Violence towards children in the United States and Sweden

Child Abuse & Neglect. Vol. IO. pp. 501-510. 1986 Printed in the U.S.A. All rights reserved.

0145.2134186 $3.00 + 00 Copyright 0 1986 Pergamon Journals Ltd.

VIOLENCE TOWARDS CHILDREN IN THE UNITED

STATES AND SWEDEN

RICHARD J. GELLES, PH.D.

College of Arts and Sciences, University of Rhode Island, Kingston, RI 02881

AKE W. EDFELDT. PH.D.

Department of Education, University of Stockholm, 106 91 Stockholm

Abstract-This paper reports the results of a cross-cultural comparison of violence towards children in the United States and Sweden. Data from the United States are based on interviews with a nationally representative sample of 1,146 households with at least one child between the ages of 3 and 17 years living at home. Data from Sweden are based on interviews with a nationally representative sample of 1,168 households with a child 3 to 17 years of age at home. Violence and abuse were measured using the Conflict Tactics Scales. In general, Swedish parents reported using less violence than did parents in the United States. There was no significant difference between the two countries in the rate of reported severe or abusive violence. The paper compares factors found associated with violence towards children in the two countries, including age, marital status, education, and parents’ background. The results are analyzed by considering methodological and cultural factors that explain the similarities and differences in the use of violence towards children in the two countries.

Resume-Les auteurs presentent les resultats d’une enquete transcuiturehe comparant la violence a i’egard d’enfants aux Etats-Unis et la violence a I’egard d’enfants en Suede. Les don&es recoltees aux Etats-Unis sont fondles sur des interviews au sein d’un collectif representatif sur le plan national de 1’146 menages ayant au moins un enfant entre l’age de 3 et 17 ans residant a la maison. Les don&es recoltees en Suede sont fondees sur des interviews avec un collectif egalement representatif sur le plan national de 1’168 menages ayant un enfant entre 3 et 17 ans vivant a la maison. La violence et la maltraitance ont ete Cvaluees a l’aide des Cchelles tactiques conflictuelles (Conflict Tactics Scales). De facon g&r&ale il apparait d’apres leurs dires que les parents suedois recourent moins a la violence que les parents des Etats-Unis. En ce qui concerne la violence et la maltraitance graves, on ne constate pas de difference entre les deux pays, telle qu’elle est signalee. Les auteurs cornparent les facteurs associts a la maltraitance d’enfants dans les deux pays; cela comprend I’age, I’etat civil, le niveau d’education, et les traditions et I’origine familiale (background). Les auteurs analysent les resultats de leur enquete a la lumiere des facteurs methodologiques et culturels qui peuvent expliquer les similitudes et les diver- gences dans le recours a la violence a I’egard d’enfants dans ces deux pays.

MUCH OF THE RESEARCH carried out on child abuse and violence towards children has focused on the problem in the United States. Yet, students of child maltreatment have long been aware that American children are not the only, or even the most likely, victims [l-3].

While there has been a tremendous growth of an international perspective on abuse and neglect, there have been few actual cross-cultural studies conducted or published [4]. Problems with definitions, costs, and access have limited the ability of researchers to field cross-cultural investigations. This paper reports the results of a cross-cultural study of violence towards children in the United States and Sweden. A comparison of the incidence and patterns of violence towards children in the United States and Sweden is presented.

This research was supported by NIMH Grants MH 27557 and MH 40027 and the Tercentenary Foundation of the Swedish Bank.

501

502 Richard J. Gelle\ and Ake W. Edfeldt

BACKGROUND

Sweden and the United States offer important contrasts when it comes to cultural and social factors which may be related to violence and abuse. Child abuse has not generally appeared to be an overwhelming problem in Scandinavian countries, including Sweden. Vesterdal lists four reasons [S]. First, social conditions are generally good. Second, there is widespread use of contraceptives and free abortions, reducing the number of unwanted babies. Third, many mothers work and leave their children in day-care institutions. Last, premature babies are kept in a neonatal ward until they are a certain weight. The babies are released only when their parents are taught to handle a newborn.

