Variation in Acceptable Child Discipline Practices by Child Age: Perceptions of Community Norms by Medical and Legal Professionals

Abstract

Mandated child abuse reporters may judge specific disciplinary practices as unacceptable for young children, whereas child law professionals arbitrating allegations may be less inclusive. Do the views of these groups diverge, by child age, regarding physical discipline? Judgments of community norms across a wide range of children’s ages were obtained from 380 medical and legal professionals. Because the Parent-Child Conflict Tactics Scale (PC-CTS) can be used to assess the epidemiology of child disciplinary behaviors and as a proxy to examine the incidence or prevalence of child abuse, the disciplinary practices described on the PC-CTS were presented as triggers for questions. Significant child age effects were found for disciplinary practices classified as “harsh.” The consistencies between legal and medical professionals were striking. Both groups reflected changes in United States norms, as non-physical approaches were the most approved. We conclude that instruments estimating the prevalence of child maltreatment by parent-report should consider modifying how specific disciplinary practices are classified.

All children require discipline because lessons regarding proper behavior need to be taught and behavior corrected. Acceptable parental disciplinary practices are culturally defined; actual parental practices vary widely across cultures (Runyan et al., 2010). Studies of the frequency of disciplinary practices and the use of harsher forms of physical or emotional discipline have been used to estimate the occurrence of child physical abuse, although harsh punishment may not meet legal definitions of abuse (Runyan et al., 2009; Runyan et al., 2010; Theodore et al., 2005). This proxy measurement of child abuse has a significant advantage in that only the parent and the child may be present when maltreatment occurs and no report to social services or the police may follow. Ample evidence exists that parents will self-report harsh punishment of their children at rates far greater than recorded rates of physical abuse (Theodore et al., 2005).

Some legally mandated reporters of child discipline, such as pediatricians, are in a position to ask about or provide guidance on child discipline. In the context of clinical care, doctors may inquire about child discipline practices or be told about specific acts in the context of evaluation of marks or bruises on a child’s body. These professionals will need to make a decision about whether acts reported by parents are unacceptable and thus abusive. Final determination of whether specific disciplinary acts are abusive in nature is often a legal determination.

In the context of developing epidemiological estimates of the use of physical and emotional punishment of children and approximating the “true” rates of maltreatment, we were faced with deciding whether the “harsh” categorization of some disciplinary practices in the Parent-Child Conflict Tactics Scale (PC-CTS) should take child age into account. Specifically, we wondered whether there is a consensus among both physicians, who are mandatory reporters, and lawyers, who deal with the legal determination of abuse, as to which child disciplinary practices, at which ages, might be considered harsh and potentially unacceptable. As the PC-CTS does not ascertain whether marks, bruising, or physical harm has occurred, we wondered which acts admitted by parents of children, at different ages, might be interpreted as harsh by professionals working in child abuse. We hypothesized that physicians, with a duty to report, might choose a lower threshold for harshness, erring on the side of classifying acts as harsh so that children might be better protected, and thereby valuing sensitivity of definition. We further hypothesized that attorneys in the field of child abuse and neglect might either be inured to harsh punishment or choose to err on the side of avoiding labeling questionable behaviors as illegal behaviors. Using the disciplinary practices described in the PC-CTS as the triggers for questions, we undertook surveys of child abuse pediatricians and legal professionals through professional organization listservs to assess their perceptions of acceptability of specific disciplinary practices with children in their communities.

Parent/Child Conflict Tactics Scale

In the field of public health, surveillance is undertaken to understand the distribution of a disease or condition or the burden (prevalence) of a condition in the population. The PC-CTS has been used as one tool to assess the epidemiology of child disciplinary behaviors and as a proxy in studies that examine the incidence or prevalence of child abuse (Straus, Hamby, Finkelhor, Moore, & Runyan, 1998a, 1998b). It is clear that most maltreatment is done in private and official report data are inadequate to estimate the burden or prevalence of child abuse (Theodore et al., 2003). The PC-CTS obtains information directly from a participant unfiltered by child abuse laws, reporting biases, and decisions in the investigation and legal processes that follow a report (Theodore et al., 2005). Although this scale has been important in estimating the proportion of children exposed to behaviors generally considered to be physically or emotionally abusive, the PC-CTS classification of behaviors as “harsh” does not consider child age, with the single exception that shaking a child younger than two years of age is part of the PC-CTS harsh category. For example, many might consider one specific parenting behavior - using an object to spank a child on the buttocks - as acceptable physical discipline for children in general, but view the same act as “harsh” physical discipline when administered to a 3-month-old child.

Corporal Punishment

Several items within the PC-CTS address corporal punishment and harsh corporal punishment. In her meta-analysis, Gershoff (2002) reported significant associations between corporal punishment and multiple negative outcomes for its recipients. Among the concerning outcomes were aggression, antisocial behavior, delinquency, mental health problems, parent-child relationship issues, and increased risks of commission and victimization of physical abuse (see Gershoff & Bitensky, 2007; Gershoff, 2010, 2013).

