Effects of type of maltreatment, severity of physical consequence, and respondent variables on the perceptions of medica...

MISSING IMAGE

Material Information

Title:
Effects of type of maltreatment, severity of physical consequence, and respondent variables on the perceptions of medical neglect and physical abuse
Physical Description:
viii, 120 leaves : ; 29 cm.
Language:
English
Creator:
Abner, John-Paul
Publication Date:

Subjects

Subjects / Keywords:
Research   ( mesh )
Child Abuse   ( mesh )
Perception   ( mesh )
Health Personnel   ( mesh )
Occupations   ( mesh )
Chronic Disease   ( mesh )
Questionnaires   ( mesh )
Psychological Tests   ( mesh )
Data Collection   ( mesh )
Data Interpretation, Statistical   ( mesh )
Department of Clinical and Health Psychology thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Health Related Professions -- Department of Clinical and Health Psychology -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1996.
Bibliography:
Bibliography: leaves 112-119.
Statement of Responsibility:
by John-Paul Abner.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 027966356
oclc - 49016257
System ID:
AA00022645:00001

Full Text










EFFECTS OF TYPE OF MALTREATMENT, SEVERITY OF PHYSICAL
CONSEQUENCE, AND RESPONDENT VARIABLES ON THE
PERCEPTIONS OF MEDICAL NEGLECT AND PHYSICAL ABUSE












By

JOHN-PAUL ABNER


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY



UNIVERSITY OF FLORIDA


1996









TABLE OF CONTENTS


page

LIST OF TABLES....................................... .. iv

ABSTRACT.............................................. vi

CHAPTERS

1 LITERATURE REVIEW................................ 1

Introduction.................................... . 1
Medical Neglect of the Chronically Ill .... 2
Neglect of Neglect.......................... 5
Children's Health in Foster Care
Situations.............................. 7
Etiology and Effects of Child Neglect: A
Brief Review................................. 8
Etiology of Child Neglect................. 8
Parental Personality Factors.............. 8
Economic Factors............................ 9
Ecological Factors......................... 10
Effects of Child Neglect Upon Children.... 11
Perceptions of Seriousness of Child
Maltreatment ................................ 13
Maltreatment Type.......................... 14
Severity of Injury......................... 17
Child and Family Variables................ 20
Rater Variables............................. 25
Review of Author's Previous Study.............. 29
Hypotheses of Current Study.................... 33
Primary Hypotheses......................... 34
Secondary Hypotheses....................... 35
Current Study's Contribution to Child
Maltreatment Research........................ 36

2 METHODS........................................ 38

Measure........................................... 38
Questionnaire Revisions......................... 39
Subjects....................................... 39
Procedures..................................... 40

3 ANALYSIS AND RESULTS 43

Comparisons of Response Rate, Sex of Respondent
and Age by Occupation........................ 43
Respondents Eliminated from the Study due to
Lack of Variability of Responses............ 43









Reduction of 18 Vignettes Into Within Subject
Variables................................... 44
Overview of Analysis............................ 45
Order Effects........... ......................... 46
Repeated Measures Analysis of Variance Results. 46
Maltreatment Type and Physical Consequence
Severity.................................... 48
Occupational Effects............................... 51
History of Reporting Child Maltreatment........ 55
Vignette Order.................................. 55
Other Interactions.............................. 58
Comparison of Student and Professional
Populations................................ 58

4 DISCUSSION........................................ 62

Effects of Physical Consequence Severity and
Maltreatment Type............................ 62
Medical Neglect of Chronic Vs. Acute Illness... 66
Sex of Child..................................... 67
Age and Sex of Respondents................... 68
Vignette Order.................................. 68
Occupation Effects.............................. 69
Report History.................................... 70
Parenting History, Training, and Experience
with Maltreated Children..................... 71
Student Vs. Professionals....................... 72
Conclusions ....................................... 74
Future Research Directions...................... 74

APPENDICES

A MEASUREMENT DEVELOPMENT ........................ 77

B QUESTIONNAIRE USED IN THE FIRST PILOT STUDY... 87

C QUESTIONNAIRE USED IN THESECOND PILOT STUDY... 100

D QUESTIONNAIRE USED IN THE CURRENT STUDY....... 106

REFERENCES... ......................................... 112

BIOGRAPHICAL SKETCH..................................... 120


iii









LIST OF TABLES


Table pane

1 Means and Standard Deviation by Maltreatment
Type and Severity of Physical Consequences... 32

2 Sample Characteristics By Participant's
Occupation................................... 41

3 Respondents Eliminated from the Study due to
Lack of Response Variability Classified by
Occupation................................... 44

4 Within Subject Variable Means.................. 45

5 Repeated Measures Analysis: Final Model........ 48

6 Means for Main Effect: Severity of Physical
Consequences and Type of Maltreatment........ 49

7 Duncan Multiple Comparison Results for Main
Effect of Occupation.......................... 51

8 Means and Standard Deviations of Maltreatment
Type and Physical Consequence Severity by
Occupation................................... 52

9 Duncan Multiple Range Test for Differences in
Seriousness Ratings for Maltreatment Type
Across Occupation............................. 53

10 Duncan Multiple Range Test for Differences in
Seriousness Ratings for Physical Consequence
Severity Across Occupation................... 54

11 Means for Report History Maltreatment Type... 55

12 Means of Vignette Order Type of Maltreatment
Interaction .................................. 56

13 Means of Vignette Order Consequence Severity
Interaction .................................. 57

14 Means for Vignette order Consequence Severity
Maltreatment Type........................... 58

15 Comparison of Percentage of Questionnaires
Retained or Eliminated by Professional
Status....................................... 59









LIST OF TABLES continued

16 Table Comparing Means of Professionals and
Student Across Physical Consequence Severity
and Maltreatment Type......................... 61

Al Result of the First Pilot Study: Students Rate
Seriousness of Physical Consequences (n=71). 79

A2 Results of the Second Pilot Study: Physicians
rate seriousness of physical consequences.... 81

A3 Students' and Physicians' Mean Seriousness
Ratings of Physical Consequences Used to
Construct Vignettes........................... 83

A4 Results of the Bonferroni T-Tests.............. 85











Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of
the Requirements for the Degree of Doctor of Philosophy


EFFECTS OF TYPE OF MALTREATMENT, SEVERITY OF PHYSICAL
CONSEQUENCE, AND RESPONDENT VARIABLES ON THE
PERCEPTIONS OF MEDICAL NEGLECT AND PHYSICAL ABUSE

By

John-Paul Abner

August, 1996


Chairman: Suzanne B. Johnson
Major Department: Clinical and Health Psychology

Child neglect, particularly neglect of children's

medical needs, is a serious but under researched topic. In

this study, a questionnaire was used to examine the

differences in child maltreatment perceptions. This

questionnaire consisted of 18 two-sentence vignettes, each

describing a type of child maltreatment and the resulting

physical consequences. The vignettes varied by

maltreatment type (medical neglect of an acute illness,

medical neglect of a chronic illness, and physical abuse)

and physical consequence severity (mild, moderate, and

severe). A variety of secondary variables were examined

for their possible effect on child maltreatment

perceptions. These included the respondent's age, sex,

parenting experience, experience with suspected

maltreatment, training in child maltreatment issues, and

history of reporting suspected maltreatment, and the sex of









child. Respondents to the questionnaire were 395

professionals likely to have contact with maltreated

children (Health and Rehabilitative Service protective

investigators and protective service workers, journalists,

lawyers, pediatricians, pediatric nurses, child and/or

family psychologists, social workers, and elementary school

teachers). Respondents read each vignette and rated it on

a nine-point Likert scale as to the seriousness of child

maltreatment. Both maltreatment type and physical

consequence severity affected respondents' ratings.

Vignettes involving more severe physical consequences

and/or physical abuse were ranked as significantly more

serious. Medical neglect of chronic illness was ranked

more serious than medical neglect of an acute illness.

Maltreatment type and physical consequence severity

significantly interacted as physical abuse with moderate

consequences was rated more seriously than both forms of

medical neglect with severe consequences. Also, whereas

the severity effect was consistent across both medical

neglect categories (mild < moderate < severe), there was no

significant difference in the perception of physical abuse

with moderate consequences and physical abuse with severe

consequences. Occupation had a significant effect as

teachers and journalists rated child maltreatment more

seriously than psychologists and lawyers. Respondents who

had a history of reporting child maltreatment rated

maltreatment less seriously than respondents without a









history of reporting maltreatment. No significant effects

were found for respondent's sex, parental status, age,

training in child maltreatment issues, and contact with

suspected child maltreatment, or for sex of child.














CHAPTER 1

LITERATURE REVIEW

Introduction

Child neglect has been defined as a condition in

which the child's caretaker "... either deliberately

or by extraordinary inattentiveness permits the child

to experience avoidable present suffering and/or fails

to provide one of the ingredients generally deemed

essential for developing a person's physical,

intellectual, and emotional capacities" (Polansky,

Hally, & Polansky, 1975, p. 5). One important aspect

in a child's development is the provision of adequate

medical care. Negligent or deliberate failure to

provide adequate health care resulting in foreseeable

injury or impairment to the child has been termed

medical neglect (Giovannoni & Becerra, 1979).

Nationally, there are approximately 109,300 children

who experience medical neglect per year.

Approximately 72,000 of these children are refused

health care (needed care is either refused or not

provided despite recommendation from a health care

professional) and another 37,700 receive delayed

health care (failure to seek timely and appropriate

medical care which any reasonable layman would have










recognized as requiring professional medical

attention). Medical neglect therefore accounts for

6.51% of the estimated incidence of child abuse

(National Center for Child Abuse and Neglect, 1988)

and it is a subtype of neglect which merits public and

professional attention. Yet, there is very little

research on this subject.

Bross (1992) reviewed legal precedent in the area of

medical care neglect, particularly the right of parents to

refuse to give consent to treatment for their children. A

variety of factors were observed to influence the court's

decision to overrule the parent's right to consent

including severity of prognosis, few contraindications to

the medical procedure; and impairment of the child's

quality of life. It is significant to note that Bross

observed that if the outcome is only disfigurement or a

marginally disabling condition, the court is likely to

deter legal orders and not intervene. Bross (1992)

concluded that whereas parental rights are important, court

intervention should be considered if no minimally

reasonable parent would choose to refuse treatment.

Medical Neglect of the Chronically Ill.

Much of the existing medical neglect literature

addresses the medical neglect of the chronically ill.

Long term effects of medical neglect of chronically ill

children can be different and more severe than its effects










on healthy children (Jaudes & Diamond 1986). Further,

medical neglect may actually be more prevalent in children

with chronic illnesses than in other children.

In one descriptive study, Jaudes and Diamond (1986)

conducted a retrospective chart review of neglected and

abused children during a 6-year time period at La Rabida

Children's Hospital in Chicago. A total of 61 chronically

ill children were identified as neglected, and a total of

114 individual occurrences of neglect were documented for

these children. Seventy-five occurrences were of medical

care neglect, two of which resulted in the child's death.

Other instances of death resulting from medical neglect

have been noted in the literature including a case study in

which one child died from medical neglect of diabetes

(Geffken, Johnson, Silverstein, & Rosenbloom, 1992). The

authors concluded that neglect can be a serious problem for

chronically ill children and that health care professionals

should be aware of the difficulties imposed by this type of

maltreatment.

Boxer, Carson, and Miller (1988) observed that many

chronically ill children have experienced an unstable

course of illness with frequent or long-term

hospitalizations resulting not from extraordinary disease

but rather because the children come from dysfunctional and

neglectful families.










Higher abuse and neglect incidence rates for children

with disabilities and chronic illnesses compared with

physically healthy children have been reported (American

Association for Protecting Children, 1984). Several

authors have noted that these incidence rates may be higher

because chronically ill or handicapped children are at risk

for abusive behavior (Glaser & Bentovim, 1979, Jaudes &

Diamond, 1986, Morgan, 1987). Ammerman, Van Hasselt,

Hersen, McGonigle, & Lubetsky, (1989) found that 39% of the

sample had experienced a history or at least suspicion of

abuse and/or neglect, a statistic that is much greater than

the national child maltreatment incident rate.

Several possible explanations for the over-

representation of maltreatment in disabled and chronically

ill populations have been offered. Chronically ill

children may be more vulnerable to maltreatment, because

they exhibit characteristics often associated with child

abuse: frequent separations from their mothers resulting in

poor mother-child attachment; increased parental stress due

to care requirements and medical needs, (Ammerman, Van

Hasselt, & Hersen, 1987) parental pain associated with the

loss of the expected normal child, family social stigma,

and increased demands of caring for chronically ill

children which overwhelm some parents who may have

adequately coped with a normal child (Jaudes & Diamond,

1986). Both Jaudes and Diamond (1986) and Morgan (1987)










note that the incidence of abuse and neglect may be higher

in this population partially because abuse and neglect have

created these handicaps or conditions.

Neglect of Neglect.

Medical neglect of chronically ill and acutely ill

children is a frequently occurring phenomenon needing

research and public attention. Yet, compared to physical

and sexual abuse, all forms of neglect, including medical

neglect, receive relatively little attention. Wolock and

Horowitz (1984) have described this as the "neglect of

neglect." A national study conducted by the American

Association for Protecting Children (1986) estimated that

in 1981, 1,108,500 reports of child maltreatment were filed

with various child protection service agencies and an

estimated 470,600 of these reports were substantiated.

