Men and Their Father Figures: Exploring Racial and Ethnic Differences in Mental Health Outcomes

Abstract

Though gender, racial, and ethnic disparities in health in the United States are well documented, it is less clear how these factors intersect to produce patterns of mental health outcomes among men. This study examined the presence of father figures in the lives of African American, Caribbean black and non-Hispanic white American males until the age of 16; assessed the current socio-demographic factors of these men as adults; and explored whether these factors lead to variations in mental health outcomes. Regression models were used to examine the correlates of socio-demographic, psychosocial, and retrospective father figure measures for depressive symptoms and non-specific psychological distress among African American (n = 999), Caribbean black (n = 506), and non-Hispanic white men (n = 193) from the National Survey of American Life. Findings revealed racial and ethnic group differences by age, employment status, education, and household income on depressive symptoms (measured using the CES-D scale) and non-specific psychological distress (using the Kessler-6 scale). Findings suggested that being raised by a grandfather placed both African American and Caribbean black men at greater risk for depressive symptoms and non-specific psychological distress under certain socio-demographic conditions. This study is unique in that it considers the influence of father figures on the mental health outcomes of adult males across three racial and ethnic groups. We conclude with a discussion of the implications for future mental health research and practice with men of color.

Introduction

Many health problems that are specific to men are modifiable because the causes are primarily social and behavioral, not biological (Featherstone et al. 2007). More important than any specific disease, men’s health focuses on the influence of the social construction of maleness, manhood, and masculinities (Lee and Owens 2002). Men’s health cannot be understood without an appreciation for men’s social and cultural contexts, though men’s health research traditionally has been decontextualized and removed from this context (Rieker and Bird 2005). One of the key yet understudied aspects of the social, cultural, and life course context of men is the role of other men, particularly father figures, in men’s health.

Men play a critical role in the health and development of children, particularly young men (Lu 2010), and positive father involvement has been associated with improved child outcomes (Caldwell et al. 2004; Levine and Pitt 1995; Zimmerman et al. 2000). Though biological fathers have been found to play critical roles in the lives of youth, research on African American youth has found that “father figures” may fulfill many of the same roles and responsibilities as biological fathers (Caldwell et al. 2011). These father figures may be members of the child’s extended family network or they may be augmented relationships that have significance for family functioning and adaptation (Caldwell et al. 2011). Regardless, the quality, not quantity, of time father figures spend with their children has been associated with young men’s health and improved well-being (Caldwell et al. 2004). Yet, it is unclear how father figures of young men may affect their mental health as adults.

Men, Socialization, and Mental Health

Studies have reported differential mental health patterns for African American, Caribbean black, and non-Hispanic white men. As such, Caribbean black men have been shown to have higher rates of mood and anxiety disorders when compared with African American men (Joe et al. 2006; Williams et al. 2007a, b), and rates of 12-month major depressive disorder for non-Hispanic white men (7.9%) are more than those for black men (4.0%) (Gavin et al. 2010). In this paper, we use the terms “black Americans” and “blacks” interchangeably to characterize a racial group in the United States that includes considerable ethnic heterogeneity and includes African Americans, Caribbean blacks, blacks from Latin America, and blacks from the continent of Africa (Ford and Harawa 2010; Griffith et al. 2011a, b).

Phenomena associated with gender and racial socialization strongly influence the health of men across racial and ethnic groups. There are separate literatures on male gender socialization (Courtenay 2000; Nicholas 2000; Pleck 1981) and racial socialization (Sellers et al. 1998; Stevenson 1997), though most of the racial socialization literature has focused on black Americans. Men’s lives, health, and health behaviors are shaped by both male gender socialization (the process by which men learn the gender and culturally ascribed behaviors that characterize masculinity in a particular society) and racial socialization (the process by which people’s sense of racial identity is shaped by families and communities through oppressive and affirming experiences throughout the life span). Yet, what is less common in the literature is a description of the intersection of these processes and how they affect the mental health of men across race and ethnicity.

African American men’s sense of racial and gender identity are shaped by cultural socialization or the broader context for understanding gender, ethnicity, sexual orientation, and racial identity development (Mahalik et al. 2006; Pierre 2002; Pierre and Mahalik 2005; Pierre et al. 2001; Pieterse and Carter 2007; Sellers et al. 1998; Stevenson 1997; Wade 2008). A meta-study of black male mental health and well-being found that black men are more likely to experience poor mental health, at least in part, from adhering to traditional masculine norms that promote the avoidance of medical care (Watkins et al. 2010). Similarly, Majors and Billson (1992) have discussed the mental health implications of using “cool pose”— behaviors by black males that deliver a message of strength and control—to conceal feelings of self-doubt and insecurity.

The socialization experience of Caribbean black men in the United States may be different than that of African American men and non-Hispanic white men. The racial and ethnic differences among these men may be influenced by social and cultural differences between groups, the nativity status of Caribbean blacks, and length of time Caribbean black men have been in the United States (Griffith et al. 2011a; Williams et al. 2007a, b). For instance, a qualitative study of Jamaican and Guyanese families found a strong gender differentiation in work–family task performance with both lower-class and middle-class Jamaican men. Findings suggested that Jamaican men were less likely than men in North America to participate in tasks commonly performed by women, such as domestic responsibilities or child care (Smith 1988). However, more recent studies have found that Jamaican men in the United States were much more likely to participate in domestic responsibilities and provide assistance with basic household chores than non-US men (Foner 2005). Among Caribbean black men, length of marriage and immigration status have been linked to marital satisfaction (Bryant et al. 2008), another determinant of mental health status for men. Recent studies on African American and Caribbean black men have also found ethnic differences with regard to suicidal ideation (Joe et al. 2006), depression prevalence (Williams et al. 2007a, b), and help-seeking behaviors (Neighbors et al. 2007; Woodward et al. 2008).

