Forgiveness by God, Forgiveness of Others, and Psychological Well-Being in Late Life

Abstract

The purpose of this study is to examine the relationships among forgiveness by God, forgiveness of others, and psychological well-being with data provided by a nationwide survey of older adults. Three main findings emerge from the analyses. First, the data suggest that forgiving others tends to enhance psychological well-being, and these salubrious effects are greater than those associated with forgiveness by God. Second, the findings indicate that how older people go about forgiving others is important: older adults who require transgressors to perform acts of contrition experience more psychological distress than those who forgive unconditionally. Third, the results reveal that forgiveness by God may be involved in this process because older people who feel they are forgiven by God are less likely to expect transgressors to perform acts of contrition.

Introduction

Forgiveness figures prominently in the Christian religion. The basic tenets of this faith place an emphasis on repentance and seeking forgiveness from God (Brown 1997). Moreover, a central message in the New Testament involves the importance of forgiving other people for things they have done (Rye et al. 2001). Recently, the study of forgiveness has attracted a great deal of attention because some scholars believe it may enhance physical as well as psychological well-being across the life course (Thoresen, Harris, and Luskin 2001; Worthington, Berry, and Parrott 2001). A small cluster of studies, which focus primarily on forgiveness of others, provide empirical support for this view. More specifically, this research suggests that people who are able to forgive others for things they have done are less likely to experience symptoms of psychological distress than individuals who are unable or unwilling to forgive (Maltby, Macaskill, and Day 2001; Toussaint et al. 2001). Further support for the potentially beneficial effects of forgiveness may be found in the literature on clinical psychology, where a number of psychotherapeutic programs have been developed to explicitly enhance a person’s ability to forgive others (Enright and Coyle 1998; Hargrave and Anderson 1992).

Although research on forgiveness and psychological well-being is intriguing, a good deal of work remains to be done. The purpose of the present study is to address three issues that have been largely overlooked by other investigators. First, most studies on forgiveness focus solely on forgiving other people; far fewer compare and contrast the effects of forgiving others with receiving forgiveness from God (for a notable exception, see Toussaint et al. 2001). This is unfortunate because simultaneously examining the effects of forgiveness from different sources may begin to shed light on why forgiveness may be associated with psychological well-being. More specifically, forgiving other people provides the opportunity to translate belief into actions and behaviors that may provide direct tangible benefits in daily life, such as the mending of strained social ties. In contrast, relationships with God are based largely on faith and belief and, as a result, the benefits of receiving forgiveness from God may primarily involve cognitive and affective factors.

A number of investigators maintain that the process of forgiving others is quite complex because people go about forgiving others in different ways (Enright and North 1998). For some individuals, forgiveness is gift-like or unconditional. In this instance, transgressors do not need to do anything before they are forgiven (Scobie and Scobie 1998). In contrast, other people require transgressors to perform acts of contrition (e.g., making restitution) before they are willing to forgive them (Frankel 1998). The fact that there is more than one way to forgive others raises some important research questions. In particular, we need to know if both ways of forgiving promote well-being, and if they do, whether one approach is more effective than the other. To the best of our knowledge, no one has assessed whether forgiving others unconditionally exerts a more beneficial impact on psychological well-being than requiring transgressors to perform acts of contrition. The second goal of the present study is to examine this important issue.

The third aim of the present study is to delve more deeply into the process of forgiveness by probing the factors that influence the decision to require transgressors to perform acts of contrition. We suspect that religious factors may be especially important in this respect. More specifically, based on findings from our qualitative research on forgiveness (Krause and Ingersoll-Dayton 2001), we explore the notion that people who feel they are forgiven by God are more likely to forgive others unconditionally than individuals who do not believe that God has forgiven them for things they have done. Again, this appears to be the first time this issue has been examined empirically in the literature.

The data for this study come from a nationwide probability survey of older adults. There are two reasons why this sample is especially well suited for the study of religion, forgiveness, and well-being. First, many studies on forgiveness are based on convenience samples (Mullet et al. 1998) or samples consisting of college students (Maltby, Macaskill, and Day 2001; McCullough et al. 2001). As a result, it is not clear whether the findings from this work may be generalized to a wider population. The nationwide sampling frame used in the present study helps alleviate these concerns. Second, the sample used in this study consists solely of older adults. This is an especially important population to focus on when studying forgiveness and religion because research consistently shows that older people are more forgiving than younger individuals (Girard and Mullet 1997). Moreover, recent evidence indicates that people tend to become more religious as they grow older (Wink and Dillon 2001). It follows from this work that if forgiveness is related to well-being, and if religion plays a role in this process, then the relationships among these constructs should be especially evident in samples comprised of older people.

The discussion that follows is divided into three main sections. First, the theoretical underpinnings of the study are developed in greater detail. Following this, the study sample and measures are presented. In the final section, the study findings are reviewed and discussed.

Forgiveness and Well-Being in Late Life

This section begins with a discussion of our plan to expand the scope of research on forgiveness by God and forgiving other people by examining an outcome measure that has not been used by other researchers (i.e., death anxiety). Following this, we examine why different approaches to forgiving others may be associated with psychological well-being. Finally, a preliminary first step is taken to probe the process of forgiveness more deeply by identifying factors that may encourage older people to adopt different strategies for forgiving others.

