Parenting influences from the pulpit: Religious affiliation as a determinant of parental corporal punishment (original) (raw)

Parenting Influences From the Pulpit: Religious Affiliation as a Determinant of Parental Corporal Punishment

Elizabeth Thompson Gershoff, Pamela C. Miller, and George W. Holden
University of Texas at Austin

Abstract

This study examined religious affiliation as a source of differences in beliefs about and reported use of corporal punishment by 132 mothers and fathers of 3-year-old children. Conservative Protestants reported using corporal punishment more than parents of other religious groups, but no religious differences were found in parents’ reported use of 8 other disciplinary techniques. Conservative Protestants’ belief in the instrumental benefits of corporal punishment was associated with their frequency of corporal punishment use. More than parents of other religious affiliations, Conservative Protestants intended to use corporal punishment for children’s moral, social, prudential, and escalated misbehaviors and expected it to prevent future transgressions. Religious affiliation, particularly a Conservative Protestant one, appears to have a strong and consistent effect on child rearing.

One of the most important and most salient tasks for parents of young children is discipline. How parents discipline their children predicts crucial aspects of children’s positive and negative interpersonal behaviors, including the extent to which children comply with parental directives (Minton, Kagan, & Levine, 1971;

[1]Power & Chapieski, 1986), internalize parental values (Grusec & Goodnow, 1994; Hoffman, 1983), behave altruistically (Zahn-Waxler, RadkeYarrow, & King, 1979), and act disrespectfully or aggressively (Strassberg, Dodge, Pettit, & Bates, 1994). It is because parents’ behaviors have such profound consequences for children’s future behavior that understanding why parents choose certain discipline practices over others is an important task for psychological research.

One source of disciplinary practices and beliefs that has rarely been examined is religion. An individual’s religious beliefs often provide a guide for human interactions in general and parent-child interactions in particular (Holden, 1997; Wiehe, 1990). Religion has been a source of parenting beliefs in the United States since its Colonial days (Greven, 1991), and in some denominations American Christians are admonished to use the Bible as their most important parenting manual (e.g., Fugate, 1980). Despite the reticence of researchers to study religion as a possible determinant of parenting behavior, parents themselves often readily acknowledge the role played by religious teachings in helping them make child-rearing, and particularly disciplinary, decisions. Indeed, some parents use the wrath of God as a tool for disciplinary control, threatening their children that “God will punish” them for their misbehaviors (Nelsen &

1. Elizabeth Thompson Gershoff, Department of Human Ecology, University of Texas at Austin; Pamela C. Miller and George W. Holden, Department of Psychology, University of Texas at Austin. Pamela C. Miller is now at the Department of Psychology, University of Houston.
   This work was supported in part by a National Institute of Child Health and Human Development grant to George W. Holden (1 R01 HD26574-OIAI). Additional support came from a National Institute of Mental Health-sponsored postdoctoral fellowship through the Program for Prevention Research awarded to Elizabeth Thompson Gershoff. Portions of this article were presented at the biennial meetings of the Society for Research in Child Development, Washington, DC, April 1997. We thank Christopher Ellison for his helpful comments on a draft of this article.
   Correspondence concerning this article should be addressed to Elizabeth Thompson Gershoff, who is now at the Department of Psychology, Arizona State University, P.O. Box 871104, Tempe, Arizona 852871104. Electronic mail may be sent to liztg@imap4. asu.edu. ↩︎

Kroliczak, 1984). Of all parenting behavior, discipline is particularly conducive to the influence of religious teachings, because it is often through discipline that parents teach their children moral and social norms (Hoffman, 1983), issues that typically fall under the province of religion.

Up to now, interest in religious affiliation as a predictor of parenting practices has been restricted to the prediction of corporal punishment. Indeed, Catholics, Protestants, and Jews alike cite the Old Testament as support for their use of corporal punishment, although specific directives to spank children in the text are rare (Carey, 1994; Greven, 1991). In particular, Conservative Protestant parents consistently report using corporal punishment more often than parents of other religious affiliations (Day, Peterson, & McCracken, 1998; Ellison, Bartkowski, & Segal, 1996; Giles-Sims, Straus, & Sugarman, 1995). One goal of the present study is to determine whether the religious differences found in parents’ frequency of using corporal punishment extend to other types of parental discipline. Furthermore, this study examines parents’ beliefs about corporal punishment as a potential source of the religious difference in parents’ use of corporal punishment. These include parents’ beliefs about the appropriateness, effectiveness, and effects of corporal punishment in general and for specific child misbehaviors.

Corporal punishment is only one among a range of techniques parents use to punish and discipline their children’s misbehaviors. In their attempts to secure compliance, parents reason with, withdraw privileges from, and issue time-outs to their children. Given that parents correct their young children’s behavior as often as every 6 to 8 min (Minton et al., 1971) and that corporal punishment is used on average less than once a month with young children (Straus, 1994), corporal punishment constitutes only a small portion of the discipline parents use on a daily basis. The religious difference found for the infrequent behavior of corporal punishment may extend to parents’ preferences for the techniques they use on a more regular basis. We predicted that Conservative Protestant parents’ more frequent use of corporal punishment is indicative of a tendency to use powerassertive techniques in general, including yell-
ing at or threatening children, to obtain children’s obedience.

A likely source of religious differences in parents’ use of corporal punishment is their beliefs about when and why it should be used as well as when and how it is effective in achieving child-rearing goals. Consistent with their frequency of using corporal punishment, parents who report a Conservative Protestant religious affiliation believe corporal punishment should be used more than parents of other religious affiliations (Ellison & Sherkat, 1993; Grasmick, Morgan, & Kennedy, 1992; Wiehe, 1990). Conservative Protestants’ belief in the acceptability of corporal punishment appears to be an extension of their beliefs in human nature as evil, human sins as requiring punishment, and the Bible as an infallible guide for parenting practices (Ellison & Sherkat, 1993; Wiehe, 1990). These beliefs are codified in conservative parenting advice books that advocate the use of corporal punishment to ensure children’s obedience to authority (e.g., Dobson, 1970; Fugate, 1980; Lessin, 1979; see review by Bartkowski & Ellison, 1995).

