Journal of Interpersonal Violence http://jiv.sagepub.com
Predictors of Adult Attitudes Toward Corporal Punishment of Children Marie-Hélène Gagné, Marc Tourigny, Jacques Joly and Joëlle Pouliot-Lapointe J Interpers Violence 2007; 22; 1285 DOI: 10.1177/0886260507304550 The online version of this article can be found at: http://jiv.sagepub.com/cgi/content/abstract/22/10/1285
Published by: http://www.sagepublications.com
On behalf of: American Professional Society on the Abuse of Children
Additional services and information for Journal of Interpersonal Violence can be found at: Email Alerts: http://jiv.sagepub.com/cgi/alerts Subscriptions: http://jiv.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations (this article cites 37 articles hosted on the SAGE Journals Online and HighWire Press platforms): http://jiv.sagepub.com/cgi/content/refs/22/10/1285 Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Predictors of Adult Attitudes Toward Corporal Punishment of Children
Journal of Interpersonal Violence Volume 22 Number 10 October 2007 1285-1304 © 2007 Sage Publications 10.1177/0886260507304550 http://jiv.sagepub.com hosted at http://online.sagepub.com
Marie-Hélène Gagné Laval University, Quebec, Canada
Marc Tourigny Jacques Joly Sherbrooke University, Quebec, Canada
Joëlle Pouliot-Lapointe Laval University, Quebec, Canada
This study identifies predictors of favorable attitudes toward spanking. Analyses were performed with survey data collected from a representative sample of 1,000 adults from Quebec, Canada. According to this survey, a majority of respondents endorsed spanking, despite their recognition of potential harm associated with corporal punishment (CP) of children. The prediction model of attitudes toward spanking included demographics, experiencing or witnessing various forms of family violence and abuse in childhood, and perceived frequency of physical injuries resulting from CP. Spanking was the most reported childhood experience (66.4%), and most violence and abuse predictors were significantly and positively correlated. Older respondents who were spanked in childhood and who believed that spanking never or seldom results in physical injuries were the most in favor of spanking. On the other hand, respondents who reported more severe physical violence or psychological abuse in childhood were less in favor of spanking. Findings are discussed in terms of prevention of CP and family coercion cycle. Keywords: attitudes; child abuse; child discipline; corporal punishment; parenting practices; spanking; survey
Authors’ Note: This investigation used data collected in a population survey sponsored by the Marie-Vincent Foundation. We would like to thank the Foundation for allowing access to these data. Please address correspondence to Marie-Hélène Gagné, Associate Professor, School of Psychology, Laval University, Quebec City (Quebec), G1K 7P4, Canada; e-mail:
[email protected]. 1285 Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1286
Journal of Interpersonal Violence
C
orporal punishment (CP) of children is a frequent parenting practice in several cultures, though its level of acceptance varies widely. Many Scandinavian countries are known for their laws banning CP by parents (Larzelere & Johnson, 1999). In other cultures, however, harsh disciplining of children is the norm: Smith and Mosby (2003) presented the case of Jamaica to illustrate this point. It is recognized that cultural values and childrearing standards play an important role in defining what is child abuse and what it is not (Korbin, 1991). Individual attitudes toward CP reflect these values and standards. Research has shown that favorable parental attitudes toward CP are a significant predictor for its use with children (Durrant, Rose-Krasnor, & Broberg, 2003; Holden, Coleman, & Schmidt, 1995; Pinderhughes, Dodge, Bates, Pettit, & Zelli, 2000). However, less research has tried to understand what variables can predict the adoption of such attitudes in parents and adults in general. This is the aim of the current study.
Controversy About CP According to Straus (1994), CP refers to “the use of physical force with the intention of causing a child to experience pain, but not injury, for the purpose of correction or control of the child’s behavior” (p. 4). In Northern American cultures, CP is a highly controversial, but still legal, parenting practice.1 The main question sustaining this controversy is, “Is CP really detrimental for children?” For some authors (Gershoff, 2002a, 2002b; Rizzo, 2002; Straus & Donnelly, 2001), the answer is clearly yes. Their opinion is supported by an important body of research showing the links between CP and a wide range of youth problems and by the lack of demonstrated positive effects of CP. Other authors (Baumrind, Larzelere, & Cowan, 2002; Larzelere, 2000; Saadeh, 2002) argued that the detrimental effects of CP on children have not been demonstrated and that CP has some positive outcome in terms of child’s compliance to parents and control of child’s misbehavior. Given that CP is used adequately (only with children between age 2 and 12 years, not with frustration or anger, not in the head or face, etc.), these experts believe that it is an effective and legitimate form of discipline. Unfortunately, the use of CP does not always follow these guidelines: Frustration with the child is an important predictor for the use of CP by parents (Wissow, 2002), and many infants and very young children experience CP (Clément, Bouchard, Jetté, & Laferrière, 2000; Straus & Stewart, 1999). Benjet and Kazdin (2003) underlined other disturbing facts that reach consensus among experts. First, it is recognized that CP is correlated with more severe physical violence, psychological abuse, and other familial and
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Gagné et al. / Attitudes Toward Corporal Punishment
1287
contextual risks. Second, frequent CP (1-3 times a week or more) is clearly associated with negative outcomes in children (externalizing and mental health problems, poor competencies). Recent research suggests that frequent spanking is also associated with children’s hormonal reactivity to stress (Bugental, Martorell, & Barraza, 2003). Finally, according to the cultural spillover hypothesis, this believed-tobe minor form of violence could act as a distal, macrosystemic variable contributing to the general level of societal violence (Straus & Donnelly, 2001). To support this hypothesis, Straus (1996) underscored that the level of cultural acceptance of CP is associated with the overall homicide rate and with the homicide rate for children in 10 European countries. All these acknowledgments and hypotheses suggest that CP, even if neither necessary nor sufficient to explain behavioral or emotional problems in youths (Gershoff, 2002a; Holden, 2002), is not benign and deserves attention.
