Lay Theories of Suicide: An Examination of Culturally Relevant Suicide Beliefs and Attributions Among African Americans and European Americans

Abstract

The purpose of this study was to examine African Americans’ lay beliefs and attributions toward suicide. The Attitudes Toward Suicide Scale, Life Ownership Orientation Questionnaire, Stigma Questionnaire, and Suicide Ideation Questionnaire were administered to 251 undergraduate college students. Beliefs about stigma associated with suicide were comparable across ethnic groups. However, African American college students were significantly less likely than European American college students were to attribute suicide to interpersonal problems and to report that the individual or government is responsible for life. African American students were significantly more likely to report that God is responsible for life. These findings have important implications for suicide risk and also for developing culturally appropriate interventions.

The low suicide rate among African Americans has been attributed in part to long-standing cultural values and beliefs that resist suicidal behavior as a problem-solving alternative (e.g., Gibbs, 1997). However, suicide risk, along with other public health problems, is typically understudied in African American populations, and empirical support for widely held assumptions is sparse. Challenges to understanding suicide in the African American community are compounded by limitations (e.g., false positives) that plague suicide science and prediction of high-risk behavior in the United States.¹ Despite widely held speculation that suicide among African Americans is buffered by cultural beliefs, empirical literature that addresses African Americans’ lay attributions for suicide is limited.

The purpose of this article is to explore African American lay beliefs about suicide. An interesting question is to ascertain whether these lay beliefs are culturally specific to African Americans (emic) or whether they represent a broader universal (etic) view of suicide beliefs regarding suicide. Given the potential for cultural differences, a comparative research framework provides an empirical test of constructs that are not likely to be universal. Azibo (1988) argued that the comparative research framework is good practice when groups are equated on a variable (e.g., suicide attributions), although the construct might be expressed differently across groups. Potentially divergent findings for suicide beliefs and attributions, variables that are of universal interest, underscore the need for culturally specific prevention efforts in the African American community.

AFRICAN AMERICAN SUICIDE: A PUBLIC HEALTH PROBLEM

Epidemiologists observed an increase in Black suicide deaths unparalleled by other ethnic groups in the United States from the 1970s through the 1990s (Centers for Disease Control and Prevention [CDC], 1998; Joe & Marcus, 2003; National Center for Injury Prevention and Control, 1995). African American college students as a subgroup have been said to attempt suicide more often than their White counterparts (CDC, 1997) but disclose suicidality less readily (Morrison & Downey, 2000). Suicide attempts (including those that require serious medical attention) among African American males continue to surpass attempts for both White males and White females in the United States (CDC, 2004).

These statistics draw strong reactions from the community as African Americans have been believed to be “protected” from suicide, and Black families have been said to immunize Black youth against suicidal behavior (see Gibbs, 1988). African American suicide rates have been relatively low despite overrepresentation of suicidogenic factors (e.g., marginalization, chronic social stressors, economic strain) in the African American community. Gibbs (1997) poignantly characterized the “paradox” of Black suicide in light of lesser suicide mortality despite a long history of disadvantages afforded to African American people. Some reports have indicated that African Americans aged 15 to 24 years may be more likely to report problems with physical health rather than emotional instability (Smith & Carter, 1986). Such dilemmas critically undermine both help-seeking behavior and meaningful efforts for prevention.

Stigma associated with mental illness has been identified as a significant barrier to help seeking in troubled African American people (Poussaint & Alexander, 2000). Some research suggests that ethnic minorities may hold stronger stigmatizing attitudes toward mental illness than do European Americans (U.S. Department of Health and Human Services, 2001). For example, Cooper-Patrick et al. (1999, p. 436) found that African American patients voiced more concern about stigma than did European Americans and noted that seeking help for mental health problems was not “culturally acceptable” among family and peers. Thompson, Bazile, and Akbar (2004) found that not only do African Americans consider stigma a salient barrier to seeking professional treatment for mental health concerns but also that they may hold more stigmatizing attitudes toward mental illness than other racial and ethnic groups. Among African Americans in particular, because therapy and other professional mental health resources are often associated with character weakness and being “crazy,” there is stigma associated with any kind of formal/professional treatment in this community (Brown, Abe-Kim, & Barrio, 2003; Franklin, 1992).