Between 1976 and 1982, deprivation of necessities accounted

for 64% of substantiated child maltreatment, compared with

25% physical abuse, 6% sexual maltreatment, and 17%

emotional maltreatment. In 1987, the reported cases of

neglect outnumbered physical abuse reports by over eleven

to one in New York City (Green, 1991). Neglect was

responsible for 51% of deaths due to child maltreatment

during this time (American Humane Association, 1984).

Even though neglect accounts for a higher number of

reports of child maltreatment and a higher number of

deaths, it does not generate the same amount of public










concern or research as does physical abuse (Mayhall &

Norgard, 1986). Across the nation, interviews of child

protection service caseworkers indicate that child neglect

cases are given a lower priority than cases involving

sexual or physical abuse (Mayhall & Norgard, 1986). In a

sample of 1,874 cases of child maltreatment reports

gathered in New York State in 1985, 48% of sexual abuse

allegations and 39% of physical abuse allegations were

substantiated as opposed to 28% of neglect cases

(Eckenrode, Powers, Doris, Munsch, & Bolger, 1988). Also,

a lower percentage of reported neglect cases receive

protective services. Fifty-three percent of neglect cases

received protective services as opposed to 55% of cases

which involved minor or unspecified injury, 61% of cases

involving major physical injury, and 72% of sexual

maltreatment cases (American Humane Association, 1984).

Mayhall and Norgard (1986) discussed several reasons

why neglect receives much less attention than abuse.

Neglect appears to be much more difficult to define,

document and substantiate. Johnson (1993) noted that legal

definitions may be too complicated or too abstract to

provide physicians with clear guidelines for recognizing

and reporting neglect. Also, the definition of neglect may

be open to a broader interpretation and more dependent on

community standards of child care because neglect involves

acts of omission rather than acts of commission (Green,










1991). However, Craft and Staudt (1991) found no

difference in the definition of neglect when comparing

urban and rural communities and black and white respondents

and they suggested that this may indicate that there is a

closer consensus of what constitutes child neglect than is

normally believed.

Wolock and Horowitz (1984) noted that the issue of

child maltreatment first entered the realm of public

awareness as a social problem, mainly through the wide

publication of the battered child syndrome, a syndrome that

did not include neglect. Therefore, "neglect was virtually

excluded from the initial phase in which child maltreatment

as a contemporary social problem was recognized" (Wolock &

Horowitz, 1984, p. 530). They speculated that the

extremely high correlation between neglect and poverty may

also be a factor in this prejudicial treatment. Also,

abuse is a dramatic and newsworthy event whereas neglect

receives scant media attention (Paget, Phillip, &

Abramczyk, 1993).

Children's Health in Foster Care Situations.

Another factor that might explain the lack of

attention devoted to medical neglect is the general lack of

concern for health issues within child care systems.

Several researchers have found that a significant

proportion of children in the foster care system have

incomplete immunization records (Swire & Kavalier, 1977;










Schor, 1981; and Hochstat, Jaudes, Zimo & Schacter, 1987),

identified disorders that are left untreated (Schor 1981,

Swire and Kavalier, 1977), and chronic health problems that

are often undiagnosed (Schor, 1981). Since child

protective networks have difficulties in caring for the

basic health needs of children involved in their system, it

may be that such a system may also have trouble identifying

and treating problems of medical neglect.

Etiology and Effects of Child Neglect: A Brief Review

Etiology of Child Neglect.

Generally, there are three different etiological

approaches to explain the causes of neglect: parental

personality, economic, and ecological factors. (Polansky,

Ammons, & Gaudin, 1985).

Parental Personality Factors.

The personality approach attempts to explain neglect

by noting general deficits in the parent's personality

structures (Polansky, Ammons, & Gaudin, 1985). Parental

factors that have been associated with neglect include

depression, intellectual inadequacy, alcoholism (Hegar and

Yungman, 1989), a history of neglect and abuse in the

parent's developmental background (Gelardo & Sanford,

1987), apathy, a lack of impulse control (Polansky,

Chalmers, Buttenwieser, & Williams, 1981), and unrealistic

developmental expectations (Twentyman & Plotkin, 1982).

Burgess and Conger (1978) found that parental interactions









with their children were significantly associated with

child neglect. Neglectful mothers' behaviors were more

extreme than the behaviors of abusive and control mothers;

neglectful mothers had more negative behaviors and fewer

positive behaviors with their children as well as less

interaction with other family members.

Economic Factors.

The economic approach is a sociological explanation

which focuses on the deprivations and shortages that

poverty-stricken families share with their children

(Polansky, Ammons, & Gaudin, 1985). Bath and Haapala (1993)

found neglectful families were poorer, more reliant on

public income, and more likely to consist of a single

parent household. Generally, income and instances of

physical neglect are inversely related (Hampton, 1987,

Garbarino, 1978). Lower income families and African-

Americans are over represented nationally in statistical

reports while whites are under represented (Hampton, 1987).

Several reasons have been posited for the higher

frequency of neglect in poorer populations. One theory is

that poverty results in increased parental (Garbarino,

1976) and community (Garbarino, 1978) stress that is

expressed in child maltreatment. Another explanation is

that child neglect is defined and prosecuted by the white

middle class who have different cultural child-rearing

values than other subpopulations (Polansky & Williams,









1978). However, Polansky, Ammons &, Weathersby (1985)

found that there is a common American standard of minimal

child care to which both poor and middle class ascribe.

Ecological Factors.

The ecological explanation is a social-psychological

approach. This approach focuses on the strong impact that

the environment has upon family functioning and the

interaction that the family has with the environment

(Polansky, Ammons, & Gaudin, 1985). A variety of

characteristics have been associated with neglectful

parents and their children. Neglectful families are

characterized by: more children, less income, inadequate

sleeping and housing arrangements (Dubowitz, Black, Starr,

and Zuravin, 1993), household density above 1.51 persons

per room (Zuravin, 1986), less education (Polansky, Gaudin,

Ammons, & Davis, 1985), less likelihood of fuli-time

employment, younger mothers (Jones And McCurdy, 1992) and

greater likelihood of having medical, mental-health and

substance abuse problems (Bath and Haapala, 1993).

Neglected children are often younger than physical abuse

victims and more likely to be female (Jones and McCurdy,

1992)

Social support has been a frequent ecological variable

examined in this literature. Neglectful parents were

significantly more likely to be isolated and have strained

relations with their relatives, (Dubowitz, Black, Starr,









and Zuravin, 1993) Vondra (1990) stated that this pattern

of social isolation includes short term friendships, few

friends, and dissatisfaction with affiliations with

relatives. Furthermore, neglectful families may exhibit an

over-reliance on formal support services (welfare,

substance abuse centers, etc.) and an under-reliance on

informal support services such as youth groups, churches,

and other neighborhood networks (Vondra, 1990). Neglectful

mothers were significantly more lonely than control mothers

who did not neglect their children (Polansky, Ammons, &

Guadin, 1985). Neighbors of neglectful families tend to

see the neglecting mothers as deviant and tend to distance

themselves from them (Polansky, Ammons & Guadin, 1985;

Polansky, & Guadin, 1983; Guadin & Polansky 1986). Seagull

(1987) stated that the connection between social support

and physical neglect currently lacks solid empirical

evidence whereas there is strong evidence indicating that

neglectful parents are socially isolated. However, Seagull

(1987) does not view this social isolation as causal but

rather the product of characterologicall deficits of the

parents" (p. 49).

Effects of Child Neglect Upon Children.

The impact of neglect upon children is substantial.

Child neglect negatively impacts cognitive functioning,

leading to lower IQ scores (Rogerness, Amrung, Macedo,

Harris, & Fisher, 1986) and more academic difficulties










(Croucn & Milner, 1993). Neglected children performed

significantly worse than abused children and normal

controls on auditory and verbal language tests (Law and

Conway, 1992) as well as language comprehension tests

(Fox, Long, & Langlois, 1988). Child neglect also has

significant developmental repercussions including failure

to thrive, (Crouch and Millner, 1993) and higher

frequencies of resistant and avoidant attachments to the

primary caretaker (Lamb, Gaensbauer, Malkin, & Schultz,

1985). Neglected preschool and school age children were

found to be more passive, to produce less frequent play

behavior initiatives with their mothers, and to be less

affectionate than non-maltreated children (Crittenden,

1992).

Neglected children experience more emotional and

behavioral difficulties than their non-maltreated peers.

They are more impulsiveness (Green, 1991), have fewer pro-

social behaviors (Hoffman-Plotkin, Twentyman, 1984), are

more aggressive in school interactions (Reidy, 1977), have

more self-reported psychiatric symptoms, and have more

parent-reported behavior problems (Williamson, Bourdin, and

Howe, 1991). Childhood experiences of neglect have also

been positively related to delinquency (Brown, 1984; Koski,

1988).

In summary, a wide range of consequences has been

associated with child neglect. In comparing neglected










children with other types of maltreated children, Egeland,

Stroufe, & Erickson (1983) described their neglected

subjects as the most unhappy of the maltreated children and

as experiencing significant difficulties in effectively

coping with their environments. In addition to all other

possible consequences, neglect can result in serious

physical injury or death for the maltreated child.


Perceptions of Seriousness of Child Maltreatment

This section shall address four different

characteristics which may impact perceptions of the

seriousness of child maltreatment: type of maltreatment,

severity of the physical consequences associated with the

maltreatment, characteristics of the child or maltreating

agent, and characteristics of the rater. Several studies

reviewed include ratings of seriousness. However, many

others examined how the four factors cited above impacted

reporting or substantiation rates. Thus, reporting and

substantiation rates will be addressed as a rough measure

of the seriousness perception. However, it should be noted

that other factors may come into play in reporting and

substantiation rates other than perceptions of seriousness.

For example, a reporter may perceive an incident of neglect

to be just as serious as an incident of physical abuse, but

the reporter may choose not to report the neglect because

it is more difficult to prove.










Maltreatment Type.

A few studies have addressed whether the type of

maltreatment impacts professional and lay persons'

assessment of the seriousness of various forms of child

maltreatment. Giovannoi and Becerra (1978) constructed

maltreatment vignettes describing thirteen different

maltreatment categories including medical neglect. The

vignettes were given to 1,065 lay respondents, 71 lawyers,

113 social workers, 70 pediatricians, and 50 policemen, all

of whom were asked to rate on a nine-point scale how

serious they thought each incident would be from the

standpoint of the welfare of the child. Vignettes of

medical neglect were consistently rated as less serious

than physical abuse by all classes of raters even though

the actual physical damage may have been more severe

(although a comparison of medical neglect versus physical

abuse vignettes was not a focus of the study and was not

tested for significance).

Fox and Dingwall (1985) presented a subset of 20

Giovannoni and Becerra vignettes to 20 social workers and

20 lay persons in Southern England. Lay persons and social

workers agreed about the relative seriousness of the

described incidents. However, even though statistical

comparisons of maltreatment type were not conducted in this

study, it is notable that episodes of medical neglect were

generally perceived as less serious than physical abuse.









This is particularly interesting because the lowest ranked

physical abuse vignette--"The parents usually punish their

child by spanking him with a leather strap leaving red

marks on his skin" (p. 470)--was ranked almost equally with

the most serious medical neglect scenario--"The parents

ignored the fact that their child was obviously ill, crying

constantly and not eating. When they finally brought the

child to a hospital he was found to be seriously

dehydrated" (p. 471). Although the actual physical damage

to the children is quite vague in these vignettes, the

child who experienced medical neglect required

hospitalization while the physically abused child did not.

Therefore, even though the bodily damage to the medically

neglected child may have been greater than that of the

physically abused child, there was no difference in the

perception of seriousness.

Misener (1986) conducted a mail survey in which 596

nurses from eight different specialty groups responded to

the Giovannoni and Becerra's vignettes. Also included was

a scale on which the nurses rank-ordered 11 operationally

defined categories of child maltreatment. Medical neglect,

defined as "the failure to obtain medical care for a sick

child or failure to provide routine medical, dental, or

optical care," was ranked fourth in seriousness below

physical abuse, sexual abuse, and emotional maltreatment.










Nightingale and Walker (1986) administered a

questionnaire on child maltreatment identification

reporting to 143 Head Start personnel. The authors

hypothesized that because of its prevalence and conceptual

ambiguity, neglect would be the least reported whereas

sexual abuse, because of its repulsiveness, would be more

likely to be reported. Respondents were asked to state the

number and type of incidents in which they suspected child

maltreatment. Then, they were asked to report the number

of incidents that they referred to protective services.

The authors reported that neglect, though identified more,

was reported significantly less than sexual abuse and there

was a tendency to report neglect less than all the other

forms of maltreatment. The authors found that neglect may

not be reported as frequently because the Head Start

workers are often exposed to incidents of neglect, possibly

leading to some degree of desensitization. Also, because

neglect is so difficult to define and there is no

prescribed treatment, child care workers may be less likely

to report it.

Ards and Harrell (1993) analyzed data from the 1980

and 1986 National Study of the Incidence and Prevalence of

Child Abuse and Neglect. They found that child protection

services workers were more likely to be aware of sexually

abused children than physically abused children. Also,

child protection services workers were less likely to be










aware of children who were physically or emotionally

neglected than those children who experienced physical

abuse.