Using national data from a 1995 Gallup Organization poll, Straus and Stewart (1999) measured the prevalence of a variety of corporal punishment methods. They found that the percentage of United States parents who reported having used any form of corporal punishment rose from 35% for infants to 94% for children who were between 3 and 4 years old. The prevalence fell for children from the ages of 5 to 17 years old, with slightly more than 40% of children who were 13 years old receiving corporal punishment. In addition, the severity of punishment varied by child age. Mild corporal punishments (barehanded spanking of the buttocks and slapping of the limbs) were common for children between the ages of 2 and 4 years. More intense and less normative punishments (spanking with an object, slapping of the face, head, or ears, and pinching) were most common for children between the ages of 5 and 12. Among United States children, pinching and slapping of the face, head, or ears were the least common methods of punishment. Noting that shaking a child under 2 years old can be fatal, they reported that 10.3% of children between 12 and 23 months old were shaken, although no children younger than 1 year in this sample were reported to have been shaken.

A number of other researchers have demonstrated that corporal punishment practices vary by the age of the recipient children (Bradley, Corwyn, McAdoo, & Coll, 2001; Dietz, 2000; Gershoff, Lansford, Sexton, Davis-Kean, & Sameroff, 2012; MacKenzie, Nicklas, Brooks-Gunn, & Waldfogel, 2015). Dietz (2000) examined the PC-CTS findings by child age and degree of punishment severity. With data from the Gallup Organization 1995 poll, she found that the majority of children between the ages of 2 and 8 years old had received corporal punishment within the past year; parents were less often found to corporally punish older, adolescent children. She reported that 57% of children had received ordinary corporal punishment (spanking or slapping of the limbs), and 26% received severe corporal punishment (shaking children who were 2 or more years old, hitting with an object, pinching, and slapping of the face, head, or ears). Using data from the National Longitudinal Survey of Youth, Bradley et al. (2001) found an inverse relation between spanking and child age. They noted that roughly 29% of children who were under 5 years old were reported to have been spanked at least three times within the past 7 days. They added that the percentage dropped to 15% for children between the ages of 6 and 9 years, and to 4% for children between the ages of 10 and 14 years. A more recent analysis of data from the Early Childhood Longitudinal Study’s Kindergarten Cohort of 1998–1999 similarly found that the percentage of mothers who reported using spanking within the past week decreased from 27% when the children were in kindergarten to 15% when the children were in third grade (Gershoff et al., 2012). Finding less variation, MacKenzie et al. (2015) reported, based on data from the Fragile Families and Child Well-Being Study, that 28% of 1-year-old children, 57% of 3-year-old children, 53% of 5-year-old children, and 49% of 9-year-old children were reported to have been spanked by mothers.

Community Norms and Professional Decisions

Community norms can be an important basis for what parents and professionals actually do (see for example, Taylor, Hamvas, & Paris, 2011). Health and medical professionals working in the area of child abuse and professionals who work with the law may have different perspectives on what constitutes acceptable physical discipline of young children by their parents. Previous research has been conducted on differences between prosecutors and defense attorneys in their modes of questioning in child abuse cases (Evans, Lee, & Lyon, 2009) and their opinions of abuse (Hartman, Karlson, & Hibbard, 1994). In a study of prosecutor versus defense attorney attitudes on what constitutes sexual abuse in the state of Indiana, findings showed that prosecutors were more likely to rate a behavior as abusive than were defense attorneys (Hartman et al., 1994). Ho and Gross (2015) analyzed pediatric nurses’ perceptions of whether a variety of physical punishments were acceptable for children between the ages of 3 and 6 years. Although they found that there was an overall consensus in perceptions of what was most and least acceptable, their results also indicated some variation as to what the nurses perceived to be acceptable physical punishment. It is important to study these professionals specifically because they work closely with children.

Current Study

This study examined the PC-CTS by examining professional perspectives of community norms on whether specific child disciplinary behaviors should be classified as harsh physical discipline based on child age. We compared medical and legal professionals’ judgments of community norms regarding discipline of children. In this regard, we hypothesized that medical professionals would be more likely to classify a behavior as harsh physical discipline at an earlier child age than legal professionals. The imperative for health professionals is to identify those who might benefit from intervention and eschew missing a person in need of services. If a medical professional prescribes an examination or additional services based on the false belief of abuse, the individuals involved will not suffer life-changing consequences except with relation to the actions that might be taken within the legal system. In contrast, legal professionals might be more concerned with false positives and the legal implications of a label. If a legal professional presents a case on a false belief that abuse has occurred, the individuals involved have a high probability of suffering from life-changing consequences such as removal of the child from the home, fines, and even incarceration. Blackstone’s formulation guides the law: “It is better to let ten guilty persons escape than that one innocent suffer” (Blackstone, 1765–1769).