Forgiving Others and Being Forgiven by God

As several investigators point out, there are a number of different dimensions, or sources, of forgiveness (Toussaint et al. 2001). More specifically, people may forgive themselves, people may forgive others, and people hold different beliefs about whether they have been forgiven by God. It is somewhat surprising to find that relatively few studies in the literature compare and contrast the effects of these different dimensions of forgiveness on psychological distress and well-being. As noted earlier, Toussaint et al. (2001) conducted one of the few studies that examines this issue. Their research reveals that forgiveness of others tends to exert a more beneficial effect on psychological distress and life satisfaction than forgiveness by God. Since an extensive amount of research has not been done on this issue, we begin our analyses by seeing whether the important findings reported by Toussaint et al. (2001) can be replicated.

However, instead of merely verifying results reported by other investigators, we contribute to the literature by including an outcome measure that has not been considered by other researchers. In particular, an effort is made to compare and contrast the effects of forgiveness by God and forgiveness of others on death anxiety. The theoretical work of Erikson (1959) and Butler and Lewis (1982) shows why death anxiety may be an especially relevant outcome measure in research on forgiveness in late life. The final stage in Erikson’s (1959) well-known theory of life-span development deals with resolving the crisis of integrity versus despair. This is a time of deep introspection, where a person looks back over the life he or she has lived and attempts to accept the way things have turned out. If this stage is not resolved successfully, Erikson (1959) maintains that a person can slip into despair, which often involves a “fear of death” and “a show of disgust, a misanthropy, or chronic contemptuous displeasure with particular institutions and particular people” (Erikson 1959:98).

Similar views may be found in Butler’s compelling work on the life-review process (Butler and Lewis 1982). He argues that as people enter late life, they invest a significant amount of time reviewing experiences they have had with an eye toward weaving their life stories into a more coherent whole. Themes of forgiveness figure prominently in this process. In particular, Butler maintains that one of the key developmental tasks in the life-review process is the “expiation of guilt, the resolution of intrapsychic conflicts, and the resolution of family relationships” (Butler and Lewis 1982:326). As Tomer (1994) suggests, older people who have been unable to resolve these critical challenges may grow increasingly more anxious as death draws near. If the alleviation of guilt and the resolution of longstanding interpersonal tensions are critical factors in late-life development, it follows that forgiveness by God, as well as forgiveness of others, may play a key role in this process by lowering levels of death anxiety.

Approaches to Forgiving Others

Requiring Acts of Contrition

As noted earlier, research indicates that some people are willing to forgive others, but only after transgressors perform one or more acts of contrition. Although researchers have identified a range of specific acts of contrition, one useful scheme was developed by Frankel (1998). Based on her study of the Talmud, she argues that in order to be forgiven, transgressors must: (1) be aware of and reflect upon their wrongdoing; (2) make an explicit admission of regret or remorse; (3) make a resolution not to repeat the offense; and (4) make restitution.

There are several reasons why requiring transgressors to perform acts of contrition may enhance the psychological well-being of the victim. Going through steps similar to those identified by Frankel (1998) ensures that the hurt is acknowledged explicitly. This is important because sometimes a transgressor will commit a hurtful act without even being aware of what he or she has done. In addition, performing acts of contrition provides a set of procedures that allow both the transgressor and the victim to openly express their negative feelings and work toward a potentially constructive resolution of their difficulties. Finally, performing acts of contrition may give the victim assurance that the offense will not be repeated in the future. This is important because, as Baumeister, Exline, and Sommer (1998) maintain, victims may feel vulnerable to the same offense in the future if they let transgressors off the hook without requiring them to pay in some way for what they have done.

Instead of being helpful, however, it is also possible to argue that requiring transgressors to perform acts of contrition may have a deleterious effect on the psychological well-being of the victim. More specifically, expecting the transgressor to make amends may set in motion a vicious cycle of recrimination and revenge because transgressors may not perform the anticipated acts of contrition, or they may not perform them in a way that is satisfactory to the victim. For example, restitution might involve financial compensation for damage caused by a transgression. But if the transgressor fails to pay, or does not pay an amount that is acceptable to the victim, then a second hurtful act becomes joined with the initial offense, thereby adding to the bitterness and resentment that is already present.

Requiring others to perform acts of contrition may also exert a deleterious effect on well-being because there is some evidence that these expectations may be driven by ulterior motives. This point is demonstrated forcefully in a series of qualitative studies that were conducted as part of our research program on religion in late life (Krause 2002; Krause and Ingersoll-Dayton 2001). The purpose of this qualitative work was to learn more about how older adults practice various aspects of religion, including forgiveness, in the course of daily life. In addition, this qualitative work was designed to develop quantitative closed-ended questions about religion, including some of the measures used in the present study. This approach is noteworthy because it appears to be the first time that both qualitative and quantitative methods have been simultaneously brought to bear on the study of forgiveness. In response to a question about acts of contrition, one subject in our qualitative research stated that “if you have to bare your soul to somebody, they have more than a solution in mind. They want the person to suffer” (Krause and Ingersoll-Dayton 2001:262). As this study participant went on to discuss, making transgressors suffer does not bring peace of mind, solace, or a sense of well-being to the victim.