Tjaden and Thoennes (1992) used information procured

from criminal court and child protective services recorded

in Denver (CO), Los Angeles (CA), and Newcastle (DE) in

1985-86, to identify factors associated with the decision

to file criminal charges in child maltreatment cases. The

type of maltreatment was the strongest determinant of the

decision to file criminal charges. Sexual abuse was

significantly more likely to result in criminal prosecution

whereas cases involving neglect were significantly less

likely to involve criminal charges. The authors

hypothesized that caseworkers may tend to view neglectful

parents as emotionally or mentally incompetent rather than

malevolent or they may feel that they are simply the victim

of difficult circumstances.

In four of these studies, (Misener, 1986; Nightingale

& Walker, 1986; Tjaden & Thoennes, 1992; and Ards &

Harrell, 1993) statistical methods were used to compare

different types of maltreatment with one other. In each

case, neglect was reported less often or ranked less

seriously than abuse.

Severity of Injury

Surprisingly, severity of injury associated with

maltreatment has been considered in only a handful of









studies. Turbett and O'Toole (1983) administered vignettes

of five physical abuse cases to 91 teachers, 178 registered

nurses, and 76 pediatricians. Race, occupational status of

the family, and level of injury were varied in the

vignettes. All three groups responded to the level of

injury, with more serious injuries receiving higher

ratings. In another study, over 1,196 mandated reporters

completed a vignette measure; the more severe the incident,

the more likely it was to be regarded as abuse and reported

(Zellman, 1992).

Alter (1985) found that when sufficient evidence

exists that severe abuse has occurred, child care workers

almost always choose to substantiate. When evidence exists

for moderate abuse, other factors play an important role in

the substantiation decision. Tjaden and Thoennes (1992)

also found that the likelihood of criminal prosecution

increased with the severity of the maltreatment.

Winefield and Bradley (1992) reviewed all child

protection reports in South Australia from 1988-89

(N=3,228) searching for determinants of child protection

workers' decisions to register a reported incident as child

abuse or neglect. They analyzed a variety of factors

including age, source of report, sex, ethnicity, type of

abuse, severity of maltreatment and previous report using

logistic regression. Only two factors significantly










contributed to registration as younger children and more

severe maltreatment were more likely to be registered.

In all of these studies, severity of the maltreatment

was considered but it was not clearly defined. For

example, in the Zellman (1992) study, the severity in one

vignette was defined by the injury sustained whereas in

another vignette it was defined by the perpetrator's

behavior. Tjaden and Thoennes (1992) defined maltreatment

severity arbitrarily: "We categorized physical abuse as

severe if they were found to involve severe cuts or

bruises, burns, bone fractures, skull fractures, brain

damage, or congenital drug addiction...neglect cases as

severe if the victims were abandoned, exposed to a

dangerous behavior or a life-threatening environment." (p.

813). Winefield and Bradley (1992) noted that while the

strongest predictor of child abuse substantiation was the

severity of maltreatment, there were no clear cut

guidelines as to how the workers rated the maltreatment.

They illustrated the inconsistencies by noting that verbal

abuse was ticked as the most serious sign of abuse in 32

cases. However, four of the five cases with skull

fractures were rated as moderate in severity. The authors

concluded that the severity variable needs more

investigation because despite the important role that it

plays in report substantiation, there is no information as

to how the caseworkers form a judgment about severity.










Further, no data addressing the reliability of the

maltreatment severity were provided in any of the studies

reviewed.

Child and Family Variables

Various authors have suggested that demographic

variables play a role in child abuse/neglect reporting.

Reported cases of child maltreatment tend to over-

represent lower socioeconomic status groups and under-

represent the more affluent (McPherson & Garcia, 1983).

Kim's (1985) survey of 120 physicians in the mid-South and

Eastern Seaboard regions of the United States revealed that

physicians reported significantly more cases when dealing

with poorer or minority groups. In Turbett's and O'Toole's

study (1983), socioeconomic status and ethnic status had

no significant effect on nurse and teacher recognition and

reporting of child abuse, but physicians were significantly

more likely to identify and report abuse if the family was

identified as being black or poor.

McPherson and Garcia (1983) explored the effects of

socioeconomic status and patient familiarity on

pediatricians' child abuse responses. Vignette

questionnaires were mailed to 160 randomly selected

pediatricians; 109 completed the questionnaires.

Pediatricians ranked the low socioeconomic status families

as significantly more likely to be chronic offenders than

higher socioeconomic status families. Families with whom










the pediatrician was more familiar were less likely to

receive an abuse report than families less familiar to the

pediatrician. The authors speculated that the under-

reporting of affluent families may result from the

correlation of income and familiarity of the family to the

physician, as lower income patients often do not have a

consistent family physician.

Participants, from 15 different states, whose

occupations are mandated to report child maltreatment if

encountered (n=1,196) completed a vignette measure

detailing 12 different maltreatment vignettes including

physical abuse, sexual abuse, and neglect (Zellman, 1992).

Respondents ranked the seriousness of the incident, whether

the incident was abusive, and several reporting

characteristics. Abuse-relevant judgments and reporting

rates varied as a function of case characteristics. A

record of previous abuse, recantation by the child, and the

severity of abuse were significant and powerful predictors

of the subjects' responses. History of a prior abuse led

to ratings of greater seriousness and an increase in the

likelihood to report. When a child recanted the alleged

abuse, there was a decrease in the likelihood to report.

Cases involving younger children were also more likely to

be reported, and lower socioeconomic status families were

more likely to be reported. The intent of the abuser also

impacted the raters' judgment, with angry or lazy










perpetrators more likely to be reported. The authors

concluded that most of the effects of case characteristics

were sensible and even admirable but that some data

indicated that workers may be misinformed or biased

especially when issues of race, SES, and child recantations

arise.

Like the reporting literature, demographic variables

as well as other characteristics affect substantiation

rates. In their review of child protection reports from

South Australia, Winefield and Bradley (1992) found that

older children's reports were substantiated significantly

more often than reports concerning younger children. It

was hypothesized that this may be due to the older child's

increased verbal faculties or the confounding presence of

prior reports.

Eckenrode et al. (1988) reviewed a representative

sample of abuse and neglect reports filed in New York state

in 1985. Their final sample consisted of 880 neglect

reports, 796 sexual abuse reports, and 198 physical abuse

reports. A number of variables were studied as possible

predictors of substantiation: whether the source of the

report was a professional or a nonprofessional, number of

contacts with the subject, number of contacts with the

reporter, number of allegations, length of investigation,

and a variety of demographic variables. For all types of

maltreatment, reports from a professional were










significantly correlated with a higher substantiation rate

than reports from a lay person. Rates of substantiation

were higher if the report involved: an African-American or

Hispanic child as opposed to a Caucasian; a younger rather

than older child; a child from New York City (rather than

any other place in the state), or a prior report of

neglect.

Other studies have reflected the findings of Eckenrode

et al. (1988). Frequency distributions from several states

suggested that reports submitted by professionals are more

likely to be substantiated than those by non-professionals

(Groenveld & Giovannoi, 1977). A large scale study in

Georgia also documented this substantiation trend (Jason,

Andereck, Marks, & Tyler, 1982) and data collected by the

American Humane Society have provided information

suggesting that this is a nationwide trend (American

Association for Protecting Children, 1986). Likewise,

Jason et al. (1982) and the American Association for

Protecting Children (1986) have also noted that the

substantiation rates for reports involving black children

are higher than for white children.

Alter (1985) used semi-structured interviews with 12

child care workers to assess the influence of various

factors on substantiation rates. The author identified

four concepts which are used to aid the decision-making

process when there is evidence of only moderate abuse.









These factors include: (1) willful behavior of the parent

(i.e., parents have adequate financial, intellectual, and

emotional ability but fail to provide adequate care); (2)

positive or negative parent-child relationship (i.e., do

the parents view the child as good or bad; can the parents

separate the child's identity from their own, do the

parents feel able to meet the child's needs); (3) low or

high parental social deviance (i.e., do the parents exhibit

any criminal activity; do they generally conform to

community standards or is there social isolation or drug

usage); (4) parental desire to change (i.e., do the parents

exhibit a positive attitude toward the investigator,

admitting the existence of a problem or asking for help;

are the parents negative with the examiner and exhibit

denial, passivity or hostility). A repeated-measure

analogue experimental study was then conducted using full-

time protective service workers (N=73) as subjects. A two

page vignette questionnaire was administered. The subjects

were informed as to the age of the child, the frequency of

the neglectful behavior, and that this neglectful behavior

was moderate in nature. On the second page, the previously

mentioned factors were presented in sixteen different

arrangements which included all possible combinations. All

four factors were significantly correlated with a decision

whether or not to substantiate. Willful parental behavior










and poor parent-child relationships were the most

influential factors.

In summary, several demographic variables associated

with the child or the abusive parent have been identified

as influential in the perceived seriousness, reporting,

and/or substantiation of child maltreatment allegations.

Several studies have found that child care professionals,

most notably pediatricians, are more likely to report

black, lower socioeconomic status patients, and patients

with whom they are less familiar. Likewise, lower

socioeconomic status and minority groups are more likely to

have reports of child maltreatment substantiated. A

variety of other factors influence substantiation rates

including: the presence of a prior report, the source of

the report, parent-child relationship, willful parental

behavior, parental social deviance, and perceived parental

desire to change.

Rater Variables

Demographic variables associated with child

maltreatment reporters themselves have been found to

influence seriousness perceptions and reporting rates.

Nightingale and Walker (1986) gathered demographic data on

Head Start personnel, including socioeconomic status,

education, parenting experience, and training in child

maltreatment identification. The personnel responded to

vignettes of child neglect, physical abuse and sexual










abuse. The most influential variable was that of prior

training in child maltreatment identification. Workers who

had received such training tended to rate vignettes more

seriously. Likewise, personnel with more work experience

were more likely to report the incident (Nightingale and

Walker, 1986). Similarly, in a survey of 440 teachers in

Illinois, teachers who reported greater awareness of signs

of child maltreatment had significantly more exposure to

training in neglect and abuse detection than other teachers

(McIntyre, 1987).

Kean and Dikes (1991) administered a Vignette survey

to 160 jurors and 176 child protective service team members

in Colorado. Subjects were given three separate vignettes

representing physical abuse, psychological maltreatment,

and neglect of a child. Subjects were asked to rate each

incident on three 7- point Likert scales: to what extent

an abusive situation had taken place, to what extent was

the abuse the child's fault, and to what extent did the

parent need help. The child's sex and the sex of the

abusing parent was manipulated by the experimenter. Female

jurors were more likely to say that they would report

maltreatment than male jurors and they were less likely to

report the abuse of a male victim than a female victim.

Other groups did not show this reporting bias. Jurors born

before 1945 were significantly less likely to regard an

incident as abusive than younger jurors. Both male and










female subjects stated that they would report an abusive

father more often than an abusive mother.

Craft and Staudt (1991) used a questionnaire

containing 30 vignettes of child neglect to examine the

differences between rural and urban communities in

scenarios that would be reported as neglect by a lay

population (n = 38 rural; n = 40 urban). They also

examined which scenarios were likely to be reported and

substantiated by protective service workers (n = 25).

Protective team members had four choices in responding to

the vignettes: "(1) not likely to substantiate in either

community; (2) more likely to substantiate in a rural than

urban community (3) more likely to substantiate in an urban

that rural community and (4) likely to substantiate in

either community. Among lay respondents, only one case

produced a statistically significant difference in

reporting between urban and rural respondents. However,

there was a significant tendency for protective team

workers to choose the "more likely to substantiate cases in

an urban" setting than in rural settings. The authors

concluded that rural and urban communities have similar

values in defining child neglect but that professionals in

the child protective services may treat child neglect

differently depending upon the setting of the maltreatment.

The authors suggest that the working definition of neglect









varies because of factors such as worker caseload, services

available, and juvenile court expectations.

Ringwalt and Caye (1989) administered a questionnaire

consisting of 27 vignettes of minimal to moderate child

neglect to a convenience sample of 471 residents of a rural

county in North Carolina. Their study examined the effects

of demographic variables (age, sex, race, education,

occupation, income, and child-rearing experience). Women

were more likely than men to rate the vignettes as severe,

and African-Americans were more likely to judge the

vignettes as severe than were white people. People with

white collar jobs rated the vignettes as more severe than

did blue collar job holders. However, one finding that

the authors found confusing was that education level was

inversely related to the rankings of seriousness. The

authors did not comment on the variability of the responses

but it is possible that the higher-educated population gave

more variable responses, using a greater range on the five

point scale. One problem with this study was that a large

majority of the respondents were overwhelmingly female

(79.8%) and it is possible that the males who responded did

not represent the population.

Thus, various factors associated with the child care

workers themselves may influence perceptions of seriousness

as well as substantiation and reporting practices. Studies

to date suggest that women are more likely to report child









maltreatment than men (Ringwalt & Caye, 1989). Females may

be especially more likely to report if the child is female

(Kean & Dikes, 1991). Work experience and training in

neglect and abuse detection appear to lead to higher

reporting rates, (Nightingale & Walker, 1986; McIntyre,

1987) and persons born before 1945 tend to view child

maltreatment less seriously than their younger counterparts

(Kean & Dikes, 1991). People with white collar jobs are

more likely to rate maltreatment more severely than people

with blue collar jobs, although academic education level

has been found to be inversely related to seriousness

rankings (Ringwalt & Caye, 1989).