Prevalence of CP The high prevalence of CP is another matter of concern. In the United States, a Gallup survey conducted in 1995 with parents of a representative sample of 1,000 children shows that 61.4% of children younger than age 18 years sustain minor forms of physical violence from their parents (Straus, Hamby, Finkelhor, Moore, & Runyan, 1998). The CP scale used in this survey included six items from the Parent-Child Conflict Tactics Scales: spanked on bottom with bare hand; slapped on the hand, arm, or leg; hit on the bottom with an object; pinched; slapped on the face, head, or ears; and shook for children older than age 2 years. Using the same data, Straus and Stewart (1999) underlined that almost all toddlers and preschool children get physically punished by parents and that more than one third of CP starts as an infant. In a similar survey conducted in Quebec, Canada, with 2,469 mothers representative of the general population, the overall rate of CP was 47.8% (Clément et al, 2000).2 Again, children between age 3 and 6 years were the most frequently punished, and almost one half of the children age 0 and 2 years had experienced CP. According to Wissow (2002) however, studies are inconsistent on whether boys are more likely to be spanked than girls.
The Importance of Attitudes Favorable attitudes toward CP have consistently been associated with the use of this behavior in recent research. In a small sample of mother and 3-year-old children dyads, Holden et al. (1995) found that maternal attitudes toward spanking, but not maternal mood, was a significant predictor Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1288
Journal of Interpersonal Violence
of the frequency of spanking. Among all other predictors, attitudes had the clearest association with the frequency of spanking. In a cross-cultural study, Durrant et al. (2003) found that Canadian mothers expressed more favorable attitudes toward CP than Swedish mothers, even if both groups were not in favor of CP in general. Also, the former used CP more often than the latter, and attitudes toward CP were significantly and positively related to the frequency of CP in both groups. Pinderhughes et al. (2000) studied the precursors to parents’ harsh discipline in 585 families (423 mothers and 300 fathers). Belief in spanking was significantly and positively correlated with indicators of physical discipline (r = .12-.30). Moreover, parents’ belief in spanking mediated the negative relation between socioeconomic status (SES) and discipline. These findings were similar for mothers and fathers. Moreover, with a small convenience sample of 22 general population parents and 9 at-risk parents (77.8% female), Crouch and Behl (2001) found that level of parenting stress was positively associated with physical child abuse potential, but only among parents who reported high levels of belief in the value of CP. All these findings support the cognitive-instrumental view of spanking: Parents spank because they believe it is a useful disciplinary technique for teaching the child how to behave. In fact, the major reason why parents spank is disobedience (Andero & Stewart, 2002).
Predictors of Favorable Attitudes Toward CP With regards to CP, the link between attitudes and behavior suggests that modification of attitudes in parents and adults in general could be an effective mean to prevent the use of CP, especially frequent CP, and its potentially harmful effects. In this respect, it is important to identify the factors associated with the adoption of favorable attitudes toward CP. Jackson et al. (1999) used the 1995 Gallup survey data described above to identify predictors of parents’ attitudes toward physical discipline with children age between 0 and 17 years.3 Significant predictors were child’s age (+), attitudes that devalue children (+), conservative ideology (+), use of physical discipline (+), use of nonphysical discipline (–), and importance of religion (–). Childhood physical abuse was not a significant predictor of parents’ attitudes toward CP. However, these authors did not make a difference between minor and severe physical violence. On the other hand, parents’ attitudes toward CP are a significant predictor of parents’ use of physical discipline in the current study. With a sample of 436 undergraduate students, Ateah and Parkin (2002) studied the link between childhood experiences with CP and current attitudes toward the use of CP. Of the respondents, 40% thought that CP is necessary as a means of discipline. Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Gagné et al. / Attitudes Toward Corporal Punishment
1289
Respondents who reported being spanked as a child were more likely to agree with this idea than respondents who reported other and possibly more severe forms of CP (slapped on body, slapped on head, shaken, whipped, or hit with an object). In this study, age, gender, and family income did not make any difference regarding attitudes toward CP. Finally, Deater-Deckard, Lansford, Dodge, Pettit, and Bates (2003) used longitudinal data to predict endorsement of parental use of CP among 425 adolescents from the general population. Respondents whose mother previously reported using spanking were more approving of this discipline method, regardless of the overall frequency, timing, or chronicity of CP they had received. However, this was not the case for adolescents who were suspected of physical maltreatment in early or middle childhood, which supports Ateah and Parkin’s findings. SES and ethnicity were other significant predictors of endorsement of CP: In this study, poorer adolescents, as well as African Americans, were more acceptant of this discipline method.
Objectives and Hypotheses The current study is a secondary analysis of data collected in a population survey sponsored by the Marie-Vincent Foundation, a private organization devoted to child abuse prevention. It aims at identifying predictors of favorable attitudes toward spanking in a representative adult sample. The first objective is to test the links between childhood experiences with CP and physical abuse, and favorable attitudes toward CP. Based on the literature review, it is hypothesized that spanking in childhood, but not severe physical abuse, will predict favorable attitudes toward spanking, whereas other childhood experiences with violence and abuse (exposure to domestic violence, psychological abuse, and sexual abuse) and sociodemographic characteristics of the respondents (age, gender, and education) are taken into account. The second objective is to explore the link between the level of perceived frequency of physical injuries resulting from CP and the adoption of favorable attitudes toward CP. The inclusion of this factor in the prediction model, and the inclusion or various childhood violence and abuse variables, are original elements of the current study. Originality also lies in the use of a representative sample of adults.