Following from these religious differences in general support for corporal punishment, we sought to establish whether there are religious differences in parents’ specific beliefs about when corporal punishment should be used and when it is most effective. In particular, whether parents select corporal punishment because of their belief in its instrumental utility or out of their emotional arousal in response to child misbehaviors may predict the frequency with which they use corporal punishment. This instrumental-emotional dimension of parents’ beliefs about corporal punishment has only recently been identified (Holden & Miller, 1997; Straus & Mouradian, 1998). However, the instrumental-emotional distinction has important implications for the likelihood of physical abuse, as a combination of instrumental corporal punishment and negative affective arousal can escalate corporal punishment into physical abuse (Vasta,-1982). Indeed, the use of corporal punishment for emotional rather than instrumental reasons has been associated with increases in negative child behaviors (Straus & Mouradian, 1998). In this study, we examined whether parents of different religious groups vary in their inclinations toward instrumental or emotional use of corporal punishment. Because of their

strong belief in corporal punishment as a tool for teaching authority, we predicted that Conservative Protestant parents would be higher in instrumental incentives and lower in emotional impulses to use corporal punishment than parents of other religious groups.

By extension, a belief in the instrumental effectiveness of corporal punishment also should manifest as an expectation of more positive and fewer negative outcomes when corporal punishment is used. Although securing child compliance is typically at the top of parents’ minds when they use corporal punishment or any form of discipline, parents are also trying to achieve other outcomes, such as having children learn not to repeat dangerous behaviors in the future. We thus were interested in whether parents of different religious groups agree on the particular positive and negative effects of corporal punishment on their children. We predicted that Conservative Protestant parents, in keeping with their endorsement of corporal punishment’s instrumental usefulness, would believe that corporal punishment yields more positive and fewer negative consequences for children than do parents of other religious groups.

In addition to their beliefs about the effects of corporal punishment in general, parents’ use of corporal punishment is likely affected by their beliefs about the appropriateness and effectiveness of corporal punishment for specific types of child misbehaviors. Although difficult child temperaments can sometimes account for differences in the frequencies with which parents use corporal punishment (Day et al., 1998), the finding that difficult temperaments do not appear to account for increases in corporal punishment used by Conservative Protestant parents (Ellison et al., 1996) suggests that Conservative Protestant parents consider more child misbehaviors to be deserving of corporal punishment than do other parents. Parents select their discipline in general (Dix, Ruble, & Zambarano, 1989; Grusec, Dix, & Mills, 1982) and their corporal punishment in particular (Catron & Masters, 1993; Chilamkurti & Milner, 1993) based on the type and severity of children’s misbehaviors. However, we do not know if there are religious affiliation differences in the likelihood that parents respond to various child misbehaviors with corporal punishment (Ellison et.al., 1996).

Two specific dimensions of misbehaviors that
may elicit differences among religious groups are the type of misbehavior and whether or not the defiance is escalated. Parents have been found to use discipline techniques differentially for three types of misbehaviors, namely prudential, moral, and social (Catron & Masters, 1993; Gralinski & Kopp, 1993). Prudential transgressions pose a danger or threat to the child, whereas moral transgressions involve harming others, violating the rights of others, or negatively affecting others’ welfare. In social transgressions, children fail to act in accordance with their surrounding social system (Smetana, 1997; Tisak & Turiel, 1984; Turiel, 1977). Following from their tendency to use corporal punishment more overall, we expected that parents with a Conservative Protestant affiliation would find all of the misbehaviors as more deserving of corporal punishment than would parents of other religious groups.

Escalation, when children misbehave after being told specifically not to, has received particular attention from conservative parenting advisors. Often called “willful disobedience,” escalated misbehaviors are particularly challenging to Conservative Protestants (e.g., Dobson, 1970; Lessin, 1979), because they challenge parents’ ultimate authority over their children. Conservative parenting advisors warn that in order to avoid problems with children in the future, escalated misbehaviors must be punished firmly. We thus expected that when presented with examples of children’s escalated disobedience, Conservative Protestant parents would be more likely to use corporal punishment and less likely to use reasoning than parents of other religious groups.

As a follow-up to asking parents’ beliefs about the effects of corporal punishment in general, we also asked parents to predict the likelihood of positive and negative outcomes when corporal punishment and then reasoning are used in response to specific child misbehaviors. We anticipated that Conservative Protestant parents would predict more positive and fewer negative consequences when using corporal punishment than would other parents.

Method

Participants

A total of 93 mothers and 39 fathers were recruited from a database derived from birth announcements in

a midsize southwestern U.S. city. Mothers and fathers were invited to participate in one or more of the three components of this project; all were parents of 3 -year-old children (range: 35 to 41 months), an age found to be associated with the highest frequency of spanking by parents (Straus, 1994). Seventy-four of the parents’ children ( 56%56 \% ) were female. The mothers ranged in age from 25 to 45 years with a mean of 34 years; fathers were on average 36 years old and ranged in age from 29 to 43 years. One hundred and seven of the parents ( 81%81 \% ) were European American, 18 ( 14%14 \% ) were Hispanic American, 3 ( 2%2 \% ) were either African American or Asian American, and 4 ( 3%3 \% ) parents did not report their ethnic affiliation.

The majority of the parents ( 65%65 \% ) had at least a college degree; the highest level of education of the remaining participants ( 35%35 \% ) was either high school or some college. The participants’ average Hollingshead occupational score was 6.8(SD=1.56.8(S D=1.5; range =3−9=3-9 ), interpreted as minor professionals (managers, small-business owners; Hollingshead, 1975). Twelve percent of the parents reported combined family incomes of between 15,00015,000\15,000 15,000 and 29,000,4229,000,42%\29,000,42 29,000,42 \% reported incomes of 30,00030,000\30,000 30,000 to 59,00059,000\59,000 59,000, and 46%46 \% reported incomes of 60,00060,000\60,000 60,000 and above.