Forgiving Unconditionally

In direct opposition to the steps outlined by Frankel (1998), research by Scobie and Scobie (1998) suggests that many people instead follow what they call the “Christian Model” of forgiveness. More specifically, this model is based on the belief that people should forgive as God does—unconditionally—requiring neither compensation nor even a promise to avoid repeating the act in the future. Although it is evident that all Christians do not endorse this approach (Krause and Ingersoll-Dayton 2001; Marty 1998), there is some evidence that many do because it is a central tenet of the Christian faith (Rye et al. 2001).

Perhaps the Christian Model identified by Scobie and Scobie (1998) enhances psychological well-being because it allows the victim to let go of the hurt and resentment associated with a transgression, thereby avoiding dark ruminations that are sometimes fostered by egregious offenses. This is important because research indicates that chronic rumination (e.g., reliving a hurt again and again) may be associated with greater psychological distress (Roberts, Gilboa, and Gotlib 1998). In addition, the Christian Model may avoid problems that are likely to arise when victims seek retribution. As Murphy (1997) points out, retribution should be based on the level of suffering that a transgressor deserves to experience because of what he or she has done. But he goes on to argue that seeking to extract retribution is risky because it is often difficult to determine with any precision what a transgressor actually deserves. If the anticipated punishment is too strict, the transgressor may feel justified in abandoning all efforts at making amends and may instead engage in further hurtful acts. However, if the punishment is too lax, the victim may continue to experience feelings of resentment and may also feel more vulnerable to further offenses.

Although forgiving others unconditionally would seem appealing because it allows victims to get on with the more positive aspects of life, some investigators have expressed reservations about whether this lofty goal can actually be attained. More specifically, Krause and Ingersoll-Dayton (2001) question whether it is truly possible to forgive others through sheer acts of will alone. Instead, forgiving unconditionally may lead to what Baumeister, Exline, and Sommer (1998) call “hollow forgiveness.” This means that victims may merely say they forgive a transgressor right away, but privately continue to harbor deep resentment and anger.

As the discussion provided up to this point reveals, there are benefits as well as disadvantages associated with each way that people may go about forgiving others. Consequently, it is important to determine whether forgiving others unconditionally or requiring transgressors to perform acts of contrition best enhances feelings of well-being in late life. Although addressing this issue is important from a theoretical point of view, it also has a potentially important bearing on how to design and evaluate therapeutic programs that encourage people to forgive others.

Factors that Encourage Elders to Require Acts of Contrition

If requiring transgressors to perform acts of contrition is related to psychological well-being, then it is important to identify the factors that make some victims more inclined to require transgressors to earn forgiveness in this manner. Fortunately, some insight into this issue may be obtained by returning to our qualitative work (Krause and Ingersoll-Dayton 2001). Analysis of these in-depth interviews revealed that a number of subjects felt quite strongly that if God forgives them for their own transgressions, then they were duty-bound to forgive other people unconditionally when these significant others commit a hurtful act. These strongly held beliefs are evident in the following excerpts from our in-depth interviews. The first comes from an older African-American man, who indicated that “He forgives you for all things, regardless … and if He forgives, I most certainly can forgive (others).” This man went on to point out that not only must he forgive others, he must do so unconditionally. The implications of not doing so were dramatically expressed by an older white man, who indicated that because God had forgiven him, his failure to forgive another right away would be “almost an unpardonable sin.”

Based on these insights, as well as similar comments from a number of other respondents in this qualitative study, a set of supplementary analyses are performed below to see if older people who feel they are forgiven by God are less likely to require transgressors to perform acts of contrition. These analyses are noteworthy because they point to a fundamental issue that has not been examined in the literature. More specifically, these analyses may reveal that there are potentially important linkages among the different dimensions or sources of forgiveness.

Methods

Sample

The data for this study come from a nationwide survey of older whites and older African Americans. The study population is defined as all household residents who were either African American or white, noninstitutionalized, English-speaking, and 66 years of age or older. 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 was restricted to individuals who were currently practicing Christians, people who were Christians in the past but no longer practice any religion, and individuals who were not affiliated with any faith at any point in their lifetime. People who practice a religion other than Christianity (e.g., Jews or Muslims) were excluded because it would be difficult to devise a set of religion measures and construct conceptual models that are suitable for persons of all faiths.

The sampling frame consisted of all eligible persons contained in the Health Care Finance Administration (HCFA) Medicare Beneficiary Eligibility List (HCFA is now called the Center for Medicare and Medicaid Services). This list contains the name, sex, and race of virtually every older person in the United States. It should be emphasized that people are included in this list even if they are not receiving Social Security benefits. Nevertheless, some older people are not included in this database because they do not have a Social Security number. This may be due to a number of factors, including illegal immigration.