Review of Author's Previous Study

For a variety of reasons, medical neglect has been

subjected to relatively few empirical studies. However,

the available literature does suggest that medical neglect

is perceived as less serious than other forms of child

maltreatment. Unfortunately, studies that have compared

various types of child maltreatment, including medical

neglect, have not adequately controlled severity of injury

that resulted from this maltreatment.

In an effort to address this, Abner (1992) explored

students' perceptions of different types of child

maltreatment using vignettes in which severity of the

physical consequence of the maltreatment was controlled.

Two pilot studies were conducted to gather reliable









empirical estimates of severity of various forms of

physical consequences that were independent of the child

maltreatment context. Three reliable statistically

different levels of severity of physical consequence were

identified: mild (mean between 2-4 on a 9 point Likert

scale), moderate (mean between 5-6), and severe (mean

between 7-9). A questionnaire consisting of 18 two

sentence vignettes was then constructed.

Three types of child maltreatment were included

(physical abuse, medical neglect of an acutely ill child,

and medical neglect of a chronically ill child) with three

levels of severity of physical consequences (mild,

moderate, and severe) two vignettes represented each type

of maltreatment and physical consequence severity

condition. The respondent was asked to rate the

seriousness of each child maltreatment vignette on a 9

point Likert Scale. Four different forms of the

questionnaire were developed to control for order effects

and to vary the sex of the child described in the

vignettes.

The questionnaire was administered to seven different

groups of students likely to encounter situations involving

child maltreatment in their future occupations:

journalism, elementary education, medicine, law, nursing,

nursing assistant, and social work.









A substantial number of subjects had little or no

variability in their responses (used only 1 or 2 out of 9

ranks) and were excluded from subsequent analysis, reducing

the sample size from 511 to 436. Chi-square analyses

indicated that law and medical students had a significantly

lower percentage of questionnaires eliminated than other

student groups. No significant sex difference in

questionnaire elimination was found.

Because each cell in the 3 (Type of Maltreatment) X 3

(Severity of Physical Consequences) design was comprised of

two vignettes, the mean of the two vignette ratings served

as the dependent variable. Relevant means and standard

deviations are presented in Table 1.

Vignette order, sex of child, and sex of rater had no

significant effect on seriousness ratings and were dropped

from consideration. Powerful main effects were found for

both within subject variables (type of maltreatment and

severity of physical consequences). Across maltreatment

types, severe consequence vignettes received higher ratings

than moderate consequence vignettes, followed by mild

consequence vignettes. Physical abuse was rated as more

serious than both medical neglect categories. Medical

neglect of a chronically ill child was rated as more

serious than medical neglect of an acutely ill child.

There was also a significant interaction between type

of maltreatment and severity of physical consequence.










Subsequent repeated measures analyses of variance indicated

that the main effect for maltreatment type was significant

in both the mild and moderate categories. Physical abuse

categories were rated as the most serious followed in

descending order by medical neglect of a chronically ill

child and medical neglect of an acutely ill child. At the

severe physical consequence level, physical abuse was

ranked higher than both medical neglect categories, which

were not significantly different from each other (See Table

1).

Table 1

Means and Standard Deviation by Maltreatment Type and

Severity of Physical Consequences


~____________ ___Maltreatment Type_______
Consequence Medical Neglect Medical Neglect Physical Abuse
Severity Acute Illness Chronic Illness
______Mean S.D. Mean S.D. Mean S.D.
Mild 1__ 6.29 1.43 6.58 1.43 6.76 1.56
Moderate 7.04 1.47 7.38 1.37 8.33 0.89
Severe 8.05 1.11 8.02 1.14 8.63 0.68


Totals


Physical Consequence Severity
Mild Moderate Severe
Mean S.D. Mean S.D. Mean S.D.
6.54 1.31 7.58 1.06 8.23 7.46

Maltreatment Type
Medical Neglect of Medical Neglect of Physical Abuse
Acute Illness Chronic Illness
Mean S.D. Mean I S.D. Mean S.D.
7.13 1.27 7.33 1.18 7.91 0.89










Furthermore, dependent T-tests indicated that medical

neglect of an acutely ill child and medical neglect of a

chronically ill child with severe physical consequences was

rated as less serious than physical abuse of a child with

moderate physical consequences.

There was also a significant three way interaction,

student major x maltreatment X physical consequence

severity. Medical students' ranked the vignettes

significantly less seriously than other groups of students

except for law students. The exception was that there were

no differences in student's perceptions of mild physical

abuse, hence the three way interaction.

Hypotheses of Current Study

Abner's previous research illustrated that college

students have different perceptions of the seriousness of

child maltreatment based on the type of maltreatment and

the severity of the physical consequence. However, college

students have little or no actual contact with maltreated

children and may not be representative of the population

who makes judgments about child maltreatment. Kean and

Dukes (1991) suggest that using university students as

subjects in studying perceptions of child maltreatment make

efforts to generalize difficult because of age and

socialization differences. Therefore, it would be

advantageous to collect similar data from professionals. In

the current study, the Abner (1992) vignettes were










administered to teachers, family lawyers, social workers,

journalists, child protection team workers, psychologists,

nurses, nursing assistants, and pediatricians. Since the

same questionnaire was used in the Abner (1992) study of

students and the current study of professionals, the

similarities and differences between a student population

and a professional population could be compared.

Since previous research has illustrated the importance

of the rater's demographic characteristics on maltreatment

reporting and substantiation rate, in the current study,

the effects of rater demographic characteristics were also

considered. These included the respondent's age, sex,

parenting experience, previous training in child

maltreatment, previous experience with suspected child

abuse or neglect, and previous history of reporting child

abuse or neglect.

Primary Hypotheses.

Main effects for severity of physical consequence and

maltreatment type as well as a significant interaction

between these variables were expected. More specific

hypotheses include: (1) physical abuse will be rated more

serious than both forms of medical neglect when their

physical consequences are of the same severity; (2) medical

neglect will be rated as less serious than physical abuse

even when the physical consequence suffered in the medical

neglect vignette is more severe; (3) similar to the student










participants in the previous study, professionals will

rate medical neglect of chronically ill children as more

serious than medical neglect of acutely ill children.

When comparing the student population to the

professional population, professionals were expected to

give more varied ratings than students because they have

had more experience with child maltreatment. The greater

variance may result in professionals rating the vignettes

less seriously.

Secondary Hypotheses.

A variety of respondent variables that have been found

to be influential in child maltreatment seriousness

perceptions were examined in this study: the respondent's

age, sex, parenting experience, training in child

maltreatment identification and reporting, experience with

suspected child abuse or neglect, and history of reporting

child maltreatment. Also the effect of the maltreated

child's sex was examined. Older subjects born before 1945

have rated maltreatment vignettes less seriously than

younger adults (Kean and Dikes, 1991). In some studies,

women have rated maltreatment vignettes more seriously than

men (Ringwalt & Caye, 1989; Kean & Dikes, 1991), but in

other studies this has not proved to be the case (Abner,

1992). Parent have stated that they would be more likely

to report than non-parents (Nightingale & Walker, 1986).

Training in child maltreatment identification and










reporting, experience with suspected child abuse or

neglect, and history of reporting child maltreatment, all

have resulted in higher ratings of seriousness (McIntyre,

1987; Nightingale and Walker, 1986).

In some studies, maltreatment of female children has

been rated more seriously than maltreatment of males (Kean

& Dikes 1991), but other studies have not found this to

be the case (Abner, 1992).

Current Study's Contribution to Child Maltreatment Research

The current study represents an addition to the

research reviewed in several ways. Only three previously

published studies have compared the effect of maltreatment

type upon seriousness perceptions (Nightingale'& Walker,

1986; Tjaden & Thoennes, 1992; and Ards & Harrell, 1993)

and no other study has used vignettes to compare medical

neglect with physical abuse. Likewise, no other study has

compared ratings of medical neglect of chronic versus acute

illness.

Although severity of consequences has been considered

in several studies (Turbett and O'Toole 1983; Alter, 1985,

Tjaden and Thoennes, 1992; Winefield & Bradley, 1992) it

has not been objectively defined or reliably quantified.

The current study controlled for severity of consequence by

obtaining objective, reliable ratings of each physical

consequence, independent of the type of maltreatment. This





37


reliable quantification of consequence severity represents

a significant contribution to the literature.

Only Tjaden and Thoennes (1992) have researched the

effect of both maltreatment type and severity of

consequence in one study. However, in their study,

severity of consequence was not objectively defined or

reliably quantified and the type of maltreatment did not

include medical neglect.














CHAPTER 2

METHODS

Measure

The vignette measure assembled by Abner (1992) was

used in this study. Abner (1992) constructed vignettes

describing one of three types of child maltreatment

(medical neglect of an acutely ill child, medical neglect

of a chronically ill child, and physical abuse) as well as

one of three levels of physical consequence (mild,

moderate, or severe injury or illness) that resulted from

the maltreatment. Through two pilot studies, Abner

obtained empirical estimates of the severity of various

forms of physical consequences independent of maltreatment

contexts. These independent estimates were necessary in

order to quantify physical consequence severity into three

categories: mild, moderate, and severe. The final

questionnaire consisted of 18 vignettes, with two vignettes

representing each cell in a 3 (Maltreatment Type) X 3

(Severity of Physical Consequence) design. Each vignette

was followed by a nine point Likert scale where the reader

rated the seriousness of the incident. A detailed account

of the pilot studies and questionnaire development











excerpted from the author's previous study (Abner, 1992)

can be found in Appendix A.

Four different forms of the questionnaire'were

created, using two different random order vignette

presentations describing either a male or female child.

Questionnaire Revisions

There were some minor modifications made to the

original questionnaire for the purposes of the present

study. Previous directions presented orally were included

as written directions. In order to examine the effects of

participant demographic data, requests for additional

information were added: the participant's occupation and

age, the participant's parenting experience, past training

in child maltreatment, experience in working with a child

suspected of being abused or neglected, and history of

reporting a child as being neglected or abused. A copy of

the modified questionnaire can be found in Appendix B.

Subjects

The subjects for this study were professional

counterparts of the student population collected in the

previous sample: doctors (pediatricians or family

practice), nurses (pediatric), journalists, social

workers, and elementary school teachers. Three additional

groups were added who did not have student counterparts in

the previous study but who often have experience with child









40

maltreatment: psychologists, child protective service team

workers, and child protective investigators. Please see

Table 2 for subject characteristics.

Procedures

Questionnaires were distributed by the author to

participants by personal contact or through inter-office

mail at the participant's place of employment after

receiving approval from the employee's supervisor.

Participant groups who received the questionnaire via

inter-office mail included nurses, journalists, child

protective services workers, child protective

investigations, social workers, and elementary school

teachers. Psychologists, lawyers, and physicians were

contacted either personally or through contact with their

administrative assistant or receptionist. Participants

were informed either orally or in writing that this was a

study regarding perceptions of child maltreatment.

Participants either returned the questionnaire to a common

collection site or mailed the questionnaires to the

examiner via a self-addressed stamped envelope provided by

the author. All answers were confidential and no names

were used or requested. Locations of questionnaire

distribution and response rates are located in Table 2.












Table 2

Sample Characteristics By Participant's Occupation

_______ _Questionnaires____ __ Sex ___
Occupation Location Number Number Return Male Female Mean
Dis- Re- Rate Age
~__________ tribute turned ____
Child Protective HRS* Gainesville, 64 41 64% 8 31 40
Services Ocala, Jacksonville ______________ (21%) (79%)____
Child Protective HRS* Gainesville, 80 54 64% 22 32 42
Investigations Ocala, Jacksonville ____ ______ (41%) (59%)
Journalist Gainesville 150 53 35% 31 21 39
Sun/Florida Times (60%) (40%)
Union __ ____ _____
Lawyers Private Offices 125 40 30% 23 14 42
Gainesville, Ocala, (62%) (38%)
Jacksonville, Tampa ________ _______
Physicians Shands Hospital 65 42 64% 31 8 38
Private offices _______ ____(79%) (29%)____
Nurses Shands Teaching 136 40 29% 0 36 37
Hospital pediatric (0%) (100%)
floors ____ _______
Psychologists Private Offices, 101 41 40% 21 17 47
Ocala, Jacksonville, (55%) (45/%)
____________Tampa ______ ______ _____
Social Workers Mental Health 63 40 63% 8 32 41
Services/Shands (20%) (80%)
_____~_____~___Hospital ________ ____ ______
Elementary Alachua County 82 44 53% 3 38 40
School Public Schools (7%) (93%)
Teachers___ ________________ _______ ____ _____ ___

*HRS = Health and Rehabilitative Services


Participants were directed by the written instructions

to read the questionnaire containing the 18 maltreatment

vignettes, to imagine that the child described was a 7 year

old child (boy or girl depending on the questionnaire), and

to rate the seriousness of the incident described in each

vignette on a scale from 1 to 9 where 1 represents a rating








42

of low seriousness and a 9 represents a rating of most

serious. The participants were allowed to change the

answers on the questionnaire as they went along if they so

desired. Following the vignettes the participants were

directed to answer nine brief demographic questions.














CHAPTER 3

ANALYSIS AND RESULTS

Comparisons of Response Rate, Sex of Respondent and Age by
Occupation

Chi-square analysis indicated there were occupational

differences in the percentage of questionnaires returned

(2 = 71.49, p < .001) and the sex of respondents (X2 =

100.7, p < .001). Also, there were occupational difference

in the age of respondents ((F = 3.93, p < .001). Since the

occupational groups differed in sex and age, these

demographic variables were included in subsequent analyses.