There are stark contrasts between the United States and Sweden in terms of social attitudes towards violence. While in the United States the majority of states permit the corporal punishment of school children, corporal punishment has been outlawed in Sweden since 19.52. Legislation prohibiting the spanking of children was passed in 1979, and has since been adopted in all other Scandinavian countries. Firearm ownership is rigorously controlled in Sweden. While nearly half of all American households contain guns, mostly handguns, gun ownership in Sweden is mostly limited to weapons used for hunting. Television violence offers another important contrast. American children witness as many as 15,000 killings each year on television; violent programming in Sweden is severely restricted. (Actually, Swedish television is barely on the air as many hours as the average American child watches television in a week.) The level of concern for children’s programming in Sweden can be seen in the decision to limit the popular American movie, ET, to audiences over I I years of age. A final contrast is in public violence. Capital punishment is banned in Sweden. but it is allowed and growing in use and popular support in the United States.

There is considerable difference in one area of social complexity. While the United States is a heterogeneous nation which is made up of numerous ethnic, religious, and cultural groups and communities, Sweden is more homogeneous in terms of such cultural and social attributes. There are, of course, innumerable similarities in the lives of children and families in the United States and in Sweden. However, the contrasts. especially in the area of attitudes towards violence, make a compelling case for comparing the level of violence towards children in the two countries.

Rescwtrch in thr United Stutcs. Various techniques have been used to measure the extent of child abuse and violence towards children in the United States [6. 7). Two of the most widely cited studies are the National Center on Child Abuse and Neglect report titled “Recognition and Reporting of Child Maltreatment” [X] and the “National Analysis of

Official Child Neglect and Abuse Reporting,” published yearly by the American Humane Association. Both studies provide interesting data on officially recognized cases of abuse and neglect that come to public attention. However, the studies do not provide a direct measure of the true incidence of the occurrence of abusive events. As Burgdorf admits, a study of reported cases of maltreatment only identifies the “tip of the iceberg of child maltreatment” [8]. Measures of reported child abuse tend to focus on the more injurious forms of violence towards children. Such research cannot be used to examine the total range of violent acts experienced by children.

Not only do official report data fail to inform us about the true extent of violence towards children, such data also do not offer an accurate representation of the factors re-

Violence towards children SO3

lated to violence and abuse of children. Official reports of child abuse include a number of biases. Certain individuals and families are more likely to be reported due to the nature of the injury, social status of the alleged abuser, and social characteristics of the victim. Minority, poor, and single parents are more likely to be correctly and incorrectly reported as abusers than are wealthy, white and intact families [9-l I]. Infants are more likely to be identified as victims than are teenagers [9].

Research in S~vden. The State Board of Social Welfare in Sweden undertook two national investigations in which they tried to establish the extent and character of child abuse in Sweden. The first study, which was reported in 1969, examined the period from 1957 to 1966. Data were collected from all general hospitals and all surgical, children’s surgical, neurological, pediatric, and children’s and youth psychiatric clinics. In addition, data were collected from all medico-legal centers in Sweden. In all, I78 places were con- tacted as part of the survey. Two cases were reported in 1957; by 1966 there were 66, with an average of 1 I .9 cases annually for the IO-year period.

A second report was published in 1975 and described the conditions during the period 1969 to 1970. For this period, there was an average of 147 cases of bodily abuse each year and 7 cases titled “sexual outrage.”

Sweden, like all countries with the exception of the United States and Canada, does not have laws mandating the reporting of child abuse and neglect [12]. Thus, most of the iceberg of child maltreatment remains below the surface of official recognition in Sweden.

METHODS

One source of data which is not biased by official reports or awareness and which ex- amines a range of violent acts experienced by children is the National Family Violence Survey carried out in 1976 by Murray Straus, Richard J. Gelles, and Suzanne Steinmetz [ 131. This survey served as the comparison data for a national survey of violence toward Swedish children conducted by Ake Edfeldt.

A full description of the National Violence Survey in the United States, and specifically the portion of the survey which focused on violence towards children, has been published elsewhere [13]. This section briefly reviews the definitions, measurement, and sample used in the survey.

Defining \Biolencc trnd crhrrse. Violence was nominally defined as “an act carried out with the intention, or perceived intention, of physically injuring another person.” The injury could range from slight pain, as in a slap, to murder. The motivation might range from concern for a child’s safety (as when a child is spanked for going into the street) to hostility so intense that the death of the child is desired [14]. Abusive acts were those acts which had a high probability of causing an injury to the child (an injury did not have to actually occur).