Although the stigma associated with suicidality may contribute to incalculable underreporting of suicide death in African American adults and youth, there are few studies that have specifically studied stigma as it relates to suicidal behaviors in this community. Despite the fact that experts suggest that suicide may be more stigmatized in the African American community than in other communities, a search of PsychInfo and Medline databases (April 27, 2005) for “suicide,” together with “stigma” and “African American” did not yield any studies. A modified search of “suicide,” “stigma,” and “Black” netted two studies. Of note, most studies have focused more broadly on stigma and “mental health,” as a PsychInfo search for these terms, along with “African American,” yielded 34 studies for 1996 through 2005. No studies existed prior to this time period. To date, investigations in African American suicide schema have been more prone to exploring religious beliefs and attitudes.

RELIGIOSITY AND ATTITUDES TOWARD SUICIDE

Some researchers argue that African American suicide rates are low because of cultural beliefs that include religious or spiritual deference, resilience to overwhelming strain, and stigmatization. Early and Akers’s (1993) interviews of 30 Black pastors revealed assumptions that suicide is outside “the Black experience” (p. 283) and perceived as a “denial of Black identity and culture” (p. 287). The Black experience was characterized by struggle, and Black people were believed to be more resilient than White people because of the history of economic, political, and social deprivation. The pastors were interviewed as “informants,” believed to have reliable access to the community given the traditionally pivotal role of the Black church. Although a community survey was not conducted, Early and Akers speculated that the pastors’ attitudes were representative of the larger community. Future research should substantiate these beliefs in larger African American community samples.

Studies show that as religiosity increases, attitudes toward suicide become less favorable. Anglin, Gabriel, and Kaslow (2005) found that suicide acceptability and religious well-being were associated with suicide-attempt status such that nonattempters were more likely than attempters to report religious well-being and less likely to endorse suicide as acceptable. Ellison and Smith (1991) conducted a study among college students that supported their hypothesis that an individual’s faith in God and this individual’s moral objection to suicide are strongly related. Studies from previous decades suggest a similar association between religiosity and attitudes toward suicide. Minear and Brush (1980), for example, conducted studies of college students using the Suicide Belief and Suicide Values Scales, which revealed that students with weak or no religious ties had more favorable and more accepting attitudes toward suicide both for others and themselves when compared to students with stronger religious ties. In addition to religious attachment, Minear and Brush found that church attendance was associated with attitudes toward suicide such that increased frequency of attendance was associated with decreased consideration of suicide as an option for self or others.

Studies of religion and suicide have demonstrated that religious coping, activities, and attitudes do not fully account for the low suicide death rates for African Americans. Stack (1998) found that the indicator of religiosity that significantly lowered suicide acceptability for both Black men and women was church attendance. However, Western residence for Black men and education level for Black women were more compelling predictors of suicide acceptability than were religious factors. Neeleman and colleagues (1998) found that although educational and social differences did contribute, less favorable attitudes toward suicide were mostly attributable to the comparatively high levels of orthodox religious belief and personal devotion. Results from these studies and the general lack of empirical research on specific racial and ethnic groups demonstrate the need for more research in this area. Definitive investigations that understand African Americans’ attitudes toward suicide are vital for effective prevention and intervention efforts.