Racial socialization for white men is rarely discussed in the literature. Rather, studies suggest that hegemonic masculinity and traditional gender norms are historically framed with white, middle-class, heterosexual men in mind (Mankowski and Manton 2010). Traditional gender norms teach men that they should be confident, dominant, and not exude feminine characteristics (Addis and Mahalik 2003; Courtenay 2000). Thus, responding to stressful situations and mental health issues can be a mysterious process for young men whose role models conform to traditional gender roles. The endorsement of traditional gender norms by black men compared with non-black men is consistent across studies (Courtenay 2000; Lease et al. 2010; Levant et al. 1998), and African American men tend to endorse more traditional perspective when compared with European American men (Courtenay 2000; Levant and Majors 1997; Levant et al. 1998). The endorsement of traditional norms of masculinity also changes depending on the age and professional or non-professional status of men. In addition, men who are young, have lower educational levels, and have lower family incomes tend to endorse more traditional, dominant norms of masculinity (Courtenay 2000).

Father Figures and the Mental Health of Males

Despite the evolution of gender roles over time, the family provider role continues to be a salient aspect of men’s identity (Bowman 1988, 1989; Williams 2003). Traditional gender roles have deep roots in fathers’ responsibilities as “breadwinners” and mothers’ roles as primary childcare providers (Marks and Palkovitz 2004). Yet, black men, when compared with white men, are less likely to successfully fulfill the provider role (Hammond and Mattis 2005; Neighbors et al. 1983; Roopnarine et al. 2009). Sociologist David Williams highlights how middle-class status is often recent, tenuous, and marginal for many black Americans with regard to positive health outcomes. Williams (2003) suggested that college-educated black Americans are four times more likely to experience unemployment than white Americans (Wilhem 1987); middle-class black Americans when compared with middle-class white Americans have markedly lower levels of wealth (Davern and Fisher 1995); and middle-class black Americans are less likely than middle-class white Americans to be able to translate similar levels of education and income into desirable housing and neighborhood conditions (Alba et al. 2000). These economic frustrations may help to explain why socioeconomic status is inversely related to suicide for white American men but positively related to suicide for black American men (Stack 1998; Williams 2003). Additionally, poverty and education have a significant impact on suicide rates among black American males, with poverty reducing the risk of suicide while educational attainment increasing the risk of suicide (Fernquist 2004).

One of the primary ways that young men are socialized is by observing others, particularly their male role models such as father figures (Bandura 2004; Gavanas 2002). In Social Cognitive Theory, Bandura (2004) suggests that learning from others is a critical factor that influences health behavior. Young men learn from observing their father figures and other male role models about what it means to be a man and how men respond to stressful situations and mental health issues. Often the behavioral norms associated with masculine identity evoke stressors in the lives of men (Bowman 1989; Harland 2008; Watkins et al. 2010). Sometimes, these stressors emerge during youth and can shape the mental health and well-being of men over the life course. In recent years, studies on masculinity and mental health have identified complex, multifaceted impressions of what it means to be a man (Addis and Mahalik 2003; Brooks 2001; Harland 2008). Yet, additional complexities surface when attempting to understand how men are conditioned to embody manhood across different racial and ethnic groups (Hurtado and Sinha 2008; Lease et al. 2010; Mahalik et al. 2006; Royster 2007; Wester 2008) and how exposure to father figures in their youth influence their mental health outcomes in adulthood.

Purpose of this Study

Parenting has lasting effects on an individual’s well-being from childhood to adulthood (Mallers et al. 2010). Previous research on the implications of parenting across the life course has focused solely on mother–child relationships (Flouri and Buchanan 2003), leaving a gap in the literature on the role of father figures in the mental health of adults. To date, few studies have considered how father involvement may protect against both emotional and behavioral problems in adolescence and psychological distress in adult life. Even fewer studies probe into how the father figure’s presence may affect the mental health of the male child in adulthood (Flouri 2006; Flouri and Buchanan 2003; Mallers et al. 2010). Therefore, the primary goals of this study were to (a) examine the presence of father figures in the lives of African American, Caribbean black, and non-Hispanic white American males until the age of 16; (b) assess the current socio-demographic factors of these men as adults; and (c) explore whether these factors lead to variations in mental health outcomes of the men in adulthood. Specifically, this study explores the father figures of African American, Caribbean black, and non-Hispanic white male respondents (until the age of 16) and the variations in depressive symptoms and non-specific psychological distress that exist within the context of their socio-demographic differences. Comparing these three groups allows for the examination of ethnic differences among black Americans—African Americans versus Caribbean blacks—and racial differences—black Americans versus non-Hispanic white Americans.

This study considers the findings from previous studies such as Flouri (2006) and Flouri and Buchanan (2003) who only found effects for females and not males, assuming gender to be the major determinant in adult mental health. Yet, based on previous studies about cultural differences in fathers and fatherhood (Williams 2007), we decided to exclude females and consider the racial and ethnic differences of the men from a national sample. We anticipated racial and ethnic differences across the men with regard to their mental health outcomes, but due to the exploratory nature of this study we were unsure about the details of those differences. In addition, our pursuit of retrospective family history information such as those related to childhood father figures (i.e., stepfathers, grandfathers, and uncles) may reveal new information about the mental health of adult men, particularly uncovering racial and ethnic group differences never before uncovered.

Method

Participants

This study used data on adults from the National Survey of American Life (NSAL; Jackson et al. 2004; Sweetman et al. 2006). The NSAL was an integrated national household probability sample of 3,570 African Americans, 891 non-Hispanic whites, and 1,621 blacks of Caribbean descent (Caribbean blacks), for a total sample of 6,082 individuals 18 years and older. Of the total sample, 2,286 were men, and 2,277 reported their primary race or ethnicity as African American, Caribbean black, or non-Hispanic white. The core of the NSAL was the African American sample, which is a nationally representative sample of households in the 48 contiguous states with a black adult in the household aged 18 or older. The non-Hispanic white sample represented households from census tracts and blocks with a 10% or greater African American population. Data were collected between February 2001 and June 2003 using face-to-face interviews and computer assisted instruments. The average interview lasted approximately 2 h and 20 min. Additional details about the NSAL sampling and study procedures have been published elsewhere (Jackson et al. 2004; Sweetman et al. 2006). Weights accounted for the disproportionate sample, non-response, and population representation based on multiple socio-demographic characteristics across all racial and ethnic samples. The sample used in this study was restricted to the 2,010 men (of the total 2,277 African American, Caribbean black, or non-Hispanic white men) with complete data on age, marital status, household income, education, employment status, father figures, Center for Epidemiologic Studies-Depression (CES-D) scale scores, and Kessler-6 (K-6) non-specific psychological distress scale scores. The final self-identified samples included the following: 1,196 African American, 607 Caribbean black, and 207 non-Hispanic white men.