Respondents Eliminated from the Study due to Lack of
Variability of Responses.

Respondents who used only one or two ranking points

out of nine ranks were excluded from the analysis due to

lack of response variability. This reduced the sample size

from 395 respondents to 356 respondents. A table of the

respondents eliminated by their occupational status is

located in table 3.

Chi square analyses indicate that elementary school

teachers had a significantly higher number of responses

eliminated than all other professions (X2= 9.90, p < .021).

Child protective investigators had a significantly lower









Table 3

Respondents Eliminated from the Study due to Lack of
Response Variability Classified by Occupation

Occupation Total number Questionnaires Percentage of
of Eliminated questionnaires
questionnaires from the eliminated from
analysis the analysis
Protective 54 0 0.0
Investigators
Physicians 42 1 2.4
Protective 41 2 4.9
Services Workers
Psychologists 41 2 4.9
Social Workers 40 3 7.5
Nurses 40 4 10.0
Lawyers 46 4 10.0
Journalists 53 6 11.3
Elementary 44 17 38.6
School Teachers
Total 395 39 9.8


percentage of their responses eliminated than social

workers (X2= 4.18, p < .043), nurses (X2 = 5.640, p<0.018),

lawyers (X2 = 5.640, p < 0.018), and journalists (X2 =

6.47, p < .012) All other professions exhibited no

significant difference in the percentage of responses that

were eliminated.

Reduction of 18 Vignettes Into Within Subject Variables

Since each cell in the 3 X 3 design (physical

consequence severity X maltreatment type) consisted of two

vignettes, the mean of the two vignettes served as the

dependent variables. Relevant means and standard deviations

are presented in Table 4.









Table 4

Within Subject Variable Means


Maltreatment Type
Consequence Medical Medical Neglect Physical Abuse
Severity Neglect Acute Chronic Illness
S Illness
Mean S.D. Mean S.D. Mean S.D.
Mild 6.00 1.89 6.28 1.54 6.59 1.61
Moderate 6.72 1.57 7.27 1.31 8.39" 0.84
Severe 7.97a 1.02 T8.03 1.15 8.400 0.98
a_
a = Items are not significantly different from each other.
= Items are not significantly different from each other.


Overview of Analysis

Initially, there was a preliminary examination of

vignette order effects on the within subject variables.

Following this analysis, a large model repeated measures

analysis of variance was run to examine the effects of the

primary between subject variables of interest (sex of

respondent, sex of maltreated child, occupation, and

vignette order), the within subject variables (maltreatment

type and physical consequence severity), as well as their

interactions. Non-significant variables and interactions

were then eliminated from the model. After the primary

model was established, each of the secondary variables

(parental status, age group, former training in child abuse

and neglect issues, contact with suspected child abuse or

neglect and history of reporting child abuse/neglect) were

added individually to the model. Only significant

interactions and main effects were retained. Due to the

large number of subjects, the level of significance was set









at .01. Each significant interaction and main effect was

examined by appropriate follow-up statistical tests.

Finally Abner's (1992), original study's student responses

were compared with the responses of their professional

counterparts.

Order Effects

A preliminary repeated measures analysis of variance

was conducted examining the effect of the between subject

variable order and the two within subject variables

(physical consequence severity and maltreatment type). The

analysis indicated that vignette order had a significant

effect (F = 5.11, p < .01). Therefore the order of

vignette presentations was retained as a between subject

variable in all subsequent analyses.

Repeated Measures Analysis of Variance Results

A large model repeated measures ANOVA was conducted to

examine the effects of the between subject variables (sex

of respondent, sex of maltreated child, occupation,

vignette order), the within subject variables (maltreatment

type and physical consequence severity), and their

interactions. Sex of child and sex of respondent had no

significant effects and are subsequently dropped from the

model.









Secondary variables (parental status, age group

defined as adults born before 1945 were compared to

adultsborn after 1945, former training in child abuse and

neglect issues, contact with suspected child abuse or

neglect and history of reporting child abuse or neglect)

were then added individually into the model. Four of the

five secondary variables, parental status, age group,

contact with suspected child abuse or neglect, and child

maltreatment training had no significant effects. However,

history of reporting child abuse or neglect significantly

interacted with the within subject variables and was

retained.

Thus, the final repeated measures analysis of variance

examined the effects of occupation, vignette order, history

of reporting child abuse and neglect, and the within

subject variables, maltreatment type and physical

consequence severity. Powerful main effects were present

for maltreatment type and physical consequence severity as

well as a significant interaction between maltreatment type

and physical consequence severity. There was a main effect

for occupation and vignette order. No other main effects

were present for any between subject variables. There

were eleven significant interactions. The model was





Age groupings were determined in a manner consistent with prior
research (Kean and Dikes, 1991). A correlational analysis found no
significant correlations between age and seriousness ratings.










further trimmed by eliminating nonsignificant interactions.

The final model is presented in Table 5.

Table 5

Repeated Measures Analysis: Final Model


Mean
Source DF Square F Value Pr > F
Between Subject Variables
Vignette Order (Order) 1 53.98 6.47 0.0114
Occupation 8 29.33 3.52 0.0007
Report History 1 14.95 1.79 0.1815
Order*Occupation 8 10.32 1.24 0.2761
Error 315 8.34

Within Subject Variables _____ ________
Severity of Physical. Consequence 2 629.04 566.11 0.0001
(Severity) _____ ________
Severity Order 2 23.24 20.92 0.0001
Severity Occupation 16 6.83 6.15 0.0001
Severity Report History 2 4.69 4.17 0.0268
Severity Order Occupation 16 3.78 3.41 0.0001
Error 630 1.11

Maltreatment Type (Maltreatment) 2 146.47 140.42 0.0001
Maltreatment*Order 2 8.04 7.71 0.0007
Maltreatment Occupation 16 3.22 3.09 0.0001
Maltreatment Report History 2 5.55 5.32 0.0063
Maltreatment Order Occupation 16 1.20 1.15 0.3047
Error 630 1.04

Severity*Maltreatment 4 32.94 61.73 0.0001
Severity*Maltreatment*Order 4 14.34 26.88 0.0001
Severity*Maltreatment* 32 3.36 6.30 0.0001
Occupation _____ ________
Severity*Maltreatment*Report 4 1.07 2.01 0.0936
History________________
Severity*Maltreatment*Order* 32 1.84 3.46 0.0001
Occupation___________ _________ _____
Error 1260 0.53


Note: N may vary from original number
data in Report History variable.


of respondents due to missing


Maltreatment Type and Physical Consequence Severity

The final repeated measures ANOVA revealed potent main

effects for the within subject variables maltreatment type









and physical consequence severity. Paired comparison T-

tests were used to explore the main effects. Across

severity types, vignettes with severe physical consequences

were perceived as significantly more serious than moderate

consequences (T = 20.21, p < .001)and mild consequences (T

= 27.60, p < .001). Vignettes with moderate consequences

were perceived as more serious than mild consequences (T =

24.15, p < .001). (Please see Table 6).

Regarding maltreatment types, medical neglect of

chronically ill children received significantly more

serious ratings than medical neglect of an acutely illchild

(T = 7.36, p < .001). Physical abuse received

significantly higher seriousness ratings than either

medical neglect of a chronically ill child (T = 12.94, p <

.001) or medical neglect of an acutely ill child (T =

17.02, p < .001). (Please see table 6.)

Table 6

Means for Main Effect: Severity of Physical Consequences

and Type of Maltreatment


Physical Consequence Severity
Mild Moderate Severe
Mean S.D. Mean S.D. Mean S.D.
6.28 1.44 7.47 1.03 8.14 0.88

Maltreatment Type
Medical Neglect of Medical Neglect of Physical Abuse
Acute Illness Chronic Illness
Mean S.D. Mean S.D. Mean S.D.
6.92 1.16 7.18 1.28 7.79 0.86









There was a significant interaction between the

physical consequence severity and maltreatment type.

Subsequent analyses indicated that the main effect for

maltreatment type was significant in both the mild and

moderate categories of consequence severity. Medical

neglect of an acutely ill child was rated as the least

serious followed by medical neglect of a chronically ill

child. Physical abuse was rated as the most serious. At

the severe physical consequence level, physical abuse was

rated more seriously than both medical neglect categories

but there was no significant difference between the two

types of medical neglect. The main effect for consequence

severity was consistent in both medical neglect categories

with severe consequences being rated as the most serious

followed in descending order by moderate consequences and

mild consequences. In the physical abuse category, mild

consequences severity was rated as the least serious but

there were no significant rating differences between

moderate and severe physical consequences (See Table 4).

As predicted, medical neglect of chronic or acute

illness with severe consequences was rated significantly

less seriously than physical abuse with moderate

consequences (T = 7.93, p < .0001); (T = 7.28, p < .0001,

respectively).










Occupational Effects

The final model revealed a significant between

subjects effect for occupation as well as significant

interactions between occupation and physical consequence

severity and occupation and maltreatment type. Duncan

multiple comparison results for main effect of occupation

can be found in Table 7. Means and standard deviations

presented by occupation are located in Table 8. Duncan

multiple range tests were used to explore these differences

and the results can be found in Tables 9, and 10

respectively. Overall, teachers and journalists tended to

rate child maltreatment more seriously than other

professions while psychologists, lawyers, and physicians

tended to have lower ratings. Exceptions to this trend

were protective investigators who had the lowest ratings on

mild physical consequence severity and physical abuse.

Table 7

Duncan Multiple Comparison Results for Main Effect of

Occupation

Occupation N Mean Duncan
Serious- Grouping
ness
Teachers 47 7.81 A
Journalists 27 7.69 A
Social Workers 36 7.47 A B
Nurses 39 7.38 C A B
Protective Services 33 7.37 C A B
Protective Investigators 53 7.11 C B
Physicians 40 7.05 C B
Psychologists 38 6.96 C
Lawyers 36 6.90 C

Note: Means with the same letter are not significantly different










Table 8


Means and Standard Deviations of Maltreatment


Type and


Physical Consequence Severity by Occupation

Occupation Maltreatment Type______
Medical Neglect Medical Neglect Physical Abuse
of Acute Illness of Chronic
Illness __________
Mean S.D. Mean S.D. Mean S.D.
HRS P.I. 6.99 1.26 7.35 0.85 7.76 0.76
HRS P.S. 6.67 1.27 7.18 1.05 7.46 0.94
Journalists 7.50 1.01 7.55 1.08 8.04 0.71
Lawyers 6.40 1.15 6.77 1.53 7.54 1.13
Physicians 6.66 0.97 6.68 1.32 7.80 0.96
Nurses 6.89 0.83 7.21 1.11 8.20 0.74
Psychologists 6.60 1.40 6.80 1.40 7.52 0.88
Social Workers 7.17 1.14 7.37 1.11 7.86 0.79
Teachers 7.07 1.40 7.90 1.20 8.43 0.78

Occupation Physical Consequence Severity
Mild Moderate Severe
Mean S.D. Mean S.D. Mean S.D.
HRS P.I. 6.06 1.40 7.53 0.93 8.52 0.54
HRS P.S. 5.67 1.40 7.29 1.07 8.33 0.73
Journalists 6.94 1.10 7.84 0.89 8.30 0.75
Lawyers 5.76 1.53 7.21 1.13 7.74 1.14
Physicians 6.47 1.32 7.07 0.96 7.60 0.97
Nurses 6.45 1.11 7.52 0.83 8.19 0.73
Psychologists 5.96 1.60 7.18 1.15 7.74 0.98
Social Workers 6.41 1.49 7.69 0.94 8.31 0.76
Teachers 7.07 1.40 7.90 1.20 8.44 0.78










Table 9

Duncan Multiple Rance Test for Differences in Seriousness

Ratings for Maltreatment Type across Occupation


Medical Neglect of Acute Illness


Occupation N Mean Duncan Grouping
Journalists 47 7.49 A
Teachers 27 7.40 A
Social Workers 36 7.18 A B
Protective Services 39 6.99 C A B
Nurses 33 6.89 C A B
Protective Investigators 53 6.67 C B
Physicians 40 6.66 C B
Psychologists 38 6.57 C B
Lawyers 36 6.40 C

Medical Neglect of a Chronic Illness


Occupation N Mean Duncan Grouping
Teachers 27 7.72 A
Journalists 47 7.55 A
Social Workers 36 7.37 A
Protective Services 39 7.35 A B
Nurses 33 7.21 C A B
Protective Investigators 53 7.17 C A B
Psychologists 38 6.79 C B
Lawyers 36 6.77 C
Physicians 40 6.67 C

Physical Abuse


Occupation N Mean Duncan Grouping
Teachers 27 8.27 A
Journalists 47 8.03 A B
Nurses 33 8.02 A B
Social Workers 36 7.86 C B
Physicians 40 7.79 C B
Protective Services 39 7.76 C B
Lawyers 36 7.54 C
Psychologists 38 7.51 C
Protective Investigators 53 7.45 C

Note: Means with the same letter are not significantly different










Table 10

Duncan Multiple Range Test for Differences in Seriousness


iirr@ frw Phv~i r~ 1 ~


~~vm-itv Arro~~ Oc~cunation


Mild Physical Consequence Severity


Occupation
Teachers
Journalists
Physicians
Nurses
Social Workers
Protective Services
Psychologists
Lawyers
Protective Investigators


Mean
7.07
6.93
6.47
6.46
6.41
6.06
5.96
5.76
5.68


Duncan Grouping
A
A
A B
A B
A B


Moderate Physical Consequence Severity


Occupation N Mean Duncan Grouping
Teachers 27 7.90 A
Journalists 47 7.84 A
Social Workers 36 7.69 A B
Protective Services 39 7.52 C A B
Nurses 33 7.50 C A B
Protective Investigators 53 7.30 C B
Lawyers 36 7.21 C B
Psychologists 38 7.18 C B
Physicians 40 7.07 C

Severe Physical Consequence Severity


Occupation N Mean Duncan Grouping
Protective Services 39 8.51 A
Teachers 27 8.43 A
Protective Investigators 53 8.32 A
Social Workers 36 8.30 A
Journalists 47 8.30 A
Nurses 33 8.17 A
Psychologists 38 7.74 B
Lawyers 36 7.74 B
Physicians 40 7.57 B

Note: Means with the same letter are not significantly different


a ,^^^y f nv^- Ph aing j.-^j s- em..-- e-.---- -e er t cros --n-"---- ---r-io -









History of Reporting Child Maltreatment

The final model revealed a significant interaction of

report history and maltreatment type. Participants who had

no history of reporting a child to protective services

ranked medical neglect of an acutely ill child (T(213.5) =

2.16, p < .05) and of a chronically ill child (T(223.3) =

2.63, p < .01) as significantly more serious than

participants who had previously reported child maltreatment

to authorities. These differences were not significant

within the physical abuse category but the pattern of

results remained the same(non-reporters gave higher ratings

than reporters). Relevant means are located in Table 11.