Method

Participants

Three hundred and eighty legal and medical professionals participated in this study (Table 1). Most participants were between 31 to 60 years of age. There were 74 males and 267 females surveyed (39 participants did not identify their gender). Participants were from a variety of ethnic backgrounds (75% being Caucasian). Participants were contacted via e-mail listservs composed of physicians from the Ray E. Helfer Society (www.helfersociety.org/), physicians participating in the North Carolina Medical Evaluation program (www.med.unc.edu/cmep), members of the American Bar Association (the ABA Center on Children and the Law), North Carolina Guardian ad Litems, and social workers (North Carolina specific listservs for these professionals). These groups were selected for study because they are professionals who are likely to have given considerable thought to the issue of child discipline.

Table 1.

Sample Description.

Variable	n
Gender
Male	74
Female	267
Did Not Specify	39
Total	380
Age
20–30yrs	13
31–40yrs	80
41–50yrs	84
51–60yrs	99
61–70yrs	47
71–80yrs	4
Did Not Specify	53
Total	380
Ethnicity
Caucasian	285
African American	27
Hispanic/Latino(a)	10
Asian/Pacific Islander	5
American Indian	3
Other	7
Did Not Specify	43
Total	380
Primary Occupation
Legal Professionals
Attorney	108
Prosecutor	3
Defense Attorney	7
Judge	1
CASA/GAL	83
Social Worker	16
Paralegal	2
Medical Professionals
Pediatrician	73
Research Psychologist	1
Psychiatrist	1
Pathologist	1
Not included
Parent	10
Other	22
Did Not Specify	52
Total	380
Do You Have Children?
Yes	269
No	68
Did Not Specify	43
Total	380

The participants were divided into two groups: legal professionals and medical professionals. Legal professionals (n = 220) consisted primarily of lawyers (e.g., prosecutors and defense attorneys), social workers, Court Appointed Special Advocates (CASAs), Guardian ad Litems (GALs) (note that CASAs and GALS typically consist of trained, court-appointed volunteers who advocate for children who are victims of child abuse and/or neglect), and paralegals. Legal professionals completed the survey after receiving it from the ABA, social worker, or GAL listservs. Medical professionals (n = 76) received the survey from the Helfer and North Carolina listervs, and consisted primarily of pediatricians. Those who identified their occupation only as parents (n = 10) or “other” (n = 22), or who failed to identify their occupation (n = 52) were omitted from analyses.

Survey

Participants each received an e-mail containing a link to the online survey used in this study. The survey was created via Survey Monkey (http://www.surveymonkey.com) and then distributed to participants through e-mail listservs obtained through contacts within the legal and medical system. The survey included a consent page that described the voluntary nature of the study as well as information about possible risks and contacts for questions or concerns. Participants were asked to answer questions about forms of child discipline based on the norms of their community and were asked to respond based upon what they believed to be acceptable in the town or city in which they lived. Participants were given a list of 19 specific discipline tactics or behaviors and were asked whether they felt that their community would find such a behavior acceptable by specific age group categories for the children. A subset of (15) PC-CTS items more relevant to younger children were selected in an effort to shorten our survey so that professionals would be more willing to complete the questionnaire. In addition, we added four questions about child discipline developed from international focus groups queried during the process of adapting the PC-CTS for international studies (Runyan et al., 2010). We selected items that were culturally common in the United States and tactics that would be less likely to be limited by the physical age of the child (e.g., you would not tell an infant to stop doing something). A Likert-style scale was provided for each behavior and age with the options of: 1 = Never OK/Never use, 2 = Rarely use/In extenuating circumstances, 3 = Occasional use, and 4 = OK/General use. Thus, a higher score on the scale indicated greater acceptability of the item. A physical discipline method was defined as harsh when 90% or more of all legal and medical professionals rated that tactic as one that they believe the community norm was to “never use” or “rarely use.”

Participants were asked if each form of child discipline would be acceptable to members of their community to use for each of the following age groups of children: 0–3 months, 4–8 months, 9–13 months, 14–18 months, 19–24 months, 3–5 years, 6–10 years, 11–18 years. These ages were chosen because the behavioral capabilities of infants and children and their ability to understand disciplinary tactics change remarkably across childhood. We expected that many tactics would not be acceptable for use with children during the first two years of life but might be more acceptable for children in middle to late childhood. We selected three age ranges in the first year of life, two in the second year, and broader ranges in later childhood and early adolescence. We included several age ranges in the first two years of life because we thought respondent ratings might change more rapidly as a function of child age or be associated with more notable between-group differences during this age range as opposed to older ranges of child age. These questions were then followed by nine demographic questions about the participants. When the survey was completed, participants received a message thanking them for their involvement and providing them with contact information if they had any questions or concerns.

Results

Percent Rating Tactic as Harsh

Table 2 lists the 19 parental discipline tactics along with the overall percentage of the sample of professionals who rated the discipline method as harsh based on community standards. That is, in Table 2, we list the percentage of all professionals who rated the discipline/punishment method as one that they believe, based on community norms, would rarely or never be used (i.e., gave ratings of 1 or 2). Emphasis (bold italicized) is added to items that were rated as unacceptable by 90% or more of participants. Means and standard deviations for ratings made on the 1–4 scale for each professional group are provided with higher scores indicating more acceptable use, in regard to community norms, of the discipline tactic.