LAY BELIEFS AS BUFFERS FOR AFRICAN AMERICAN SUICIDE

It is important that the African American community assists in defining suicide meaning for African American people (cf. Smith, 2004). Lay theories, those intrinsic or “common sense” beliefs held by lay persons, have been critically examined for addiction (Furnham, 1996) suicide (Knight, Furnham, & Lester, 2000), psychopathology (Furnham & Buck, 2003), and depression (Falcota, Nordt, & Rossler, 2003). Kluger and Tikochinsky (2001) argued that “although common sense beliefs are not scientific theories, they are likely to contain a valid kernel, which if discovered, could provide a generalization of facts to be explained by scientific theories” (p. 408). Knight et al. (2000) found that certain personality characteristics (e.g., psychoticism) were positively associated with a belief in the right to die by suicide and that suicide is normal. The authors concluded that because suicidal behavior may be affected by attitudes toward suicide, assessment of these attitudes may affect suicide-risk assessment. Although lay beliefs have been examined for suicide and also for cross-cultural group differences (Angermeyer & Matschinger, 1999), no study to our knowledge has systematically investigated African American attributions for suicide attempts or deaths. Even as studies continue to examine suicide attitudes in nonclinical populations, African Americans are frequently underrepresented (see, e.g., Gibb, Andover, & Beach, 2006).

The purpose of this study was to examine lay beliefs about suicide, its stigmatization, and cases when it is acceptable in addition to who the right to life is given (i.e., the individual, God, or state). In the present study, African American and European American college students were expected to endorse disparate attitudes with respect to suicide attributions and “ownership of life.” That is, beliefs that God (rather than the individual) regulates day-to-day life were expected to be endorsed more so among African Americans given the strong relationship between religious beliefs and reported objections toward suicide. We expected that African Americans and European Americans would also differ on suicide stigmatization.

METHOD

PARTICIPANTS

The participants were 251 undergraduate students enrolled in a large university in the southeastern United States. The students participated in this study to partially fulfill a requirement for an introductory psychology class or to gain some other academic credit. Mean age for the total sample was 21.00 years (SD = 3.95 years; see Table 1), with ages ranging from 18 to 53 years. The ethnic composition of the sample was 48% African American (n = 120) and 52% European American (n = 131). The majority of the sample (80%) was female (n = 201).

TABLE 1.

Demographic Characteristics of Participants

Characteristic	n	%	M	SD
Sex
Female	201	80%
Male	50	20%
Ethnicity
African American	120	48%
European American	131	52%
Age			21.00	3.95

MEASURES

Attitudes Toward Suicide Scale (ATTS)

Beliefs about causes for suicide were assessed via the ATTS (Lester & Bean, 1992; Knight et al., 2000), an 18-item self-report inventory. Each item was rated on a Likert-type scale ranging from 1 (strongly disagree) to 6 (strongly agree). Thus, possible inventory scores ranged from 18 to 108, whereby higher scores were indicative of positive responses in the direction of the construct being assessed. Higher scores reflect increased attribution toward certain causes for suicide. These causes are measured via the three subscales of the ATTS. The first subscale represents intrapsychic problems. An example item is, “People who commit suicide are usually mentally ill.” The second subscale, interpersonal causes, is reflected by the item “Suicide is usually an attempt to get empathy from others.” The dimension of suicide that is attributed to societal causes is represented by “Those who are oppressed in a society are more likely to commit suicide.” Although Lester and Bean (1992) did not report reliability data, they asserted that the ATTS is a reliable measure for defining dimensions of attitudes and attributions for suicide. In this sample, alpha = .72.

Stigma Questionnaire

Suicide stigma and prejudice was assessed via a modified 8-item measure of social distance (Kalish, 1966; Lester, 1988). Each item required a “yes” or “no” response and yielded possible total scores of 0 to 8, in which lower scores were indicative of more stigmatized beliefs. Participants were instructed to “answer questions for a person who has attempted suicide in the past year.” Questions included the following: “If you had met and liked this person, would you be willing to become friends with him/her?” and “Do you think that it is all right to permit this person to visit the United States for a 2-week holiday?” Thus, participants considered an unknown person’s suicidal status. Although reliability and validity estimates are not available for previous studies, in the current study, alpha = .74.