Table 11

Means for Report History Maltreatment Type

Medical Neglect of Medical Neglect of Physical Abuse


Chronic Illness*


Acute Illness


Report
History X SD n X SD n X SD n

No 7.17 1.30 113 7.35 1.19 111 7.88 0.84 113

Yes 6.79 1.28 227 7.07 1.14 223 7.74 0.87 224


.Denotes significant difference between those who did and did not have
a history of reporting child maltreatment


Vignette Order

The final model revealed a main effect for vignette

order as well as significant two and three way interactions

between vignette order, consequence severity, and

maltreatment type. Respondents who received questionnaire









order 1 (mean seriousness rating = 7.16, S.D. = 1.00)

provided lower ratings than respondents receiving

questionnaire order 2 (mean seriousness rating = 7.41, S.D.

= .98).

The vignette order*maltreatment type and vignette

order*consequence severity was further analyzed by using t-

tests across vignette order and paired comparison t-tests

across maltreatment type and across consequence severity.

The main effect for maltreatment was consistent across both

sets of questionnaires. However, the vignette order effect

was significant only for the medical neglect of acute

illness category, hence, the two way interaction. Means for

this interaction are located in table 12.

Table 12

Means of Vignette Order Type of Maltreatment Interaction

Medical Neglect of Medical Neglect of Physical Abuse
Acute Illness** Chronic Illness
Vignette
Order X SD n X SD n X SD n

1 6.72 1.33 177 7.07 1.13 177 7.73 0.87 176

2 7.11 1.20 184 7.28 1.14 180 7.85 0.83 182


Order differences are statistically different.


Similarly, while the main effect for consequence

severity was consistent across both sets of questionnaires,

difference between questionnaires was significant for only

two (mild and moderate) of three consequence severity









categories; hence the two-way interaction. Means for this

interaction are located in Table 13.

Table 13

Means of Vignette Order Consequence Severity Interaction

Mild Consequence Moderate Severe
Severity** Consequence Consequence
Severity** Severity
Vignette
Order X SD n X SD n X SD n

1 6.00 1.49 175 7.37 1.02 175 8.16 0.79 176

2 6.56 1.32 182 7.56 1.04 183 8.11 0.97 181


Order differences are statistically different.


The vignette order*consequence severity*maltreatment

type was examined using t-tests for between questionnaire

order comparisons and repeated measure ANOVA's for between

maltreatment X severity comparisons. Effects were

consistent across vignette order except that at the mild

level there was no significant difference between medical

neglect of a chronically ill child and physical abuse in

questionnaire 1 whereas this difference was significant for

questionnaire 2. There was also no significant difference

between medical neglect of chronic and acute illness

vignettes for mild consequence severity in questionnaire 2

whereas this difference was significant for questionnaire

1. At the severe physical consequence level, questionnaire

2 respondents had no significant difference between

physical abuse and medical neglect of acute illness, while










these were significantly different for the questionnaire 1

respondents. Means for these interactions are contained in

Table 14.

Table 14

Means for Vicnette order Consequence Severity *

Maltreatment Type


Consequence Maltreatment Vignette Order 1 Vignette Order
Severity Type 2
Mean S.D. Mean S.D.
Mild Med. Neg. Acute 5.56 1.95 6.43c 1.74
Mild Med. Neg. Chronic 6.21b 1.51 6.35ac 1.58
Mild Physical Abuse 6.22b 1.65 6.95 1.48
Moderate Med. Neg. Acute 6.74d 1.48 6.70d 1.65
Moderate Med. Neg Chronic 6.98 1.35 7.56 1.20
Moderate Physical Abuse 8.39e 0.80 8.38e 0.88
Severe Med. Neg Acute 7.86 1.07 8.199 0.94
Severe Med. Neg. Chronic 8.01f 1.03 7.92' 1.25
Severe Physical Abuse 8.60 0.63 18.20 1.20


a Indicates
b Indicates
c Indicates
d Indicates
e Indicates
f Indicates
9 Indicates


that cells are not significantly different
that cells are not significantly different
that cells are not significantly different
that cells are not significantly different
that cells are not significantly different
that cells are not significantly different
that cells are not significantly different


Other Interactions


The final model repeated measures ANOVA included

several other complex three and four way interactions.

However, none proved to be meaningful and will not be

discussed further.

Comparison of Student and Professional Populations

An initial chi-square analysis was performed comparing

the number of questionnaires eliminated from the student

population with the number of questionnaires eliminated









from the professional population. Students had a

significantly larger number of questionnaires eliminated

from the study (x2(1) = 4.379, p < .05). However, a second

chi-square analysis was conducted that was restricted to

those occupations for which there was a student

counterpart. In this analysis, there was no difference

between students and professionals in the number of

questionnaires eliminated from the study (x2 =.005, n.s.).

Please see table 15 for the comparison of questionnaires

eliminated by professional status after the three

occupations without student counterparts were removed.

Table 15

Comparison of Percentage of Questionnaires Retained or

Eliminated by Professional Status



Questionnaires Retained or Eliminated
Professional % Retained % Eliminated
Status

Student 86.31% (n = 416) 86.49% (n = 224)

Professional 13.69% (n =66) 13.51% (n= 35)


Student participant responses collected in the

author's former study (Abner, 1992) were then compared to

their matching professional counterparts respondents

collected in the current study. Six of the occupations

sampled had student counterparts: lawyers, teachers,

physicians, nurses, social workers, and journalists. The









data bases from the two studies were combined and a new

class variable "Professional" was designated with students

receiving a "No" Professional classification and the

professional counterparts receiving a "Yes". The

professional variable was then entered into the main model.

Report history was eliminated from the model as data on

this variable was not available from the student sample

(Abner, 1992). There was a significant three way

interaction involving consequence severity, maltreatment

type and professional (F= 2.71, P <0.001). There was no

main effect for professional and no other interaction

present. Means are located in table 16

The three way interaction occurred as students ranked

medical neglect (acute illness) with moderate physical

consequences(T(396.1) = 1.98 p < .05) and physical abuse

with severe physical consequences (T(311.4) = 3.18 p < .05)

significantly more seriously than professionals.

Professionals ranked physical abuse with moderate physical

consequences more seriously than students (T(481.9) = 2.20

p < .05).






61




Table 16

Table Comparing Means of Professionals and Student Across

Physical Consequence Severity and Maltreatment Type.


Mild Moderate Consequence Severe Consequence
Consequence Severity Severity Severity
Pro Student Pro Student Pro Student

X SD X SD X SD X SD X SD X SD
MN 6.3 1.8 6.2 1.7 6.7 1.6 7.0 1.4 8.0 1.0 8.0 1.1
Acu ___ _____ ___ ___ _____ _____ ____ ____
MN 6.3 1.5 6.5 1.4 7.3 1.2 7.3 1.3 7.8 1.2 8.0 1.1
Chr _________________________ ________________
Phy 6.9 1.5 6.7 1.6 8.4 0.7 8.0 0.8 8.4 1.0 8.6 0.6
Abu_______________________ ____ ___________

Physical Consequence Severity Means: Totals

Mild Moderate Severe
Consequence Consequence Consequence
Severity Severity Severity

Group X SD n .SD n ..SD n

Professionals 6.79 1.49 254 7.70 1.06 255 8.18 .90 255

Students 6.82 1.44 480 7.75 1.10 482 8.31 .84 482


Maltreatment Type: Totals

Medical Neglect Medical Neglect of Physical Abuse
of Acute Illness Chronic Illness

R SD n
......up.......... .S .............n......

Professionals 7.23 1.34 257 7.41 1.23 253 8.04 .84 256

Students 7.34 1.34 482 7.51 1.23 480 8.03 .91 481














CHAPTER 4

DISCUSSION

This study's purpose was to evaluate the effect of

physical consequence severity and maltreatment type upon

professional's perceptions of seriousness of child

maltreatment. This chapter discusses the hypothesis

examined in the study along with speculation concerning the

results.

Effects of Physical Consequence Severity and Maltreatment

Type

Physical consequence severity had powerful effects

upon perceptions of seriousness of child maltreatment.

Overall, vignettes with mild physical consequences were

perceived less seriously than vignettes with moderate

consequences. Moderate consequence severity vignettes were

perceived less seriously than vignettes with severe

consequences.

Likewise, there were powerful effects for maltreatment

type. Physical abuse was rated significantly more

seriously than medical neglect of a chronic illness

followed by medical neglect of an acutely illness.

The powerful effects of physical consequence severity

and maltreatment type in rating maltreatment vignettes









suggest that both play a significant role in how

professionals determine the seriousness of a maltreatment

incident.

An interesting interaction occurred between physical

consequence severity and maltreatment type. The severity

effect was consistent across both medical neglect

categories (mild < moderate < severe). However, there was

no significant difference in the perception of physical

abuse with moderate consequences and physical abuse with

severe consequences. Both instances received extremely

high ratings. It is possible that participants viewed them

so extremely that there was not enough upper end

measurement on the scale to provide discrimination. It is

also possible that the maltreatment effect overwhelmed the

consequence severity effect at this level. Thus, physical

abuse resulting in anything more than mild physical

consequences is perceived as the most serious occurrence of

maltreatment.

Across all severity levels, participants rated

physical abuse more seriously than both forms of medical

neglect. The hypothesis that the effect of maltreatment

type would be so potent that subjects will rate medical

neglect less serious even when the physical consequence of

the abuse is less severe was partially supported by the

data. Physical abuse with moderate consequences was rated

more seriously than both forms of medical neglect with









severe consequences. However, physical abuse with mild

consequences was not perceived more seriously than medical

neglect with moderate consequences. As noted in the

author's previous study (Abner, 1992), one of the mild

physical consequences used in the physical maltreatment

category, had been rated significantly less seriously than

the other mild consequences. Therefore, the maltreatment

effect may not have been able to overwhelm the severity

effect at the mild consequence level.

As predicted, both types of medical neglect were seen

as less serious than physical abuse across every level of

consequence severity. This effect was consistent not only

in the present study, but in the author's former study

(1992). Such replicated findings strongly indicate that

physical abuse is perceived more seriously than medical

neglect when the physical consequences are equal or, in

some instances, when the consequences of medical neglect

are greater.

This maltreatment effect may have serious

ramifications. Significant neglect instances may be

ignored or treated with less urgency and may be partially

responsible for lower reporting (Mcpherson and Garcia,

1983) and substantiation rates (Kim, 1985) of neglect.

This effect could also be somewhat responsible for the

rather limited amount of research conducted about child

neglect (Wolock and Horowitz, 1984).









There are several possible reasons for this

maltreatment effect. Alter (1985) noted that whether or

not a parent's behavior was perceived as intentional had an

effect upon perceptions of abuse. Intentional behavior was

associated with an increased tendency to report

maltreatment. It is possible that the participants viewed

physical abuse as being a more intentional act than

neglect. Physical abuse may be seen as an act of

commission whereas neglect may be viewed as an act of

omission. It is interesting to note that while neglect may

be seen as an unintentional act, neglect tends to be more

chronic in nature. It is unclear whether chronic neglect

is viewed as more intentional. The vignettes used in this

study described single events and as a consequence did not

capture examples of chronic neglect.

Polansky (1984) noted that the dramatic nature of

abuse may result in increased attention for the phenomenon.

Physical abuse may be more dramatic than ambiguous episodes

of child neglect and thus may arouse more societal and

research attention.

Nightingale and Walker (1986) noted that conceptual

ambiguity may give neglect a low public priority. People

may simply have a difficult time defining what constitutes

neglect. Medical neglect may further this conceptual

ambivalence as it may generate questions concerning the

parent's adequate knowledge of medical care. However, in









the current study, the vignettes were not ambivalent; both

types of maltreatment were succinctly described and

severity of physical consequence was controlled.

Therefore, conceptual ambiguity can not completely explain

the maltreatment effect.

