Revision of ENTIRE Essay
michaeln
SecondSource.pdf
10.1177/0192513X03258313 ARTICLEJOURNAL OF FAMILY ISSUES / September 2004Turner, Muller / LONG-TERM EFFECTS OF CORPORAL PUNISHMENT
Long-Term Effects of Child Corporal Punishment on Depressive Symptoms
in Young Adults Potential Moderators and Mediators
HEATHER A. TURNER PAUL A. MULLER
University of New Hampshire
Based on a sample of 649 students from 3 New England colleges, this study examined the long-term effects of childhood corporal punishment on symptoms of depression and consid- ered factors that may moderate or mediate the association. Similar to national studies, ap- proximately 40% of the sample reported experiencing some level of corporal punishment when they were 13 years old. Findings indicated that level of corporal punishment is posi- tively related to depressive symptoms, independent of any history of abuse and the frequency of other forms of punishment. Although parental monitoring and perceived norms regarding corporal punishment had no direct or moderating effects, level of parental anger during cor- poral punishment was the strongest predictor of depression. The association between “angry corporal punishment” and symptoms of depression in young adulthood is partially mediated by mastery and self-esteem. Implications of these finding are discussed.
Keywords: corporal punishment; depression; parental anger; self-concept
It is widely accepted, and there is much empirical evidence, that children’s exposure to physical abuse is related to a host of negative outcomes. Some of the problems associated with abuse include post-traumatic stress disor- der, depression, and alcoholism (Carlin et al., 1994; Gordon et al., 1989; Holmes & Robins, 1988; Kolko, 1992), aggressive and violent behavior (Dodge, Bates, & Pettit, 1990; Howing, Wodarsky, Kurtz, Gaudin, & Herbst, 1990; Patterson, 1986), and academic and interpersonal difficul- ties (Kolko, 1992; Malinosky-Rummell & Hansen, 1993). Although ad- verse outcomes of child abuse are well established, not as much research has focused on a less severe and more normative form of violence toward
761
Authors’ Note: This research was supported by a grant from the National Institute of Mental Health (R03#MH56169). All correspondence related to this article should be directed to Heather A. Turner, Department of Sociology, Horton Social Science Center, University of New Hampshire, Durham, NH 03824; e-mail: [email protected]
JOURNAL OF FAMILY ISSUES, Vol. 25 No. 6, September 2004 761-782 DOI: 10.1177/0192513X03258313 © 2004 Sage Publications
children—corporal punishment. Moreover, to the extent that corporal punishment has long-term effects on well-being, the mechanisms by which this occurs and the conditions that influence the association are still unclear.
The current study adds to the existing research on corporal punishment by examining the impact of corporal punishment in childhood on later de- pression among young adults, and by considering mediating and moderat- ing factors that may operate in this process. Specifically, this research (a) examines the impact of corporal punishment on depressive symptoms in young adults (ages 18 to 29 years), independent of sociodemographic fac- tors (age, sex, race, parental education), frequency of other types of pun- ishment, and any incidence of physical or sexual abuse; (b) assesses the extent to which race, parental monitoring, perceptions of corporal punish- ment norms, parental anger, and frequency of other types of punishment moderate the impact of corporal punishment on depression; and (c) deter- mines whether mastery and self-esteem mediate the association between corporal punishment and depressive symptoms.
CORPORAL PUNISHMENT AND NEGATIVE OUTCOMES
Corporal punishment refers to “the use of physical force with the inten- tion of causing a child to experience pain, but not injury, for the purpose of correction or control of the child’s behavior” (Straus, 1994, p. 4). Corpo- ral punishment is an accepted child-rearing practice in the United States (and most other countries as well). Straus (1991) found that more than 90% of parents in the United States employ some form of corporal punish- ment in disciplining their children. Moreover, two thirds of parents agree that, “It is sometimes necessary to discipline a child with a good hard spanking” (Straus & Mathur, 1996, p. 95).
It has been suggested (e.g., Straus, 1994) that corporal punishment should be considered a form of violence against children. Although “legit- imate” corporal punishment does not result in injury, it is possible that some of the same negative outcomes associated with abuse—violence that does result in injury—will be exhibited by children who are exposed to corporal punishment. Indeed, work by Straus and Kaufman Kantor (1994), led these authors to conclude that there is persuasive empirical ev- idence showing corporal punishment to be a risk factor for later spousal abuse, child abuse, depression, suicide, and alcohol abuse.
762 JOURNAL OF FAMILY ISSUES / September 2004
A number of studies have supported the notion that corporal punish- ment contributes to aggressive behavior of children and subsequent vio- lence in adulthood.
For example, research has shown that childhood exposure to corporal punishment increases the likelihood that those individuals will later hit their own children or their spouse (Simons, Whitbeck, Conger, & Wu, 1991; Straus, Gelles, & Steinmetz, 1980). Others have shown a significant relationship between frequency of corporal punishment and children’s aggressive behavior toward family members (e.g., Larzalere, 1986). In- deed, Straus (1991) reported that children who received “ordinary” corpo- ral punishment, as compared with those who received none, were 3 times as likely to engage in sibling assaults. An association between exposure to corporal punishment and antisocial and impulsive behavior has also been reported (Straus & Mouradian, 1998).
In another study, Simons, Lin, and Gordon (1998) sought to better un- derstand the causes of dating violence among male adolescents by consid- ering the potential effects of certain parenting behaviors (use of corporal punishment, interparental violence, and supportive parenting). The re- searchers suggested three possible processes by which parental behavior may affect dating violence. First, male children may learn to be violent to- ward dating partners by modeling their behavior after the interparental vi- olence they witness as children. Second, exposure to corporal punishment may teach children that physical aggression is a useful tactic for modify- ing the behavior of others, such as their dating partners. Finally, lack of in- volved and supportive parenting may contribute, in general, to children’s antisocial behavior—including dating violence. Of the three parenting constructs considered, only exposure to corporal punishment was directly related to subsequent male dating violence. As the authors stated, “This suggests that corporal punishment provides a type of training that increases the probability of violence toward dating partners” (p. 475).