Method Survey Sample Data were collected by telephone in a survey conducted with a sample of 1,000 adults from the Quebec, Canada, population (maximum margin of
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1290
Journal of Interpersonal Violence
error is near 3%, alpha = .05). Households were selected by random digit dialing (Kish, 1965) among those having a telephone all over the Quebec territory. In each selected household, a respondent was chosen using a random selection schedule. The respondent had to be at least age 18 years and able to answer either in French or English. The overall response rate was 51.3%. This rate accounts for refusals to participate and for a proportion of randomly selected households that remained impossible to join after five attempts on different days and times.4 From the initial sample, 21 respondents who answered fewer than 50% of the questions were eliminated. Compared to the rest of the sample, these 21 respondents were older, poorer, and proportionally more francophone. Data were then weighted by region, age, and gender based on the 2001 Canada census data. Afterward, each weighting coefficient was multiplied by 0.84 (1/design effect) to correct for the weighting effect on statistical accuracy. In doing so, the sample size was statistically reduced to 822. Table 1 reports the characteristics of (a) the initial sample (N = 1000), (b) the final unweighted sample (N = 979), and (c) the final weighted and corrected sample (N = 822). In the last, data on gender, age, and region correspond to those of the population. Further analyses were performed with this representative sample. Table 1 also presents data about household income, education, and primary language. Because of the high rate of missing income data, the level of education was used as the indicator of SES in further analyses.
Questionnaire In addition of sociodemographic variables shown in Table 1, questions were asked to measure, among others, (a) attitudes toward spanking, (b) perception of how often physical injuries result from CP, and (c) various experiences with violence and abuse in childhood, including psychological abuse, spanking, and more severe physical violence, exposure to domestic violence, and sexual abuse before age 18 years. Most items were taken from available family violence surveys conducted in North America. Dependent variable. Attitudes toward spanking were measured by four of five items developed by Clément et al. (2000) for the purpose of their own survey. Respondents indicated their level of agreement with each statement on a 4-point Likert-type scale. A score was calculated by adding all item responses; the higher it is, the more favorable the attitudes toward spanking. In the current study, this score was fairly normally distributed,
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Gagné et al. / Attitudes Toward Corporal Punishment
1291
Table 1 Sociodemographic Characteristics of Initial, Final Unweighted, and Final Weighted Samples Initial N = 1000 Characteristics Gender Male Female Age 18-24 years 25-34 years 35-44 years 45-54 years 55-64 years 65 years and older Region Montreal Montreal suburbs Quebec City & suburbs Rest of Quebec Household income Less than CAN$ 20,000 $20,000 – 39,999 $40,000 – 59,999 $60,000 – 79,999 $80,000 and over Refusal Years of education 0-7 8-12 13-15 16 and more Refusal Primary language French English Other Refusal
Unweighted N = 979
Weighted N = 822
n
%
n
%
n
%
430 570
43.0 57.0
422 557
43.1 56.9
396 426
48.2 51.8
121 187 229 169 147 147
12.1 18.7 22.9 16.9 14.7 14.7
121 186 225 168 140 139
12.4 19.0 23.0 17.2 14.3 14.2
98 132 182 145 126 139
11.9 16.1 22.1 17.7 15.3 16.9
306 194 201 299
30.6 19.4 20.1 29.9
299 191 199 290
30.5 19.5 20.3 29.6
213 181 75 354
25.9 22.0 9.1 43.0
160 259 175 115 128 163
19.1 30.9 20.9 13.7 15.3 —
155 259 173 115 128 149
18.7 31.2 20.8 13.9 15.4 —
132 224 143 94 106 124
18.8 32.1 20.5 13.5 15.2 —
90 375 244 288 3
9.0 37.6 24.5 28.9 —
83 368 241 284 3
8.5 37.7 24.7 29.1 —
84 329 184 223 2
10.3 40.1 22.4 27.2 —
861 56 79 4
86.4 5.6 7.9 —
842 55 78 4
86.4 5.6 8.0 —
714 45 61 3
87.1 5.4 7.4 —
M = 10.46, SD = 2.83, and internal consistency was satisfying for research purposes (alpha = .73). Items and their proportion of endorsement are listed in Table 2.
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1292
Journal of Interpersonal Violence
Table 2 Attitudes Toward Spanking Scale % Agreement (N = 822)
Item 1. There should be a law forbidding parents to slap their children. 2. It’s not really pleasant, but parents have a duty to slap their children if it’s necessary. 3. It’s wrong for parents to slap their children (i.e., it’s not right to do it). 4. Some children need to be slapped so they will learn to behave themselves (i.e., learn good behavior).