Opcr.ationcllizing \siolcncxJ clnd rrbrrsc. Violence was operationalized through the use of the Conflict Tactics Scales (CTS). First developed at the University of New Hampshire in

so4 Richard J. Gelles and Ake W. Edfeldt

197 1, this technique has been modified extensively and used in numerous studies of family violence [15-201. The Conflict Tactics ScaIes contain items to measure three vari- ables: (1) use of rational discussion and argument (discussed the issue calmly; got infor- mation to back up one’s side: brought in/tried to bring in someone to help settle things); (2) use of verbal and nonverbal expressions of hostility (insulted or swore at the other; sulked or refused to talk about it; stomped out of the room or house; did or said something to spite the other; threatened to hit or throw something at the other; or threw, smashed, hit or kicked something); and (3) use of physical force or violence as a means of managing the conflict (threw something at the other; pushed, grabbed, shoved the other; slapped or spanked; kicked, bit or hit with a fist; hit or tried to hit with something; beat up the other; threatened with a knife or gun; or used knife or gun). The abuse items were those acts that had a high probability of causing an injury: kicked, bit or hit with a fist; beat up; threatened with a knife or gun: or used a knife or gun.

The Conflict Tactics Scales were presented to subjects in the order enumerated, and the subjects were asked to say how often they used each technique when they had a disagreement or were angry with a child, both in the previous year and in the course of the relationship with the child.

The Conflict Tactics Scales’ definition of violence and abuse is different from traditional definitions of child abuse. First, in one sense, the definition is broader than most detini- tions since it is not limited to acts of violence that only result in injury (operational definitions of child abuse that rely on official reports almost invariably are limited to instances of violence that result in a diagnosable injury). In another sense, the Conflict Tactics Scales are narrower than many definitions of abuse because the Scales are limited to a specific list of violent acts. Omitted, for example, are burnings or scaldings of children. Another limitation of the Scales is that they rely on the parents’ recollection and wiliing- ness to report specific acts of violence.

Reliability rnd vaiidity. The reliability and validity of the Conflict Tacts Scales has been assessed over the 15-year period of its development and modification. A full discussion of reliability and validity can be found elsewhere [13, 211. There is evidence of adequate internal consistency reliability, concurrent validity, and construct validity.

Sample. A national probability sample of 2,143 households comprised the National Family Violence Survey 1131. In each family where there were two caretakers present and there was at least one child at home between the ages of 3 and 17, a referent child was selected using a random procedure. An important limitation of the sample was the omis- sion of referent children under the age of 3 years and single parents. Many experts beheve that children under 3 years of age and children of single parents are at the greatest risk of physical abuse [22-241, Because the examination of violence towards children was im- bedded in a larger survey designed to measure all forms of family violence, including violence between couples and violence between siblings, single parents and children less than 3 years of age were excluded from the sample (acts perpetrated by children younger than 3 years of age could not be considered violence in the same way as acts committed by older children).

Of the 2,143 families interviewed, 1,146 had two caretakers and children between the ages of 3 and 17 living at home. The data on parent-to-child violence in the United States are based on the analysis of these 1,146 parent-child relationships.

Violence towards children 505

The Survey in Sweden

SumpEe. A basic sample of 1,618 individuals representative of Swedish children aged 3 to 17 was purchased by the research team in Sweden. The list contained the name, address, and national registration number of the children as well as an identifier which indicated who was officially responsible for the household. The head of the household’s name, address, and national registration number were obtained also. (In Sweden, the male partner in a two-parent household is automatically identified as the head of the household; thus, a female name automatically indicated a single parent household.) As the Swedish survey was not concerned with violence between partners, the Swedish sample included two-parent and single-parent households.

There was an initial loss of 3.3% of the sample due to incorrect or insufficient identifi- cation data. Forty-one households (2.5%) refused to participate in the survey, although telephone interviews were completed with 1 I of these individuals. Additional cases were lost due to language barriers. interviews were completed with 78.9% of the sample-a higher completion rate than the United States survey. However, a random subsample of 171 completed interviews were lost in transit between two offices of the Swedish State Data Processing Unit. Due to the complete anonymity of the survey, researchers were unable to identify and recontact these individuals.

A final completion rate of 68% was obtained for 1,105 usable interviews. There were 255 single-parent households (only 3 of which were headed by men) and 8.50 two-parent households. The age and regional distribution of the sample approximated the distribution in Sweden.