Life Ownership Orientation Questionnaire (LOOQ)

Beliefs about who has control over life events were assessed via the LOOQ (Ross & Kaplan, 1993–1994), a 21-item self-report instrument. The Likert-type scale items ranged from 1 (disagree) to 5 (agree) and yielded a total score range of 21 to 105. Ross and Kaplan (1993–1994) reported Cronbach alphas for three factors—God (alpha = 0.88), individual (alpha = 0.67), and state (alpha = 0.50). Example items are “Only God has the right to decide whether an individual should live or die,” “I feel secure when I rely on myself,” and “I feel secure when I live in a society that takes care of me,” respectively. The items are said to reflect decision making, security, responsibility, reliability, control, problem solving, and on what or whom to rely. The subscales were found to be more predictive than measures of religiosity (Ross & Kaplan, 1993–1994). Correlations with measures of other variables (e.g., attitudes toward suicide) were significant and provided construct validity such that high attributions toward God reflected opposition to abortion and suicide. Also, high individual attributions correlated with more support for abortion and suicide (Ross & Kaplan, 1993–1994). In this sample, alpha = .52.

Suicide Ideation Question (SIQ)

Suicidal ideation was measured via the SIQ (Reynolds, 1987), a 30-item self-report inventory for assessment of suicidal thoughts. Each item was rated on a Likert-type scale ranging from 0 (never had the thought) to 6 (having the thought almost every day), with a range of possible scores of 0 to 180. Higher scores were indicative of higher frequency of suicidal thoughts. Internal consistency has been estimated as .90 and higher (Reynolds, 1987; Ritter, 1990). Reynolds (1987) reported construct validity in relation to depression, hopelessness, and anxiety. In this sample, alpha reliability = .95.

PROCEDURE

The university’s Institutional Review Board approved the present study. Each participant was informed that a questionnaire packet would be administered that included questions about behavior, views, and feelings with regard to suicide in addition to a brief demographic form. Consent for participation in the study was assumed on completion of the anonymous questionnaire packet. Approximately 20 minutes were required to complete the questionnaire packet. Students were informed that participation in the study could cease at any time and referral to the university counseling center or psychology clinic for free services would be available if needed. None of the participants discontinued participation or requested referral. Participants were debriefed and also provided with a form that included the principle investigator’s phone and e-mail contact information. Participants were informed that the principle investigator is a licensed clinical psychologist available for follow-up or to answer questions as needed.

RESULTS

Intercorrelations between all measures are summarized in Table 2. Means and standard deviations for all measures are presented for African American and European American participants in Table 3. All values were within expected limits. As Table 2 shows for African American and European American college students, suicide stigma was positively correlated with beliefs about intrapsychic problems as a cause for suicide (r = .26, p < .01). It was also correlated with beliefs about God’s ownership of life (r = .13, p < .05) such that suicide was more stigmatized if the cause was believed to be attributed to psychological challenges and also if God was believed to be responsible for life. Stigma was negatively correlated with self-reported suicidal ideation (r = –.24, p < .01) such that the more that suicide was stigmatized, the less that suicidal thoughts were reported among African American and European American college students.

TABLE 2.

Intercorrelations for Life Ownership, Attribution, Stigma, Age, and Sex

Measure	1	2	3	4	5	6	7	8	9
1. LO_God	—
2. LO_indiv	–.27**	—
3. LO_state	–.52**	–.43**	—
4. Att_psych	.22**	.05	–.06	—
5. Att_pers	.09	.10	.07	.38**	—
6. Att_soc	.01	.14*	.10	.49**	.50**	—
7. Stigma	.13**	–.12	–.12	.26**	.08	.09	—
8. Suicide Ideation	–.27**	.16*	.24**	–.14*	–.01	.06	–.24**	—
9. Age	–.01	–.10	–.15*	–.05	–.12	–.11	.10	–.05	—
10. Sex	–.16*	.13*	.13	.03	.09	.02	–.12	.07	.00

NOTE: LO_God = life ownership—God; LO_indiv = life ownership—individual; LO_state = life ownership—state; Att_psych = attribution—intrapsychic; Att_pers = attribution—personal; Att_soc = attribution—societal. Sex is coded 0 = female, 1 = male.