29.6 62.4 35.8 42.2
Independent variables. Table 3 presents items used to measure independent variables. First, perception of how often physical harm results from CP was measured by a single question adapted from the work of Daro (1999). Respondents had to tell at which frequency they believed that CP leads to some form of physical injury for the child. Three questions were used as retrospective measures of childhood psychological abuse, spanking, and more severe physical violence. Respondents were asked how often their parents treated them like that during childhood. They indicated their answer on a 4-point scale. Using the same frequency scale, exposure to domestic violence was initially measured by two questions to reflect violence to mother and violence to father separately. Because responses to those two questions were highly correlated, they were merged into a single indicator to avoid collinearity problems. All these questions came from Clément et al. (2000). Some of them were initially inspired by the Emotional and Physical Abuse Questionnaire (Buntain-Ricklefs, Kemper, Bell, & Babonis, 1994). Finally, two indicators of sexual abuse were measured: unwanted sexual touching and unwanted sexual intercourse before age 18 years. First, respondents indicated if any of these experiences ever happened to them. If yes, they were asked if it happened once or more than once. For each experience, a variable was computed indicating if the respondent reported no abuse, only one episode of abuse, or many episodes of abuse. These questions were used in an American survey conducted by Finkelhor, Hotaling, Lewis, and Smith (1990) and were originally borrowed from the work of Badgley et al. (1984). For violence and abuse variables in general, “I don’t know” answers were recoded as “never happened,” supposing that if it had happened, the
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Gagné et al. / Attitudes Toward Corporal Punishment
1293
Table 3 Survey Questions About Violence and Abuse in Childhood Perception of physical injuries frequency: “How often do you think children are physically injured after their parents administer physical (i.e., corporal) punishment to discipline them?” (very often, often, occasionally, rarely, or never) Psychological abuse: “Would you say that you were threatened, humiliated, or ridiculed by your parents when you were a child?” (never, rarely, often, or very often) Spanking: “Were you spanked by your parents when you were a child?” (never, rarely, often, or very often) Severe physical violence: “Were you hit harder than a spanking when you were a child?” (never, rarely, often, or very often) Exposure to domestic violence: “Did you witness any violence toward your mother (or maternal figure) when you were a child?” (never, rarely, often, or very often) “Did you witness any violence toward your father (or paternal figure) when you were a child?” (never, rarely, often, or very often n) Sexual abuse – rape: “Before the age of 18, were you ever forced to have sex (including any sexual activity with oral, vaginal, or anal penetration) by an adult or older child (at least 3 years older than you), when you did not want to? It includes anyone who was a member of your family, or anyone outside your family.” (yes or no) “Did it happen only once or more than once?” Sexual abuse – touching: “Before the age of 18 (and excluding any sexual activity mentioned before), were you personally ever touched in a sexual way by an adult or older child (at least 3 years older than you), when you did not want to be touched that way, or were you ever forced to touch an adult or older child in a sexual way? It includes anyone who was a member of your family, or anyone outside your family.” (yes or no) “Did it happen only once or more than once?”
respondent would remember it. Therefore, our measures might slightly underestimate the prevalence of child abuse and spanking. However, because this rarely occurred, the impact on findings is thought to be negligible. Procedure The telephone survey was conducted between the March 14 and March 24 of 2002 by an established survey firm (CROP). Verbal consent was solicited at the time of the interview. To ensure the most honest answers
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1294
Journal of Interpersonal Violence
possible, confidentiality was guaranteed and participants were not asked about their own violent or abusive behavior toward children. Secondary analyses of survey data were approved by the Sherbrooke University Institutional Review Board. Data Analysis Data were entered and verified by the personnel of the professional survey firm. In general, missing data were not replaced; however, the score of attitudes toward spanking was calculated with only three items if only one answer was missing. First, descriptive statistics for violence and abuse variables were computed to estimate their prevalence. Second, the correlations between all independent variables were calculated. Finally, a multiple regression analysis (standard method) was performed to predict the score of attitudes toward spanking using a set of demographic, perception of injuries frequency, and childhood victimization predictors.5 This technique is appropriate to estimate the degree of relationship between a continuous dependent variable (DV) and many independent variables (IVs) that can be interrelated to some extent. It also indicates the relative importance of each IV to predict DV (Tabachnick & Fidell, 2001). Given the large sample size, and because most missing data occurred on a single predictor, missing data were taken into account by using the “pairwise” option. Collinearity statistics showed appropriate tolerance levels for each predictor, and residuals showed normality, linearity, and homoscedasticity. Two participants with standardized residuals greater than 3 were excluded from regression analysis. All analyses were performed with SPSS 12.0.
Results Prevalence Rates Figure 1 shows the prevalence rates for each form of childhood violence and abuse, as reported retrospectively by adults. Spanking appears as the most reported form of violence. On the whole, 66.4% of the participants experienced spanking in childhood, and more than 1 of 10 participants reported frequent spanking. Other forms of abuse were reported less frequently, especially sexual abuse. However, most participants who were sexually abused reported more than one unwanted sexual experience before age 18 years.
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Gagné et al. / Attitudes Toward Corporal Punishment
1295
Figure 1 Prevalence Rates for Various Forms of Violence and Abuse in Childhood 88,4
90 80
82 76,5
70 55,8
60 50
Never
40
33,6
Rarely
30 20 10
Often/Very often
15,9 10,6
7,6
13 8,3
5
3,4
0 Psychological Spanking Abuse
Exposure to DV
95,4
100 90
Severe physical violence
87
80 70 60 Never 50
Once More than once
40 30 20 10
5,9
7 2
2,5
0 CSA-touching
CSA-rape
Regarding perceived frequency of physical injury resulting from CP, 3.8% of participants believed that children are never injured by such disciplinary practices, 25.2% believed they are rarely injured, 39.6% believed they are occasionally injured, 22.6% believed they are often injured, and 8.9% believed they are very often injured (n = 735).