Measures

Questionnaires

Religious Affiliation Questionnaire (RAQ). This questionnaire (Ellison, 1995) asks respondents to identify their current religious affiliation according to the following categories: (a) Mainline Protestant (e.g., Episcopal, Lutheran, Methodist, Presbyterian, Unitarian); (b) Conservative Protestant (e.g., Adventist, Baptist, Pentecostal); © Roman Catholic; (d) Jewish; (e) other (e.g., Buddhist, Mormon, Hindu); and (f) no affiliation with any formal religion. Forty-seven parents reported a Mainline Protestant affiliation ( 36%36 \% ), 33 reported a Roman Catholic affiliation ( 25%25 \% ), 27 reported no religious affiliation (20%), and 25 reported a Conservative Protestant affiliation ( 19%19 \% ). The other affiliations were selected too infrequently to be included.

There was a significant difference in the religious affiliations of mothers and fathers, χ2(3,N=132)=\chi^{2}(3, N=132)= 7.87,p<.057.87, p<.05; mothers were more likely to report a Mainline Protestant ( 41%41 \% ) or Roman Catholic ( 27%27 \% ) religious affiliation, whereas fathers were most likely to declare no religious preference ( 33%33 \% ) or a Conservative Protestant affiliation ( 23%23 \% ). However, sex of the parent was a significant covariate with religious affiliation in only one analysis. As well, only one difference was found for parents’ occupational level; parents’ years of education, one of the indexes of socioeconomic status (SES), was not a significant covariate for any of the analyses.

Parental Beliefs About Spanking (PBAS). The 16-item PBAS questionnaire, developed for this study, was designed to assess parents’ beliefs about the corporal punishment they administer to their own children. Sample items include, “How effective do you think spanking is to get your child to stop misbehaving in the future?”; “After your child has misbehaved and right before you discipline him/her, how irritated do you usually feel?”; and “Do you think spanking affects your relationship with your child by promoting respect?” This questionnaire had high internal consistency with this sample (Cronbach’s α=.93\alpha=.93 ).

A factor analysis of the PBAS resulted in two factors, Instrumentality (factor loadings for 13 items: range =.52=.52 to .91 ) and Emotionality (factor loadings for 3 items: range =.53=.53 to .88 ). These factors accounted for 57%57 \% of the variance of the items. The Instrumentality factor reflects the extent to which parents agree that spanking (a) is not abusive, (b) is necessary, © is effective, (d) allows them to be less upset, (e) allows them to be in control, (f) does not cause them to be bothered a day later, (g) does cause them to think they were a good parent a day later, (h) promotes respect by child, (i) does not promote hostility in child, (j) promotes cooperation, (k) gets child to stop misbehaving immediately, (l) gets child to stop misbehaving in the future, and (m) gets child to obey authority. The Emotionality factor measures the extent to which parents report feeling (a) irritated before spanking their child, (b) frustrated before spanking their child, and © upset when seeing another parent spank a child. We inferred that the Instrumentality factor assesses parents’ beliefs about whether particular goals are met by the use of corporal punishment, whereas the Emotionality factor taps the emotions parents associate with corporal punishment.

Parental Responses to Child Misbehavior (PRCM). This questionnaire asks parents to rate how often in an average week they use nine disciplinary responses to their children’s misbehaviors (Holden, Coleman, & Schmidt, 1995). Using an 8-point interval scale from never to 9 or more times per week, parents indicated how often they reason, divert, negotiate, threaten, use time-out, spank or slap, ignore, withdraw privileges, and yell in anger. The questionnaire was found to be internally consistent with this sample, Cronbach’s α=.73\alpha=.73. Test-retest reliability was assessed with a separate sample of 21 mothers, r(21)=.64,p<.01r(21)=.64, p<.01 (range =.93=.93 [spank] to .26 [withdrawal of privileges]; Holden et al., 1995). The validity of the corporal punishment item has also been established through a high correlation with mothers’ daily reports of spanking in phone interviews, r(38)=.77,p<r(38)=.77, p< .001 (Holden et al., 1995).

Phone Interview

The phone interview, developed for this study, asked parents a series of questions regarding their beliefs about corporal punishment and its consequences for children. As part of a larger interview, mothers were asked whether their children experience any of five positive and eight negative consequences of being spanked. Regarding positive consequences, mothers were asked, “Does being spanked make your child …” (a) more obedient, (b) respectful of authority, © learn not to do dangerous things, (d) a stronger person, and (e) less rebellious. Mothers also reported the perceived negative consequences of spanking by answering, “Does being spanked make your child …” (a) feel bad about him or herself, (b) more aggressive with peers, © feel scared, (d) feel angry, (e) afraid of authority, (f) more rebellious, (g) resentful, and (h) hit back. The test-retest reliability of the entire interview was determined by reinterviewing eight mothers ( 9%9 \% of the sample) a second time, an average of 31 days after the first interview. The mean correlation between mothers’ responses across the two times was r(8)=.79,p<r(8)=.79, p< .001 (range =.62=.62 to .85 ).

Computer Vignettes

We used a computer to present to parents six vignettes of children’s misbehaviors. The misbehaviors illustrating each category of transgression were previously rated by mothers in a pilot study ( n=17n=17 ) as equally serious on a 6 -point scale ranging from not at all serious to extremely serious (prudential Ms=5.7M s=5.7 and 5.6 ; moral Ms=5.1M s=5.1 and 5.0 ; social Ms=3.4M s=3.4 and 3.3). To assess the effect of escalated disobedience by the child on parents’ discipline choice, one vignette of each type included willful disobedience. These escalated transgressions involved clear and knowing disobedience by the child after he or she understood the parent’s request. Internal consistency across the six vignettes was high, with an average Cronbach α\alpha of .85 (range =.70=.70 to .96 ).