Measures

Independent Variables

Socio-Demographic Measures

Socio-demographic measures included the following: age (18–29, 30–44, 45–59, and 60+), marital status (married, partner, separated/widowed/divorced, never married), employment status (employed, unemployed, and not in the work force), education (<11, 12, 13–15, and 16+ years), and household income. Household income was determined by adjusting for household size by dividing household income by the square root of the household size, measured in continuous US dollars and categorized into quartiles ($0–13,063.90, 13,063.91–23,922.20, 23,922.21–38,000.00, 38,000.01–310,000).

Father Figure Measure

The father figure measure was determined by assessing the retrospective information collected about the man who mostly raised the respondent until the age of 16. There were six possible response choices: biological father, stepfather, grandfather, uncle, other man, or no man. Minimal sample sizes among males with an “uncle” father figure resulted in the combination of “uncle” and “other man” responses into a single category.

Everyday Discrimination

The everyday discrimination scale assesses chronic, routine, and less overt experiences of discrimination that have occurred in the previous year (Williams et al. 1997). This measure has been found to be associated with measures of institutional racial discrimination and interpersonal prejudice (Hughes 2003; Krieger et al. 2005). Evidence suggests that it effectively captures racialized discriminatory experiences that are often part of the life experiences of racial and ethnic groups. We used the 10-item scale with the stem question: “In your day-today life, how often have any of the following things happened to you?” Sample items included “People act as if they think you are dishonest” and “You are followed around in stores.” Responses ranged from 1 (“never”) to 6 (“almost every day”), (Cronbach’s α = 0.86). Higher scores indicated a greater number of events that occurred in the previous year. Results were re-coded from 0 to 5, with a possible range of scores from 0 to 50.

Self-Esteem

The self-esteem measure was derived from Rosenberg’s (1965) 10-item, 4-point response scale used to assess the self-acceptance dimension of self-evaluations. The scale included two items measuring hopelessness, as derived from previous studies on hopelessness (Everson et al. 2000). Items were presented in both negative and positive forms. This measure has had an increasing presence in the mental health literature on men of color and has been used to assess depressive symptoms and psychological distress among racial and ethnic samples of men (Mizell 1999; Watkins et al. 2010, 2011). Responses to the negative items were reverse coded such that 4 (strongly agree) represented high self-esteem and 1 (strongly disagree) represented low self-esteem. The mean of the responses across the ten items was used to ascertain self-esteem scores, and the Cronbach’s alpha (α = 0.77) indicated internal consistency for this measure.

Mastery

Mastery was assessed using Pearlin’s Mastery Scale, a 7-item measure that uses a 4-point agree–disagree format to assess personal mastery (Pearlin et al. 1981). The mastery scale was included in this study because it has been used by other studies to assess perceived control over one’s environment and the things that happen in one’s life among samples of African American men (Mizell 1999; Watkins et al. 2011). The response scale was reverse coded so that higher scores meant that respondents had a higher sense of mastery and lower scores meant respondents had a lower sense of mastery. The mean of the responses across the seven items was used to determine final mastery scores, and the Cronbach’s alpha (α = 0.72) indicated internal consistency for the measure.

Dependent Variables

Center for Epidemiologic Studies-Depression (CES-D)

Depressive symptoms were assessed using the 12-item version of the Center for Epidemiologic Studies-Depression scale (CES-D) (Radloff 1977). This abbreviated CES-D has a similar factor structure compared to the original version. Though the CES-D has been widely used among community samples (i.e., Grunebaum et al. 2008), Love and Love (2006) have cautioned use of the four-factor model among some samples of older black men. In the current study, item responses are coded 1 (“hardly ever”) to 3 (“most of the time”). The 12 items measure the extent to which respondents had trouble keeping their mind on tasks, enjoyed life, had crying spells, could not get going, and felt depressed, hopeful, restless, happy, as good as other people, that everything was an effort, that people were unfriendly, and that people dislike them in the past 30 days. Positive valence items were reverse coded and summed resulting in a continuous measure; a high score indicated a greater number of depressive symptoms, with a total possible range of scores from 0 to 36 (Cronbach’s α = 0.75).

Non-Specific Psychological Distress (NSPD)

Non-specific psychological distress was assessed using the Kessler-6 (K-6) scale. This is a 6-item scale designed to assess nonspecific psychological distress including symptoms of depression and anxiety in the past 30 days (Kessler et al. 2002, 2003). The K-6 is a briefer version of the Kessler-10 (K-10) and contains a subset of the questions. The brevity and accuracy of the K-6 scale has made it an attractive screen for serious mental illness and has been widely used as a measure of distress among clinical and community samples (Kessler et al. 2003). It identifies persons with mental health problems severe enough to cause moderate to serious impairment in social and occupational functioning and to require treatment. The NSAL used only the first root question of the K-6; therefore, the other questions were not included in the analyses. The question used was “In the past 30 days, about how often did you feel…?” with each item (so sad that nothing could cheer you up, nervous, restless or fidgety, hopeless, that everything was an effort, and worthless) measured on a 5-point Likert scale ranging from 0 (none of the time) to 4 (all of the time). Positive valence items were reverse coded, and summed scores ranged from 0 to 24, with higher scores (≥13) reflecting higher levels of psychological distress (Cronbach’s α = 0.81).