^*p < .05.

^**p < .01.

TABLE 3.

Analysis of Variance (ANOVA) for Suicide Perception Measures

			Mean (SD)
Measure	df	F	Black	White
LO_ God	1,244	22.93**	28.16 (4.86)	23.70 (7.56)
LO_indiv	1,246	14.78**	21.65 (5.46)	24.22 (5.21)
LO_ state	1,244	21.11**	13.99 (3.93)	15.97 (3.50)
Att_psych	1,245	0.37	23.97 (4.48)	24.15 (3.91)
Att_pers	1,247	7.76**	20.97 (3.76)	22.20 (3.53)
Att_soc	1,242	2.00	20.77 (4.70)	21.43 (3.54)
Stigma	1,244	0.10	2.02 (1.99)	1.84 (1.75)

NOTE: F ratios were generated from the Hotelling’s statistic. LO_God = life ownership— God; LO_indiv = life ownership—individual; LO_state = life ownership—state; Att_psych = attribution—intrapsychic; Att_pers = attribution—personal; Att_soc = attribution—societal.

^**p < .01.

ETHNIC GROUP DIFFERENCES IN SUICIDE ATTRIBUTIONS

To analyze African American and European American patterns of lay suicide beliefs, life ownership, and suicide stigma, a two-way analysis of variance (ANOVA) was performed. In this analysis, ethnicity was the independent variable and scores for the Stigma Questionnaire in addition to subscale scores for the ATTS (i.e., interpersonal, intrapsychic problems, societal pressures) and LOOQ (i.e., God, individual, state) were dependent variables. The results of this analysis are reported in terms of an Hotelling’s Trace converted to an exact F statistic and detailed in Table 3. The ANOVA produced significant main effects for all subscales of the LOOQ as well as the interpersonal-conflicts subscale of the ATTS. Descriptive statistics revealed that, relative to European American college students, African American college students reported more beliefs that life ownership was attributed to God (Ms = 28.01 and 23.85, respectively). European Americans reported more so than African Americans that individuals (Ms = 24.07 and 21.33, respectively) and state or government (Ms = 15.88 and 13.67, respectively) are responsible for life.

DISCUSSION

As predicted, significant ethnic group differences emerged for lay attributions of who (i.e., God, individual, and state or government) controls life and what circumstances (e.g., intrapsychic, interpersonal, or societal difficulty) dictate when suicide might be attempted. African Americans were significantly less likely than European Americans to report that suicide is attributed to an interpersonal problem such that attempts are believed to be triggered by conflict, work stress, or experiencing a broken home. These findings are consistent with Early and Akers’s (1993) report of pastoral assertions that Black suicide may be low because African Americans experience a culture in which struggle is expected and endured.

African Americans were also more likely to attribute ownership of life to God, whereas European Americans indicated more so that the individual and state or government was responsible for life. These attitudes may reflect philosophical differences in worldview that potentially underlie African Americans’ seeming “protection” from suicide via religiosity and religious and spiritual well-being (e.g., Kaslow et al., 2004; Walker, Utsey, Bolden, & Williams, 2005). Future studies should continue to determine mechanisms and dimensions by which religiosity buffers risks for suicide in persons of African descent. As an example, Washington and Teague (2005) suggested that spirituality may be instrumental in developing “healthy drug attitudes” that rebuff the use of illicit substances. These attitudes would in turn be an important mediator for high-risk behaviors that include suicide.

African American lay beliefs that God (and not the individual) controls life and that suicide is not attributed to life stressors may account for the discrepant stress-suicide mortality rate for African Americans. Clark, Anderson, Clark, and Williams (1999) cited anxiety, helplessness-hopelessness, and anger among a myriad of responses that incite maladaptive coping (e.g., anger suppression, hostility, substance use, etc.) in African Americans. Utsey (2002) found that African Americans experienced more race-related stress than both Asian Americans and Latinos. Yet these difficulties do not appear to translate to high suicide-death rates. Future investigations of cultural beliefs should consider worldview perspectives that reinforce interpersonal resiliency (Obasi, Flores, & Myers, 2006) in the face of overwhelming strain.