In the prudential nonescalated vignette, parents imagined that their children were opening some medicine they found; in the prudential escalated vignette, parents were asked to imagine that their children ran away from them in a busy parking lot after the parent asked the child to hold his or her hand. In the moral nonescalated vignette, the child hit a friend while arguing; in the moral escalated vignette, the child threw a book at and kicked the parent after being reprimanded. Finally, the social nonescalated vignette involved a child who interrupted the parent having an important telephone conversation; the social escalated vignette involved a child who sneaked out of the bedroom after repeatedly being told to stop coming out of the room. After reading
each vignette, parents were asked how likely (on a 7 -point scale ranging from extremely unlikely to extremely likely) they were to respond to the misbehavior with corporal punishment and with reasoning.

After reporting the likelihood that they would use corporal punishment and reasoning in each situation, parents were then asked to imagine that they had corporally punished their children for the misbehavior and to rate the probability of three positive and four negative outcomes: (a) immediate compliance to the parent’s wishes, (b) future recurrence of the misbehavior, © long-term socialization of acceptable behavior, (d) emotional distress in their child, (e) feelings of guilt in the child over the transgression, (f) an increase in the child’s respect for parental authority, and (g) their own experience of guilt.

Procedure

There were 14 married couples in the sample, and thus for the analyses that included both mothers and fathers, we excluded these 14 mothers to ensure independent data (fathers were more difficult to recruit). We analyzed questionnaire data from 79 mothers and 39 fathers ( 89%89 \% of the total sample). We interviewed 85 of the mothers by phone ( 91%91 \% of all mothers; 8 declined to be interviewed), and a subsample of 38 mothers and 39 fathers ( 58%58 \% of total sample) came into the laboratory to participate in the computer vignette portion of the study.

Questionnaires

Parents completed the questionnaires while either at the laboratory (for those participants who came to do the computer task) or at home and returned them by mail in a provided stamped envelope (for mothers interviewed by phone).

Phone Interview

We telephoned mothers to participate in a telephone interview regarding their beliefs about and experiences with corporal punishment. The questions concerning positive or negative outcomes used in this study comprised only a few minutes of the 30−min30-\mathrm{min} interview.

Computer Vignettes

In a research lab, parents operated a computer to respond to questions following six computerpresented vignettes, one escalated and one nonescalated situation for each of three types of child transgressions: prudential, moral, and social. For this task, parents were asked to imagine themselves in each situation with their own children; the name of

the parent’s child appeared in the vignettes to facilitate the parent’s identification with the situation. The order in which the six vignettes were presented was randomized across parents for the approximately 45 -min-long task.

Results

Religion and Reported Disciplinary Practices

For the analysis of religious differences in parents’ reported use of nine disciplinary techniques, an initial 4 (religious affiliation) ×9\times 9 (discipline practice) multivariate analysis of variance (MANOVA) of the PRCM was significant, F(27,290)=1.57,p<.05F(27,290)=1.57, p<.05. When we examined follow-up analyses of variance (ANOVAs), we found a striking result (see Table 1). The four religious affiliation groups did not differ in the frequency with which they reportedly administered eight types of discipline (in descending order of frequency): reason, divert, negotiate, threaten, time-out, yell in anger, ignore, and withdraw privileges. Contrary to our prediction, no pattern of reported disciplinary practices differentiated the groups. Rather, the only significant difference between groups was the frequency with which parents reported spanking their children.

Post hoc Bonferroni tests revealed a consistent effect: Conservative Protestant parents
indicated that they spanked their children more often than Mainline Protestant parents, Roman Catholic parents, and parents with no religious affiliation. The latter three groups did not differ in the frequency with which they reported spanking their children, each averaging less than once per week, whereas Conservative Protestant parents reported using corporal punishment once or twice per week. These differences between religious groups were magnified in parents’ reports of frequent spanking: 29%29 \% of the Conservative Protestant parents spank their children three or more times per week, compared with 5%5 \% of Mainline Protestants, 3%3 \% of Roman Catholics, and 0%0 \% of parents with no affiliation.

Religion and Beliefs about Corporal Punishment

As a test for possible religious affiliation differences in parents’ beliefs about the instru-mental-emotional dimensions of corporal punishment, a 4 (religious affiliation) ×2\times 2 (Instrumentality and Emotionality factors) MANOVA of the PBAS revealed a significant main effect, F(6,216)=3.82,p<.001F(6,216)=3.82, p<.001. One-way ANOVAs identified significant group differences on the Instrumentality factor but not on the Emotionality factor (see Table 2). In follow-up Bonferroni tests, Conservative Protestant parents were

Table 1
Religious Differences in Parental Responses to Children’s Misbehaviors

Note. N=118\quad N=118. Means range from 0 (never) to 7 (nine or more times per week). Means in the same row that do not share subscripts differed at p<.05p<.05 in post hoc Bonferroni multiple comparisons. Sample sizes in individual analyses differ.
∗p<.01{ }^{*} p<.01.

Table 2
Religious Differences in Parents’ Beliefs About Spanking

Note. N=118\quad N=118. Means range from 1 (strongly disagree) to 7 (strongly agree). Means in the same row that do not share subscripts differed at p<.05p<.05 in post hoc Bonferroni multiple comparisons.
∗∗p=.001{ }^{* *} p=.001.
found to emphasize the instrumental benefits of corporal punishment significantly more than Mainline Protestants, Roman Catholics, and nonaffiliated parents. The only parent sex difference in this study was found here: Mothers with no religious preference were significantly less likely to endorse the instrumentality of corporal punishment than mothers of the three other religious affiliations, F(1,104)=6.37F(1,104)=6.37, p<.05(MNA=2.0p<.05\left(M_{\mathrm{NA}}=2.0\right. vs. MRC=3.6,MMP=3.2M_{\mathrm{RC}}=3.6, M_{\mathrm{MP}}=3.2, &MCP=4.1;NA=\& M_{\mathrm{CP}}=4.1 ; \mathrm{NA}= no affiliation, RC=\mathrm{RC}= Roman Catholic, MP = Mainline Protestant, and CP=\mathrm{CP}= Conservative Protestant). We did not find this difference for fathers.