Past studies also suggest negative emotional outcomes associated with corporal punishment. For example, Bryan and Freed (1982) found corpo- ral punishment was not only related to aggression and delinquency but also to negative psychological states, such as anxiety and depression. Similarly, Turner and Finkelhor (1996) found positive relationships be- tween frequency of corporal punishment and psychological distress and clinically relevant depression. Moreover, although children who were fre- quently exposed to corporal punishment had the highest levels of distress and depression, even moderate or low exposure to corporal punishment had a negative impact on children’s well-being.
Turner, Muller / LONG-TERM EFFECTS OF CORPORAL PUNISHMENT 763
Because much of the past research on outcomes of corporal punish- ment examines links between this form of discipline and children’s ag- gressive behavior, it is important to acknowledge the possibility that ag- gressive children may be more likely to elicit physical punishment than children who do not exhibit these problematic behaviors. In other words, it is possible that the causal direction of a cross-sectional association be- tween frequency of corporal punishment and negative outcomes runs from the child to corporal punishment. Although this concern is certainly warranted, longitudinal studies have demonstrated the influence of corpo- ral punishment on subsequent increases in antisocial behavior (Straus, Sugarman, & Giles-Sims, 1997), providing support for corporal punish- ment as a causal factor for these negative outcomes. Moreover, the alterna- tive causal interpretation is less appropriate when considering internaliz- ing behaviors as outcomes. Because depressed mood is less likely to elicit punishment than antisocial behavior, it seems very likely that a significant part of an association between corporal punishment and child distress would be due to the causal impact of corporal punishment on children’s well-being.
POTENTIAL MODERATORS
If corporal punishment has a negative impact on the well-being of chil- dren, as has often been demonstrated, it becomes important to identify factors that may moderate its adverse effects. Indeed, given the relatively high prevalence of corporal punishment as a form of discipline, it seems likely that conditions often exist to help reduce its negative impact. Con- versely, corporal punishment may be influential only under certain cir- cumstances or contexts. Thus, we suggest several factors (based on empir- ical or theoretical evidence) that may either buffer or accentuate the impact of corporal punishment: race, frequency of other forms of punish- ment, parental monitoring, perceived normative levels of corporal punishment, and parental anger.
Several studies have examined the extent to which the impact of corpo- ral punishment differs by race and ethnicity. Some of this research sug- gests that minorities and lower-income families (a) are more approving of the use of corporal punishment, (b) tend to use more corporal punishment, and (c) are less prone to the deleterious effects of corporal punishment
764 JOURNAL OF FAMILY ISSUES / September 2004
(Baumrind, 1972; Dornbusch, Ritter, Leiderman, Roberts, & Fraleigh, 1987; Heffer & Kelley, 1987).
In one study, Straus and Lauer (1992) found that use of corporal pun- ishment is associated with a greater likelihood of criminal behavior among all ethnic groups. However, the authors also pointed out that this finding is more consistently supported among Whites than among His- panics and is least consistent in the African American sample. Gunnoe and Mariner (1997) found even more dramatic group differences in the ef- fects of corporal punishment on children. Whereas corporal punishment was related to more instances of aggressive behavior (fighting at school) among White children, corporal punishment was actually related to fewer aggressive events among Black children.
Parenting behaviors can create an important context in which corporal punishment is interpreted and child outcomes manifested. For example, Larzalere (1986) found that the effect of physical punishment on aggres- sion toward parents was dependent on parental use of reasoning, such that, “The association of spanking with aggression toward the parent was elim- inated for those parents who responded frequently to parent-child conflict by discussing the issue calmly” (p. 32). Similarly, Straus and Mouradian (1998) found that exposure to corporal punishment was related to a greater likelihood of engaging in antisocial behavior. However, when in- teractions between corporal punishment and various other (e.g., parenting) variables were tested, corporal punishment was most strongly related to antisocial behavior in children when mothers administered the corporal punishment impulsively. Thus, it appears that corporal punish- ment is less damaging when it is carried out under control. We suggest that parental anger during corporal punishment likely reflects impulsiveness and loss of control; that is, it is reasonable to assume that parents who are very angry when administering discipline are more likely to spank impul- sively, exhibit less control, and are less likely to incorporate reasoning than when they administer corporal punishment without anger.
Similar to the use of reasoning and controlled discipline (e.g., without anger), monitoring of a child’s activities is a parental behavior that has been linked to positive outcomes for youth. Many studies have shown that adequate monitoring or supervision by parents is related to higher career and goal aspirations (Spencer, Dupree, Swanson, & Cunningham, 1996) and protects against the adverse psychological effects of poor socioeco- nomic conditions (Jarrett, 1997). In addition to its direct positive effects,
Turner, Muller / LONG-TERM EFFECTS OF CORPORAL PUNISHMENT 765
parental monitoring may also interact with corporal punishment to mod- erate its impact on children’s well-being; that is, corporal punishment may be less strongly related to depression for children who receive high levels of parental monitoring. Because parental monitoring is likely to create a more informed context for discipline, corporal punishment may be more expected (i.e., less impulsive) and/or more often perceived as fair or justi- fied. Greater parental involvement may also increase children’s perceptions of parental support, offsetting negative effects of punishment.
Another potentially important factor for understanding the effects of corporal punishment is the frequency of other types of punishment. At a minimum, it is important to control for the frequency of other forms of discipline when considering the effects of corporal punishment. Without such controls, one could not distinguish between the impact of corporal punishment and the effects of parental intervention more generally. How- ever, it is also possible that the frequency of other forms of discipline con- ditions the effects of corporal punishment. We suggest two alternative hypotheses concerning the direction of this interaction.
First, corporal punishment may be more detrimental in the context of minimal other punishment. This may occur because children who receive corporal punishment along with other forms of discipline interpret corpo- ral punishment in a broader context of parents setting boundaries and thus do not perceive it as especially threatening. Thus, children who are fre- quently punished in a variety of ways (including corporal punishment) may be less affected by any one form of punishment. Conversely, it is pos- sible that frequent other punishment exacerbates the deleterious effects of corporal punishment. It may be that children who are already emotionally taxed by high levels of other discipline become overwhelmed by corporal punishment, whereas children who receive low levels of other punishment are better able to cope with its potentially harmful effects. This hypothesis suggests that excessive punishment of any kind will be detrimental to the well-being of children and that such effects can be synergistic.