A five-step process was used to draw the sample. First, once each year researchers at HCFA draw a 5 percent sample of names from their master file using a simple random-sampling procedure. The sampled names include all individuals who are 65 years of age or older. However, by the time the field period for the present study began, subjects in the 5 percent file were at least 66 years of age. It is for this reason that the study population was defined earlier as including individuals who were 66 years of age or older. In the second step of the sampling procedure used in this study, the 5 percent file was split into two subfiles—one containing older whites and the other containing older African Americans. Each file was sorted by county, and then by zip codes within each county. In the third step, basic groundwork was laid for selecting individual cases in each subfile. This was accomplished in the following manner. The target sample size for this study was set at 750 older African Americans and 750 older whites. It was determined that these target numbers could be achieved more efficiently by creating 75 clusters of subjects, and aiming for 10 completed interviews per cluster. The basic groundwork involved determining the total number of cases in each subfile, and calculating a selection interval (i.e., an nth interval) to identify individual cases that would be used as the initial starting point for forming the clusters. Following a random start, 75 nth selections were made in each subfile. In the fourth step of the sampling strategy, complete clusters (i.e., primary sampling units or PSUs) were formed by selecting approximately 25 additional names above and 25 additional names below each case identified in Step 3. Finally, in the last stage, sampled persons within each PSU were recruited for an interview.

Interviewing began in March 2001, and concluded in August 2001. The data collection was performed by Louis Harris and Associates (now Harris Interactive). A total of 1,500 were completed. Older African Americans were oversampled so that sufficient statistical power would be available to fully explore race differences in religion. As a result, the sample consisted of 748 older whites and 752 older African Americans. The overall response rate for the study was 62 percent.

After using listwise deletion of missing values to deal with item nonresponse, and after taking question skip patterns into account, the number of cases used in the analyses presented below ranged from 1,316 to 1,187. Based on the group consisting of 1,316 cases, preliminary analysis revealed that 51 percent were older whites and 49 percent were older African Americans. The average age of these individuals was 74.5 years (SD = 6.4 years). Approximately 42 percent were older men, and 50 percent were married at the time of the interview. Finally, these older adults indicated that they successfully completed an average of 11.5 years of schooling (SD = 3.4 years). These descriptive statistics, as well as the results presented below, are based on weighted data.

Measures

Table 1 contains the survey items that are used in this study. The procedures used to code these indicators are provided in the footnotes of this table. Before turning to the specific measures, a word is in order about how these questions were developed. Before this nationwide survey was conducted, three years were spent conducting a series of focus groups, in-depth interviews, and cognitive interviews with a separate group of 399 older whites and older African Americans (see Krause (2002) for a detailed discussion of this item-development strategy). The goal of this intensive item-development program was to craft a set of religion measures that capture the way older whites and older African Americans experience religion in daily life. Moreover, the intent was to maximize the salience of the items by using the respondents’ own words to write the question stems.

TABLE 1.

STUDY MEASURES

Forgiveness of Othersa How often do you feel resentful toward others for things they have done? How often do you hold a grudge? How often do you forgive others for things they have done? Forgiveness by Godb I believe God has forgiven me for the things I’ve done wrong. Acts of Contritionc Before I can forgive others, they must apologize to me for the things they have done. Before I can forgive others, they must promise not to do the same thing again. Before I can forgive others, they must repay me or compensate me for what they have done. Depressed Affectd I felt like I could not shake off the blues even with the help of my family and friends. I felt depressed. I felt sad. Somatic Symptomsd I did not feel like eating, my appetite was poor. I felt like everything I did was an effort. My sleep was restless. I could not get going. Life Satisfaction As I look back on my life, I am fairly well satisfied.c I would not change the past even if I could.c Now please think about your life as a whole. How satisfied are you with it?e Death Anxietyc I find it hard to face up to the fact that I will die. Thinking about death makes me feel uneasy. I do not feel prepared to face my own death. Church Attendancef How often do you attend religious services? Private Prayerg How often do you pray by yourself?

^aThese items are scored in the following manner (coding in parentheses): never (1); once in a while (2); fairly often (3); very often (4).

^bThese items are scored in the following manner: strongly disagree (1); disagree (2); not sure (3); agree (4); strongly agree (5).

^cThese items are scored in the following manner: strongly disagree (1); disagree (2); agree (3); strongly agree (4).

^dThese items are scored in the following manner: rarely or none of the time (1); some or a little of the time (2); occasionally or a moderate amount of the time (3); most or all of the time (4).

^eThis item is scored in the following manner: not satisfied at all (1); not very satisfied (2); somewhat satisfied (3); very satisfied (4); completely satisfied (5).

^fThis item is scored in the following manner: never (1); less than once a year (2); about once or twice a year (3); several times a year (4); about once or twice a month (5); 2–3 times a month (6); nearly every week (7); every week (8); several times a week (9).

^gThis item is scored in the following manner: never (1); less than once a month (2); once a month (3); a few times a month (4); once a week (5); a few times a week (6); once a day (7); several times a day (8).

Forgiveness of Others

As shown in Table 1, forgiveness of others is measured with three items. These indicators assess whether older study participants feel resentful toward others, hold a grudge, and whether they are able to forgive people for things they have done. The measures assessing forgiveness of others are coded so that a high score indicates that an older respondent is more forgiving. The internal consistency reliability estimate for this brief composite measure is 0.63.

Forgiveness by God

Forgiveness by God is measured with a single item. This indicator is coded so that a high score means that older study participants are more likely to believe that God has forgiven them for things they have done.

Acts of Contrition

If respondents indicated they forgive other people at least once in a while, they were asked whether they first required these others to perform three acts of contrition. Put another way, the contrition items were not administered to 35 subjects because they indicated they never forgive other people. As shown in Table 1, these acts of contrition included offering an apology, promising not to do the same thing again, and repaying or compensating the respondent in some way. These indicators are coded so that a high score means a study subject felt that a transgressor must earn forgiveness by performing more acts of contrition. The reliability estimate for this short scale is 0.87.