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1296
Journal of Interpersonal Violence
Table 4 Correlation Matrix (r) Between Predictors (N = 731 to 822) 1 1. Age 2. Gender .05 3. Education –.33** 4. Perceived injuries –.04 5. Exposure to DV .02 6. Psychological abuse –.05 7. Spanking –.10** 8. Severe physical .01 violence 9. CSA - touching –.03 10. CSA - rape –.04
2
3
4
–.02 –.14** .09* –.02 –.08* –.07 .07* .01 –.04 –.11** .03 .03 –.07* –.07* –.06 .00 .07*
.00 .01
5
6
7
8
9
10
.36** .26** .45** .36** .54** .56**
–.09* .19** .32** .18** .21** –.02 .11** .21** .18** .24** .41**
Note: DV = domestic violence; CSA = child sexual abuse. *p < .05. **p < .01.
Correlations Among Predictors Table 4 shows that most correlations between factors are nonsignificant or weak (< .30). Nevertheless, many forms of violence and abuse show significant positive, moderate correlations with each other. Psychological abuse is correlated with exposure to domestic violence (r = .36), spanking (r = .45), severe physical violence (r = .54), and unwanted sexual touching (r = .32). Spanking is correlated with more severe physical violence (r = .56), which in turn shows a correlation with exposure to domestic violence (r = .36). Finally, severe and milder forms of sexual abuse are correlated (r = .41).
Regression Analysis Multiple regression analysis was significant, F(10, 690) = 12.42, p = .000. It appears that older participants are more in favor of spanking than younger ones. No other demographic predictor contributed significantly to the model; however, there is a tendency for less educated participants to be more in favor of spanking (p = .06). Interactions of gender with other variables, especially victimization experiences, have been tested; none proved significant. Table 5 summarizes these findings. The more participants reported spanking in their own childhood, the more they approved of this kind of disciplinary practice. On the other hand,
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Gagné et al. / Attitudes Toward Corporal Punishment
1297
Table 5 Summary of Regression Analysis for Variables Predicting Favorable Attitudes toward Spanking (N = 808a) Variable Age Gender Education Perceived frequency of injuriesb Exposure to domestic violence Psychological abuse Spanking Severe physical violence CSA – touchingb CSA – rapeb
B
SE B
β
.023 .138 –.203 .823 .151 –.518 .893 –.446 –.180 .081
.007 .205 .108 .102 .213 .189 .174 .216 .209 .324
.13** .02 –.07 .29** .03 –.12* .22** –.10* –.04 .01
95% CI for B .010 –.264 –.414 .623 –.267 –.888 .552 –.870 –.590 –.555
.036 .539 .009 1.024 .570 –.148 1.234 –.022 .229 .717
Note: R2 = .15; adjusted R2 = .14. a. There were 12 missing data for the dependent variable. b. There were 82 missing data for this predictor (N = 726), mostly due to “I don’t know” answers. Respondents who did not provide an answer to this question were significantly younger than other respondents, t(820) = –2.5, p = .014, but did not differ on gender, education, region, and language. Other predictors were not affected by missing data. *p ≤ .01. **p ≤ .001.
the more they have felt threatened or humiliated by their parents in childhood, and the more they reported severe physical violence from them, the less they agreed with spanking. Child sexual abuse and exposure to domestic violence did not contribute significantly to this prediction model. Finally, the more participants were convinced that CP is not physically dangerous, the more they were in favor of spanking to discipline children. In fact, this factor appears to be the best predictor of attitudes toward spanking in the current model. Interactions of this belief with victimization variables have been tested but were not significant. Overall, predictors used in the regression equation explained 15% of the variance of attitudes toward spanking.
Discussion In the current study, various intra- and extrafamilial experiences with childhood violence and abuse were examined as potential risk factors for favorable attitudes toward CP of children. The proposed model also accounted for demographic variables and the respondents’ perception of
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1298
Journal of Interpersonal Violence
how often physical injuries result from CP. Many predictors of this model, especially violence and abuse variables, appeared to be significantly correlated to each other. This is not surprising: Previous research has shown that various victimization experiences, including exposure to domestic violence, co-occur in the lives of youths (Appel & Holden, 1998; Bowen, 2000; Crittenden, Claussen, & Sugarman, 1994; Edleson, 1999; Gagné, Lavoie, & Hébert, 2005; Lessard, 2004; McGee, Wolfe, & Wilson, 1997; Vissing, Straus, Gelles, & Harrop, 1991). Finkelhor and Dziuba-Leatherman (1994) introduced the metaconcept of developmental victimology to account for the continuum of victimization experiences from early childhood to the end of adolescence. The hypothesis that spanking in childhood would predict favorable attitudes toward spanking was supported; however, the significant negative link between childhood physical abuse and attitudes was unexpected. According to the findings, adults who were frequently spanked during childhood (but not severely hit); who have not been (or did not feel) threatened, humiliated, or ridiculed by their parents; and who think that CP is not likely to cause physical injuries are the most in favor of spanking. These individuals might perpetuate a coercive ideology of child rearing, and parents among them might be at risk for spanking and harsh discipline. Consequently, they might contribute to the intergenerational cycle of coercion and to consequent youth antisocial behavior and mental health problems (Reid, Patterson, & Snyder, 2002). Prevention and education efforts should be directed at them. Programs should pay special attention to the belief that spanking does not harm children: Among all predictors used in the current study, this is the most strongly associated with favorable attitudes toward spanking. Adults, and especially parents, should be told that though spanking is not physically dangerous with older children, frequent spanking might have some physiological impacts on young children. They must also be informed of the various emotional and behavioral problems associated with frequent spanking to help them realize that physical injuries may not be the most detrimental effect of CP. Of course, they must also be taught effective alternatives to discipline their children and control their behavior. And because endorsement of CP can be observed as soon as adolescence (Deater-Deckard et al., 2003), school-based violence prevention programs should include activities on that topic. Findings also suggest that adults who sustained severe physical violence, and who have felt threatened, humiliated, or ridiculed by their parents in childhood, tend to be the least in favor of spanking. These participants undoubtedly remember the pain, physical and emotional, generated by their parents’ behavior; this might explain why they do not believe that CP is good or
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Gagné et al. / Attitudes Toward Corporal Punishment
1299