Finally, there is reason to suspect that children’s perceptions regarding the normative nature of corporal punishment will affect the extent to which corporal punishment is harmful; that is, when children believe that the level of corporal punishment they receive is similar to or less than the punishment that other children receive, it may be less detrimental to well- being. Thus, we hypothesize that the perceived normative level of corpo- ral punishment also represents a potential moderator of actual corporal punishment on later depression.
766 JOURNAL OF FAMILY ISSUES / September 2004
POTENTIAL MEDIATORS
In addition to identifying moderating conditions that may reduce or ex- acerbate the effects of corporal punishment, the identification of factors that mediate between corporal punishment and depression—the mecha- nisms by which corporal punishment in childhood becomes translated into greater depressive symptomatology in young adulthood—would fur- ther increase our understanding of this association. Although child ag- gression associated with corporal punishment may involve the modeling of parental behavior, we suggest that psychological outcomes such as de- pression likely stem from changes in self-concept. Specifically, we hy- pothesize that mastery and self-esteem represent potential mediators in the corporal punishment-depression relationship.
Corporal punishment may inhibit the development of a healthy self- concept, which, in turn, may produce negative emotional outcomes in adulthood. Research supporting this idea (Bryan & Freed, 1982) found that participants who reported high levels of corporal punishment were more likely to describe their grades as “below average” despite no actual difference in grade level. The authors interpret these findings as evidence of a damaged self-concept. Key components of the self-concept are mas- tery and self-esteem. Self-esteem has been defined as, “the evaluation which the individual makes and customarily maintains with regard to himself or herself: it expresses an attitude of approval or disapproval to- ward oneself” (Rosenberg, 1965, p. 5). Mastery is a concept related to per- ceived causal relevance and “concerns the extent to which one regards one’s life-chances as being under one’s own control in contrast to being fatalistically ruled” (Pearlin & Schooler, 1978, p. 5).
Turner and Roszell (1994) reviewed an impressive body of empirical evidence showing the importance of mastery and self-esteem for predict- ing depression. They stated that, “A substantial and rather consistent body of evidence has accumulated on the connection between mastery or con- trol and the occurrence of psychological distress” and “research has con- tinued to accumulate indicating a significant inverse correlation between self-esteem and depressive symptomatology” (pp. 184, 192). In a study of adolescent depression, Robertson and Simons (1989) found that parental rejection was directly related to depressive symptoms. In addition, the re- searchers found evidence that parental rejection affected depression indi- rectly through a reduction in self-esteem. Very similar to Bryan and Freed’s (1982) suggestion, corporal punishment experienced in child- hood may be related to negative outcomes in adulthood—such as depression—through the mediating influence of mastery and self-esteem.
Turner, Muller / LONG-TERM EFFECTS OF CORPORAL PUNISHMENT 767
The purpose of the current study is (a) to examine the relationship be- tween the corporal punishment of children and depressive symptoms in young adults, controlling for sociodemographic factors and two poten- tially confounding variables (physical or sexual abuse and the frequency of other forms of discipline); (b) to determine whether certain conditions (race, parental monitoring, perceived normative levels of corporal punish- ment, frequency of other forms of discipline) moderate the impact of cor- poral punishment on depressive symptoms; and (c) to examine the extent to which mastery and self-esteem mediate the association between child- hood exposure to corporal punishment and depression in young adulthood.
METHOD
SAMPLE AND PROCEDURE
The current study is based on a sample of 649 individuals attending one of three colleges in the New England area. These include a university con- sisting largely of White, middle-class students, many of whom come from small, semirural communities; a state college consisting of a mixture of working-class White, Hispanic, African American, and Asian students living in a medium-sized urban community; and an inner-city community college consisting of mostly lower-income African American and His- panic students who live in a large urban center. Although we acknowledge that college students are not typically representative of all young adults, we were able to increase the diversity of the sample by obtaining students from colleges that enroll individuals of differing socioeconomic statuses, racial backgrounds, and urbanicities. Of the sample, 20% are non-White, and 40% of respondents came from households where the main provider had less than a college degree. The sample included students ranging in age from 18 to 29 years, although 95% of the sample is younger than age 25 (median age = 19 years). The sample is 41% men and 59% women.
The majority of the sample (approximately 65%) was obtained through a random sample of student registration directories. The response rate for this part of the sample was 86%. The sample also includes students who were recruited through a variety of college classes within the liberal arts. Response rates within classes ranged from 60% to 95%. Given favorable response rates and our success in identifying and recruiting respondents with varied sociodemographic characteristics, we believe the sample is reasonably representative of a diverse New England college population.
768 JOURNAL OF FAMILY ISSUES / September 2004
Nevertheless, we acknowledge that any college sample is likely to under- represent the most distressed and/or the most disadvantaged individuals.
Face-to-face and telephone interview modes were used (18% in-per- son; 82% telephone). There were no significant differences in reported rates of corporal punishment across the two types of interviews, and the magnitude of the associations of interest in the current study did not sig- nificantly differ by mode of data collection. Graduate students and profes- sional survey research interviewers conducted interviews. All interview- ers attended extensive training sessions and were monitored closely throughout the survey. Respondents were paid $10 for their participation.
MEASURES
Corporal punishment. Following measurement recommendations out- lined by Straus (1998), corporal punishment was assessed using the refer- ent age of 13 years and a list of specific corporal punishment acts. Specifi- cally, corporal punishment was assessed with five items: “How many times during that year (when you were 13) did a parent or guardian . . . 1) spank or slap you on your on the bottom with his or her hand; 2) hit you on the bottom with something like a belt, hairbrush, stick, or other hard ob- ject; 3) slap you on the hand, arm, or leg with his or her hand; 4) slap you on the face, head, or ears with a hand; and 5) shake you.” Responses ranged from 0 (never) to 6 (more than 20 times). The items were summed. Given a positively skewed distribution, three categories representing level of frequency of corporal punishment were created: 0 = never, 1 = 1 to 3 times (moderate), 2 = 4 or more times (high). The alpha coefficient for this measure is .80.