When asked during phone interviews about the effects of corporal punishment, mothers of the four religious groups did not differ in the number of positive consequences they attributed to their use of corporal punishment, such as making the child more obedient or a stronger person (see Table 3). However, mothers across the four religious groups did identify significantly different numbers of negative consequences of their spanking, such as making the child more aggressive or making the child resentful (see Table 3). Post hoc Bonferroni tests determined that Mainline Protestants and Roman Catholics reported significantly more negative consequences of spanking than Conservative Protestants. Specifically, Conservative Protestant mothers were much less likely than other mothers to report that corporal punishment made their children aggressive ( 8%8 \% ), resentful (8%)(8 \%), afraid of authority ( 0%0 \% ), or more rebellious (0%)(0 \%).

Given these findings, we wondered whether differences in parents’ beliefs about the instrumentality of corporal punishment mediate the association between religious affiliation and
frequency of using corporal punishment. To test this possibility, we conducted two analyses of covariance (ANCOVAs) using as covariates (a) parents’ beliefs about the instrumental utility of corporal punishment and (b) their beliefs about the number of negative outcomes resulting from corporal punishment. In the first one-way ANCOVA predicting reported frequency of corporal punishment from religious affiliation with instrumentality as a covariate, the main effect of religion became nonsignificant, F(2F(2, 77)=2.14,ns77)=2.14, n s, whereas the Instrumentality covariate was highly significant, F(1,77)=F(1,77)= 18.05, p<.0001p<.0001. In the second ANCOVA predicting reported frequency of corporal punishment from religious affiliation and number of negative outcomes attributed to corporal punishment, the main effect of religion remained significant, F(3,56)=4.68,p<.01F(3,56)=4.68, p<.01, whereas the negative outcomes covariate was not significant, F(1,56)=0.07,nsF(1,56)=0.07, n s. These findings indicate that parents’ beliefs in the instrumental effectiveness of corporal punishment mediates the association between religious affiliation and frequency of using corporal punishment.

With regard to their beliefs about the appropriateness of corporal punishment for specific child misbehaviors in vignettes, a religious group main effect was found for parents’ intended likelihood of spanking, F(3F(3, 73)=8.79,p<.000173)=8.79, p<.0001, as well as their likelihood of reasoning, F(3,73)=3.42,p<.05F(3,73)=3.42, p<.05, across the three types of misbehaviors. Bonferroni post hoc tests revealed that Conservative Protestants were significantly more likely than the other three religious groups to say that they would spank their children ( MCP=3.3M_{\mathrm{CP}}=3.3 vs. MMP=1.9,MRC=2.0,&MNA=1.6M_{\mathrm{MP}}=1.9, M_{\mathrm{RC}}=2.0, \& M_{\mathrm{NA}}=1.6 ). Furthermore, Conservative Protestants were signifi-

cantly less likely to indicate they would reason in response to child misbehaviors ( MCP=5.3M_{\mathrm{CP}}=5.3 ) than Mainline Protestants ( MMP=6.2M_{\mathrm{MP}}=6.2 ) and parents with no religious affiliation ( MNA=6.3M_{\mathrm{NA}}=6.3 ).

These religious differences in intentions to use corporal punishment and reasoning were not moderated by whether the child had committed a prudential, moral, or social transgression: The type of transgression did not differentiate the four religious affiliation groups regarding likelihood to spank, F(6,144)=1.41,nsF(6,144)=1.41, n s, or likelihood to reason, F(6,144)=0.98,nsF(6,144)=0.98, n s. Indeed, all parents in this study agreed that moral (M=2.4)(M=2.4) and prudential (M=2.4)(M=2.4) transgressions warrant corporal punishment more than social transgressions ( M=1.5M=1.5 ).

The escalation of children’s disobedience did moderate parents’ intentions to spank or reason, however. In nonescalated situations, parents in each of the religious affiliation groups did not differ in their likelihood of using corporal punishment or reasoning. Yet in escalated situations, in which parents are confronted with an openly defiant child, parents’ intentions to spank increased, F(3,73)=5.57,p<.01F(3,73)=5.57, p<.01, and to reason decreased, F(3,73)=3.65,p<.05F(3,73)=3.65, p<.05. This was especially true of the intentions of Conservative Protestants (see Figure 1). Post hoc tests revealed that, when faced with escalated disobedience, Conservative Protestant parents would spank more ( MCP=4.3M_{\mathrm{CP}}=4.3 ) than all other parents ( MMP=2.3,MRC=2.3,&M_{\mathrm{MP}}=2.3, M_{\mathrm{RC}}=2.3, \& MNA=1.8M_{\mathrm{NA}}=1.8 ). In addition, Conservative Protestants would reason less in response to escalated disobedience ( MCP=4.6M_{\mathrm{CP}}=4.6 ) than all other parents ( MMP=6.1,MRC=5.5,&MNA=6.2M_{\mathrm{MP}}=6.1, M_{\mathrm{RC}}=5.5, \& M_{\mathrm{NA}}=6.2 ). Parents’ occupational level was a significant covariate in this instance, F(3,72)=8.52,p<.01F(3,72)=8.52, p<.01. On inspecting the correlations between occupational status and likelihood of reasoning, we found only one to be significant: as the status of their occupations increased, Roman Catholic parents were less likely to use reasoning in response to escalated misbehaviors, rRC(21)=r_{\mathrm{RC}}(21)= −.66,p<.001-.66, p<.001.