necessary for children. This type of remembrance anchored in childhood experiences could lower the level of tolerance of violence. However, this mind-set will not necessarily prevent a person from using harsh discipline, and even more abusive violence, with his or her own children. The link between attitudes and behavior is mediated by many properties of cognitions (Cooke & Sheeran, 2004), and intergenerational transmission of harsh discipline and abuse is not just a matter of attitudes. According to Egeland, Bosquet, and Chung (2002), 45% of parents who were abused in childhood end up maltreating their own children. These authors described intergenerational transmission of child maltreatment as a highly complex phenomenon determined by a number of factors pertaining to (a) parent personality, psychopathology, heredity, developmental history, and attributions and/or expectations toward the child; (b) family violence, quality of marital relationship, and inner working models of the family; (c) family’s level of involvement in and support from the community, poverty, community violence; and finally (d) society’s tolerance of violence, including CP. In maltreated individuals, aggressive and coercive behavior is adopted through mechanisms related to attachment, trauma, and social learning (Gagné, Desbiens, & Blouin, 2004). In this context, remembrance of the pain associated with parental violence and modification of attitudes toward spanking might not be sufficient by themselves to prevent harsh discipline and abuse. These could nevertheless be considered as worthwhile protection factors. On a macrosystemic level, less tolerant attitudes toward spanking could also help the elaboration of a less violent society. In the current study, that older participants are more in favor of spanking than younger ones may indicate that mentalities are already changing, which is encouraging. Finally, findings suggest that among forms of violence and abuse, only those having a disciplinary connotation (psychological abuse, spanking, and more severe physical violence) are significant predictors of attitudes toward spanking. Types of abuse that are not used as discipline strategies (exposure to domestic violence, sexual abuse) do not contribute to the prediction. It could mean that the links between childhood experiences, attitudes, and parenting practices are context specific. Perhaps it is not the fact of being a victim that influences attitudes toward spanking, but the more specific experience of coercive discipline. This hypothesis has to be supported by future research.
Limitations and Strengths The scope of our findings is limited by the cross-sectional nature of the survey design. As regards to demographic factors and perceived frequency
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1300
Journal of Interpersonal Violence
of physical injuries, the model that is tested is best described as an association model and does not allow causal inference. However, the temporal sequence of childhood experiences and adult attitudes implies causality, although interpretation of this causal link is limited by the reliance on retrospective recall of childhood violence and abuse experiences. Although recognizing the reality of this problem, Rutter, Pickles, Murray, and Eaves (2001) relied on recent research to state that the biases tend to be minor. It should also be emphasized that the use of behaviorally specific items to measure childhood experience with violence (see Table 3) may contribute to reduce recall biases. Finally, it was not possible to compare parents to nonparents with regards to their attitudes toward spanking because this distinction was not available in the data set. These limits are compensated by the use of a large, representative sample and by the originality of the model that has been tested, especially with the inclusion of the perception of how often physical injuries result from CP and the consideration of multiple forms of direct and indirect childhood victimization and exposure to violence, including psychological, physical, and sexual components.6
Conclusion The current study identified factors that are associated with adoption of favorable attitudes toward spanking of children in a representative sample of adults from the general population. It should be recalled however that our model explains only 15% of the variance of attitudes toward spanking. Obviously, many other factors influence the adoption of favorable attitudes toward spanking; recent research (Deater-Deckard et al., 2003; Jackson et al., 1999) identified some of them (child’s age, attitudes that devalue children, conservative ideology, religion, SES, and ethnicity). Also, our findings suggest that affective and cognitive processes anchored in the experience of being treated harshly in childhood (e.g., remembrance of pain associated with violence and abuse, but maybe also schemas, attributions, information processing, perception of self as a victim or not, feelings of insecurity, need for control, hostility, etc.) could act as important mediators or moderators of the link between childhood experiences and adult attitudes. Clearly, more research is needed to not only fully understand the adoption of favorable attitudes toward CP and how these attitudes contribute to CP, but also to other disturbing social processes. For example, attitudes toward CP seem to be related to physicians’ reporting of abuse. A study conducted by Tirosh, Shechter, Cohen, and Jaffe (2003) showed that CP was approved by 58% of
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Gagné et al. / Attitudes Toward Corporal Punishment
1301
their Israeli physicians sample (95 pediatricians and 12 family practitioners), regardless of gender and specialty (with regards to gender, it should be underscored that, in the current study and in other research using multivariate statistics, gender was not significantly associated with attitudes toward CP). Furthermore, physicians more likely to accept CP were found less willing to report children to child protection agencies. Such unwillingness to report could be harmful for children who need protection. These findings suggest that parents should not be the only target for education and prevention.
Notes 1. For instance, the Supreme Court of Canada recently legitimized corporal punishment (CP) of children, recognizing the right of parents to use reasonable physical force to discipline their children between age 2 and 12 years. 2. In the Quebec survey, some behaviors regarded as CP in the U.S. survey (hitting with an object and slapping in the face, head, or ears) were excluded from the CP score because they were considered too severe. 3. In this article, the terms predictors and prediction are used in their statistical sense and do not infer causality. 4. This proportion is based on households that have been joined successfully. It represents the number of households corresponding to the selection criteria divided by the total number of randomly selected households. 5. Region and primary language were not used as predictors because there was no rationale to it. On a bivariate level, these variables showed no significant link with attitudes toward spanking. 6. Findings regarding this variable require cautious interpretation: They might be less generalizable because of missing data, although bias seems to be minor.