A number of investigators have documented reliability problems asso- ciated with recall data (Kessler & Wethington, 1991; Prusoff, Merikangas, & Weissman, 1988). In the current study, we asked respon- dents to recall the number of times a parent disciplined them with corporal punishment during a focal year that occurred an average of 6 years prior to the interview year. Although we acknowledge the potential for bias in such reports, there are grounds for believing that these data are reasonably accurate. First, interviewers worked with respondents to establish a con- text for the recall year by using a modified “life calendar approach” (Freedman, Thornton, Camburn, Alwin, & Young-DeMarco, 1988). Year in school (the eighth grade for most respondents), place of residence, and household composition during the focal year (age 13) was established to help respondents recall events occurring in that particular year. This tech- nique has been used successfully in another study of young adults (Turner
Turner, Muller / LONG-TERM EFFECTS OF CORPORAL PUNISHMENT 769
& Lloyd, 2002) and has been shown to increase the reliability of recall and the temporal ordering of events (Freedman et al., 1988).
We are also encouraged by past studies showing consistency between data provided by adults recalling the frequency of their childhood corpo- ral punishment and contemporary data on rates of corporal punishment provided by parents (Straus, 1994). Indeed, the rates of corporal punish- ment reported by respondents in the current study are virtually identical to current national prevalence rates for 13-year-olds reported by parents (see Results).
Depressive symptomatology. Symptoms of depression were assessed by the Center for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977). Respondents indicated how often over the preceding 2 weeks they had experienced each of 20 symptoms on a 4-point scale rang- ing from 0 (rarely or none of the time) to 3 (most or all of the time). A sum- mary of the 20 items was constructed. The validity and reliability of this scale are well established (Radloff, 1977). In the current study, the reli- ability coefficient for the CES-D is .89.
Other discipline. Frequency of other discipline was measured by the sum of four items. Respondents were asked to recall the frequency of vari- ous other forms of punishment they received when they were around 13 years old. Specifically, respondents indicated how many times a parent or guardian responded to misbehavior in the following ways: (a) “talked to you calmly about something you did wrong”; (b) “took something (like allowance or television) because you did something wrong”; (c) “grounded you, or not let you go someplace, because you did something wrong”; and (d) “yelled or swore at you because you did something wrong.” Responses ranged from 0 (never) to 6 (more than 20 times).
Child abuse. A series of questions were asked concerning 32 poten- tially traumatic life events that may have occurred anytime during the re- spondent’s life, including in childhood. Six of these items were used in constructing a measure of parental child abuse: “In your whole life, were you ever forced or threatened into having sexual intercourse when you didn’t want to?”; “[Other than that/those time(s)] has there ever been a time (including when you were a child or teenager) when someone touched your genitals [or breasts] or made you touch their private parts when you didn’t want him or her to?”; “In your whole life, have you ever been badly beaten up—punched, kicked, or hit very hard—by a family member, like a parent, stepparent, sibling, or other relative?”; “In your
770 JOURNAL OF FAMILY ISSUES / September 2004
whole life, have you ever been actually shot with a gun or injured with some other weapon, like a knife or bat?”; “In your whole life, has someone (including friends, family members, or strangers) ever threatened or at- tacked you with a gun, knife, or some other weapon even though you were not injured?”; and “In your whole life, have you ever been chased, but not caught, by a gang, ‘bully,’ or someone you were frightened of, when you thought you could really get hurt?” If respondents indicated having expe- rienced an event, they were then asked a series of detailed probes, includ- ing who was involved in the incident. If any of these events had ever oc- curred in which a parent or guardian was the perpetrator, respondents received a 1 on the child abuse measure. All other respondents were given a 0. Thus, parental child abuse represents a dichotomous variable (1 = his- tory of abuse; 0 = no history of abuse). A total of 30 respondents (4.6%) reported a history of abuse by a parent or guardian.
Monitoring. A measure of parental monitoring based on the sum of four items was created. Participants were asked to recall the level of moni- toring they received around age 13 by indicating how often a parent or guardian (a) knew where they were throughout the course of the day; (b) knew who they were with when they were away from home; (c) talked to them about what was going on in their life; and (d) knew if they had come home or were in bed by the set time. Responses ranged from never to very often (0 = never; 1 = seldom; 2 = sometimes; 3 = very often). The alpha co- efficient for this scale is .60.
Normativeness. A single item was used to measure the level to which respondents felt that the corporal punishment they received at about age 13 was normative. They were asked, “When you were 13, did you think that the kind of spanking, slapping, or hitting you received was worse, the same, or better than the kind of punishment other kids received from their parents?” The responses were (a) “Worse than other kids”; (b) “Same as other kids”; and (c) “Better than other kids.” Thus, a higher score reflects greater perceived norms favoring corporal punishment.
Parental anger. Parental anger during corporal punishment was as- sessed with a single item. Respondents who indicated experiencing at least one incidence of corporal punishment were asked the following question: “When you were punished in any of these ways, how often did a parent or guardian act very angry?” They responded on a 4-point scale ranging from 0 (never) to 3 (very often). Those respondents who had never received corporal punishment were also given a score of 0.
Turner, Muller / LONG-TERM EFFECTS OF CORPORAL PUNISHMENT 771
Mastery. Mastery was assessed using the mean score of an eight-item scale developed by Pearlin and Schooler (1978). Respondents rated each item of a 4-point scale ranging from strongly agree to strongly disagree. This scale has been used successfully in numerous studies, and its psychometric properties are well established. In the current study, the al- pha coefficient is .71.
Self-esteem. Self-esteem was measured with a summary score of an in- strument developed by Rosenberg (1965). This scale is also well estab- lished in the literature. It is composed of seven items reflecting different “self-statements,” or beliefs. Respondents rate each statement on a 5-point scale ranging from 1 (strongly agree) to 5 (strongly disagree). The internal reliability for this scale is .81.
Sociodemographics. Gender is a dichotomous variable (0 = male, 1 = female), while age is a continuous variable ranging from 18 to 29. Given relatively small numbers within minority subgroups in this sample, mi- nority status was collapsed into a dichotomous variable (0 = White, 1 = non-White). Respondents coded as 1 on this variable (n = 130), include Hispanic Whites (13%), Hispanic Blacks (8%), African Americans (28%), Asians (17%), and other (38%). Respondents who placed them- selves in the other category were largely non-Hispanic Caribbean Blacks and mixed-ethnicity respondents who claimed to have no dominant iden- tity. Respondents were also asked the highest level of education com- pleted by the parent who “provided the major financial support for the family or household.” Respondents answered on an 11-point scale rang- ing from grade school only to doctorate degree.