Depressive Symptoms

Seven indicators were taken from the Center for Epidemiologic Studies Depression Scale (CES-D) to assess depressive symptoms (Radloff 1977). A confirmatory factor analysis of these measures (not shown here) revealed that these indicators reflect two underlying factors or dimensions. The first, which is referred to as a depressed affect, assesses the cognitive-affective aspects of depressive symptoms, including feeling sad, blue, and depressed. The second factor (somatic symptoms) captures physiological manifestations of psychological distress, including difficulty sleeping, having a poor appetite, and having little energy. The results of this confirmatory factor analysis are consistent with findings from other studies that have administered the CES-D Scale to samples of older people (e.g., Hertzog et al. 1990). Two separate outcome measures were created based on these results. A high score on either measure denotes greater depressive symptomatology. The reliability of the depressed affect measure is 0.84; the reliability estimate for the brief composite assessing somatic symptoms of depression is 0.77.

Life Satisfaction

To evaluate the positive aspects of mental health, a brief three-item measure of life satisfaction was administered to the study subjects. Life satisfaction refers to an assessment of the overall conditions of life that is based on comparing one’s aspirations to one’s achievements (George 1981). Simply put, it is a cognitive evaluation of progress toward desired goals in life. The first two indicators used to assess this conceptual domain come from the Life Satisfaction Index A (Neugarten, Havighurst, and Tobin 1961). The third-listed item assesses satisfaction with life as a whole. A high score on these measures means that study participants feel more satisfied with the way their lives have turned out. The internal consistency reliability estimate for this short scale is 0.72.

Death Anxiety

Three items assessing anxiety about death were taken from scales that are in the literature (see Neimeyer (1994) for a list of these scales). These indicators are coded so that a high score reflects more anxiety about dying. The reliability estimate for this scale is 0.82.

Religious Control Measures

As a number of investigators have observed, religion is a complex, multidimensional phenomenon (Ellison and Levin 1998). Moreover, a number of the dimensions subsumed within this broad conceptual domain are correlated fairly highly. Consequently, two additional religion measures are included in the analyses presented below to ensure that the findings reflect the potential effects of forgiveness, per se, and not some other component of religion that is correlated with it. As shown in Table 1, the first assesses the frequency of church attendance, while the second measures the frequency of private prayer. These indicators are coded so that a high score reflects more frequent church attendance or more frequent private prayer.

Demographic Control Measures

The relationships among the measures of religion, forgiveness, and psychological well-being were evaluated after the effects of age, sex, race, education, and marital status were controlled statistically. Age is scored in a continuous format. Similarly, education is coded in a continuous format reflecting the total number of years of completed schooling. Sex is a binary variable contrasting older men (scored 1) with older women (scored 0), while race is also a binary indicator contrasting older whites (scored 1) with older African Americans (scored 0). Finally, marital status is coded as a binary variable where a score of 1 denotes older adults who are currently married and a score of 0 stands for study participants who were not married at the time of the interview.

Results

The findings from this study are presented below in three sections. The relationship between forgiveness of others, forgiveness by God, and the measures of psychological well-being are examined first. Following this, the relationship between acts of contrition and the well-being outcomes are explored. Finally, the relationship between forgiveness by God and the tendency to require others to perform acts of contrition is reviewed. Except where noted otherwise, ordinary least squares multiple regression analyses are used to explore these relationships.

Forgiveness of Others, Forgiveness by God, and Well-Being

Table 2 contains the findings from the analyses that were designed to examine the relationships among forgiveness by God, forgiving others, and well-being in late life. Consistent with the results reported by Toussaint et al. (2001), these data suggest that the relationship between forgiving others and psychological well-being is stronger than the relationship between forgiveness by God and psychological well-being. More specifically, the findings reveal that older people who forgive others report they experience fewer symptoms associated with a depressed affect than older people who are unable or unwilling to forgive other people for things they have done (Beta = −0.180; b = −0.231; p < 0.001). In contrast, forgiveness by God is also associated with fewer depressed affect symptoms, but the magnitude of this relationship is much smaller (Beta = −0.078; b = −0.217; p < 0.01). Similarly, the data in Table 2 suggest that forgiving others is associated with fewer somatic symptoms of depression (Beta = −0.181; b = −0.310; p < 0.001), while forgiveness by God fails to exert a statistically significant effect on this outcome measure (Beta = −0.016; b = −0.058; n.s.).

TABLE 2.