Table 2.

Percent of Sample that Rated Each Behavior as Harsh Physical Discipline and Mean Rated Acceptability by Medical and Legal Subgroups

	Statistic	0–3 months	4–8 months	9–13 months	14–18 months	19–24 months	3–5 years	6–10 years	11–18 years
Shout, yell, or scream at child (N = 350)	Percent	94	91.4	86.6	78.3	72.1	58.4	41.9	39.9
Medical (n = 75)	Mean (SD)	1.18 (.65)	1.28 (.74)	1.61 (.93)	2.04 (1.11)	2.27 (1.18)	2.62 (1.13)	2.82 (.98)	2.86 (.98)
Legal (n = 217)	Mean (SD)	1.23 (.69)	1.28 (.74)	1.49 (.95)	1.66 (.99)	1.86 (1.02)	2.35 (1.04)	2.68 (.99)	2.77 (.97)
Take away privileges (N = 344)	Percent	97.4	96.5	88.7	73.7	54.9	10.1	1.4	1.1
Medical (n = 74)	Mean (SD)	1.01 (.12)	1.03 (.23)	1.26 (.72)	1.55 (.98)	2.32 (1.15)	3.55 (.77)	3.92 (.32)	3.95 (.28)
Legal (n = 216)	Mean (SD)	1.14 (.59)	1.17 (.63)	1.44 (.95)	1.93 (1.18)	2.49 (1.18)	3.62 (.72)	3.91 (.32)	3.94 (.25)
Spank child on the buttocks with hand only (N = 347)	Percent	98	97.1	87.4	73.5	62.8	45.3	42.7	62.5
Medical (n = 76)	Mean (SD)	1.08 (.43)	1.16 (.52)	1.49 (.87)	1.87 (1.10)	2.26 (1.20)	2.70 (1.11)	2.43 (1.17)	2.03 (1.15)
Legal (n = 215)	Mean (SD)	1.07 (.38)	1.15 (.51)	1.48 (1.12)	1.88 (1.05)	2.20 (1.05)	2.69 (.95)	2.75 (.95)	2.25 (1.06)
Hit child’s buttocks with an object such as a belt or switch (N = 358)	Percent	99.4	99.7	99.2	84.4	89.8	77.5	71.4	78.3
Medical (n = 74)	Mean (SD)	1.01 (.12)	1.01 (.12)	1.09 (.33)	1.24 (.56)	1.42 (.79)	1.82 (1.13)	1.89 (1.15)	1.75 (1.06)
Legal (n = 214)	Mean (SD)	1.00 (.07)	1.02 (.14)	1.07 (.27)	1.17 (.46)	1.34 (.68)	1.69 (.96)	1.91 (1.01)	1.81 (.98)
Curse or swear at child (N = 346)	Percent	94.2	93.4	91.6	89.1	85.6	81	76	69.8
Medical (n = 74)	Mean (SD)	1.14 (.58)	1.20 (.65)	1.24 (.71)	1.32 (.80)	1.47 (.90)	1.63 (1.06)	1.76 (1.07)	1.96 (1.13)
Legal (n = 216)	Mean (SD)	1.24 (.71)	1.25 (.72)	1.31 (.78)	1.38 (.87)	1.44 (.92)	1.56 (1.00)	1.79 (1.10)	1.98 (1.12)
Giving a child a time out or sending him/her to his/her room (N = 362)	Percent	97.8	97	92.1	68.5	34.4	5.1	2.9	13.6
Medical (n = 74)	Mean (SD)	1.07 (.30)	1.09 (.38)	1.38 (.84)	2.16 (1.05)	3.09 (.96)	3.83 (.47)	3.88 (.37)	3.45 (.84)
Legal (n = 209)	Mean (SD)	1.08 (.42)	1.12 (.49)	1.36 (.74)	2.10 (1.10)	2.80 (1.05)	3.70 (.58)	3.85 (.42)	3.56 (.79)
Pinch child (N = 343)	Percent	98.5	97.7	95.9	92.7	90.1	84	81.2	83.6
Medical (n = 75)	Mean (SD)	1.04 (.35)	1.09 (.41)	1.21 (.55)	1.31 (.68)	1.47 (.83)	1.70 (1.02)	1.68 (1.02)	1.57 (.96)
Legal (n = 217)	Mean (SD)	1.06 (.38)	1.11 (.50)	1.17 (.60)	1.29 (.72)	1.38 (.80)	1.59 (.88)	1.68 (.93)	1.67 (.93)
Refuse to speak to a child as punishment (N = 344)	Percent	98.3	98.3	97.1	94.5	91.6	82.1	71.1	63.7
Medical (n = 74)	Mean (SD)	1.08 (.43)	1.08 (.43)	1.15 (.57)	1.22 (.65)	1.38 (.75)	1.76 (1.01)	2.07 (1.12)	2.29 (1.08)
Legal (n = 216)	Mean (SD)	1.06 (.39)	1.07 (.40)	1.11 (.51)	1.19 (.64)	1.30 (.79)	1.54 (.96)	1.89 (1.09)	2.11 (1.09)
Call child names like stupid, ugly, or useless (N = 346)	Percent	97.7	97.1	96.5	94.5	93.4	89	84.8	83.6
Medical (n = 75)	Mean (SD)	1.07 (.41)	1.07 (.41)	1.09 (.44)	1.13 (.53)	1.25 (.68)	1.44 (.87)	1.57 (.97)	1.67 (1.06)
Legal (n = 218)	Mean (SD)	1.09 (.48)	1.11 (.53)	1.13 (.59)	1.19 (.65)	1.22 (.68)	1.33 (.81)	1.44 (.88)	1.47 (.92)
Withhold food from child as punishment (send child to bed without supper or deprive child of a full meal) (N = 343)	Percent	99.7	99.7	99.4	98.6	96.2	87.5	76.9	73.8