We found few religious differences in parents’ ratings of the likelihood of outcomes when corporal punishment is used for specific child misbehaviors. The only significant main effect was for parental guilt (see Table 4). Conservative Protestant parents indicated they would feel less guilt after spanking than did all other parents. In addition, we found one significant
interaction between religious affiliation and type of transgression, F(6,144)=3.09,p<.01F(6,144)=3.09, p<.01, for parents’ predictions about the future recurrence of the transgression. Although all parents believed spanking would decrease the likelihood of future recurrence for prudential transgressions, the Conservative Protestants predicted that moral misbehaviors would not recur ( MCP=4.1M_{\mathrm{CP}}=4.1 ) more than parents of the other religious groups ( MMP=4.9,MRC=5.2,&M_{\mathrm{MP}}=4.9, M_{\mathrm{RC}}=5.2, \& MNA=5.0M_{\mathrm{NA}}=5.0 ). Similarly, Conservative Protestant parents predicted fewer recurrences of social transgressions after corporal punishment ( MCP=3.9M_{\mathrm{CP}}=3.9 ) than did parents from the other three religious groups ( MMP=5.4,MRC=4.8M_{\mathrm{MP}}=5.4, M_{\mathrm{RC}}=4.8, &MNA=4.6\& M_{\mathrm{NA}}=4.6 ).

Discussion

Using several methods in this study, we examined religious affiliation as a source of differences in parents’ reported use of and beliefs about corporal punishment and other forms of discipline. This study has extended the literature on religious affiliation and parenting by testing for religious differences in practices other than corporal punishment. In addition, we have examined in some depth parents’ beliefs about the appropriateness, effectiveness, and effects of corporal punishment in general and for specific child misbehaviors. Throughout, we found a consistent, yet circumscribed, effect of religious affiliation: Conservative Protestants differed from other parents primarily in terms of their use of and beliefs about corporal punishment.

Religious Differences in Parents’ Disciplinary Practices and Beliefs

Perhaps the most striking finding of this study is that although religious groups differ in the frequency with which they report using corporal punishment, they do not differ in how much they reason, divert, negotiate with, threaten, use time-out with, ignore, withdraw privileges from, or yell at their children. The specificity of this religious difference in parenting practices suggests that Conservative Protestant doctrine singles out corporal punishment as crucial to effective parenting. It is encouraging that all parents equally use nonpunitive discipline such as reasoning, diversion, and negotiation. It may be that Conservative Protestant parents are more

Table 4
Religious Differences in Outcome Expectancies for Corporal Punishment

Note. N=77\quad N=77. Means range from 1 (extremely unlikely) to 7 (extremely likely). Means in the same row that do not share subscripts differed at p<.05p<.05 in post hoc Bonferroni multiple comparisons.
likely than other parents to use corporal punishment as a backup when other disciplinary techniques fail, a strategy that has proven effective in securing child compliance (Latzelere, Sather, Schneider, Larson, & Pike, 1998). Future studies that examine religious group differences in the combinations of discipline techniques used by parents would specify when Conservative Protestant parents are using corporal punishment.

In light of this finding of religious differences only in parents’ use of corporal punishment, it is especially important to understand the role of parents’ beliefs in determining the frequency of their corporal punishment. The extent of parents’ beliefs in the instrumental effectiveness of corporal punishment proved to be an important factor accounting for religious differences in parents’ beliefs about corporal punishment. As predicted, Conservative Protestant parents believe in the instrumental effectiveness of corporal punishment more than other parents. When examined further, it is Conservative Protestant parents’ strong beliefs in the instrumentality of corporal punishment in particular that is linked with the frequency of their corporal punishment use. It appears that a parent’s belief in the instrumental usefulness of corporal punishment, rather than his or her religious affiliation per se, helps predict his or her likelihood of using corporal punishment.

Parents of all four religious affiliations generally concurred about the positive outcomes
they achieve when using corporal punishment, yet they differed in the negative outcomes they perceived. In particular, Conservative Protestants were least likely to think that negative outcomes such as children’s aggression, anger, or resentment would accrue from corporal punishment. Interestingly, none of the Conservative Protestant parents believed that corporal punishment made their children afraid of authority or more rebellious. This finding is consistent with the proscriptions of conservative parenting literature (e.g., Dobson, 1970; Fugate, 1980; Lessin, 1979) emphasizing corporal punishment as a particularly effective means of securing the outcome they view as paramount: children’s obedience to parental authority.

In surprising contrast to the apparent role of instrumental beliefs in predicting parents’ corporal punishment, the number of negative effects attributed to corporal punishment did not affect the association between parents’ religious affiliation and their frequency of using corporal punishment. Attributing negative effects to corporal punishment is thus not the antithesis of believing corporal punishment is instrumentally effective. It may be that parents attend to the positive effects and discount the negative effects of disciplinary techniques that they use with good intentions. For parents of all religious affiliations, the effectiveness of corporal punishment in securing children’s compliance and in teaching them to behave appropriately in the

future may outweigh any possible negative outcomes that may result.

Interestingly, parents from the four religious affiliation groups did not differ in their evaluations of which transgressions deserve corporal punishment. Consistent with previous findings (Catron & Masters, 1993), parents from all four religious groups agreed that they would use corporal punishment for moral and prudential transgressions more than for social transgressions. Thus, like other parents, Conservative Protestant parents do not use corporal punishment indiscriminately but rather differentiate some misbehaviors as more deserving of firm punishment than others.

Although the type of misbehavior did not differentiate parents, the presence of escalation did: Conservative Protestants were more likely to use corporal punishment and less likely to use reasoning if their children openly defied them. This is consistent with the recommendations by some religious authors that willful disobedience warrants a strong reaction from parents (e.g., Dobson, 1970; Fugate, 1980). Conservative Protestant parents were more likely than all other parents to respond to children’s escalated disobedience with corporal punishment, appearing to view escalated disobedience as a dangerous challenge to their parental authority that must be quelled as swiftly and strongly as possible.