THE RELATIONSHIP BETWEEN FORGIVENESS BY GOD, FORGIVENESS OF OTHERS, AND PSYCHOLOGICAL WELL-BEING

Independent Variables	Dependent Variables
	Depressed Affect (N = 1,303)	Somatic Symptoms (N = 1,316)	Life Satisfaction (N = 1,263)	Death Anxiety (N = 1,213)
Age	−0.007a (−0.002)b	0.012 (0.005)	0.007 (0.002)	−0.034 (−0.010)
Sex	−0.050 (−0.215)	−0.055 (−0.313)	−0.016 (−0.059)	0.023 (0.084)
Education	−0.091*** (−0.060)	−0.078** (−0.065)	0.005 (0.003)	−0.060* (−0.033)
Marital status	−0.083** (−0.057)	−0.100*** (−0.135)	0.097*** (0.359)	−0.028 (−0.138)
Race	−0.018 (−0.076)	−0.024 (−0.135)	0.010 (0.037)	−0.038 (−0.138)
Private prayer	0.042 (0.049)	0.094** (0.148)	0.072* (0.074)	−0.004 (−0.004)
Church attendance	−0.121*** (−0.096)	−0.166*** (−0.177)	0.078* (0.054)	−0.073* (−0.050)
Forgiveness of others	−0.180*** (−0.231)	−0.181*** (−0.310)	0.217*** (0.242)	−0.194*** (−0.215)
Forgiveness by God	−0.078** (−0.217)	−0.016 (−0.058)	0.070* (0.171)	−0.055 (−0.130)
Multiple R2	0.079	0.086	0.096	0.061

^aStandardized regression coefficient.

^bMetric (unstandardized) regression coefficient.

^*p < 0.05;

^**p < 0.01;

^***p < 0.001.

In addition to depressive symptoms, the findings indicate that forgiveness also appears to be associated with life satisfaction and death anxiety. More specifically, as the data in the third column of Table 2 reveal, older people who forgive others tend to be more satisfied with their lives (Beta = 0.217; b = 0.242; p < 0.001). In contrast, forgiveness by God is only weakly associated with life satisfaction (Beta = 0.070; b = 0.171; p < 0.05). Finally, as hypothesized, the data in the far right column of Table 2 suggest that older people who are able to forgive others report they are less anxious about dying than older people who do not forgive other people (Beta = −0.194; b = −0.215; p < 0.001). Surprisingly, forgiveness by God appears to be unrelated to death anxiety (Beta = −0.055; b = −0.130; n.s.).¹

Automatic Forgiveness Versus Acts of Contrition

A major contribution of this study arises from the fact that we move beyond evaluating whether people forgive others to assessing variations in how they go about doing so. As the data in Table 3 reveal, this is a potentially important step. Viewed broadly, the findings indicate that requiring transgressors to perform acts of contrition is consistently associated with greater psychological distress and diminished feelings of well-being.

TABLE 3.

THE RELATIONSHIP BETWEEN ACTS OF CONTRITION AND PSYCHOLOGICAL WELL-BEING

Independent Variables	Dependent Variables
	Depressed Affect (N = 1,256)	Somatic Symptoms (N = 1,269)	Life Satisfaction (N = 1,219)	Death Anxiety (N = 1,187)
Age	−0.036a (−0.012)	−0.013 (−0.006)	0.032 (0.009)	−0.049 (−0.014)
Sex	−0.031 (−0.129)	−0.051 (−0.288)	−0.037 (−0.136)	0.022 (0.081)
Education	−0.053 (−0.033)	−0.074* (−0.061)	−0.043 (−0.023)	−0.028 (−0.015)
Marital status	−0.072* (−0.300)	−0.083*** (−0.468)	0.080*** (0.293)	−0.003 (−0.010)
Race	−0.013 (−0.053)	−0.005 (−0.029)	−0.014 (−0.050)	0.017 (0.063)
Private prayer	0.015 (0.017)	0.060 (0.092)	0.097*** (0.096)	0.009 (0.009)
Church attendance	−0.129*** (−0.101)	−0.175*** (−0.184)	0.105 (0.072)	−0.084*** (−0.057)
Acts of contrition	0.130*** (0.150)	0.071*** (0.110)	−0.160*** (−0.160)	0.276*** (0.281)
Multiple R2	0.053	0.059	0.073	0.094

^aStandardized regression coefficient.

^bMetric (unstandardized) regression coefficient.

^*p < 0.05;

^**p < 0.01;

^***p < 0.001.

The data in the first column of Table 3 suggest that older people who expect transgressors to perform acts of contrition have higher depressed affect scores (Beta = 0.130; b = 0.150; p < 0.001). Similarly, the findings indicate that requiring acts of contrition is also associated with more somatic symptoms of depression, but the relationship is not as strong (Beta = 0.071; b = 0.110; p < 0.01). The data in Table 3 further reveal that levels of life satisfaction are lower among older people who require transgressors to earn their forgiveness by performing acts of contrition (Beta = −0.160; b = −0.160; p < 0.001). Finally, the results suggest that older people who expect transgressors to perform acts of contrition are more anxious about dying than older people who are less inclined to require transgressors to earn forgiveness in this manner (Beta = 0.276; b = 0.281; p < 0.001).²

Factors that Encourage Acts of Contrition

The findings that have been discussed up to this point create the impression that forgiveness by God plays a relatively minor role in the process of forgiveness. However, the results presented in this section provide a different view. As Pargament (1997) and others have argued, a number of religious factors may be involved in shaping a process like forgiveness, with some occupying a more proximal role while others exert a more distal influence. Cast within this context, forgiveness by God appears to play a more distal role in the process of forgiveness. However, care must be taken to make sure the best data analytic procedures are used to evaluate the potentially important effects of forgiveness by God. Consequently, before turning to the empirical findings, it is important to briefly review the data analytic strategy that was used to examine the relationship between forgiveness by God and requiring transgressors to perform acts of contrition.