Medical (n = 74)	Mean (SD)	1.05 (.37)	1.05 (.36)	1.07 (.38)	1.11 (.42)	1.23 (.65)	1.59 (.89)	1.84 (.94)	1.92 (.96)
Legal (n = 217)	Mean (SD)	1.00 (.00)	1.00 (.00)	1.03 (.23)	1.09 (.42)	1.16 (.54)	1.48 (.81)	1.80 (.98)	1.93 (1.01)
Threaten to leave or abandon child (N = 349)	Percent	98.3	98.3	96.9	96	94.6	89.4	84.9	82.1
Medical (n = 73)	Mean (SD)	1.03 (.16)	1.03 (.16)	1.07 (.30)	1.15 (.43)	1.23 (.56)	1.41 (.76)	1.55 (.91)	1.68 (.92)
Legal (n = 215)	Mean (SD)	1.08 (.38)	1.08 (.38)	1.13 (.51)	1.18 (.57)	1.24 (.65)	1.37 (.76)	1.47 (.83)	1.59 (.92)
Slap child on the face or back of the head (N = 345)	Percent	100	99.7	98.3	97.1	94.8	89	83	78.7
Medical (n = 75)	Mean (SD)	1.01 (.12)	1.01 (.12)	1.09 (.44)	1.19 (.56)	1.31 (.70)	1.51 (.89)	1.66 (.99)	1.78 (1.08)
Legal (n = 219)	Mean (SD)	1.01 (.10)	1.02 (.22)	1.08 (.38)	1.12 (.45)	1.21 (.58)	1.37 (.78)	1.60 (.94)	1.72 (.99)
Hit child elsewhere (not buttocks) with an object such as a belt, hairbrush, stick, or some other hard object (N = 356)	Percent	99.7	99.7	99.4	96.9	95.3	88.9	85	84.7
Medical (n = 75)	Mean (SD)	1.01 (.12)	1.01 (.12)	1.07 (.25)	1.19 (.54)	1.25 (.57)	1.47 (.86)	1.57 (.90)	1.55 (.92)
Legal (n = 217)	Mean (SD)	1.00 (.00)	1.00 (.07)	1.03 (.16)	1.09 (.39)	1.17 (.53)	1.33 (.74)	1.45 (.89)	1.47 (.89)
Threaten to kick child out of the house or send away (N = 347)	Percent	99.7	99.7	99.7	99.4	98.9	96.6	87.6	76.8
Medical (n = 76)	Mean (SD)	1.03 (.23)	1.03 (.23)	1.04 (.26)	1.05 (.36)	1.08 (.39)	1.16 (.49)	1.41 (.82)	1.87 (1.00)
Legal (n = 218)	Mean (SD)	1.01 (.10)	1.01 (.10)	1.02 (.14)	1.04 (.21)	1.07 (.29)	1.17 (.46)	1.41 (.81)	1.94 (1.00)
Lock child out of house (N = 345)	Percent	100	100	99.7	99.7	99.4	98.6	94.5	86.5
Medical (n = 75)	Mean (SD)	1.00 (.00)	1.00 (.00)	1.04 (.35)	1.04 (.35)	1.04 (.35)	1.08 (.40)	1.20 (.59)	1.56 (.86)
Legal (n = 217)	Mean (SD)	1.00 (.00)	1.00 (.00)	1.00 (.00)	1.00 (.00)	1.02 (.21)	1.08 (.35)	1.23 (.61)	1.62 (.89)
Put chili pepper, hot pepper, or spicy food in child’s mouth as punishment (N = 348)	Percent	99.7	99.7	99.4	99.1	98	94	92	92
Medical (n = 73)	Mean (SD)	1.04 (.35)	1.04 (.35)	1.05 (.36)	1.11 (.42)	1.16 (.49)	1.26 (.60)	1.24 (.57)	1.20 (.55)
Legal (n = 218)	Mean (SD)	1.00 (.00)	1.00 (.00)	1.02 (.13)	1.05 (.28)	1.11 (.43)	1.28 (.62)	1.39 (.73)	1.38 (.73)
Beat child, that is, hit over and over again with an object or fist (N = 346)	Percent	99.4	99.4	99.4	99.1	98.8	97.7	95.4	94.5
Medical (n = 76)	Mean (SD)	1.04 (.34)	1.04 (.34)	1.04 (.34)	1.05 (.36)	1.11 (.45)	1.16 (.59)	1.18 (.61)	1.25 (.64)
Legal (n = 218)	Mean (SD)	1.02 (.17)	1.02 (.17)	1.02 (.17)	1.04 (.28)	1.05 (.29)	1.08 (.39)	1.16 (.55)	1.23 (.65)
Shake a child (N = 368)	Percent	99.5	99.5	99.7	99.7	99.2	97.6	96.2	96.2
Medical (n = 74)	Mean (SD)	1.00 (.00)	1.00 (.00)	1.00 (.00)	1.03 (.16)	1.05 (.23)	1.05 (.23)	1.13 (.38)	1.14 (.39)
Legal (n = 216)	Mean (SD)	1.01 (.14)	1.01 (.14)	1.00 (.07)	1.02 (.13)	1.04 (.23)	1.12 (.43)	1.20 (.51)	1.25 (.54)
Hit hard or kick child (N = 350)	Percent	100	100	99.7	99.7	99.4	98	96.6	95.7
Medical (n = 76)	Mean (SD)	1.01 (.12)	1.01 (.12)	1.03 (.23)	1.04 (.26)	1.08 (.32)	1.16 (.49)	1.20 (.54)	1.21 (.55)
Legal (n = 218)	Mean (SD)	1.00 (.00)	1.00 (.00)	1.00 (.07)	1.01 (.10)	1.03 (.16)	1.08 (.32)	1.12 (.42)	1.18 (.49)