Wolock and Horowitz (1984) speculated that the high

correlation between poverty and neglect may be a factor in

the prejudicial treatment of neglect. It is possible that

poverty is a perceived cause of neglect. Neglect is

therefore excused because the offender is not perceived as

having adequate resources to meet the needs of a child.

This may be particularly relevant to medical neglect, as

the poor may be seen as unable to afford appropriate health

care for their children. Poverty may also be perceived to

be associated with ignorance of appropriate medical care.

Medical neglect may be "excused" in part because it is

viewed as an unintentional consequence of ignorance and

inadequate resources. Once again, however, it is

significant to note that poverty was not a variable in this

study and yet the maltreatment effect still occurred.

Therefore, poverty alone can not account for the

maltreatment effect.

Medical Neglect of Chronic Vs. Acute Illness

As hypothesized, the results of this study were

consistent with the author's previous study (Abner 1992) in

that participants rated medical neglect of chronically ill









children as more serious than medical neglect of acutely

ill children. This occurred at the mild and moderate

severity level. At the severe physical consequence level,

there was no significant difference between both forms of

medical neglect.

Several causes for this finding may exist.

Chronically ill children may be seen as more deserving of

sympathy and in need of medical care. Neglect of chronic

illness may also be perceived as more dramatic than the

mundane nature of day to day acute illness. Participants

may assume that parents of chronically ill children may be

more educated regarding their children's illness. Thus,

neglect of chronically ill children may be seen as more

intentional while neglect of acute illness may be perceived

as a result of ignorance. Finally, chronically ill

children may be perceived as at greater risk for serious

injury or complications.

Sex of Child

Similar to the findings of the author's previous study

the sex of child had no significant effect upon seriousness

perceptions. Although the child's gender had effects in

one study which controlled for the child's gender (Keans &

Dikes, 1991), their study used three large and detailed

case studies rather than a series of multiple two sentence

vignettes. The additional detail may have elicited more

sympathy for the female child. Also, perhaps as our









society has become more educated regarding issues of child

maltreatment, the abuse and neglect of either gender is

being perceived more similarly.

Age and Sex of Respondents

There was no significant effect of age of participants

and seriousness ratings, unlike the Kean and Dikes (1991)

study. Although some studies have reported females rate

maltreatment more seriously than males (Kean & Dikes, 1991)

in Abner's (1992) previous study and in the current

investigation, there was no significant difference in male

and female perceptions of seriousness. Thus, despite

occupational differences in gender and age of respondents,

these factors had no significant impact upon seriousness

ratings.

Vignette Order

Respondents to questionnaires with vignette order 1

provided generally lower ratings than respondents to

questionnaire order 2. In a number of cases the difference

between questionnaire responses was significantly different

(e.g. vignettes with mild physical consequence severity,

vignettes with moderate consequence severity; vignettes

with medical neglect of acute illness).

This order effect seemed to have occurred because

questionnaire 1 had more severe physical consequence

severity items randomly placed at the beginning of the

questionnaire whereas questionnaire 2 had more mild









physical consequence severity items randomly placed at the

beginning of the questionnaire. Items at the beginning of

the questionnaire appeared to establish an "anchor"

determining how the remaining items would be rated.

Respondents to questionnaire 2 selected more severe ratings

as they were introduced to more severe consequences whereas

respondents to questionnaire 1 provided less serious

ratings as the were introduced to less severe items. In

addition, on both questionnaires, items placed at the end

of the questionnaire tended to be rated less seriously.

Although it would be presumptuous to over analyze such a

finding as anything more than a measurement effect, it

could be argued that there was some degree of

desensitization to maltreatment that was occurring in the

questionnaire itself. As respondents were exposed to more

episodes of maltreatment, their seriousness ratings tended

to decrease.

Occupation Effects

Participant's occupational status significantly

affected their responses, interacting with physical

consequence severity and maltreatment type. Teachers and

journalists tended to rate the vignettes significantly more

seriously than physicians, lawyers, and psychologists.

Teachers and journalists also had more questionnaires

eliminated due to rating all of the vignettes with either

an 8 or 9 rating, a finding which confirms their tendency









to rank vignettes more seriously. Nurses, social workers,

protective services and protective investigation tended to

ranked the vignettes as less serious than journalists and

teachers, but as more serious than psychologists, lawyers

and physicians.

As in the author's previous study (Abner, 1992),

people with more education perceived maltreatment less

seriously than other participants. It may be that these

professionals give more variable responses to the measure,

therefore reducing their mean seriousness perception

rating. Also, the quantity of the education may not be as

important as the model or pattern of the education itself.

Psychologists, physicians and lawyers may receive more

training in differential classification methods, possibly

resulting in a more variable response pattern.

Different professions attract may also attract

different personalities which may in turn impact

perceptions. Lawyers, psychologists, and physicians may

approach the situation with a certain clinical detachment

whereas teachers may react more emotionally to the plight

of the child.

Report History

The participant's history of reporting child

maltreatment appeared to influence their seriousness

perceptions. However, the direction of the relationship

was opposite to that previously reported (Nightingale &









Walker 1986). Across maltreatment types, participants who

had previously reported child maltreatment rated medical

neglect of acute illness less seriously than non-reporters.

Since journalists and teachers gave the highest serious

rankings and were among the occupational groups with the

lower reporting rates, there appears to be some

correspondence between occupation and history or reporting

child maltreatment. There may also be some degree of

desensitization that occurs when professionals have

frequent contact with child maltreatment. This experience

may lessen the shock value of the vignettes and thus the

maltreatment is not rated as seriously. Anecdotally,

during data collection, elementary school teachers were

quite disturbed by the questionnaire and several

journalists wrote comments on their questionnaires

regarding the "gruesome" or "horrifying nature of the

vignette." Meanwhile, other professionals seemed to take

the questionnaire in stride. Psychologists tended to

critique the measure and one lawyer sent a letter detailing

legal ramifications of several vignettes.

Parenting History. Training, and Experience with Maltreated
Children

Contrary to prior research (Nightingale & Walker 1986)

neither training in child maltreatment or experience with

maltreated children resulted in higher ratings of

seriousness. Several possibilities exist for the failure









of this study to find an association. Both variables were

only summarily dealt by yes/no questions whereas

Nightingale and Walker (1986) used a more detailed,

quantitative measure. Therefore, the study's measure may

not have been sensitive enough to discriminate any

differences. This limited sensitivity may also explain why

there was a significant effect for report history and not

for experience with suspected abuse or neglect. Another

possibility is that the powerful main effects of

consequence severity and maltreatment type overwhelmed less

potent participant and training effects.

Although in previous research, parenting experience

has been associated with a higher likelihood of reporting

child maltreatment, it had no effect upon seriousness

ratings. Perhaps parents may see maltreatment similarly to

non-parents, but they may be more likely to report this

maltreatment due to an enhanced sense of empathy or

responsibility.

Student vs. Professionals

The hypothesis that professionals would give more

variable responses resulting in lower mean serious ratings

was not supported by the data. When the three professional

groups that had no student counterparts were removed from

the comparison, the percentage of questionnaires removed

due to lack of variability was virtually identical in the

professional and student groups. Also, although there was








a nonsignificant trend for professionals to rate

maltreatment less seriously than students, there were no

significant main effects or two way interactions for

professional status. The three way interaction of

professional status, physical consequence severity,

maltreatment type was due to a handful of mixed findings

that do not form a consistent pattern supportive of the

study's hypothesis that professionals would rank

maltreatment less seriously.

Most striking was the remarkable similarity between

the professional and student groups. Both maltreatment

type and consequence severity had potent effects with each

group of respondents, who responded in the same manner.

Indeed, both groups of respondents perceived moderate

physical abuse more seriously than both forms of medical

neglect with severe consequences, suggesting the finding is

highly reliable.

A considerable amount of psychological research uses

college students as an analog population. The use of such

respondents have frequently been criticized for its lack of

applicability to a more general audiences. However, in

this study there were no meaningful differences between the

student and professional populations in their ratings of

child maltreatment seriousness. This suggests that in

certain situations (i.e. when looking at populations that

have been college educated) results from college students









may be very predictive of other populations. Likewise, the

similarity in response pattern of students and their

professional counterparts suggest that there are either

personality characteristics or training experience which

account for this similarity.

Conclusions

Medical neglect generates less attention than physical

abuse in part because it is perceived less seriously.

Medical neglect that results in more severe physical

consequences is sometimes perceived less seriously than

physical abuse with less severe consequences. As noted by

several researchers, neglect is a more prevalent and more

deadly phenomenon. The diminished attention given to

neglect in general and particularly medical neglect may

hinder appropriate identification and treatment. Efforts

to raise professionals' awareness and knowledge of medical

neglect appear warranted.

Future Research Directions

It would be beneficial to discern if the findings

regarding medical neglect found in this study and in the

Abner (1992) study generalize to other forms of neglect.

Is medical neglect viewed as similar or different from

other forms of neglect? Likewise, it would be beneficial

to compare medical neglect with sexual abuse, emotional

abuse, and other forms of child maltreatment. However, it









may be difficult to control for physical consequence

severity with these other forms of child maltreatment.

Research into the reasons underlying the differences

in perception of medical neglect and physical abuse appear

useful. Parental intentions seems to be an important area

of inquiry. Would medical neglect, if perceived as equally

intentional as physical abuse, still result in lower

seriousness ratings? One factor that may be related to

intentionality is chronicity. Neglect of the chronically

ill was rated more seriously than neglect of the acutely

ill. However, each vignette was written as an isolated

incident. Further research into this area could consider

the effect of chronic child maltreatment rather than just a

single maltreatment episode.

Previous research (Kean & Dikes, 1991) found that

jurors would be more likely to report abusing fathers more

than abusing mothers. Sex of offending parent was not

considered in the Abner (1992) questionnaire or in the

current study and this could be an area for future

research.

The similarity in college students and professionals

responses to child maltreatment has implications for

psychological and sociological research as a whole. This

research suggests that college students can serve as a

useful analog population for further study of perception

and attitude toward child maltreatment.





76


Intervention research should assess whether education

is useful in raising awareness of medical neglect among

medical and non-medical professions. It is vital that

medical neglect and child neglect in general become a more

common focus of research and attention.















APPENDIX A

MEASUREMENT DEVELOPMENT1

Measurement Development

A questionnaire consisting of 18 two sentence

vignettes was constructed and administered. Each vignette

described a form of child maltreatment (physical abuse,

medical neglect, medical neglect of a chronically ill

child) and the physical injury or illness (physical

consequences) which resulted from the maltreatment. In

order to test the hypothesis that subjects will rate

medical neglect reports less serious than physical abuse

reports, despite the degree of physical damage involved, it

was necessary to first obtain empirical estimates of

severity of various forms of physical consequences

independent of the child maltreatment context. Two pilot

studies were conducted for this purpose.

The First Pilot Study

The first pilot study consisted of administering a 79

provided a brief description of a physical consequence such




'This information is excerpted directly from Abner, 1992; pgs 32-42 with minimal changes to direct the
reader to appropriate table and appendices in this text..










as a "broken leg," or "chronic headaches." The studentswere

asked to imagine that these physical consequences had

affected a seven year old boy or girl and were asked to rate

the items on a nine point Likert scale with 1-3 representing

mild seriousness, 4-6 representing moderate seriousness, and

7-9 representing severe seriousness. Four questionnaires

were handed out. Order of the items were randomized and two

different randomized vignette orders were constructed. Then

for each of the randomized orders half of the students were

asked to rate a boy affected with these physical

consequences, and the other half was asked to rate a girl

with these physical consequences. The questionnaires used

in this pilot study can be found in Appendix B.

The means and standard deviations of all the items were

calculated and arranged in descending order (Please see

Table Al). Two-way anova's were used to examine sex

effects. No significant sex effects were found.