DijJerences Between the Two Surveys

Although the Swedish survey was designed as a replication of the study in the United States, there were some important differences. First, the United States survey was conducted in 1976 while the Swedish survey was completed four years later in 1980. Second, while both studies coflected data by using in-person interviews in the homes of the respondents, the United States investigators employed a professional survey research com- pany and professional interviewers. The Swedish interviews were conducted by trained psychologists.

The Conflict Tactics Scales were translated into Swedish. In the process, there were two changes from the original scales. Whereas the American research group asked how often did you threaten/use a gun or knife, the Swedish project asked how often did you threaten/use a weapon. The American version of the Scales asked about slapping or spanking, while the Swedish version asked about hitting. In America, respondents were asked how often they hit or tried to hit their children with objects. These items could include attempted and/or completed hits. In Sweden the interviewers qualified this item by measuring only completed hits.

RESULTS

Because of the contrasts between the United States and Sweden, especially with regard to social and cultural attitudes about violence and because the Swedish survey was conducted a year after the passage of the anti-spanking law, we expected that the reported

506 Richard J. Gelles and Ake W. Edfeldt

rates of violence towards children would be lower in Sweden than in the United States. The results, however, do not generally support this hypothesis.

Physical Punishment

In addition to the administration of the Conflict Tactics Scales, respondents in the United States and Sweden were asked whether they had used any form of physical punishment towards their children in the previous year. Slightly more than half (51.3%) of the Swedish parents reported using physical punishment compared to more than three-quarters (79.2%) of the Americans.

Violence Towards Children

The responses to the Conflict Tactics items offer a more precise comparison as shown in Table I. In general, Swedish parents reported less use of violence towards their children during 1980 than the American respondents reported in 1975 (29.8% vs. 63%). Swedish parents reported less violent conflict tactics during the course of raising their children than the Americans, but the difference here is smaller than for the one-year recall period.

There is no significant difference between Swedish and American parents in the reported use of severe or abusive violence (kicking, biting, hitting with a fist, beating up, threatening or using a weapon-gun or knife). If one includes hitting (or trying to hit) with an object, there are differences; but these differences are not reliable since the Swedish item only examined completed hits while the item in the United States asked about trying to hit as well.

A rate of abusive violence of 4% in Sweden means that each year 62,000 of the more than 1.5 million Swedish children 3 to 17 years of age are at risk of physical abuse (considering sampling error, the range is from 39,000 to 84,000). In the United States, between 1.4 and 1.9 million children 3 to 17 years of age living with two caretakers were at risk for physical abuse during the survey year.

An item by item examination of the two surveys reveals that Swedish parents report more pushing, grabbing or shoving than American parents (which accounts for the difference between the Swedish previous year figures) and double the rate of beating children both in the last year and over the course of raising their children. American parents report more spanking (about double the rate of Swedish parents in the previous year, a bit less

Table 1. Violence Towards Children in the United States and Sweden

United States Sweden

Type of Violence

I Threw things at 2. Pushed, grabbed, or shoved 3. Hit (spanked or slapped) 4. Kicked. bit, or hit with a fi\t S. Hit with an object’ 6. Beat up 7. Threatened with a weapon 8. Used a weapon All forms of Violence (3-8) Severe Violence (Index A: 4 thru 8) Severe Violence (Index B: 4. 6 thru 8)

1975

5.4% 40.5 5x.2

32 I;:4

I.3 0. I 0. I

63.0 14.2 3.6

Ever

9.0% 46.0 71 .o

x.0 20.0

4.0 2.x 2.9

73.0

I980 Ever

3.6% 13.2% 49.4 63.3 27.5 51.2

2.2 x.4 2.4 7.7 3.0 X.0 0.4 I .5 0.4 I.3

29.8 66.0 4.6 4.1

1 In the United States this item referred to attempted or completed hit\. In Sweden. the item referred only to completed hits.


than 50% more over the course of the child’s life). In general, there were far more similarities in the two countries than there were differences.

Factors Associated itpith Violence Towards Children

A complete profile of the factors associated with violence towards children in the United States has been previously published [13, 251. This section briefly compares the factors found related to violence and abusive violence which are comparable in the two countries.

Age. In both the United States and Sweden, younger parents are more likely to use violence towards their children than older parents.

Marital status. Only intact families were surveyed in the United States. However, much of the child abuse literature claims that single parents are at a greater risk for abusing and striking their children due to higher stress and low income [25-271. There were no significant differences in the rates of violence towards children between Swedish parents who were married and those who were single parents at the time of the interview.