Data Analysis

Weighting the data used for these analyses corrected for unequal probabilities of selection, non-response, and for population representation across various socio-demographic characteristics (Heeringa and Liu 2004). The means and standard errors were included to represent the univariate distributions of the study variables. Tests for differences in measures by race/ethnicity were assessed using Rao-Scott χ² tests for categorical variables and survey-weighted t tests for continuous variables. Linear regression for complex survey datasets was performed using SAS software, Version 9.2 of the SAS system for Windows (SAS Institute Inc., Cary, NC, USA, 2000–2004). Linear regression models of CES-D scores and NSPD scores were run. Yet, due to the highly skewed nature of both CES-D scores and NSPD scores, the square root transformation was applied to each measure to minimize violation of the normality assumption associated with linear regression. Linear regression models were adjusted for age, marital status, education, employment status, and household income. Multiple race- and ethnicity-specific regression models were used to evaluate the associations between the father figure characteristics and CES-D and NSPD scores separately. Models additionally examined the impact of self-esteem, mastery, and discrimination on these associations.

Results

Sample Characteristics

Missing Data Comparisons

The 2,010 men (88.27%) included in this sample were drawn from the 2,277 African American, Caribbean black, and non-Hispanic white men included in the NSAL study. A total of 267 men (11.73%) were excluded, of which 258 (11.33%) were excluded for not having complete data on all of the components for CES-D scores and NSPD scores. The remaining 9 (0.40%) were excluded due to missing father figure information, and one additional respondent was excluded due to missing data on employment status. Survey-weighted comparisons of those included versus the 267 men excluded in these analyses revealed that the majority of the missing data were among non-Hispanic white men (n = 164, P < 0.01), although African Americans represented 59.5% (n = 1,196) of the included sample. No differences were detected in the employment status (P = 0.14) or marital status (P = 0.22) for those included and excluded. Differences were detected in lifetime discrimination scores (P < 0.01), but not mastery scores (P = 0.12), self-esteem score (P = 0.20), CES-D scores (P = 0.57), NSPD scores (P = 0.64), household income (P = 0.73), years of education (P = 0.47), or age (P = 0.57) based on two-sample t tests that accounted for the complex survey design.

Socio-Demographic Characteristics

Sample characteristics by racial and ethnic groups are listed in Table 1. Results for the demographic measures suggest the age distribution differed by racial and ethnic groups (P < 0.01), with the majority of African American men (36.14%) aged 30–44, the majority of non-Hispanic white men (31.45%) aged 45–59, and the majority of Caribbean black men (36.06%) aged 18–29. Marital status did not differ across racial and ethnic groups, as the majority of individuals reporting being married. Differences in employment status were detected across racial and ethnic groups (P = 0.01). Though the majority of men in each group reported being employed, a significantly higher proportion of non-Hispanic white men reported not being in the work force (23.40%) compared with African American (19.11%) and Caribbean black (12.20%) men. Education level differed across the groups as well (P < 0.01), such that more African American men had 12 years of education (39.45%) compared with Caribbean black men (27.56%) and non-Hispanic white men (29.21%), while more non-Hispanic white men had 16+ years of education (32.09%). The distribution of income across the groups differed (P < 0.01), as the majority of African American men had household incomes in the lowest two quartiles (54.51%), while most Caribbean black men and non-Hispanic white men had incomes above the median.

Table 1.

Sample characteristics of men by race and ethnicity (n = 2,010)

	African American men (n = 1,196)		Caribbean black men (n = 607)		White men (n = 207)		P value
	N	Weighted %	N	Weighted %	N	Weighted %
Age group
18–29	266	24.81	178	36.06	30	14.09	<0.01
30–44	415	36.14	206	29.65	62	31.41
45–59	309	24.23	124	19.82	65	32.45
60+	206	14.82	99	14.47	50	22.05
Marital status
Married	424	40.64	260	43.77	111	63.49	0.22
Partner	105	9.50	61	15.55	11	19.84
Separated/widowed/divorced	300	19.81	99	10.36	46	51.27
Never married	367	30.05	187	30.32	39	22.40
Employment
Employed	841	71.75	463	80.85	144	73.51	0.01
Unemployed	101	9.13	55	6.94	9	3.09
Not in work force	254	19.11	89	12.20	54	23.40
Education (years)
<11	295	22.61	129	22.94	33	14.56	<0.01
12	467	39.45	178	27.56	63	29.21
13–15	274	23.23	161	25.73	46	24.14
16+	160	14.71	139	23.77	65	32.09
Income quartile
1 ($0–13,063.90)	344	27.31	123	21.02	35	16.42	<0.01
2 ($13,063.91–23,922.20)	311	27.20	149	21.87	43	19.38
3 ($23,922.21–38,000)	299	24.27	161	25.17	48	22.80
4 ($38,000.01–310,000)	242	21.22	174	31.94	81	41.40
Father figure until age 16
Biological father	734	62.91	402	69.69	176	84.94	<0.01
Stepfather	116	9.41	48	5.95	7	2.94
Grandfather	76	6.28	24	4.15	3	0.69
Other male	83	6.72	39	7.11	7	3.82
None	187	14.68	94	13.10	14	7.61
	Mean	SE	Mean	SE	Mean	SE	P value
ED score (range = 0–50)	13.76	0.47	14.75	1.01	8.86	0.56	<0.01
SELF-ES score (n = 2,004, range = 0–40)	3.63	0.02	3.61	0.04	3.52	0.05	0.20
M score (n = 2,006, range = 0–28)	3.38	0.02	3.15	0.11	3.27	0.03	0.12
CES-D score (range = 0–36)	6.13	0.23	6.66	0.87	8.70	0.85	0.57
NSPD score (range = 0–24)	3.38	0.18	3.55	0.34	4.31	0.55	0.64

SE standard error, ED score everyday discrimination score, SELF-ES score self-esteem score, M score mastery score

