical approach. 4th edition. New York: Guilford; 2009.
  • Ingersoll-Dayton B, Torges C, Krause N. Unforgiveness, rumination, and depressive symptoms among older adults. Aging and Mental Health. 2010;14:439–449. [PMC free article] [PubMed]
  • Kaplan J. Bartlett’s familiar quotations. 16th edition. Boston: Little, Brown, and Company; 1992.
  • Kasprow MC, Scotton BW. A review of transpersonal theory and its application to the practice of psychotherapy. Journal of Pyschotherapy Practice and Research. 1999;8:12–23. [PMC free article] [PubMed]
  • Kelloway EK. Using LISREL for structural equation modeling. Thousand Oaks, CA: Sage; 1998.
  • Kline RB. Principles and practice of structural equation modeling. New York: Guilford; 2005.
  • Krause N. Church-based social support and health in old age: Exploring variations by race. Journal of Gerontology: Social Sciences. 2002a;57B:S332–S347. [PubMed]
  • Krause N. A comprehensive strategy for developing closed-ended survey items for use in studies of older adults. Journal of Gerontology: Social Sciences. 2002b;57B:S263–S274. [PMC free article] [PubMed]
  • Krause N. Exploring the stress-buffering effects of church-based social support and secular social support on health in late life. Journal of Gerontology: Social Sciences. 2006b;61B:S35–S43. [PubMed]
  • Krause N. Aging in the church: How social relationships affect health. West Conshohocken, PA: Templeton Foundation Press; 2008.
  • Krause N, Bastida E. Core religious beliefs and providing support to others in late life. Mental Health, Religion & Culture. 2009;12:114–123. [PMC free article] [PubMed]
  • Krause N, Ellison CG. Forgiveness by God, forgiveness of others, and psychological well-being in late life. Journal for the Scientific Study of Religion. 2003;42:77–93. [PMC free article] [PubMed]
  • Krause N, Ellison CG. Parental religious socialization practices and self-esteem in late life. Review of Religious Research. 2007;49:109–127.
  • Lawler-Row KA. Forgiveness as a mediator of the religiosity-health relationship. Psychology of Religion and Spirituality. 2010;2:1–16.
  • McCullough ME, Pargament KI, Thoresen CE. Forgiveness: Theory, research, and practice. New York: Guilford; 2000.
  • Menard S. Longitudinal research. Newbury Park, CA: Sage; 1991. (Sage University Paper Series on Quantitative Applications in the Social Sciences, 76).
  • Mills JS. Three essays on religion. New York: Cosimo Classics; 1874/2008.
  • Oveis C, Horberg EJ, Keltner D. Compassion, pride, and social intuitions of self-other similarity. Journal of Personality and Social Psychology. 2010;98:618–630. [PubMed]
  • Penner LA, Dovidio JF, Piliavin JA, Schroeder DA. Prosocial behavior: Multilevel perspectives. Annual Review of Psychology. 2005;56:365–392. [PubMed]
  • Piolino P, Coste C, Martinelli P, Mace AL, Quinette P, Guillery-Girard B, Belleville S. Reduced specificity of autobiographical memory in aging: Do the executive and feature binding functions of working memory have a role? Neuropsychologia. 2010;48:429–440. [PubMed]
  • Radloff LS. The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1:385–401.
  • Reed GL, Enright RD. The effects of forgiveness therapy on depression, anxiety, and posttraumatic stress for women after spousal emotional abuse. Journal of Consulting and Clinical Psychology. 2006;74:920–929. [PubMed]
  • Regnerus MD, Smith C, Smith B. Social Context in the Development of Adolescent Religiosity. Adult Development Science. 2004;8:27–38.
  • Reise SP, Widaman KF, Pugh RH. Confirmatory factor analysis and item response theory: Two approaches for exploring measurement invariance. Psychological Bulletin. 1993;114:552–566. [PubMed]
  • Ross EA. Social control V. American Sociological Review. 1896;2:433–445.
  • Rye MS, Pargament KI, Ali MA, Beck GL, Dorff EN, Hallisey C, Narayanan V, Williams JG. Religious perspectives on forgiveness. In: McCullough ME, Pargament KI, Thoresen CE, editors. Forgiveness: Theory, research, and practice. New York: Guilford; 2000. pp. 17–40.
  • Schleiermacher F. On religion: Speeches to its cultured despisers. Louisville, KY: John Knox Press; 1799/1994.
  • Sherkat DE. Counterculture or Continuity? Competing Influences on Baby Boomers� Religious Orientations and Participation. Social Forces. 1998;76:1087–1114.
  • Simmel G. A contribution to the sociology of religion. In: Helle HJ, editor. Essays on religion - Georg Simmel. New Haven, CT: Yale University Press; 1898/1997. pp. 101–120.
  • Smith H. The world's religions. New York: Harper Collins; 1991.
  • Spilka B, Hood RW, Hunsberger B, Gorsuch R. The psychology of religion. 3rd Edition. New York: Guilford; 2003.
  • Stark R, Finke R. Acts of faith: Explaining the human side of religion. Berkeley, CA: University of California Press; 2000.
  • Touossaint LL, Williams DR, Musick MA, Everson-Rose SA. Why forgiveness may protect against depression: Hoplessness as an explanatory mechanism. Personality and Mental Health. 2008;2:89–103.
  • Worthington EL. Dimensions of forgiveness: Psychological research and theological perspectives. Philadelphia: Templeton Foundation Press; 1998.
  • Ximenez D. A monte carlo study of recovery of weak factor loadings in confirmatory factor analysis. Structural Equation Modeling. 2006;13:587–614.
  • Zeller RA, Carmines EG. Measurement in the social sciences: The link between theory and data. New York: Cambridge University Press; 1980.
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Links

  • MedGen
    MedGen
    Related information in MedGen
  • PubMed
    PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...