Education. A curvilinear relation was found between violence towards children and education in the United States-those parents with the highest (at least some college) and lowest (no high school) levels of education were the least likely to use violence. In Sweden, education was unrelated to the rate of abusive violence towards children. Those who had the highest levels of education actually were more likely to report using some form of violence towards their children.

Parents’ backgrounds. Violence begets violence, we are told. And, to a certain degree this is borne out in both surveys. Being spanked as a teenager and observing one’s parents hit one another raises the chances that a person will be a violent parent both in the United States and in Sweden.

DISCUSSION

The results of the comparison of violence towards children in the United States and Sweden are complex and do not generally support the hypothesis that Swedish parents are less likely to use violence than American parents. However, Swedish parents tend to be less likely to use the least harmful forms of violence. There is less spanking and slapping in Sweden than there is in the United States. There were no appreciable differences in the rates of abusive violent acts between the two countries.

What accounts for the differences and similarities? First, there are methodological arti- facts that may have entered into the comparison study.

Methodological Factors

There were important differences in the qualifications of the interviews employed in the

508 Kichaxi J. Gelles and Ake W. E&et&

United States and Sweden. The Swedish interviewers were trained psychologists. We know that they were able to obtain a higher response rate than the American survey research firm. They may have been more successful in eliciting more truthful answers from the respondents. Thus, the rate of violence towards children may be larger in Sweden because the Swedes were more inclined to provide accurate reports of their use of violence, while American respondents systematically under-reported violence.

If one continues to assume that Swedish parents are indeed less violent. then this too could have produced more accurate reporting. If violence is indeed unusual in Sweden, then Swedish parents may have been more likely to remember these unusual behaviors. If violence is “as American as apple pie,” then American parents may have been more likely to under-report the routine use of all forms of violence. Thus. Swedish parents may be less violent towards their children than American parents even though a combination of methodological biases produced similar rates for the abusive forms of violent behavior.

Cr,ltrrml Foc~tors

In the absence of certain known biases. researchers are trained to believe their data. Thus, even though methodological explanations are a plausible explanation for the findings. we still must accept another plausible explanation that the similarities and differences are real. This being the case. why are Swedish parents more moderate in their use of spanking, slapping. general hitting and throwing things than American parents, but as likely to beat, kick, bite, punch and even use weapons‘?

A close examination of the yearly rates of violence compared to the “ever” reports, provides some evidence that the anti-spanking law has had some impact on the lives and

behavior of Swedish parents. While the differences between the yearly rates of spanking, slapping, pushing, grabbing and throwing things are small in the United States, the differences are much greater in Sweden. Since 1980 was the first year that the anti-spanking law was in force (the year of the Swedish survey). parental use of the less harmful forms of violence may have begun to decline.

Spanking, slapping, pushing. and grabbing tend to be the more instrumental forms of violence used by parents. Thus. when hitting children became illegitimate in Sweden. parents may have found it easier to moderate their use of violence as an instrumental technique. Kicking. biting, punching. beating. and using weapons could be considered either more expressive forms of violence (i.e., expressed out of anger or rage) or forms of violence consciously designed to inflict serious injury on a child. These behaviors may be less amenable to control by imposing a cultural standard that spanking is wrong (note the anti-spanking law carries with it no punishment for violation).

At the risk of over-generalizing from a cross-sectional survey, Swedish parents may be responding to the anti-spanking law and lowering their use of the less harmful forms of physical punishment. The more abusive forms of violence appear to be less sensitive to change, at least in the short run.

The fact that there are essentially no differences in the use of abusive violence is still surprising given the differences in cultural attitudes. cultural complexity and social support in the two countries. One simple explanation is that the attitude changes and levels of social support (e.g., day care. health care) have not gone far enough in Sweden to curtail the use of abusive violence. A more complex formulation is that the differences between the two countries are not nearly as major as one would think. The similarities are those factors which are related to the use of abusive violence towards children. Both societies are western, industrial nations-Sweden tending more towards a socialist economy than


the United States. Family life in both countries conforms to the norm of western, industrial nations. Nuclear residence and a premium on family privacy and family rights are the cultural norm for both nations. An exchange/social control explanation of family violence 1281 explains that privacy and primacy of the family limits the amount of external social control (formal and informal) that is exerted on family members. With limited social control, the costs of using abusive violence are frequently quite minimal-even if there is a public anti-spanking law. In the extensive report from the Swedish study another more sophisticated explanation is given to the obvious similarities in the two national samples v91.