Contrary to expectations, African Americans and European Americans did not differ in report of stigma associated with suicide. This sample of African American and European American college students reported comparable stigma associated with suicidal thoughts. This is surprising because several reports have indicated that suicide is more so stigmatized in the African American community, and this stigmatization may translate to underreporting and misreporting of suicides as “accidents” (see Joe & Kaplan, 2001; Kaslow et al., 2004). Suicide may be further disguised and thus underreported in cases of “victim-precipitated homicide” in which victims, in theory, instigate others toward their own fatality (Wolfgang, 1959). Future studies might explore suicide beliefs and attitudes in areas in which reported homicide rates are highest.

These results have several important implications for prevention and intervention efforts. That is, when African Americans are seriously suicidal in response to difficult life events, family or friends who believe that life’s difficulties do not warrant a suicidal act may not respond or intervene actively. Thus, when signs of suicide are presented (i.e., giving away valuables, talking about suicide), there is no response. This implication warrants critical attention particularly in educating the lay African American community in the realism of suicide risk and resiliency. Willis, Coombs, Drentea, and Cockerman (2003) found that African Americans were as likely as European Americans to exhibit suicide-warning signs. In addition, for those who are reluctant to disclose suicide-related thinking and/or planning in an environment in which the cognitive schema reject suicide, prevention is critically compromised. Of critical importance is the consensus that African Americans are significantly less likely than members of other ethnic groups to seek professional mental health treatment (Snowden, 1999; Young, Griffith, & Williams, 2003). Attention to this constellation of factors, taken together in considering suicide prevention efforts, offers considerable promise for reducing preventable suicide deaths and severe suicide attempts.

African American family and friends potentially promote varying non-traditional (see Young et al., 2003) and traditional treatment recommendations based on beliefs and attributions for suicide etiology. Angermeyer and Matschinger’s (1999) examination of sociopolitically distinct groups found that groups were divergent on causal attributions and also treatment recommendations for schizophrenia and depression. In African Americans, beliefs that God is responsible for life or that interpersonal problems do not impact suicidal behavior may impact suggestions and also compliance with traditional recommendations for psychotherapy.

Further investigation is needed to determine whether attitudes toward suicide increase vulnerability in clinical populations. An understanding of the influence of religion/spirituality on those attitudes, in addition to some identification for potential sources of support within the community to sustain such efforts, is needed. Jones (2003) posited that, for persons who internalize an African worldview, spirituality is the cornerstone of culture. One might ask: Are there fewer tendencies to seek help if African American individuals at risk for suicide believe that “God will take care of it”? If suicide is determined by God, then religious education is important. On the contrary, if mental health is important, then psychoeducation provides some utility. In either event, tailoring interventions to the needs of specific cultural groups is paramount (Goldsmith, Pellmar, Kleinman, & Bunney, 2002).

Although we provide empirical evidence for important, culturally relevant phenomena, this study is not without limitations. One limitation is the student sample. Even though suicide risk is important to examine in college samples, an investigation that broadly examines a representative community sample would be ideal and more generalizable. The sample was also largely female, which may limit its interpretability for males, although sex was minimally related to life ownership and not at all related to other variables. Also, reliability properties for the life-ownership measure were marginal. Despite the fact that the instrument generated significant results, future studies should investigate the psychometric properties for this instrument. As previously noted, future studies might also explore suicide attitudes in clinical samples to advance models for suicide prediction and the contribution of lay beliefs in suicide science.

In sum, there is a need for careful elucidation of African American lay beliefs about suicide. Although lay theories are not as complex as psychological and sociological theories of suicide, they potentially provide important insights to suicide, a poorly understood public health problem, particularly for African Americans. Culturally relevant models informed by African American lay beliefs will advance psychosocial models, prevention, and treatment efforts.