The Second Pilot Study

The second pilot study consisted of administering a

questionnaire composed of 29 items to 17 pediatric residents

and 4 pediatric faculty members at Shands Hospital at the

University of Florida. Physicians were recruited to provide

a professional view of severity of various physical

consequences. The second pilot questionnaire consisted of

29 items from the original pilot measure. The pilot measure

was shortened to decrease repetitiveness, and to facilitate










Table Al


Result of the First Pilot Study: Students Rate Seriousness

of Physical Consequences (n=71)

Situation Mean S.D.

cardiac arrest (heart stoppage) 8.62 0.78
inability to breath/ 8.14 1.26
blocked air pathways
permanent damage to lungs 7.96 1.33
*diabetic ketoacidosis 7.87 1.71
*severe retardation 7.7 1.53
inability to breathe due to 7.56 1.34
blocked pathways
blindness 7.54 1.66
*crippled (can not walk) 7.41 1.53
*convulsions 7.39 1.41
*kidney damage 7.14 1.48
*severe concussion 7.11 1.55
inability to breath/ 7.04 2.02
blocked air paths (5 min)
*2nd degree burns (25%) 7.03 1.72
inability to breath/ 7.02 1.99
blocked air paths (2 min)
deafness 6.94 1.87
rabies 6.91 1.7
diabetic hyperglycemia 6.77 1.49
crippled 6.72 1.72
labored breathing (12 Hours) 6.7 1.94
diabetic hypoglycemia 6.65 1.84
*repeated attacks of breathing 6.49 1.57
difficulty
tetanus 6.41 1.74
2nd degree burns (10%) 6.31 1.79
crippled (can walk with crutches) 6.1 1.71
moderate retardation 6.07 1.86
labored breathing (3 hours) 5.96 1.79
severe deformity of knees 5.91 1.76
*chronic abdominal pain 5.91 1.33
*crippled (can walk unaided) 5.87 1.79
partial loss of vision 5.87 1.56
*damage to the inner ear 5.85 1.59
*chronic pain in joints 5.8 1.6
severe abdominal pain 5.79 1.63
pneumonia 5.73 1.68
*concussion 5.7 2.06
chronic pain in limbs 5.7 1.67
severe pain in limbs 5.65 1.63
mild retardation 5.63 1.97










Table Al (continued)

Situation Mean S.D.
severe deformity of the elbow 5.55 1.78
partial loss of hearing (30%) 5.55 1.8
*second degree burns 5.55 1.9
*chronic headaches 5.48 1.62
loss of strength in hand 5.48 1.77
severe pain in joints 5.48 1.52
labored breathing 5.45 1.77
second degree burns (3%) 5.34 2.03
partial loss of vision (15%) 5.2 1.96
labored breathing (15 minutes) 5.04 1.89
deformity of a hand 5.01 2.03
partial loss of hearing (15%) 4.91 1.71
crippled hand 4.87 1.90
*dehydration 4.87 2.13
mild deformity of knee 4.83 1.51
stunted growth 4.79 2.13
abdominal pain 4.75 1.95
repeated wheezing 4.73 1.97
*measles 4.71 2.00
*bruises on multiple body locations 4.69 2.14
extreme weight loss 4.54 2.03
*broken leg 4.51 1.93
mild concussion 4.48 1.9
*pain in joints 4.41 1.71
*severe headaches 4.40 1.77
pain in limbs 4.4 1.8
mild deformity of the elbow 4.37 2.05
chronic colds 4.27 1.78
*broken nose 4.2 2.02
frequent urination 4.19 2.04
ear infection 4.01 1.98
*broken arm 3.8 1.68
*chicken pox 3.77 1.96
*throat infection 3.7 1.94
red marks on the child's skin 3.43 1.96
tooth knocked out 3.32 2.05
flu 3.17 1.67
*broken tooth 3.14 1.88
*facial injuries (black eye; cut lip) 3.13 1.86
*repeated sneezing attacks 3.11 1.48
toothache 2.96 1.92
*bruises in one location 2.23 1.72

* = selected for inclusion in second pilot study









ease of completion. Items were selected from the first

pilot study based on low variability, ease of including

physical consequence represented into a maltreatment

vignette, and equal representation of mild, moderate, and

severe rankings from the first pilot study. The

physicianswere asked to imagine that the following physical

consequences have affected a seven year old boy and then to

rate the seriousness on a scale from 1-9, with 1

representing the least serious and 9 representing the most

serious. A copy of this questionnaire can be found in

Appendix C. The means and standard deviations of the

responses are arranged in descending order of seriousness in

Table A2.

Construction of Vignettes.

Results of both pilot studies were used to select 18

physical consequences of maltreatment to be used in the

final study. All items included in both pilot studies were

examined. The first pilot study was used to tentatively

define three conditions of severity of physical consequence:

mild, moderate, and severe. Severe was defined as

consequences with a mean between 7-9, moderate consequences

were defined as having a mean between 5-6, and mild

consequences were defined as having a mean between 2-4.

Items which fell between, 6-7 and 4-5 were eliminated from

consideration. Student ratings were then compared to

physician ratings in order to confirm that students'










Table A2

Results of the Second Pilot Study (n = 21): Physicians rate

seriousness of physical consequences

Situation Mean S.D.
Diabetic ketoacidosis 8.00 1.09
Kidney damage 7.62 1.02
Second degree burns (25%) 7.24 1.54
Convulsions 6.95 0.8
Crippled (can not walk) 6.9 1.48
Severe concussion 6.86 1.82
Severe headache 6.57 1.69
Severe retardation 6.38 2.01
Multiple bruises 6.33 1.24
Damage to the inner ear 6.28 1.3
Repeated attacks: breathing difficulty 6.15 1.35
Dehydration 5.9 1.99
Chronic pain in joints 5.71 1.34
Second degree burns 5.48 1.8
Crippled (can walk unaided) 5.25 1.33
Chronic headaches 5.24 1.37
Pain in joints 4.9 1.44
Concussion 4.9 1.95
Broken Leg 4.71 1.1
Measles 4.67 1.46
Facial injuries (black eye, cut lip) 4.67 1.86
Broken arm 4.57 1.03
Chronic abdominal pain 4.52 1.36
Throat infection 3.76 1.57
Broken nose 3.67 1.35
Broken tooth 3.28 1.41
Chicken pox 3.14 1.23
Bruises in one location 2.8 1.33
Repeated sneezing attacks 2.76 1.04

perceptions were similar to professional opinion. When

student-physician ratings differed by more than .8, the item

was dropped from further consideration. Twenty-one items

remained. Three items were excluded because it would be

difficult to include them in realistic maltreatment

vignettes. The final eighteen items included six items from

each severity condition (mild, moderate, severe). There was










less than a .5 discrepancy between students and physicians

for 13 of the 18 items (see Table A3).

Table A3

Students' and Physicians' Mean Seriousness Ratings of

Physical Consequences Used to Construct Vignettes


Stud
Severe Physical Mean
Consequences
Diabetic Ketoacidosis 7.87

Crippled (Can't walk) 7.41

Convulsions 7.39

Kidney Damage 7.14

Severe Concussion 7.11

Second Degree burns 7.02
(25%)

Moderate Physical Consequences
Crippled (can walk) 5.87

Damage in the inner ear 5.84

Chronic pain in joints 5.80

Concussion 5.70

Second degree burns 5.55

Chronic headaches 5.48

Mild Physical Consequences
Broken arm 3.8

Chicken pox 3.78

Throat infection 3.70

Broken tooth 3.14

Repeated sneezing att. 3.11

Bruises in one locale 2.22


cents Physicians
S.D. Mean S.D.

1.71 8.00 1.09

1.53 6.9 1.48

1.41 6.95 0.8

1.48 7.62 1.02

1.55 6.86 1.82

1.72 7.24 1.54


1.79

1.59

1.6

2.06

1.9

5.48


1.68

1.96

1.94

1.88

1.48

1.72


5.25

6.28

5.71

4.9

5.48

5.24


4.57

3.14

3.76

3.28

2.76

2.8


1.33

1.30

1.34

1.95

1.8

1.37


1.03

1.23

1.57

1.41

1.04

1.33









The Bonferroni method of multiple comparisons was

then used to test for differences within and between the

mild, moderate, and severe groups of each six conditions.

Data from only students' responses (the first pilot sample)

were used for this analysis because the respondents to the

final questionnaire were also students. The Bonferroni

method indicated that all physical consequences in the

severe group were not significantly different from each

other, and all physical consequences in the moderate group

were not significantly different from each other. In the

mild group, four of the six physical consequences were not

significantly different from each other. The exception was

"bruises in one location," which was rated significantly

less severe than three other physical consequences: "broken

arm," "chicken pox," and "throat infection." All physical

consequences in each group (severe, moderate, and mild) were

significantly different from all physical consequences in

the other groups. Therefore, it was concluded that grouping

of physical consequences into mild, moderate and severe

groups was appropriate. The only condition that did not fit

concisely within a category was the "bruises in one

location" item. Since this physical consequence was used in

a vignette describing physical abuse, it was decided that it

could remain within the mild group since it would provide a

conservative test of the study's hypotheses (i.e. if the

vignette was ranked higher than a mild medical neglect










vignette it would support the study's hypothesis. Please

see table A4 for the results of the Bonferroni T-Tests.).

Table A4

Results of the Bonferroni T-Tests


Note: Means with the same letter are
alpha = .05 level for the students'
Bonferroni grouping


Situation

Diabetic Ketoacidosis

Crippled (Can't walk)

Convulsions

Kidney Damage

Severe Concussion

Second Degree burns
(25%)

Crippled (can walk)

Damage in the inner ear

Chronic pain in joints

Concussion

Second degree burns

Chronic headaches

Broken arm

Chicken pox

Throat infection

Broken tooth

Repeated sneezing att.

Bruises in one locale


not significantly different at the


Students'
Mean S.D.

7.87 1.71

7.41 1.53

7.39 1.41

7.14 1.48

7.11 1.55

7.02 1.72


5.87

5.84

5.80

5.70

5.55

5.48

3.8

3.78

3.70

3.14

3.11

2.22


1.79

1.59

1.6

2.06

1.9

5.48

1.68

1.96

1.94

1.88

1.48

1.72










Eighteen two sentence vignettes were constructed.

Each described a form of child maltreatment and the physical

consequence of that maltreatment. Descriptions of child

maltreatment included several vignettes previously presented

in the literature (Giovannoi and Becerra, 1978) and other

vignettes constructed by the present author. Three types of

child maltreatment were included (physical abuse, medical

neglect, and medical neglect of a chronically ill child)

with three types of physical consequences (mild, moderate,

and severe). For each type of maltreatment, two vignettes

described mild consequences, two moderate consequence, and

two severe consequences.














APPENDIX B


QUESTIONNAIRE USED IN THE FIRST PILOT STUDY

Listed below are a variety of physical conditions,
injuries, or illness. Please imagine that each of these
conditions, injuries or illness have affected a seven year
old boy. Then rank the seriousness of each condition on a
scale from 1 to 9 with 1 representing the lowest severity
and 9 representing the highest severity. Please indicate
your rank by circling the appropriate number.

1. severe pain in joints.

1 2 3 4 5 6 7 8 9
1I---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

2. convulsions

1 2 3 4 5 6 7 8 9
S------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

3. second degree burns (covering 3% of child's body)

1 2 3 4 5 6 7 8 9
I ---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

4. damage to the inner ear

1 2 3 4 5 6 7 8 9
S---------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

5. second degree burns (covering 25% of child's body)

1 2 3 4 5 6 7 8 9
| -------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness









6. crippled (can walk unaided)

1 2 3 4 5 6 7 8 9
------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

7. second degree burns (covering 10% of child's body)

1 2 3 4 5 6 7 8 9
i -------------------------------I
Mild Moderate Severe
Seriousness Seriousness Seriousness

8. partial loss of vision (30%)

1 2 3 4 5 6 7 8 9

Mild Moderate Severe
Seriousness Seriousness Seriousness

9. severe abdominal pain

1 2 3 4 5 6 7 8 9
1I---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

10. pain in joints

1 2 3 4 5 6 7 8 9
1I---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

11. toothache

1 2 3 4 5 6 7 8 9
I ------------------------------I
Mild Moderate Severe
Seriousness Seriousness Seriousness










12. pneumonia

1 2 3 4 5 6 7 8 9
I ---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

13. repeated sneezing attacks

1 2 3 4 5 6 7 8 9
1I---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

14. mild concussion

1 2 3 4 5 6 7 8 9
1I---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

15. chronic pain in joints

1 2 3 4 5 6 7 8 9
-----------------------------------------I
Mild Moderate Severe
Seriousness Seriousness Seriousness

16. inability to breath due to blocked air pathways (5
minutes)

1 2 3 4 5 6 7 8 9
1 ---------------------------------------------I
Mild Moderate Severe
Seriousness Seriousness Seriousness

17. stunted growth

1 2 3 4 5 6 7 8 9
1I---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness










18. moderate retardation

1 2 3 4 5 6 7 8 9
I ------------------------------ I
Mild Moderate Severe
Seriousness Seriousness Seriousness

19. bruises on multiple body locations

1 2 3 4 5 6 7 8 9
1 ---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

20. crippled (can not walk)

1 2 3 4 5 6 7 8 9
1I---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

21. broken leg

1 2 3 4 5 6 7 8 9
I ---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

22. bruises on one location

1 2 3 4 5 6 7 8 9
-----------------------------------------I
Mild Moderate Severe
Seriousness Seriousness Seriousness

23. inability to breathe due to blocked passageways (5
minutes)

1 2 3 4 5 6 7 8 9
I ------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness









24. broken nose

1 2 3 4 5 6 7 8 9
1I---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

25. blindness

1 2 3 4 5 6 7 8 9
I -------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

26. partial loss of hearing (15%)

1 2 3 4 5 6 7 8 9
1I---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

27. deafness

1 2 3 4 5 6 7 8 9
I --------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

28. loss of strength in hands

1 2 3 4 5 6 7 8 9
S-----------------------------------I
Mild Moderate Severe
Seriousness Seriousness Seriousness

29. throat infection

1 2 3 4 5 6 7 8 9
------------------------------------ ----I
Mild Moderate Severe
Seriousness Seriousness Seriousness










30. flu

1 2 3 4 5 6 7 8 9
S-------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

31. tetanus (lockjaw)

1 2 3 4 5 6 7 8 9
I ---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

32. severe retardation

1 2 3 4 5 6 7 8 9
I ------------------------------I
Mild Moderate Severe
Seriousness Seriousness Seriousness

33. red marks on the child's skin

1 2 3 4 5 6 7 8 9
I -------------------------------I
Mild Moderate Severe
Seriousness Seriousness Seriousness

34. crippled (can walk with crutches)

1 2 3 4 5 6 7 8 9
1I---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness

35. inability to breath due to blocked air pathways (2
minutes)

1 2 3 4 5 6 7 8 9
1I---------------------------------------- I
Mild Moderate Severe
Seriousness Seriousness Seriousness