Injurious Physical Violence

One of the important limitations of the Conflict Tactics Scales used to measure violence towards children in the United States and Sweden is that it does not measure whether a child was actually ever injured or harmed by the physical violence. While we can measure violent behavior in the two countries, we have little insight into the actual harm inflicted on children. Official reporting records on child maltreatment in the United States are useful means of measuring the harm inflicted on children (at least the harm that comes to public attention). Unfortunately, there is no official reporting of child maltreatment in Sweden and, therefore, there is no opportunity to compare levels of harm experienced by children.

We do have at least some anecdotal data that would seem to imply that, although Swedish and American children are about equally likely to experience violence that could cause harm, there is less of what one would call classic child abuse in Sweden.

At the U.S.and Swedish Symposium on Physical and Sexual Abuse of Children, held in Satra Bruk, Sweden, in June 1985, the American pediatrician, Eli Newberger, presented a report on hospital management of physical abuse. He described a number of cases, including slides of various physically abused children. All Swedish participants agreed that, while they had occasionally seen similar abuse cases in the past, they were no longer seeing these types of cases in 1985.

Thus, although we cannot empirically verify this claim, the changes in social attitudes and social support in Sweden may have been successful in reducing the outrageous, grievous injuries experienced by children. Although Swedish parents are still violent, the violence is not allowed to escalate to the point of injury. However, it is just as possible that, since data on physical abuse are not routinely reported and tabulated in Sweden, the rates of injurious violence are the same in both countries (as our data suggest) but abusive violence in Sweden is simply not recognized or reported as such.

CONCLUSION

Research often tends to raise as many questions as are answered. Such was the case with this first cross-cultural comparison of the rate of violence towards children. We thought that the rates of violence would be lower in Sweden than in the United States, but we found a more complex pattern. Clearly, the relationship between social and cultural attitudes, social support, and societal complexity and the use of physical violence towards children is not simple or unitary. Only with continued and expanded cross-cultural research will we be able to refine our understanding of the socio-cultural factors that relate to the physical abuse of children.

5 IO Kichard J. Gelles and Ake W. Edfeldl

REFERENCES

I. GELLES, R. J. and CORNELL, C. in~~rnarional Per~cp~c~rivr.~ on Fami/~ Vioknw. Lexington Books. Lexington, MA (1983).

2. KEMPE, C. H. Kecent developments in the field of child abuse. CiziId Ahrc.tc 6i Napircf 2:261-267 (1978). 3. KORBIN, J. (Ed.). Child A/xr.sc and Neglec.~: Cross Culrrrml Pcr.sper litvs. University of California Press.

Berkeley i I981 ). 4. STRAUS, M. A. methodology of collaborative cross-national research on child abuse. tinpubl~5hed paper

presented at the Swedi~h/Ameri~~n Symposium on Physical and Sexual Abuse of Children, Satra Bruk. Sweden (1985).

5. VESTERDAL. J. Handling of child abuse in Denmark. C~f~l~iA~~~.\~ & Nc~,glc~c~t 1:193-I98 (1977). 6. GELLES. R. J. Violence toward\ children in the United State\. Anrrricon J~rntd of‘ Ol.rh~1i7?.f.irirctrv

48:580-X? (197X). 7. GELLES, R. J. and CORNELL, C. intimtrrc Vinlcrrr.t~ in F‘crmitic~v. Sage, Beverly Hill\. CA (1985). 8. BURGDORF. K. f?rc,ofi!nition cind f?<*r)o,-rin~ t$Chiid Mufrretrtvrent. Westat. Rockville. MD ( 1980) 9. GELL,ES. R. J. The s&l construction of’child abusz Arncri~an Jowrrtr/ ~f’Orfhop,\~(.hi~rt,~~ 45:363-371

t 1975). IO. NEWBERGER, E., REED, R. B.. DANIEL. J. H.. HYDE. J. N. and KOTELCHUCK. M. Pediatric \ociat

illness: Toward an etiologic classification. Pedictrric.r 60:178-I85 (1977). I I. TURBETT, J. P. and O’TOOLE, R. PhysicIan’\ recognition of child abuse. Unpublished paper presented at

the annual meetings of the American Sociological Association, New York (IY80). 12. KAMERMAN. S. Eight countrie\: Cros\ national perspective5 on child abuse and neglect. Children 7‘otltr>

4:34-37 (1975). 13. STRAUS, M. A., GELLES, R. J. and STEINMETZ, S. K. b<,llind Clo.\c,d LIoo,.\: I:io/enc,c~ in tllc Americ.un

Fuoni/y. Anchor/Doubleday, Garden City, NY (1980). 14. GELLES, R. J. and STRAUS. M. Determinants of violence in the family: ‘Ibwurd u theoretical integration.