Based on findings from our qualitative studies, we hypothesized that people are more likely to forgive others unconditionally if they believe God has forgiven them for their own transgressions (Krause and Ingersoll-Dayton 2001). If people forgive others unconditionally, they do not require them to perform any acts of contrition. The measure used in this study to assess acts of contrition is continuous. If the hypothesis we propose is correct, then forgiveness by God should be most evident when those with the lowest contrition scores (i.e., those who feel strongly that acts of contrition are unnecessary) are contrasted with respondents who feel it is appropriate to have transgressors perform one or more acts of contrition. Consistent with this rationale, a binary outcome measure was created by contrasting those who feel strongly that acts of contrition are not necessary (scored 1) with study participants who believe that transgressors must earn forgiveness by performing acts of contrition (scored 0). A total of 262 study subjects (i.e., 20 percent of the sample used in this phase of the analysis) indicated that they do not require transgressors to perform any acts of contrition. Then, using logistic regression, this binary outcome was regressed on forgiveness by God and the other control measures that have been used throughout this study.

The results of these logistic regression analyses are presented in Table 4. Three coefficients are provided for each independent variable so that the relationships may be described more fully. In addition to the unstandardized logistic regression coefficients (b) and odds ratios, Table 4 also contains standardized logistic regression coefficients (Beta). Before turning to the substantive results, it is important to briefly examine why standardized estimates are important and how they are computed.

TABLE 4.

THE RELATIONSHIP BETWEEN FORGIVENESS BY GOD AND ACTS OF CONTRITION (N = 1,283)

Independent Variable	ba	Betab	Odds Ratio
Age	−0.006	−0.039	0.994
Sex	−0.126	−0.062	0.882
Education	0.013	0.044	1.013
Marital status	0.118	0.059	1.125
Race	0.252	0.126	1.286
Private prayer	0.036	0.065	1.036
Church attendance	0.073*	0.194	1.076
Forgiveness by God	0.949**	0.715	2.582

^aMetric (unstandardized) logistic regression coefficient.

^bStandardized logistic regression coefficient computed by multiplying the unstandardized logistic regression coefficient by the standard deviation of the independent variable.

^*p < 0.05;

^***p < 0.001.

One problem with logistic regression analysis is that it is difficult to determine whether the impact of one independent variable is greater than another. Comparing the odds ratios does not help because, as Selvin (1991) points out, these coefficients are influenced by the metric of the independent variable. Fortunately, Selvin (1991) provides a simple way of rank ordering (i.e., standardizing) the relative effects of the independent variables in logistic regression. This procedure involves multiplying an unstandardized logistic regression coefficient by the standard deviation of the independent variable. Cast within the context of the present study, these standardized estimates reflect the change in log-odds of forgiving others right away for a one standard deviation increase in the independent variable.

The data in Table 4 reveal that older people who feel they are forgiven by God are approximately two and a half times more likely to feel that transgressors should be forgiven unconditionally than older people who do not feel they are forgiven by God (odds ratio = 2.582). Moreover, as a comparison of the standardized coefficients reveals, the relationship between forgiveness by God and the odds of forgiving right away (Beta = 0.715; b = 0.949; p < 0.001) is substantially larger than the corresponding effects of any other independent variable in the equation, including church attendance (Beta = 0.194; b = 0.073; p < 0.05) and the frequency of private prayer (Beta = 0.065; b = 0.036; n.s.).³

Discussion

Interpersonal problems are an integral facet of social life (Rook 1984). At some point, everyone hurts those who are close to them, even if they do not intend to do so. Yet social relationships appear to persevere in the face of these seemingly inevitable challenges. We need to learn more about why this may be so. Although a number of factors are likely to be at work, the study of forgiveness appears to be especially promising because it deals directly with these issues. More specifically, forgiving others provides a way to go about repairing the damage created by egregious interpersonal acts. Similarly, obtaining forgiveness from God alleviates feelings of guilt arising from one’s own conduct, thereby helping people get on with the more positive aspects of their lives. Yet we know so little about how the potentially important benefits of forgiveness arise even though it is promoted by virtually every major religion in the world (Rye et al. 2001). The purpose of the present study was to probe more deeply into this fundamental domain by examining several issues involving forgiveness by God and forgiveness of others that have been largely overlooked in the literature.

Consistent with the work of other investigators, the findings from our nationwide survey of older people reveal that older adults who forgive others tend to enjoy a greater sense of psychological well-being than older people who are less willing to forgive transgressors for the things they have done. Although forgiveness by God is also associated with greater feelings of well-being, the magnitude of the relationships are not as strong as those observed with forgiving others. Similar findings have been reported by other investigators (e.g., Toussaint et al. 2001), but it is surprising to find that few researchers provide a rationale for why this may be so. The two types of forgiveness measures may have different effects on well-being simply because they assess different phenomena. More specifically, an older person may seek forgiveness from God because he or she has done something wrong and feels guilt or remorse for the transgression. If forgiveness from God is forthcoming, the guilt and remorse are alleviated. In contrast, forgiveness of others involves having wrong done to oneself. When this happens, a different set of challenges must be confronted. Rather than grappling with guilt and remorse, the focal elder must deal with feelings of resentment, bitterness, and even hatred toward the transgressor. When the two types of forgiveness are compared and contrasted, the data suggest that the benefits associated with forgiving others appear to outweigh the benefits arising from being forgiven by God. It is important to point out, however, that this conclusion pertains only to the outcome measures used in this study. Additional benefits of being forgiven by God may emerge when a different set of dependent variables are examined.