RESULTS
Of the current sample, 39% reported one or more incidents of corporal punishment at age 13 years. Based on a nationally representative sample, Straus and Stewart (1999) found that approximately 40% of parents of 13- year-olds report some instance of corporal punishment. Thus, despite our use of a college sample, longer recall periods, and child reports rather than parent reports, the prevalence of corporal punishment in this sample is almost identical to national estimates.
Bivariate correlations between all variables of interest in this research are presented (see appendix). As expected, level of corporal punishment is positively and significantly related to depressive symptoms (r = .136,
772 JOURNAL OF FAMILY ISSUES / September 2004
p < .01). History of child abuse, frequency of other forms of discipline, and parental anger during corporal punishment also show significant pos- itive associations with depression, while parental monitoring and per- ceived corporal punishment norms show no significant relationships to depression. As expected, mastery and self-esteem are inversely related to corporal punishment and to depressive symptomatology. It is worth not- ing that level of corporal punishment differs significantly by gender, race, and parental education. Men, minorities, and respondents with less- educated parents report higher levels of corporal punishment.
To address our research questions, regression analyses were per- formed. In Step 1 of Table 1, depression is regressed on corporal punish- ment and the demographic variables. Results indicate that corporal pun- ishment is a significant predictor of depression (b = 1.253, β = .158, p < .001) even after controlling for demographic characteristics.
In Step 2, abuse is added to the equation. As would be expected, history of child abuse is significantly related to depression (b = 3.449, β = .118, p < .01) while controlling for other factors. More important, when abuse is added (i.e., controlled for), corporal punishment remains a significant predictor of depression, and the strength of the association between cor- poral punishment and depression is only slightly reduced. In other words, normative corporal punishment is a significant predictor of depression independent of parental child abuse.
In Step 3, frequency of other types of punishment is added to the re- gression equation. Corporal punishment remains significantly related to depression (b = .665, β = .084, p < .05). However, there is an even stronger relationship between frequency of other forms of discipline and depres- sion (b = .208, β = .164, p < .001), independent of corporal punishment. Furthermore, the corporal punishment coefficient is reduced by about one third when other discipline is controlled. Thus, it appears that the fre- quency of discipline that respondents experienced in general accounts for part of the variation in depression explained by corporal punishment.
Next, we were interested in examining the direct and interactive effects of our hypothesized moderators of corporal punishment. In Step 4, paren- tal anger, monitoring, and perceived corporal punishment norms were en- tered into the equation. Only level of parental anger during corporal pun- ishment is statistically significant (b = 1.074, β = .197, p < .01). With this variable in the equation, level of corporal punishment is nonsignificant. The terms representing the interactions between corporal punishment and the four parenting variables and race were added to the regression equa- tion in Step 5. Each of the interactions was tested separately with all main
Turner, Muller / LONG-TERM EFFECTS OF CORPORAL PUNISHMENT 773
774
T A
B L
E 1
R eg
re ss
io n
o f
D ep
re ss
io n
o n
t h
e P
re d
ic to
r V
ar ia
b le
s: A
T es
t of
M od
er at
in g
E ff
ec ts
S te
p 1
S te
p 2
S te
p 3
S te
p 4
S te
p 5
a
b β
b β
b β
b β
b β
A ge
–. 03
8 –.
04 1
–. 03
6 –.
03 9
–. 03
6 –.
03 8
–. 03
1 –.
03 5
S ex
.9 00
.0 71
.9 07
.0 72
.8 83
.0 70
.8 65
.0 69
R ac
e .1
34 .0
09 .0
61 .0
04 .3
07 .0
20 1.
04 8
.0 67
C ar
eg iv
er e
du ca
ti on
.1 73
.0 70
.1 91
.0 77
.1 77
.0 71
.1 63
.0 65
C or
po ra
l pu
ni sh
m en
t (C
P )
1. 25
3* **
.1 58
** *
1. 05
3* **
.1 33
** *
.6 65
* .0
84 *
–. 47
2 –.
06 0
A bu
se 3.
44 9*
* .1
18 **
3. 15
7* *
.1 08
** 3.
53 3*
* .1
24 **
O th
er d
is ci
pl in
e .2
08 **
* .1
64 **
* .1
80 **
* .1
44 **
* A
ng er
1. 07
4* *
.1 97
** M
on it
or in
g .1
00 .0
25 N
or m
at iv
e .5
46 .0
59 C
P ×
O th
er p
un is
hm en
t .0
00 .0
00 C
P ×
A ng
er .3
98 .1
27 C
P ×
M on
it or
in g
.3 09
.4 17
C P
× N
or m
at iv
e .5
00 .1
36 C
P ×
R ac
e .0
27 .0
02 R
2 .0
30 **
.0 43
** *
.0 67
** *
.0 85
** *
N um
be r
of C
as es
63 9
63 9
63 8
56 7
N O
T E
: a.
I nt
er ac
ti on
s ar
e te
st ed
s ep
ar at
el y
w it
h al
l m
ai n
ef fe
ct v
ar ia
bl es
i n
th e
eq ua
ti on
. *p
< .0
5 **
p <
.0 1.
* **
p <
.0 01
.
effect variables in the equation. None of the interaction terms were significantly related to depression.
When interpreting the above analyses, it should be noted that some portion of corporal punishment level is incorporated in the parental anger variable, because all respondents who never received corporal punish- ment also received a 0 on this variable; that is, because anger during pun- ishment is only possible among those experiencing some level of corporal punishment, the positive association implies that any level of corporal punishment in the context of high parental anger is detrimental to psycho- logical well-being. Thus, these analyses suggest that it is corporal punish- ment when parents are very angry, rather than the level of corporal punish- ment itself, that is most important. The nonsignificant interaction between corporal punishment and parental anger suggests that high anger is equally problematic at moderate and high levels of corporal punishment.
Given the high colinearity between the corporal punishment and pa- rental anger variables, we conducted the same analyses among only the 39% of respondents who reported one or more incidents of corporal pun- ishment (analyses not shown). Thus, these analyses avoid the problem of confounding between level of corporal punishment and level of anger. Among those who reported at least one incident of corporal punishment, 35% reported that their “punishing parent” was either “never” or “sel- dom” angry, 32% reported that he or she was “sometimes” angry, and 33% reported that he or she was “very often” angry during corporal punish- ment. Consistent with the above interpretation, we found that the amount of parental anger during corporal punishment was more strongly related to depression than was high versus moderate levels of corporal punishment.