Father Figures, Discrimination, and Mental Health Measures

Data on the men who mostly raised the respondents’ until the age of 16 revealed that the majority of men had a biological father or stepfather who raised them, although the proportion was significantly higher for non-Hispanic white men (84.94%) compared with African American (62.91%) and Caribbean black (69.69%) men (P < 0.01). Among the samples, 14.68% of African American men, 13.10% of Caribbean black men, and 7.61% of non-Hispanic white men reported having no father in the home, the second most common response across race and ethnicity groups. Everyday discrimination scores ranged in value from 0 to 50, with lower mean scores reported for non-Hispanic white men (Mean = 8.86, SE = 0.56) compared with African American (Mean = 13.76, SE = 0.47) and Caribbean black men (Mean = 14.75, SE = 1.01), P < 0.01. No significant differences were detected in mean self-esteem or mastery scores. On average, CES-D scores were higher for non-Hispanic white men (Mean = 8.70, SE = 0.85) compared with Caribbean black men (Mean = 6.66, SE = 0.87) and African American men (Mean = 6.13, SE = 0.23), although the differences were non-significant. NSPD scores were also higher for the non-Hispanic white men (Mean = 4.31, SE = 0.55) compared with Caribbean black men (Mean = 3.55, SE = 0.34) and African American men (Mean = 3.38, SE = 0.18), but the differences were also non-significant.

Linear Regression Models by Race/Ethnic Group

Models including socio-demographic measures are presented in Table 2. The associations between the father figure measure and transformed CES-D scores are shown in Table 3, with NSPD scores presented in Table 4. Tables 3 and 4 are not calculated for non-Hispanic white men due to small cell frequencies among non-Hispanic white men for the father figure measure.

Table 2.

Multiple linear regression models for depressive symptoms (CES-D) and non-specific psychological distress (NSPD) scores by socio-demographic measures

	African American men (n = 1,196)				Caribbean black men (n = 607)				White men (n = 207)
	CES-D		Distress		CES-D		Distress		CES-D		Distress
	B	SE	B	SE	B	SE	B	SE	B	SE	B	SE
Age
18–29	Ref		Ref		Ref		Ref		Ref		Ref
30–44	–0.23	0.13	–0.19	0.10	0.06	0.16	–0.47*	0.20	0.12	0.29	0.08	0.25
45–59	–0.36**	0.14	–0.14	0.14	–0.55	0.28	–1.09**	0.32	0.12	0.33	0.33	0.26
60+	–0.79**	0.17	–0.80**	0.14	–0.48	0.44	–1.24**	0.31	–0.36	0.31	–0.96**	0.35
Marital status
Married	Ref		Ref		Ref		Ref		Ref		Ref
Partner	0.07	0.14	0.00	0.11	0.43	0.14	0.34**	0.14	–0.49*	0.22	0.25	0.28
Separated/widow/divorced	0.13	0.11	–0.05	0.10	0.21**	0.26	0.38	0.16	–0.37	0.40	0.23	0.30
Never married	0.00	0.11	0.03	0.09	0.41	0.22	–0.14	0.19	0.49	0.34	0.58**	0.18
Employment status
Employed	Ref		Ref		Ref		Ref		Ref		Ref
Unemployed	0.23	0.14	0.15	0.14	0.05	0.31	0.23	0.20	0.55*	0.26	1.14**	0.28
Not in work force	0.35**	0.12	0.33**	0.10	0.78**	0.30	0.92**	0.26	0.28	0.33	0.70*	0.32
Education group (years)
<11	0.43**	0.13	0.23	0.14	0.35	0.42	0.31	0.24	0.41	0.33	0.39	0.30
12	0.25*	0.11	0.12	0.12	–0.23	0.20	–0.25	0.16	0.01	0.29	–0.31	0.18
13–15	0.07	0.11	0.02	0.12	–0.32*	0.14	–0.17	0.15	0.02	0.24	–0.19	0.16
16+	Ref		Ref		Ref		Ref		Ref		Ref
Income quartile
1 ($0–13,063.90)	0.33**	0.12	0.47**	0.14	0.38	0.26	0.14	0.19	0.58*	0.28	0.08	0.32
2 ($13,063.91–23,922.20)	0.24*	0.12	0.26*	0.12	0.16	0.20	0.01	0.15	0.08	0.30	0.15	0.22
3 ($23,922.21–38,000)	0.13	0.08	0.18	0.11	–0.18	0.15	–0.12	0.17	0.08	0.20	0.23	0.18
4 ($38,000.01–310,000)	Ref		Ref		Ref		Ref		Ref		Ref
R2	0.09		0.08		0.25		0.28		0.15		0.23

CES-D and NSPD scores were transformed by taking the square root of each measure, respectively

Ref reference group, B beta, SE standard error

^*P < 0.05

^**P < 0.01

Table 3.

Multiple linear regression models for depressive symptom (CES-D) scores by father figure

	African American men				Caribbean black men
	n = 1,196		n = 1,195		n = 607		n = 606
	Model 1		Model 2		Model 1		Model 2
	B	SE	B	SE	B	SE	B	SE
Man/father figure until age 16
Biological father	Ref		Ref		Ref		Ref
Stepfather	0.13	0.11	0.11	0.10	–0.19	0.21	0.18	0.25
Grandfather	0.21	0.14	0.22*	0.09	0.86*	0.38	0.43	0.26
Other male	–0.13	0.17	–0.15	0.15	–0.13	0.24	–0.24	0.21
None	0.01	0.09	0.04	0.07	0.31	0.23	0.21	0.18
Everyday discrimination score			0.02**	0.003			0.02**	0.01
Self-esteem score			–0.84**	0.09			–0.84**	0.11
Mastery score			–0.37**	0.06			–0.45**	0.09
R2	0.09		0.34		0.28		0.49

CES-D scores were transformed by taking the square root

Ref reference group, B beta, SE standard error

^*P < 0.05

^**P < 0.01

Model 1 Father figure

Model 2 Model 1 + everyday discrimination + self-esteem + mastery

Models were adjusted for age, education, marital status, employment status, and income quartiles

Analysis was not calculated for white men due to small cell frequencies among white men for the father figure measure

Table 4.