References Andero, A. A., & Stewart, A. (2002). Issue of corporal punishment: Re-examined. Journal of Instructional Psychology, 29(2), 90-96. Appel, A. E., & Holden, G. W. (1998). The co-occurrence of spouse and physical child abuse: A review and appraisal. Journal of Family Psychology, 12, 578-599. Ateah, C. A., & Parkin, C. M. (2002). Childhood experiences with, and current attitudes toward corporal punishment. Canadian Journal of Community Mental Health, 21(1), 35-46. Baumrind, D., Larzelere, R. E., & Cowan, P. A. (2002). Ordinary physical punishment: Is it harmful? Comment on Gershoff (2002). Psychological Bulletin, 128(4), 580-589. Badgley, R., Allard, H., McCormick, N., Proudfoot, P., Fortin, D., Ogilvie, D., et al., Comité sur les infractions sexuelles à l’égard des enfants et des jeunes [Committee on Sexual Offences Against Children and Youths]. (1984). Infractions sexuelles à l’égard des enfants [Sexual offences with regard to the children] (Vol. 1). Ottawa: Health and Welfare Canada. Benjet, C., & Kazdin, A. E. (2003). Spanking children: The controversies, findings and new directions. Clinical Psychology Review, 23(2), 197-224.
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1302
Journal of Interpersonal Violence
Bowen, K. (2000). Child abuse and domestic violence in families of children seen for suspected sexual abuse. Clinical Pediatrics, 39, 33-40. Bugental, D. B., Martorell, G. A., & Barraza, V. (2003). The hormonal costs of subtle forms of infant maltreatment. Hormones and Behavior, 43, 237-244. Buntain-Ricklefs, J., Kemper, K., Bell, M., & Babonis, T. (1994). Punishment: What predicts adult approval? Child Abuse & Neglect, 18, 945-955. Clément, M.-È., Bouchard, C., Jetté, M., & Laferrière, S. (2000). La violence familiale dans la vie des enfants du Québec, 1999 [Family violence in the lives of children in Quebec]. Québec, Canada: Institut de la Statistique du Québec. Cooke, R., & Sheeran, P. (2004). Moderation of cognition-intention and cognition-behaviour relations: A meta-analysis of properties of variables from the theory of planned behaviour. British Journal of Social Psychology, 43, 159-186. Crittenden, P. M., Claussen, A. H., & Sugarman, D. B. (1994). Physical and psychological maltreatment in middle childhood and adolescence. Development and Psychopathology, 6, 145-164. Crouch, J. L., & Behl, L. E. (2001). Relationships among parental beliefs in corporal punishment, reported stress, and physical child abuse potential. Child Abuse & Neglect, 25, 413-419. Daro, D. (1999). Public opinion and behaviors regarding child abuse prevention: 1999 survey. Chicago: National Committee to Prevent Child Abuse. Deater-Deckard, K., Lansford, J. E., Dodge, K. A., Pettit, G. S., & Bates, J. E. (2003). The development of attitudes about physical punishment: An 8-year longitudinal study. Journal of Family Psychology, 17(3), 351-360. Durrant, J., Rose-Krasnor, L., & Broberg, A. G. (2003). Physical punishment and maternal beliefs in Sweden and Canada. Journal of Comparative Family Studies, 34, 585-604. Edleson, J. L. (1999). The overlap between child maltreatment and woman battering. Violence Against Women, 5, 134-154. Egeland, B., Bosquet, M., & Chung, A. L. (2002). Continuities and discontinuities in the intergenerational transmission of child maltreatment: implications for breaking the cycle of abuse. In K. D. Browne, H. Hanks, P. Stratton, & C. Hamilton (Eds.), Early prediction and prevention of child abuse: A handbook (pp. 217-232). New York: John Wiley. Finkelhor, D., & Dziuba-Leatherman, J. (1994). Victimization of children. American Psychologist, 49, 173-183. Finkelhor, D., Hotaling, G., Lewis, I. A., & Smith, C. (1990). Sexual abuse in a national survey of adult men and women: Prevalence, characteristics, and risk factors. Child Abuse & Neglect, 14, 19-28. Gagné, M.-H., Desbiens, N., & Blouin, K. (2004). Trois profils-types de jeunes affichant des problèmes de comportement sérieux [Three typical profiles of youths displayhing serious behaviour problems]. Éducation et Francophonie, 32(1), 276-311. Gagné, M.-H., Lavoie, F., & Hébert, M. (2005). Victimization during childhood and revictimization in dating relationships in adolescent girls. Child Abuse & Neglect, 29, 1155-1172. Gershoff, E. T. (2002a). Corporal punishment by parents and associated child behaviors and experiences: A meta-analytic and theoretical review. Psychological Bulletin, 128, 539-579. Gershoff, E. T. (2002b). Corporal punishment, physical abuse, and the burden of proof: Reply to Baumrind, Larzelere, and Cowan (2002), Holden (2002), and Parke (2002). Psychological Bulletin, 128, 602-611. Holden, G. W. (2002). Perspectives on the effects of corporal punishment: Comment on Gershoff (2002). Psychological Bulletin, 128, 590-595.