Note: Percent = percent of total sample that rated the behavior as reflecting a community standard of 1 = never ok/never use or 2 = rarely use or use in extenuating circumstances. Mean and SD are for the indicated subgroup on scale of 1 = never ok/never use to 4 = ok/general use. Emphasis (bold italicized) is added on percent values when 90 percent or more of the participants rated the item as unacceptable.

Certain discipline tactics were deemed harsh and unacceptable for young infants, but became viewed as more acceptable with older children and adolescents, based on the professionals’ views of community standards. As seen in Table 2, the total sample rated “take away privileges” as viewed as harsh for infants under 1 year of age, but only 10% of the sample rated this tactic as viewed as harsh for children aged 3–5 years, and still fewer rated this tactic harsh for older children and adolescents. Similar trends were seen for “shout, yell, or scream at child” and “spank child on the buttocks with hand only,” which were deemed harsh for children under the age of one year, but only about 40% of the sample rated these tactics as viewed as harsh for children between the ages of 6 and 10 years.

The next three variables on the table, “hit child on buttocks with an object such as a belt or switch,” “curse or swear at child,” and “giving a child a time out or send him/her to his/her room” were rated as harsh through 13 months of age and were rated as somewhat less harsh at older ages. The first two variables were rated as harsh through adolescence with 78% of participants rating “hitting with object…” as harsh, and 70% of participants rating the “curse or swear” variable as harsh. The “timeout” variable was considered harsh only for children under the age of three years.

The next seven items (“pinch child,” “refuse to speak to child as punishment,” “call child names like stupid, ugly, or useless,” “withhold food,” “threaten to leave or abandon child,” “slap child on the face or back of head,” “hit child elsewhere (not buttocks) with an object such as a hairbrush, stick, or other hard object”) were deemed harsh through two years of age and were then seen as somewhat less harsh for children after the age of two but still relatively harsh throughout childhood and adolescence with the clear majority of participants, ranging from 63%–85% across these items, rating these tactics as viewed as harsh for adolescents.

A few items (e.g., “threaten to kick out of the house or send away” and “lock child out of house”) were considered harsh until about 11 years of age, but it is clear that the majority of professionals thought that community norms were that these disciplinary actions should rarely or never be used.

Finally, professionals thought that the community standard was that four child discipline tactics were harsh and should never or rarely be used for children or adolescents at any age. As shown at the bottom of Table 2, these behaviors were: putting spicy food in a child’s mouth (92% or higher for all age levels), beating (94% or higher), shaking (96% or higher), and hitting hard or kicking a child (96% or higher). Professionals also thought that community members would generally agree that you should not shout, yell, or scream at infants or take away their privileges when infants are less than 9-months-old.

Acceptability of Tactics by Child Age and Respondent Profession

We performed a least squares analysis of variance on each of the 19 survey items, with Age Group (0–3 months, 4–8 months, 9–13 months, 14–18 months, 19–24 months, 3–5 years, 6–10 years, 11–18 years) as a within-subjects factor and Occupation (legal vs. medical) as a between-subjects factor. We found a significant main effect of age for each of the tested variables, Fs (7, 1974–2044) > 18.72, ps < .0001. Although the main effect of age was statistically significant for each of the 19 tactics, the magnitude of the age effects varied a great deal across tactic. For example, for the tactic “take away privileges” the mean rated acceptability went from a low value of 1.1 for children in the 0–3 month age range to an extremely high value of 3.95 for adolescents (Figure 1). Similar strong age effects were found for “giving a child a time out or send him/her to his/her room” which was rated as very unacceptable with ratings of less than 1.1 for infants at 0–3 months of age to mean ratings of 3.5 or higher for children aged 3–18 years.