Multiple linear regression models for non-specific psychological distress (NSPD) scores by father figure

	African American men				Caribbean black men
	n = 1,196		n = 1,195		n = 607		n = 606
	Model 1		Model 2		Model 1		Model 2
	B	SE	B	SE	B	SE	B	SE
Man/father figure until age 16
Biological father	Ref		Ref		Ref		Ref
Stepfather	0.09	0.14	0.07	0.13	–0.02	0.16	0.27	0.15
Grandfather	0.12	0.11	0.13	0.09	0.88**	0.27	0.56**	0.17
Other male	0.14	0.14	0.13	0.12	–0.21	0.21	–0.32	0.21
None	–0.22*	0.10	–0.19*	0.08	0.20	0.15	0.12	0.12
Everyday discrimination score			0.02**	0.004			0.03**	0.01
Self-esteem score			–0.67**	0.10			–0.54**	0.10
Mastery Score			–0.41**	0.07			–0.39**	0.08
R2	0.09		0.27		0.31		0.50

NSPD scores (also referred to as K-6 scores) were transformed by taking the square root

Ref reference group, B beta, SE standard error

^*P < 0.05

^**P < 0.01

Model 1 Father figure

Model 2 Model 1 + everyday discrimination + self-esteem + mastery

Models were adjusted for age, education, marital status, employment status, and income quartiles

Analysis was not calculated for white men due to small cell frequencies among white men for the father figure measure

Linear Regression Model Results for African American Men

Socio-Demographic Factors Associated with CES-D and NSPD

Linear regression models of the square root-transformed CES-D and NSPD scores as a function of age, marital status, employment status, education, and income for African American men are reported in Table 2. Results suggest CES-D scores were lower among African American men aged 45–59 (b = –0.36, SE = 0.14, P < 0.01) and 60+ (b = –0.79, SE = 0.17, P < 0.01) when compared with men aged 18–29, with greater decreases in CES-D scores among those 60+. Being 60 years of age or older was also associated with decreased NSPD scores (b = – 0.80, SE = 0.14, P < 0.01). Results showed that African American men not in the work force had higher CES-D scores (b = 0.35, SE = 0.12, P < 0.01) and NSPD scores (b = 0.33, SE = 0.10, P < 0.01). Graded relationships were detected between less education and lower income, such that African American men with <11 years of education had higher increases in CES-D scores (b = 0.43, SE = 0.13, P < 0.01) than those with 12 years of education (b = 0.25, SE = 0.11, P < 0.05) when compared to the reference group of individuals with 16 or more years of education. Similarly for income, African American men with household incomes in the lowest quartile had the highest increase in CES-D scores (b = 0.33, SE = 0.12, P < 0.01), while those with household incomes within the second quartile showed a smaller increase in CES-D scores (b = 0.24, SE = 0.12, P < 0.05). Similar results were found for NSPD. Multiple socio-demographic factors were included in this model, the proportion of variance explained by these measures was 9% for the linear regression model of the transformed CES-D score and 8% for the transformed NSPD score for African American men.

Father Figure Measure, CES-D, and NSPD

Linear regression models of the square root-transformed CES-D and NSPD scores as a function of father figure, adjusted for age, marital status, employment status, education, and income for African American men are reported in Tables 3 and 4. In models that examined the impact of the father figure in the household, adjusted for age, marital status, employment status, education, and income (Table 3, Model 1), results suggested that information about the father figure explained no additional variation in the model for CES-D scores (R² = 0.09) or NSPD scores (R² = 0.09) for African American men. Although the father figure measure was unassociated with CES-D scores, having no father figure was associated with decreased NSPD scores for African American men (Table 4, Model 1) (b = –0.22, SE = 0.10, P < 0.05). With the addition of discrimination, self-esteem, and mastery to the model (Table 3, Model 2), the proportion of variation explained by the model increased to 34%. These results suggested that having greater discrimination scores was associated with slightly higher CES-D scores (b = 0.02, SE = 0.003, P < 0.01), while having higher self-esteem (b = –0.84, SE = 0.09, P < 0.01) and greater mastery (b = –0.37, SE = 0.06, P < 0.01) were associated with decreased CES-D scores. Further, this model indicated that having a grandfather as a father figure was associated with higher CES-D scores for African American men (b = 0.22, SE = 0.09, P < 0.05). Similar results were found for NSPD scores in relation to lifetime discrimination, self-esteem, and mastery (Table 4, Model 2), only the association between having no father figure and decreased NSPD scores remained (b = –0.19, SE = 0.08, P < 0.05), and the proportion of variance explained with this model was 27%.

Linear Regression Model Results for Caribbean Black Men

Socio-Demographic Factors Associated with CES-D and NSPD

Table 2 models including age, marital status, employment status, education, and income suggested that these measures in combination accounted for 25% of the variation in CES-D scores and 28% of the variation in NSPD scores for Caribbean black men. Result showed that being separated/widowed/divorced was associated with higher CES-D scores (b = 0.21, SE = 0.26, P < 0.01) compared to being married. Not being in the work force was also associated with higher CES-D scores (b = 0.78, SE = 0.30, P < 0.01). Caribbean black men with 13–15 years of education tended to have lower CES-D scores than those with 16 or more years of education (b = –0.32, SE = 0.14, P < 0.05). Different associations were present with NSPD scores. Caribbean black men aged 30 and older had lower NSPD scores compared to those aged 18–29, with the decrease being more pronounced with increasing age (aged 30–44: b = –0.47, SE = 0.20, P < 0.05; aged 45–50: b = –1.09, SE = 0.32, P < 0.01; aged 60+: b = –1.24, SE = 0.31, P < 0.01). Having a partner was associated with higher NSPD scores (b = 0.34, SE = 0.14, P < 0.05), as was not being a member of the work force (b = 0.92, SE = 0.26, P < 0.01).