Only two religious differences were found in parents’ beliefs about the outcomes of corporal punishment for specific misbehaviors, namely parental guilt and future recurrence of the misbehavior. Conservative Protestant parents likely anticipate feeling little guilt if they use corporal punishment because of their religionsupported belief in the necessity and effectiveness of corporal punishment. The ambivalence about corporal punishment shared by parents of the other religious affiliations may cause them to feel guilty when they do use it. As an extension of their belief in the instrumental effectiveness of corporal punishment, Conservative Protestant parents predicted that corporal punishment would reduce the future occurrence of all three types of child misbehaviors. Although they would use corporal punishment more for moral and prudential than for social transgressions, Conservative Protestants predicted that corporal punishment would be successful at preventing future misbehavior whenever it was used. Thus,
the frequency with which Conservative Protestants use corporal punishment is supported by the consistency in their beliefs about its effectiveness.

Unlike other studies that suggest SES differences in the use and effects of corporal punishment (Grasmick et al., 1992; Gunnoe & Mariner, 1997; Kelley, Power, & Wimbush, 1992), this study did not find consistent effects of SES on parents’ beliefs about and reported use of corporal punishment. The lack of SES differences in this study may be a result of the use of a relatively homogeneous, mostly collegeeducated sample. Furthermore, the sex of the parent was not related consistently to the association between religious affiliation and disciplinary beliefs and practices. The absence of a parent sex effect could be attributed to the possibility that parent sex was confounded with religious affiliation, as mothers were more likely to profess a Mainline Protestant or Roman Catholic affiliation and fathers were more likely to report a Conservative Protestant affiliation or no affiliation at all.

Remaining Issues for Future Research

The questions of how corporal punishment actually affects children and whether its effects vary by religious group are ones that we do not directly address. However, the relatively frequent use of corporal punishment by some of the parents in this study, as often as three or more times per week, may put children at particular risk for the behavioral, psychological, and relationship problems that are associated with corporal punishment (American Academy of Pediatrics, 1998; Gershoff, 1999). Several cooccurring factors may determine whether corporal punishment has negative effects on children. Children’s acceptance of corporal punishment as deserved and fair, which is more likely to occur when parents’ use of corporal punishment is supported by the family’s religion or culture, can determine whether spanking will have positive or negative effects (Baumrind,-1997; Grusec & Goodnow, 1994; Rohner, Bourque, & Elordi,-1996). Consistency in use of corporal punishment, which is more likely to characterize spanking by parents who value its instrumental effectiveness, may also predict more positive outcomes for children because it is expected and accepted (Deater-Deckard −{ }^{-}−Dodge, 1997). Finally, the overall parenting style within which

corporal punishment is embedded may determine its effects and effectiveness more than the mere fact that it is used (Baumrind, 1997; Darling & Steinberg, 1993). These possible mitigating factors deserve future research attention.

Implications for Application and Public Policy

Understanding why parents prefer some disciplinary methods over others is a prerequisite for teaching parents how to best socialize their children in ways that positively impact their emotional and social development (Goodnow & Collins, 1990). The present study has determined that parents’ religious affiliation in part determines their beliefs about and use of corporal punishment. Thus, policy-oriented attempts to discourage parents from using corporal punishment should respect parents’ religious convictions while emphasizing the effective alternatives to corporal punishment, such as induction, natural consequences, and withdrawal of privileges. Moreover, our finding that parents’ beliefs about the instrumental effectiveness of corporal punishment is associated with their frequency of using it over and above their religious affiliation suggests that parent education programs whose goals are to reduce the incidence of corporal punishment and physical abuse should target parents’ belief that corporal punishment is instrumental in achieving their child-rearing goals.

Conclusion

Understanding the sources of parenting beliefs and practices is critical to promoting positive and preventing negative child outcomes. This work reveals the largely ignored but important influence of religion on parents’ disciplinary preferences and beliefs and represents an initial effort at documenting and evaluating the parenting messages delivered from the pulpit in the name of religion.

References

American Academy of Pediatrics. (1998). Guidance for effective discipline. Pediatrics, 101, 723-728. Bartkowski, J. P., & Ellison, C. G. (1995). Divergent models of childrearing in popular manuals: Conser-
vative Protestants vs. the mainstream experts. Sociology of Religion, 56, 21-34.
Baumrind, D. (1997). Necessary distinctions. Psychological Inquiry, 8, 176-182.
Carey, T. A. (1994). Spare the rod and spoil the child: Is this a sensible justification for the use of punishment in child rearing? Child Abuse and Neglect, 18, 1005-1010.
Catron, T. F., & Masters, J. C. (1993). Mothers’ and children’s conceptualizations of corporal punishment. Child Development, 64, 1815-1828.
Chilamkurti, C., & Milner, J. S. (1993). Perceptions and evaluations of child transgressions and disciplinary techniques in high- and low-risk mothers and their children. Child Development, 64, 1801-1814.
Darling, N., & Steinberg, L. (1993). Parenting style as context: An integrative model. Psychological Bulletin, 113, 487-496.
Day, R. D., Peterson, G. W., & McCracken, C. (1998). Predicting spanking of younger and older children by mothers and fathers. Journal of Marriage and the Family, 60, 79-94.
Deater-Deckard, K., & Dodge, K. A. (1997). Externalizing behavior problems and discipline revisited: Nonlinear effects and variation by culture, context, and gender. Psychological Inquiry, 8, 161-175.
Dix, T., Ruble, D. N., & Zambarano, R. J. (1989). Mothers’ implicit theories of discipline: Child effects, parent effects, and the attribution process. Child Development, 60, 1373-1391.
Dobson, J. (1970). Dare to discipline. Wheaton, IL: Tyndale.
Ellison, C. G. (1995). Religious Affiliation Questionnaire. Unpublished measure, University of Texas at Austin.
Ellison, C. G., Bartkowski, J. P., & Segal, M. L. (1996). Conservative Protestantism and the parental use of corporal punishment. Social Forces, 74, 1003-1028.
Ellison, C. G., & Sherkat, D. E. (1993). Conservative Protestantism and support for corporal punishment. American Sociological Review, 58, 131-144.
Fugate, J. R. (1980). What the Bible says about … child training. Tempe, AZ: Alpha Omega.
Gershoff, E. T. (1999). The short- and long-term effects of corporal punishment on children: A process model and meta-analytic review. Manuscript submitted for publication.
Giles-Sims, J., Straus, M. A., & Sugarman, D. B. (1995). Child, maternal, and family characteristics associated with spanking. Family Relations, 44, 170−176170-176.
Goodnow, J. J., & Collins, W. A. (1990). Development according to parents: The nature, sources, and consequences of parents’ ideas. Hillsdale, NJ: Erlbaum.
Gralinski, J. H., & Kopp, C. B. (1993). Everyday rules for behavior: Mothers’ requests to young children. Developmental Psychology, 29, 573-584.