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
Gagné et al. / Attitudes Toward Corporal Punishment
1303
Holden, G. W., Coleman, S. M., & Schmidt, K. L. (1995). Why 3-year-old children get spanked: Parent and child determinants as reported by college-educated mothers. Merrill Palmer Quarterly, 41, 431-452. Jackson, S., Thompson, R. A., Christiansen, E. H., Colman, R. A., Wyatt, H., Buckendahl, C. W., et al. (1999). Predicting abuse-prone parental attitudes and discipline practices in a nationally representative sample. Child Abuse & Neglect, 23, 15-29. Kish, L. (1965). Survey sampling. New York: John Wiley. Korbin, J. E. (1991). Cross-cultural perspectives and research directions for the 21st century. Child Abuse & Neglect, 15(Suppl. 1), 67-77. Larzelere, R. E. (2000). Child outcomes of nonabusive and customary physical punishment by parents: An updated literature review. Clinical Child and Family Psychology Review, 3, 199-221. Larzelere, R. E., & Johnson, B. (1999). Evaluation of the effects of Sweden’s spanking ban on physical child abuse rates: A literature review. Psychological Reports, 85, 381-392. Lessard, G. (2004). Concomitance de violence conjugale et de mauvais traitements envers les enfants: représentations des intervenants et défis de collaboration pour l’offre d’une aide cohérente [Co-occurrence of domestic violence and child maltreatment: Practitioners’ representations and collaboration challenges in order to supply coherent help]. Unpublished master’s thesis, University of Montreal, Quebec, Canada. McGee, R. A., Wolfe, D. A., & Wilson, S. K. (1997). Multiple maltreatment experiences and adolescent behavior problems: Adolescents’ perspectives. Development and Psychopathology, 9, 131-149. Pinderhughes, E. E., Dodge, K. A., Bates, J. E., Pettit, G. S., & Zelli, A. (2000). Discipline responses: Influences of parents’ socioeconomic status, ethnicity, beliefs about parenting, stress, and cognitive-emotional processes. Journal of Family Psychology, 14, 380-400. Reid, J. B., Patterson, G. R., & Snyder, J. (Eds.). (2002). Antisocial behavior in children and adolescents: A developmental analysis and model for intervention. Washington, DC: American Psychological Association. Rizzo, C. P. (2002). Spanking: Counterpoint. Clinical Pediatrics, 41(2), 89-90. Rutter, M., Pickles, A., Murray, R., & Eaves, L. (2001). Testing hypotheses on specific environmental causal effects on behavior. Psychological Bulletin, 127, 291-324. Saadeh, W. (2002). Spanking. Clinical Pediatrics, 41(2), 87-88. Smith, D., & Mosby, G. (2003). Jamaican child-rearing practices: The role of corporal punishment. Adolescence, 38, 369-381. Straus, M. A. (Ed.). (1994). Beating the devil out of them: Corporal punishment in American families. San Francisco: Jossey-Bass. Straus, M. A. (1996). Spanking and the making of a violent society. Pediatrics, 9(4), 837-844. Straus, M. A., & Donnelly, D. A. (Eds.). (2001). Beating the devil out of them: Corporal punishment in American families and its effects on children (2nd ed.). New Brunswick, NJ: Transaction Publishers. Straus, M. A., Hamby, S. L., Finkelhor, D., Moore, D. W., & Runyan, D. (1998). Identification of child maltreatment with the Parent-Child Conflict Tactics Scales: Development and psychometric data for a national sample of American parents. Child Abuse & Neglect, 22, 249-270. 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 characteristics. Clinical Child and Family Psychology Review, 2, 55-70. Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th ed.). Needham Heights, MA: Allyn & Bacon.
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.
1304
Journal of Interpersonal Violence
Tirosh, E., Shechter, S. O., Cohen, A., & Jaffe, M. (2003). Attitudes towards corporal punishment and reporting of abuse. Child Abuse & Neglect, 27, 929-937. Vissing, Y. M., Straus, M. A., Gelles, R. J., & Harrop, J. W. (1991). Verbal aggression by parents and psychosocial problems of children. Child Abuse & Neglect, 15, 223-238. Wissow, L. S. (2002). Child discipline in the first three years of life. In N. Halfon, K. T. McLearn, & M. A. Schuster (Eds.), Child rearing in America: Challenges facing parents with young children (pp. 146-177). New York: Cambridge University Press.
Marie-Hélène Gagné, PhD, is associate professor at the School of Psychology, Laval University. She is part of the Research Center on at-risk youths’ and families’ adjustment [JEFAR: Centre de recherche sur l’adaptation des jeunes et des familles à risque]. Marc Tourigny, PhD, is professor at the Faculty of Education, Department of Psychoeducation, Sherbrooke University. He holds the Marie-Vincent Chair on Sexual Aggression of Children [Chaire interuniversitaire Marie-Vincent sur les agressions sexuelles envers les enfants]. He is part of the Research Group on Childhood Social Maladjustment [GRISE: Groupe de recherche sur les inadaptations sociales de l’enfance]. Jacques Joly, PhD, is associate professor at the Faculty of Education, Department of Psychoeducation, Sherbrooke University. He is part of the Research Group on Childhood Social Maladjustment [GRISE: Groupe de recherche sur les inadaptations sociales de l’enfance]. Joëlle Pouliot-Lapointe, BA, is a PhD student in community psychology, Laval University. She is part of the Research Center on at-risk youths’ and families’ adjustment [JEFAR: Centre de recherche sur l’adaptation des jeunes et des familles à risque].
Downloaded from http://jiv.sagepub.com by guest on October 21, 2007 © 2007 SAGE Publications. All rights reserved. Not for commercial use or unauthorized distribution.