To test the hypothesis that mastery and self-esteem are mediators of the relationship between corporal punishment and depression, a second series of regression analyses was performed. In Step 2 of Table 2, only the vari- ables that were significantly related to depression in earlier analyses were added to the base model. Thus, history of child abuse, angry corporal pun- ishment, and level of other forms of discipline all have significant inde- pendent effects. It is worth noting that level of anger during corporal pun- ishment represents the strongest predictor of depression among all the parenting variables considered. In Step 3 of Table 2, mastery and self- esteem are added to the regression equation. Both of these components of self-concept have significant independent effects on depression (β = –.118, p < .05; β = –.265, p < .001, respectively). Moreover, adding mastery and self-esteem reduces the effect of angry corporal punishment and history of abuse to nonsignificance. Only frequency of other discipline remains
Turner, Muller / LONG-TERM EFFECTS OF CORPORAL PUNISHMENT 775
significantly related to depressive symptoms when self-esteem and mas- tery are controlled. This provides some evidence that mastery and self-es- teem are mediators by which parental anger during corporal punishment (and parental abuse) have long-term effects on depression. The full model accounts for 19.5% of the variance in depressive symptoms.
SUMMARY AND DISCUSSION
The purpose of the current study was to examine the relationship be- tween corporal punishment and depression, and to better specify this as- sociation by (a) identifying conditions under which corporal punishment is less (or more) strongly related to depression and (b) determining the ex- tent to which aspects of self-concept mediate this relationship.
A number of interesting findings emerged from this research. As hy- pothesized, corporal punishment experienced at age 13 years is signifi- cantly related to level of depressive symptomatology in young adulthood. This association remains significant when controlling for age, race, sex, and parent’s education. Moreover, the relationship between corporal pun- ishment and later depression cannot be explained by abuse experiences. Indeed, level of corporal punishment at age 13 years represents a stronger
776 JOURNAL OF FAMILY ISSUES / September 2004
TABLE 2 Regression of Depression on the Predictor Variables:
A Test of Mediating Effects
Step 1 Step 2 Step 3
b β b β b β
Age –.038 –.041 –.039 –.042 –.042 –.045 Sex .900 .071 .840 .067 .728 .058 Race .134 .009 .256 .017 .215 .014 Caregiver education .173 .070 .165 .066 .119 .048 Corporal punishment 1.253*** .158*** –.468 –.059 –.331 –.042 Abuse 2.821* .097* 1.816 .063 Other discipline .186*** .146*** .161*** .126*** Anger 1.028** .184** .680 .121 Mastery –1.569* –.118* Self-esteem –2.879*** –.265*** R
2 .030** .078*** .195*** Number of Cases 639 636 633
*p < .05. **p < .01. ***p < .001.
predictor of depression than having experienced an assault by a parent or guardian at some point in their childhood (Table 1, Step 2).
Although the effect of corporal punishment is reduced when the level of other forms of discipline is considered, it remains a statistically signifi- cant predictor of depression. Thus, the impact of corporal punishment is not explained by high levels of discipline more generally. However, it is important to note that the level of other types of discipline is also signifi- cantly related to depression. Given research on the negative effects of parenting practices that emphasize punishment rather than positive rein- forcement (i.e., authoritarian parenting), perhaps this finding is not sur- prising. More surprising, however, is the finding that the frequency of other forms of discipline was more strongly related to depression than was level of corporal punishment. This provides some support for the notion that it is frequent discipline in general—rather than corporal punishment per se—that is most problematic for children.
Despite a stronger relationship between other discipline and depres- sion, it would be unwarranted to disregard the finding that corporal pun- ishment has significant long-term effects on depression. In fact, we found that corporal punishment influences later depressive symptoms independ- ent of a number of contextual factors. Moreover, high parental anger dur- ing corporal punishment has particularly detrimental effects. This finding has considerable substantive significance, because parental anger is likely to be common during discipline in general, and during corporal punish- ment in particular. Because corporal punishment is often viewed as a last resort strategy (used only when other discipline has not been effective), it may often be accompanied by anger and frustration. It also seems likely that anger decreases parents’ abilities to be consistent, controlled, and nonimpulsive in their discipline—behaviors found to be important for a variety of child outcomes (Agnew, 1983; Elder & Caspi, 1988; Patterson, 1988; Straus & Mouradian, 1998).
Results from the regression of interaction terms show that none of the parenting or demographic characteristics considered in this research in- teracted with corporal punishment to differentially affect depression. Thus, unlike some previous studies (Gunnoe & Mariner, 1997; Larzalere, 1986; Straus & Mouradian, 1998; Turner & Finkelhor, 1996), we did not find evidence that corporal punishment is modified by demographic or pa- rental factors. However, as discussed earlier, the positive relationship be- tween parental anger and depression does suggest the importance of cor- poral punishment contexts. Analyses show that level of parental anger during corporal punishment is more predictive of depressive symptoms than level of corporal punishment per se. However, given how parental an-
Turner, Muller / LONG-TERM EFFECTS OF CORPORAL PUNISHMENT 777
ger is assessed in the current study, we were unable to completely separate the effects of level of corporal punishment from those of parental anger. Future research would benefit from an examination of parental anger in general (e.g., during all types of discipline) to more accurately assess its independent effects. A separate assessment of impulsiveness and consistency of discipline would further help to specify the most important parenting contexts and practices.
In our second analysis (Table 2), mastery and self-esteem were added to the regression equation in an attempt to understand the processes in- volved in the relationship between corporal punishment (and/or other punishment) and depression. Results indicate that mastery and self- esteem are significantly related to depression—an expected finding given earlier research on the relationship between self-concept and depression. Moreover, mastery and self-esteem explained the effects of angry corpo- ral punishment and history of abuse on depressive symptoms. Thus, mas- tery and self-esteem appear to represent mediators in the relationship be- tween angry corporal punishment and depression. It should be acknowledged, however, that because mastery, self-esteem, and depres- sion are all contemporaneous measures, it is also plausible that depression affects self-concept rather than the reverse. Although reciprocal effects do seem likely, the strength of past research linking mastery and self-esteem to subsequent depressive symptoms (e.g., Kaplan, Robbins, & Martin, 1983; Pearlin, Lieberman, Menaghan, & Mullan, 1981; Turner & Rozell, 1994) suggests that some significant portion of the associations is likely to reflect the influence of self-concept on depression.