Figure 1.

Mean rating of acceptability for the use of “giving a child a timeout” as a function of age and occupation. Higher scores indicate greater acceptability.

Although the age effect was statistically significant on all items, the magnitude of the effect was much smaller for certain items. For example, for the item “shake a child,” the mean level was 1.0 for infants age 0–3 months and the highest rating was for adolescents with a mean of 1.2, indicating a somewhat higher but still low level of acceptability even for adolescents. Similar trends were shown for the remaining three items (i.e., “beat,” “hit or kick,” and “putting spicy food in mouth”) that were rated as most harsh.

Within-subjects contrasts found a significant linear Age Group effect for each of the 19 items, Fs (1, 269–292) > 25.07, ps < .0001. For a small set of items, a higher-order trend for Age Group was found. For example, for the item “Giving a child a time out or sending him/her to his/her room,” there was both a significant linear trend, F(1, 269) = 3152.79, p < .0001, and a significant cubic trend, F(1, 269) = 706.13, p < .0001. As shown in Figure 1, the distribution of the mean ratings have an S-shaped curve that indicates that the use of this discipline method is initially rated as unacceptable, gradually becomes more acceptable, and then becomes somewhat less acceptable with adolescents.

Spanking a child with a hand had both a significant linear effect, F(1, 282) = 415.71, p < .0001, and a significant cubic effect, F(1, 282) = 187.10, p < .0001 (Figure 2). For spanking, there was a consistent increase in rated acceptability until 3–5 years and then it tended to drop off, especially for adolescents.

Figure 2.

Mean rating of acceptability for the use of “spanking a child with hand only” as a function of age and occupation. Higher scores indicate greater acceptability.

Between-subjects tests revealed a significant occupation main effect for only one of the 19 items. Legal and medical professionals differed significantly on how they evaluated the item “take away privileges” F(1, 283) = 4.29, p <.05. Figure 3 demonstrates that this difference arose from the fact that, for younger child ages, medical professionals rated the use of this child discipline tactic as less accepted. Both professions converged on the acceptance of this discipline method once the children were older.

Figure 3.

Mean rating of acceptability for the use of “take away privileges” as a function of age and occupation. Higher scores indicate greater acceptability.

The single item with a significant age by occupation interaction was “shout, yell, or scream at child,” F(7, 1988) = 5.49, p < .0001. The only significant age trend for this interaction was the quadratic age by profession component, F(1, 284) = 14.65, p < .0001, indicating that for medical professionals there was a quick increase in their acceptability scores as a function of child age. The legal professionals initially lagged in accepting yelling but eventually came to about the same level of acceptability as the medical professionals for the oldest age levels (Figure 4).

Figure 4.

Mean rating of acceptability for the use of “shout, yell, or scream” as a function of age and occupation. Higher scores indicate greater acceptability.

Supplementary Analyses

Defining legal professionals

Because lawyers, judges, and paralegals are most central to adjudicating child abuse, we performed a supplementary analysis by restricting the legal professionals only to those three groups, omitting the remaining individuals (e.g., social workers, GALs). We then performed analyses on all 19 items contrasting this more restricted set of legal professionals to the medical professionals. Results were unchanged in comparison with those reported above and in Table 2.

Effect of respondent age and parental status

We measured respondent age as a categorical variable and whether they had children as shown in Table 1. We tested for main effects of respondent age and parental status and whether these interacted with occupation and child age. Because this led to a large number of statistical tests across the 19 variables we adopted an alpha level of .01 to control Type I error rate. Using this level of significance, none of the main or interactive effects of these factors were significant.

Discussion

The current study sought to examine how medical and legal professionals would rate the acceptability of various child discipline tactics according to norms in their communities. We also were interested in how the age of a child might affect these ratings. Surprisingly, few significant differences were found between medical and legal professionals with regard to their ratings of which parenting behaviors were viewed as harsh, and the differences that were statistically significant were rather small in magnitude. The patterns of responses suggest that these two groups’ views of community standards mirror the increase in the trend away from corporal punishment of children in the United States (Zolotor, Theodore, Runyan, Chang, & Laskey, 2011). We found a significant occupational difference only for one item, “take away privileges,” where for younger ages, medical professionals rated the use of this discipline tactic as less acceptable, but both professions converged on the acceptance of this discipline method once the children were older. Additionally, only one item (“shout, yell, or scream at child”) revealed an age by occupation interaction where, for medical professionals, there was a quick increase in their ratings of community acceptability as a function of child age. The legal professionals initially lagged but eventually came to about the same level of acceptability ratings as the medical professionals for the oldest child age levels. Thus, trends by child age group were very similar for both legal and medical professionals.