Grasmick, H. G., Morgan, C. S., & Kennedy, M. B. (1992). Support for corporal punishment in the schools: A comparison of the effects of socioeconomic status and religion. Social Science Quarterly, 73, 177-187.
Greven, P. (1991). Spare the child: The religious roots of punishment and the psychological impact of physical abuse. New York: Knopf.
Grusec, J. E., Dix, T., & Mills, R. (1982). The effects of type, severity, and victim of children’s transgressions on maternal discipline. Canadian Journal of Behavioral Sciences, 14, 276-289.
Grusec, J. E., & Goodnow, J. J. (1994). Impact of parental discipline methods on the child’s internalization of values: A reconceptualization of current points of view. Developmental Psychology, 30, 4−194-19.
Gunnoe, M. L., & Mariner, C. L. (1997). Toward a developmental-contextual model of the effects of parental spanking on children’s aggression. Archives of Pediatric and Adolescent Medicine, 151, 768-775.
Hoffman, M. L. (1983). Affective and cognitive processes in moral internalization. In E. T. Higgins, D. N. Ruble, & W. W. Hartup (Eds.), Social cognition and social development (pp. 236-274). New York: Cambridge University Press.
Holden, G. W. (1997). Parents and the dynamics of child rearing. Boulder, CO: Westview.
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.
Holden, G. W., & Miller, P. C. (1997, April). Cognitive versus emotional parenting: Alignments between child-rearing cognitions, emotions, and reported behavior. In C. Tamis-LeMonda (Chair), The cognitive and affective sides of parenting: The roles of parenting views, emotions, and expectations in parenting interactions and children’s development. Presented at the biennial meeting of the Society for Research in Child Development, Washington, DC.
Hollingshead, A. B. (1975). Four-Factor Index of Social Status. Unpublished manuscript, Yale University, New Haven, CT.
Kelley, M. L., Power, T. G., & Wimbush, D. D. (1992). Determinants of disciplinary practices in low-income black mothers. Child Development, 63, 573-582.
Larzelere, R. E., Sather, P. R., Schneider, W. N., Larson, D. B., & Pike, P. L. (1998). Punishment enhances reasoning’s effectiveness as a disciplinary response to toddlers. Journal of Marriage and the Family, 60, 388-403.
Lessin, R. (1979). Spanking: Why, when, how? Minneapolis, MN: Bethany Fellowship.

Minton, C., Kagan, J., & Levine, J. A. (1971). Maternal control and obedience in the two-year-old. Child Development, 42, 1873-1894.
Nelsen, H. M., & Kroliczak, A. (1984). Parental use of the threat “God will punish”: Replication and extension. Journal for the Scientific Study of Religion, 23, 267-277.
Power, T. G., & Chapieski, M. L. (1986). Childrearing and impulse control in toddlers: A naturalistic investigation. Developmental Psychology, 22, 271275.

Rohner, R. P., Bourque, S. L., & Elordi, C. A. (1996). Children’s perceptions of corporal punishment, caretaker acceptance, and psychological adjustment in a poor, biracial Southern community. Journal of Marriage and the Family, 58, 842-852.
Smetana, J. G. (1997). Parenting and the development of social knowledge reconceptualized: A social domain analysis. In J. E. Grusec & L. Kuczynski (Eds.), Parenting and children’s internalization of values: A handbook of contemporary theory (pp. 162-192). New York: Wiley.
Strassberg, Z., Dodge, K. A., Pettit, G. S., & Bates, J. E. (1994). Spanking in the home and children’s subsequent aggression toward kindergarten peers. Development and Psychopathology, 6, 445-461.
Straus, M. A. (1994). Beating the devil out of them: Corporal punishment in American families. New York: Lexington Books.
Straus, M. A., & Mouradian, V. E. (1998). Impulsive corporal punishment by mothers and antisocial behavior and impulsiveness of children. Behavioral Sciences and the Law, 16, 353-374.
Tisak, M., & Turiel, E. (1984). Children’s conceptions of moral and prudential rules. Child Development, 55, 1030-1039.
Turiel, E. (1977). Distinct conceptual and developmental domains: Social convention and morality. In A. Pick (Ed.), Nebraska symposium on motivation, (Vol. 25, pp. 77-116). Lincoln: University of Nebraska Press.
Vasta, R. (1982). Physical child abuse: A dualcomponent analysis. Developmental Review, 2, 125−149125-149.
Wiehe, V. R. (1990). Religious influence on parental attitudes toward the use of corporal punishment. Journal of Family Violence, 5, 173-186.
Zahn-Waxler, C., Radke-Yarrow, M., & King, R. A. (1979). Child rearing and children’s prosocial initiations toward victims of distress. Child Development, 50, 319-330.

Received July 28, 1997
Revision received February 26, 1999
Accepted March 4, 1999