In: Conrc~mportq Thcvwic~.\ N~OUI fhc Frrmily, (Vol. I). W. R. Burr, R. Hilt. F. I. Nye, and I. Rei,, (Eds.). pp. 54%.%I. The Free Pre\\, New York (1979).

IS. ALLEN, C. and STRAUS. M. Resources. power and husband-wife violence. Unpublished paper presented to the National Council on Family Relation\ in Salt Lake City i 197.5).

16. CATE. R. M.. HENTON, J. M. CHRISTOPHER, E S. and LLOYD. S. Premarital abuse: A social psycho- logical perspective. Jotrmrri 0f Foilmil? ksrre.3 3:79-90 (1982).

17. HENTON, J., CKI‘E, R., KOVAL, .I.. LLOYD. S. and CHRIST~PH~~R. S. Romance and violence in dating relationships. Jorrrmtl ~~~~~~~?~i~~ i.urw.s 4:467-482 ( 1983).

18. HORNUNG. C.. MCCULL~OUGH~ B. and SUGl.MOTO. T. St&u\ rel~ti(~n~hip~ in marriage: Ri\k factor\ in spouse abuse. Jowmrf ~~~.~urr~~~~~, rmd the F~omily 43:67Y-692 ( 1981).

19. JORGENSEN. S. Societal cla\s heterogamy. status striving. and perception of marital conflict: A partial replication and revision of Perlin’\ contingency hypothe\i\. J~~//rt~f~~ of‘~~f~rrj~~~~~ ctnd ?hc Fumi& 43:679-692 (1977).

20. STRAUS. M. A. Leveling, civility, and violence in the family. Jcmratrt of’;Marriugc, und ri~r Family 36:13-X) (1974).

21. STRAUS, M.A. Measuring intrafamily conflict and violence: The Conflict Tactic\ (CT’, scale\. Jcwrnd of Murricfgc cod r/70 Fomi& 41:75-8X (1979).

22. FERGUSSON. D.M.. FLEMING, J. and O’NEIL, I>. Child Abwc, in New %c,cr/~rnd Rerearch Division, Department of Social Work, Wellington NZ (1972).

23. GIL, D. Vio/(,nt 0 A,yuin.\t C‘hildrrn: Plrysic~rrl Child Ahrr.sc i/r /kc, Unird S/ufr\. Harvard University Pres, Cambridge, MA (1970).

24. JOHNSON. C. C/~i/d Ahu\c in tltr .Scmrhc,u.c/: An Amc!\sis c,f’ //72 Krpor~cd C’i/.\c,\. Welfare Research. Athens, GA (1974).

25. GELLES. R. J. Violence in the Family: A review of re\carch in the seventies. Jcufrntrl of’blrrrriqqc~ cr~tl /ht Ftrmi!\~ 42:X73-XXS ( 1980).

26. PARKE. R. I). and COLl.MER, C. Child abute: An Interdisciplinary analysi\. In: Rc~?cu ofChild fk\~c/t~p- wc’nt Rcscrrrcit. (Vol. 51, M. Hetherington (Ed.). pp. I-102. University of Chicago Prrs\. Chicago (1975).

27. MADEN. M. F. and WRENCH, D. l-. Signifjcant finding> in child abuse research. Vir~timdo~y 2: 196-224 (1977).

2X. GELLES, R. J. An exchange/social control theory. in: The DurX Sick of’Fmdi~.s: C‘mwnt Fumity Violcwe Rt~\ewch, Il. Finkelhor. R. Gelle\, M. Strau\, and G. Hotaling (Eds.1. pp. 151-165. Sage. Beverly Hills. CA ( 1983).

29. EDFELDT, A. Research and theory on violence toward\ children in Sweden. ~‘npublished paper presented at the Sweden/Anieri~~n Symposium on Physical and Sexual Abuse of Children. Satre Bruk, Sweden i 19X).

Violence towards children in the United States and Sweden

Documents