Father Figure Measure, CES-D, and NSPD

Models that examined the associations of the father figure while controlling for the socio-demographic measures (Table 3, Model 1) accounted for an additional 3% of variation in CES-D scores among Caribbean black men compared to models including socio-demographic measures alone and showed that having a grandfather as a father figure was associated with higher CES-D scores (b = 0.86, SE = 0.38, P < 0.05). Similar results were found for NSPD (Table 4, Model 1). With further adjustment of the model for discrimination, self-esteem, and mastery (Table 3, Model 2), the proportion of variance in CES-D explained by the model increased to 49% and suggested no associations between the father figure and CES-D, but showed associations similar to those found among African American men. Results were similar for NSPD (Table 4, Model 1), although having a grandfather as a father figure remained associated with increased NSPD scores (b = 0.56, SE = 0.17, P < 0.01) before and after discrimination, self-esteem, and mastery were added.

Linear Regression Model Results for Non-Hispanic White Men

Socio-Demographic Factors Associated with CES-D and NSPD

Models from Table 2 including socio-demographic factors suggested that these measures accounted for 15% of the variance in transformed CES-D scores and 23% of the variation in transformed NSPD scores for non-Hispanic white men. The analyses indicated that non-Hispanic white men who were partnered had lower transformed CES-D scores (b = –0.49, SE = 0.22, P < 0.05) while those who were unemployed (b = 0.55, SE = 0.26, P < 0.05) and those with the lowest incomes (b = 0.58, SE = 0.28, P < 0.05) had higher CES-D scores. Examination of the NSPD models showed that non-Hispanic white men aged 60+ had lower NSPD scores (b = –0.96, SE = 0.35, P < 0.01) compared with their younger counterparts. Non-Hispanic white men who were never married also had higher NSPD scores (b = 0.58, SE = 0.18, P < 0.01), as did those who were unemployed (b = 1.14, SE = 0.28, P < 0.01) or not in the work force (b = 0.70, SE = 0.32, P < 0.05).

Discussion

The purpose of this study was to explore the presence of father figures (until the age of 16) in the lives of African American, Caribbean black and non-Hispanic white American men; assess the current socio-demographic factors of these men as adults; and explore whether these factors lead to variations in mental health outcomes of the men in adulthood. By running a series of regression models, we were able to explore the effects of father figures, discrimination, self-esteem, and mastery on two commonly used measures of mental health: the CES-D for depressive symptoms and the K-6 for non-specific psychological distress. Our base model (Table 2) identified statistically significant correlates that echo the current literature on men’s mental health. For instance, non-partner marital status, unemployment status, lower education, and lower income have all been associated with poorer mental health for men of various racial and ethnic groups (Joe et al. 2006; Kessler et al. 1994; Lincoln et al. 2011; Roy 2006; Watkins et al. 2006, 2011; Williams et al. 2007a, b). Additionally, we found marital partnerships, higher incomes, and higher education to be protective against the perils of poor mental health for men.

We found that everyday discrimination scores were highest for Caribbean black men, followed by African American men then non-Hispanic white men. Previous studies on psychological distress and black men have suggested that experiences of discrimination worsen their psychological health (Pieterse and Carter 2007; Utsey 1997). Over the life course, black men are more likely to encounter discrimination or experience noxious encounters in their social environments (i.e., school and the workplace); thus, threatening their mental health. How discrimination influences mental health in the context of ethnicity (African American vs. Caribbean black) is less understood in the literature (Vickerman 2001). Perhaps the mental health effects of social incongruence coupled with exposure to discrimination and cultural challenges faced by Caribbean black men make them more prone to experiencing discrimination compared to African American men. This may remain consistent for Caribbean black men beginning in their youth and continuing well to and through adulthood.

Uncovering variations in mental health outcomes based on the socio-demographic variables (i.e., marital status, household income, education level) and psychosocial factors (i.e., self-esteem, mastery, experiences with discrimination) of our sample echoes findings from previous studies (Lincoln et al. 2011; Mizell 1999; Rowell et al. 2011; Watkins et al. 2006, 2011). However, we also uncovered some unexpected racial and ethnic group differences from our models. For instance, before adding the father figure measure to the model, we found depressive symptoms and non-specific psychological distress to be highest among the non-Hispanic white men, followed by the Caribbean black men, then the African American men. These findings were aligned with previous studies that reported a higher prevalence of mental health disorders among white men (Gavin et al. 2010) and Caribbean black men (Joe et al. 2006). However, after we added the father figure measure, the African American men and the Caribbean black men experienced many of the same associational outcomes for both depressive symptoms and non-specific psychological distress.

Differences emerged once the type of father figure (i.e., stepfather, grandfather, other male, etc.) was examined and compared across the different race and ethnic groups of men. We could not locate previous studies that examined racial and ethnic group variations with regard to the presence of father figures for men—only those that discussed gender differences such as Flouri (2006) and Flouri and Buchanan (2003). Thus, our analyses with such variables may be the first of its kind. For example, among the African American male sample, the socio-demographic correlates remained virtually the same across our models until the addition of the father figure measure. It was then we noticed that being raised by a grandfather increased depressive symptoms for African American men once discrimination, mastery, and self-esteem were added. On the contrary, being raised by a grandfather also increased depressive symptoms for Caribbean black men, but this was not the case after discrimination, mastery, and self-esteem were added to the model.

When we examined non-specific psychological distress, we found that being raised by a grandfather had no statistically significant effect for African American men. Instead, being raised by no man until the age of 16 actually lowered psychological distress scores for African American men, and the same was true after discrimination, mastery, and self-esteem were added. These findings highlight implications of the presence of father figures in the African American community, where less attention has been given to male mental health outcomes and more has been given to psychosocial functioning and education (Shook et al. 2010). Earlier studies suggested that African American men raised by a single mother were at an economic and developmental disadvantage compared with African American men raised in two-parent households (Mizell 1999). Though under certain conditions this may still be the case, findings from the current study suggest that African American men (particularly those with high self-esteem and mastery) who experienced non-specific psychological distress as adults may do so as a result of several socio-demographic and psychosocial conditions. Yet, the presence of a father figure during their childhood may not be as important as the presence and/or absence of other socio-demographic factors. As such, further studies that test directional hypothesis are needed to confirm this.