Significant child age effects were found for all parental disciplinary tactics, although the magnitude of these age effects varied greatly across items. Some (e.g., “take away privileges”) went from not at all acceptable for infants to quite acceptable in adolescence. Others were not acceptable for infants and were still not acceptable for adolescents (e.g., “shake,” “beat”). Still other forms of child discipline were rated as not acceptable early on, became more acceptable in young childhood, and then were less acceptable in adolescence (“timeout”). The age norms for these tactics may reflect how we deal with these behaviors in legal and medical settings as well as in our communities.

Taking away privileges and using timeouts were rated as the most acceptable for older children, especially by 11–18 years of age. Shouting, yelling, or screaming at a child were not rated as acceptable for the youngest children and remained non-preferable, with only a 40% acceptability rating for the oldest age group.

We found that spanking a child with a hand was rated as harsh through eight months, with 89% of participants rating this item as unacceptable through the first year of life. What is remarkable about this finding is that other data suggest that 25% of United States children have been spanked by nine months of age (Zolotor, Robinson, Runyan, Barr, & Murphy, 2011), suggesting wider support for spanking of infants under 1 year of age. Then, for toddlers aged 19–24 months, 63% of professionals evaluated this behavior as harsh in their community, indicating somewhat increased acceptability of spanking. The American Academy of Pediatrics (AAP) recommends that parents do not spank their children under 18 months and further encourages parents to use other more effective disciplines tactics for children of all ages (American Academy of Pediatrics, 1998). Others have questioned the long-term impact of spanking in changing child behavior and have even found higher rates of aggression in children who were spanked. As previously mentioned, Gershoff’s (2002) meta-analysis revealed spanking to be significantly associated with, among other issues, elevated child and adult aggression. She cites previous researchers as having consistently found that children who receive physical punishment are at greater risk of behaving violently towards family in the future (Gershoff & Bitensky, 2007; Gershoff, 2010, 2013). Spanking a child with an object such as a belt or switch was rated with less acceptability than spanking with hand at all ages. Zolotor, Theodore, Chang, Berkoff, and Runyan (2008) found that reported spanking by parents was associated with the use of harsh forms of punishment by those parents, and that spanking with an object was strongly associated with the use of other reported harsh punishment behaviors. Although no research has specifically examined the adverse consequences of spanking infants, the evidence of long-term harm from spanking children, the absence of any data suggesting that infants learn any lessons from spanking, and the judgment of our respondents regarding community standards, all lend support to the AAP guidelines that spanking, especially in the first two years of life, is unacceptable.

Limitations

Because we relied on professional email listservs, we do not know the response rates for our study. Participants decided to complete our survey when queried via the listserv so this may affect the generalizability of our results to all legal and medical professionals. Another limitation of this study is that the accuracy of the perceptions of these professionals can be questioned. However, regardless of whether they are correct, they indicate what professionals believe about community norms, which likely affects their professional decision-making. Our survey asked participants to make several judgments at once. We asked each professional about the acceptability by persons in the community in which they lived of particular discipline tactics for children based on age of the child. It is possible that our results would differ if we had asked each professional about only one age group using a between-subjects design. Another limitation of this study was that we were not able to obtain sufficiently large samples of prosecutors and defense attorneys to compare the differing perceptions by these two groups. Future research should examine these two groups for potential differences. Our study relied on email listservs from both national and local state professional organizations. Because we asked professionals about what would be considered acceptable in the communities in which they live it would be helpful for future research to include more national data. Finally, we asked professionals to evaluate child disciplinary tactics in a written survey. This is different from the richer real-life cases that they evaluate, which contain more detail, and actual (not hypothetical) children and parents. The current study was nevertheless an important first-step in examining the effect of age on acceptability ratings. Future research might approach this question using vignettes or other, more realistic approaches with a between-subjects design when considering the age of the child.

Implications

As has been discussed in the literature, child maltreatment is difficult, but important, to define (see Williams & Weeks, 2014, for an overview). The present study showed that what is considered unacceptable or harsh by professionals according to their perceptions of the norms of their communities varies significantly by child age. Our results suggest that adding severity by child age as a variable in punishment scales may provide valuable information regarding punishment practices and norms. In epidemiological studies using the PC-CTS (Runyan et al., 2010) or the ISPCAN Child Abuse Screening Tools (ICAST) (Runyan et al., 2009), the category of “harsh punishment” has been used as a proxy to estimate the prevalence of child abuse. These instruments only approximate the occurrence of maltreatment, as nuances in the legal definitions of child abuse vary by state, and the measures only concern behaviors, not actual physical or emotional injury. As noted above, with the exception of “shaking of children less than 2 years of age,” prior work has not addressed child age as a factor to be considered in defining a disciplinary behavior as harsh. This work suggests that efforts to assess the prevalence of child abuse with parental questionnaires would be enhanced by considering child age when measuring behaviors that can be considered harsh.