Our findings suggest that normative corporal punishment when par- ents are angry may operate in much the same way as more severe physical or sexual abuse to affect mental health. In both cases, violence toward children by caregivers likely disrupts the normal development of self- concept, leaving children with a poorer self-image and a reduced sense of control over their lives. As Straus and Mouradian (1998) suggested with respect to impulsivity, corporal punishment when parents are very angry may lead the child to experience the act as a result of the parent’s feelings about them rather than as something done for the child’s well-being. If corporal punishment is perceived as parent centered, it may be more likely to create resentment and undermine early parent-child attachment. Thus, attributing parental violence (even normative violence) to the parent’s feelings, opinions, or beliefs may have repercussions for the child’s self- assessments. Lower self-esteem and mastery, in turn, result in greater lev- els of depressive symptomatology. In this way, corporal punishment in
778 JOURNAL OF FAMILY ISSUES / September 2004
779
A P
P E
N D
IX Z
er o-
O rd
er C
or re
la ti
on s
of A
ll R
el ev
an t
V ar
ia b
le s
1 2
3 4
5 6
7 8
9 1 0
1 1
1 2
1 3
1. D
ep re
ss io
n .—
2. A
ge –.
02 3
.— 3.
S ex
.0 55
.0 08
.— 4.
R ac
e .0
22 .1
02 **
–. 03
4 .—
5. C
ar eg
iv er
e du
ca ti
on .0
51 –.
11 0*
* –.
02 8
–. 17
8* *
.— 6.
C or
po ra
l pu
ni sh
m en
t .1
36 **
.1 41
** –.
12 8*
* .1
93 **
–. 12
9* *
.— 7.
A bu
se .1
39 **
.0 21
–. 02
8 .0
92 *
–. 09
1* .2
23 **
.— 8.
O th
er d
is ci
pl in
e .1
99 **
.0 24
–. 01
7 –.
03 7
.0 05
.2 85
** .1
16 **
.— 9.
A ng
er .1
93 **
.1 33
** –.
07 6
.1 66
** –.
07 6
.8 34
** .2
46 **
.3 27
** .—
10 . M
on it
or in
g –.
00 3
–. 08
4* .1
52 **
–. 13
4* *
.1 20
** –.
20 8*
* –.
13 2*
* .0
19 –.
22 4*
* .—
11 . N
or m
at iv
e –.
05 0
–. 10
9* *
.0 30
–. 04
4 .0
75 –.
40 5*
* –.
27 0*
* –.
19 2*
* –.
40 5*
* .1
72 **
.— 12
. M as
te ry
–. 29
3* *
–. 04
8 .0
50 –.
08 8*
.0 00
–. 17
2* *
–. 04
7 –.
11 8*
* –.
17 2*
* .0
94 *
.1 15
** .—
13 . S
el f-
es te
em –.
37 9*
* –.
02 7
–. 06
1 –.
02 0
–. 01
3 –.
11 7*
* –.
15 4*
* –.
09 5*
–. 18
8* *
.1 40
** .1
20 **
.6 10
** .—
*p <
.0 5.
* *p
< .0
1 (t
w o-
ta il
ed ).
childhood results in later depression through its damaging effect on self- conceptions.
Results suggest that mastery and self-esteem have little mediating in- fluence with respect to the association between other forms of discipline and depression. It may be that excessive punishment that is nonviolent (e.g., taking away privileges, yelling and swearing, being grounded, etc.) increases depressive symptoms, not by damaging self-concept but by damaging the ongoing parent-child relationship. Future research should attempt to better specify the conditions and mechanisms by which differ- ent parenting behaviors and discipline strategies affect psychological well-being.
In conclusion, this research demonstrates some of the complexity in- volved with these issues. The relationship between punishment (corporal and otherwise) and problem outcomes is not entirely straightforward. The current analysis suggests the importance of examining the effects of pun- ishment generally, along with the effects of corporal punishment specifi- cally. The significance of parental anger as a context for understanding the impact of corporal punishment is also suggested by this research. Further- more, results from the current study point to the processional nature of the relationship between punishment and depression, particularly through the mediating influence of elements of the self-concept.
REFERENCES
Agnew, R. (1983). Physical punishment and delinquency: A research note. Youth & Society, 15, 225-236.
Baumrind, D. (1972). An exploratory study of socialization effects on Black children: Some Black-White comparisons. Child Development, 43, 261-262.
Bryan, J. W., & Freed, F. W. (1982). Corporal punishment: Normative data and sociological and psychological correlates in a community college population. Journal of Youth and Adolescence, 11, 77-87.
Carlin, A. S., Kemper, K., Ward, N. G., Sowell, H., Gustafson, B., & Stevens, N. (1994). The effect of differences in objective and subjective definitions of childhood physical abuse on estimates of its incidence and relationship to psychopathology. Child Abuse and Ne- glect, 18, 393-399.
Dodge, K. A., Bates, J. E., & Pettit, G. S. (1990). Mechanisms in the cycle of violence. Sci- ence, 250, 1678-1683.
Dornbusch, S. M., Ritter, P. L., Leiderman, P. H., Roberts, D. J., & Fraleigh, M. J. (1987). The relation of parenting styles to adolescent performance. Child Development, 58, 1244- 1257.
Elder, G. H., & Caspi, A. (1988). Human development and social change: An emerging per- spective on the life course. In N. Bolger, A. Caspi, G. Downey, & M. Moorehouse (Eds.),
780 JOURNAL OF FAMILY ISSUES / September 2004
Persons in context: Developmental processes (pp. 77-113). Cambridge, UK: Cambridge University Press.
Freedman, D., Thornton, A., Camburn, D., Alwin, D., & Young-DeMarco, L. (1988). The life history calendar: A technique for collecting retrospective data. Sociological Method- ology, 18, 37-68.
Gordon, D., Burge, D., Hammen, C., Adrian, C., Haenick, C., & Hiroto, D. (1989). Observa- tions of interactions of depressed women with their children. American Journal of Psy- chiatry, 146, 6-18.