Instead of merely replicating the work of other investigators by contrasting the effects of forgiveness by God with the influence of forgiveness of others, the findings from the present study add to the literature in three potentially important ways. First, we examined a facet of psychological distress (death anxiety) that has not been explored previously in studies of forgiveness. As subsequent data analyses revealed, feelings of death anxiety appear to be significantly greater among older people who are unwilling to forgive others. Viewed more broadly, these results suggest that it is important to consider death anxiety when conducting studies of forgiveness because some of the more substantial effects to emerge from our data involved this outcome measure.

Second, the findings from this study are noteworthy because we compared and contrasted the effects of different ways of forgiving others. The data indicate that requiring transgressors to perform acts of contrition is associated with diminished feelings of psychological well-being. This suggests that merely knowing whether a person forgives another is not sufficient and that instead of enhancing well-being, some ways of granting forgiveness may actually have detrimental effects. This appears to be the first time the relationship between performing acts of contrition and well-being has been examined in the literature.

Third, an effort was made in the present study to delve more deeply into the process of forgiveness by exploring a factor that may encourage older people to forgive others unconditionally, and to reject the notion that transgressors must earn forgiveness by performing acts of contrition. More specifically, the findings suggest that older adults who believe they are forgiven by God are more likely to forgive others right away than are older people who do not believe God has forgiven them for their own transgressions. These results are important because they suggest there may be theoretically meaningful linkages among the different sources of forgiveness, and that instead of merely controlling for the effects of forgiveness by God when examining the effects of forgiving others, researchers need to explore how forgiveness by God may influence the way older people go about forgiving each other.

Even though the findings from this study may contribute to the knowledge base on forgiveness and well-being, a considerable amount of work remains to be done. For example, we need to know more about why requiring transgressors to perform acts of contrition may erode feelings of psychological well-being in late life. Earlier, we speculated that requiring transgressors to earn forgiveness may set in motion endless cycles of recrimination and revenge, but these secondary interpersonal problems were not assessed in this study. An important priority for the future is to do so. In the process, it may be useful to see whether acts of contrition impede the ability of victim and transgressor to arrive at a resolution or reconciliation of their problems. Although achieving a sense of reconciliation is a challenging task (McCullough, Pargament, and Thoresen 2001), devising good measures of this elusive construct may shed valuable insight into the nature and implications of forgiving other people. Finally, as noted earlier, official church doctrine advocates forgiving others and seeking forgiveness from God. Yet, we know relatively little about how these theological issues are brought into practice in daily life. Two intriguing leads are provided in the literature. First, a recent study by Wuthnow (2000) suggests that small formal groups in the church, such as prayer groups and Bible study groups, may promote the forgiveness of others. Second, research indicates that the general psychosocial climate of the congregation may have an important influence on the thought and behavior of church members (Pargament 1983). Perhaps specific aspects of the congregational climate, such as the overall level of expressiveness, empathy, and social concern among church members, may contribute to a parishioner’s willingness to forgive. Because the general psychosocial climate of the church is likely to affect the way that prayer groups and Bible study groups are run, comparing and contrasting these two institutional influences may provide valuable insight into the factors that encourage people to be more forgiving of others.

In the process of exploring these, as well as other, issues involving forgiveness, researchers would be well advised to address the limitations in the research we have conducted. Two shortcomings are examined briefly below.

The data for this study were gathered at one point in time only. As a result, the temporal ordering among the constructs in regression equations was based on theoretical considerations alone. This creates potential problems because it is possible to reverse the causal ordering and argue, for example, that older people who are initially more depressed are subsequently less likely to believe that God has forgiven them for things they have done. This, as well as other relationships involving forgiveness and psychological well-being, needs to be examined with data that have been gathered at three or more points in time.

The measures of forgiveness used in this study treat forgiveness much like a personality trait because respondents were asked, for example, whether they usually or typically forgive others for things they have done. However, as McCullough, Hoyt, and Rachal (2001) point out, it is likely that people forgive some individuals, but not everyone. More specifically, people may be less willing to forgive specific transgressors because they have a history of committing a number of different offensive acts. In addition to this, the decision to forgive someone may depend on the nature of the specific offense. Some transgressions may be so egregious that they are never forgiven, while others may be less problematic and, therefore, more likely to be forgiven. This means that the trait-like measures used in the present study may overlook rare, but potentially quite deleterious, transgressions, as well as instances where older people cannot forgive a repeat offender.

Although there are limitations in the present study, we hope the findings encourage researchers to explore the finer nuances of the process of forgiveness. Doing so should provide valuable insight into how people are able to carry on in the face of their own shortcomings as well as the shortcomings of others. Continuing with this line of research promises to show how religion has a direct impact on the daily lives of ordinary people, and provides one way of demonstrating how fundamental religious principles involving forgiveness may be integrated into therapeutic interventions aimed at alleviating mental health problems (Worthington and Wade 1999). We believe that few areas of inquiry have such far-reaching potential.