For Caribbean black men, being raised by a grandfather increased non-specific psychological distress scores both before and after discrimination, mastery, and self-esteem were added. So compared to African American and non-Hispanic white men, being raised by a grandfather as a Caribbean black man was associated with higher nonspecific psychological distress scores, despite the presence of high self-esteem and mastery. These findings draw attention to the potential ethnic and cultural differences in child rearing and gender dynamics among Caribbean black men compared with African American men and non-Hispanic white men. Are there differences in the perspectives of Caribbean black grandfathers and what they instill in their male children about manhood compared to African American and white men? And how does this translate into how Caribbean black men handle their mental health problems?

A study in the United Kingdom found the meaning of health for African–Caribbean and white fathers to be at a “functional capacity.” In other words, health was an asset that allowed fathers to meet the obligations of paid work and fathering, not for the purposes of their own health and well-being (Williams 2007). Caribbean black immigrants may also be more socially conservative than both African Americans and whites (Vickerman 2001). Therefore, the idea of exhibiting vulnerability for Caribbean black men may be uncharacteristic for them and a male gender norm that they pass on to their male youth. Caribbean black men may also undergo different rites of passage that influence their gender and cultural identity compared with African American and white men, but this needs further examination (Chevannes 1993; Cleaver 2002). Though we did not include nativity measures in this our study, the ethnic effects found in this study may be exacerbated by nativity and immigrant status (Griffith et al. 2011a). For instance, one study found that Caribbean black men who immigrated as adolescents (aged 13–17 years) had low rates of mood and anxiety disorders (Williams et al. 2007a, b). The experience of Caribbean black men who are born outside of the United States and whose father figures migrate to the United States for economic purposes is different from Caribbean black men born in the United States whose US-dwelling father figures are not involved in their lives. Issues surrounding nativity may also be influenced by generational group differences experienced by Caribbean black men compared with men from the other groups (Joe et al. 2006; Griffith et al. 2011a; Williams et al. 2007a, b). Just as the African American, Caribbean black, and non-Hispanic white men from the present study experienced the presence of their father figures differently, so too may they experience health and learned male gender norms in different ways (Connell 2005; Williams 2007).

Also noteworthy is that previous studies have described African American males’ participation in so-called “street-life” and its effects on the absence of father figures in African American families (Oliver 2006a, b; Payne 2006, in press). In this context, street-life is a socialized institution that is composed of the network of public and semipublic social settings (e.g., street corners, bars, after-hours locations, drug houses, vacant lots) that serve as important influences on the psychosocial development and life course trajectories and transitions of black males (Oliver 2006a, b). Active participation in street-life that derives from the subscription to certain street-related gender norms can diminish a black man’s role in the household and limit his abilities to fulfill the responsibilities associated with fatherhood and manhood, in general (Oliver 2006a, b; Payne 2006, in press). Our findings raise important questions regarding the characteristics of black men and if the presence (and type) of father figures during their youth influences their mental health outcomes differently than it may for men of other racial and ethnic groups. What seems to be clear, however, is that there are cultural implications of masculinity and that male norms are defined and passed down across different ethnic groups of black men.

Just as those before them, fathers and father figures often teach male children their own definitions of manhood through demonstrations (Gavanas 2002). This responsibility, while common across racial and ethnic groups, may vary with regard to specific cultural and experiential details. Our findings add to the previous studies that suggest that poor mental health outcomes of some men may result from psychosocial stressors associated with perceived cultural and gender norms in the context of their environments. This can lead to a cycle where the children see these men facing stressors and mental health problems and responding to these problems using negative coping strategies (Jackson et al. 2010). For example, men who are affiliated with street-oriented lifestyles, such as men of color, may render less desirable partners that lead to relationship disruption resulting in high divorce rates, female-headed families, out-of-wedlock births, less commitment of men to relationships, and negative perceptions of men on the part of the women (Payne 2006, in press). Under this type of gender-induced stress, many African American men are limited in their ability to serve as responsible father figures, which may include providing love, social, and financial support for their children (Oliver 2006a, b). Our findings suggest that extending the knowledge base beyond that which includes race only, but considering the implications of culture, ethnicity, and even nativity can further our understanding of mental health disparities among sub-groups of men more than our limited study of race ever could in previous studies.

The findings from this study must be interpreted in the context of its limitations. First, due to the cross-sectional nation of the design, we could not ascertain prospective information from the respondents during their youth. This information would have been helpful for probing into the involvement of respondents’ father figures as well as the extent to which respondents felt as if their father figures influenced their own mental health outcomes. Second, the amount of time that elapsed from when the study respondents were fathered to when they responded to the survey varied across respondents and, as expected, would weaken any potential effects. Third, the design of the study does not permit an analysis of the K-6 scale in its entirety. The NSAL dataset study only used the first question (including six items) of the K-6 measure, thus limiting our full interpretation of the non-specific psychological distress measure for the sample. Therefore, results from this study are preliminary and exploratory and would benefit from additional analyses. Next, our results led to very specific inferences about the sub-sample of Caribbean black men who were raised by their grandfathers. Though we compared these results to those of African American men, it should be noted that the number of African American men raised by their grandfathers was almost three times the number of Caribbean black men raised by their grandfathers. Therefore, the small sub-sample analyzed here may weaken the actual strength of these effects for Caribbean black men and should be interpreted with caution. Finally, the present study used the CES-D and the K-6 as measures of mental health outcomes for the men (Nguyen et al. 2004; Love and Love 2006). Though these measures are widely used, they may not be regarded as true measures of mental health among clinicians. Future studies should consider this as well as how cultural differences in masculine identify and conformity may influence responses to CES-D and K-6 items for men given their racial, ethnic, and cultural differences.

Despite these limitations, the study sample, methods, and findings provided a unique opportunity to explore the mental health of men across three distinct racial and ethnic groups. Our findings offer some suggestions for potential next steps for research exploring the role of father figures in the mental health outcomes of adult men. This study is unique in that we retrospectively consider the influence of father figures on the mental health outcomes of adult males. Findings highlight the need to consider both racial and ethnic heterogeneity among US men as well as the correlates of depressive symptoms and psychological distress across racial and ethnic groups of men. It is critical to consider factors associated with race and ethnicity to understand and address men’s mental health needs.