Gunnoe, M. L., & Mariner, C. L. (1997). Toward a developmental-contextual model of the ef- fects of parental spanking on childrens aggression. Archives of Pediatrics and Adolescent Medicine, 151, 768-775.
Heffer, F. W., & Kelley, M. L. (1987). Mothers acceptance of behavioral interventions for children: The influence of parent race and income. Behavior Therapy, 2, 153-163.
Holmes, S. J., & Robins, L. N. (1988). The role of parental disciplinary practices in the devel- opment of depression and alcoholism. Psychiatry, 51, 24-36.
Howing, P. T., Wodarsky, J. S., Kurtz, P. D., Gaudin, J. M., & Herbst, E. M. (1990). Child abuse and delinquency: The empirical and theoretical links. Social Work, 35, 244-249.
Jarrett, R. L. (1997). Resilience among low-income African American youth: An ethno- graphic perspective. Ethos, 25, 218-229.
Kaplan, H. B., Robins, C., & Martin, S. S. (1983). Antecedents of psychological distress in young adults: Self-rejection, deprivation of social support and life events. Journal of Health and Social Behavior, 24, 230-244.
Kessler, R. C., & Wethington, E. (1991). The reliability of life events reports in a community survey. Psychological Medicine, 21, 723-738.
Kolko, D. J. (1992). Characteristics of child victims of physical violence. Journal of Inter- personal Violence, 7, 244-276.
Larzelere, R. E. (1986). Moderate spanking: Model or deterrent of childrens aggressive be- havior in the family? Journal of Family Violence, 1, 27-36.
Malinosky-Rummel, R., & Hansen, D. J. (1993). Long-term consequences of childhood physical abuse. Psychological Bulletin, 114, 68-79.
Patterson, G. R. (1986). Performance models for antisocial boys. American Psychologist, 41, 432-444.
Patterson, G. R. (1988). Family process: Loops, levels, and linkages. In N. Bolger, A. Caspi, G. Downey, & M. Moorehouse (Eds.), Persons in context: Developmental processes (pp. 114-151). Cambridge, UK: Cambridge University Press.
Pearlin, L. I., Lieberman, M., Menaghan, E., & Mullan, J. (1981). The stress process. Journal of Health and Social Behavior, 22, 337-356.
Pearlin, L. I., & Schooler, C. (1978). The structure of coping. Journal of Health and Social Behavior, 19, 2-21.
Prusoff, B. A., Merikangas, K. R., & Weissman, M. M. (1988). Lifetime prevalence and age of onset of psychiatric disorder: Recall 4 years later. Journal of Psychiatric Research, 22, 107-117.
Radloff, L. S. (1977). The CES-D scale: A new self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385-401.
Robertson, J. F., & Simons, R. L. (1989). Family factors, self-esteem, and adolescent depres- sion. Journal of Marriage and the Family, 51, 125-138.
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton Uni- versity Press.
Turner, Muller / LONG-TERM EFFECTS OF CORPORAL PUNISHMENT 781
Simons, R. L., Lin, K., & Gordon, L. C. (1998). Socialization in the family of origin and male dating violence: A prospective study. Journal of Marriage and the Family, 60, 467-478.
Simons, R. L., Whitbeck, L. B., Conger, R. D., & Wu, C. (1991). Intergenerational transmis- sion of harsh parenting. Developmental Psychology, 27, 159-171.
Spencer, M. B., Dupree, D., Swanson, D. P., & Cunningham, M. (1996). Parental monitoring and adolescents sense of responsibility for their own learning: An examination of sex dif- ferences. Journal of Negro Education, 65, 30-43.
Straus, M. A. (1991). Discipline and deviance: Physical punishment of children and violence and other crimes in adulthood. Social Problems, 38, 133-154.
Straus, M. A. (1994). Beating the devil out of them: Corporal punishment in American fami- lies and its effect on children. Boston: Lexington.
Straus, M. A. (1998). Spanking by parents: Some ideas on measurement and analysis of a ne- glected risk factor for serious mental health problems. Behavioral Measurements Letter, 15, 3-8.
Straus, M. A., Gelles, R. J., & Steinmetz, S. K. (1980). Behind closed doors: Violence in the American family. Beverly Hills, CA: Sage.
Straus, M. A., & Kaufman Kantor, G. (1994). Corporal punishment of adolescents by par- ents: A risk factor in the epidemiology of depression, suicide, alcohol abuse, child abuse, and wife beating. Adolescence, 29, 543-561.
Straus, M. A., & Lauer, S. (1992, November). Corporal punishment of children and crime in ethnic group context. Paper presented at the meeting of the American Society of Crimi- nology, New Orleans, LA.
Straus, M. A., & Mathur, A. K. (1996). Social change and the trends in approval of corporal punishment by parents from 1968 to 1994. In D. Frehsee, W. Horn, & K. Bussman (Eds.), Family violence against children: A challenge for society (pp. 91-105). Berlin, Germany: DeGruyter.
Straus, M. A., & Mouradian, V. E. (1998). Impulsive corporal punishment by mothers and an- tisocial behavior and impulsiveness of children. Behavioral Sciences and the Law, 16, 353-374.
Straus, M. A., & Stewart, J. H. (1999). Corporal punishment by American parents: National data on prevalence, chronicity, severity, and duration, in relation to child and family char- acteristics. Clinical Child and Family Psychology Review, 2, 55-70.
Straus, M. A., Sugarman, D. B., & Giles-Sims, J. (1997). Spanking by parents and subse- quent antisocial behavior of children. Archives of Pediatrics and Adolescent Medicine, 151, 761-767.
Turner, H. A., & Finkelhor, D. (1996). Corporal punishment as a stressor among youth. Jour- nal of Marriage and the Family, 58, 155-166.
Turner, R. J., & Lloyd, D. A. (2002, April). Race/ethnic contrasts in mental health and sub- stance abuse problems: The significance of lifetime exposure to major and potentially traumatic events. Paper presented at the Eighth International Conference on Social Stress, Portsmouth, NH.
Turner, R. J., & Roszell, P. (1994). Psychosocial resources and the stress process. In W. R. Avison & I. H. Gotlib (Eds.), Stress and mental health: Contemporary issues and pros- pects for the future (pp. 179-210). New York: Plenum.
782 JOURNAL OF FAMILY ISSUES / September 2004
