Review of Psychological Assessments
The Child Sexual Behavior Inventory (CSBI) is used in assessing child sexual behavior for children ages 2 to 12 years old. It consists of a 35 item behavior checklist and this particular assessment done by William N. Friedrich and Patricia Grambsch of the Mayo Clinic in Rochester, Minnesota entitled Child Sexual Behavior Inventory Normative and Clinical Comparisons contrasted two different samples. One being a normative sample of 880 children and the other being a sample of 276 sexually abused children. It was determined that the sample of sexually abused children displayed more sexual behavior at a greater frequency than did the normative sample (Friedrich, 1992).
Likewise, the Child Abuse Potential (CAP) Inventory was used by Joel S. Milner of Northern Illinois University in his assessment of child abuse as it relates to that child having medical conditions. 1151 parents took the CAP Inventory and the results determined that abuse rates were lower for those children that suffered from illness or injury at birth (Milner, 1991).
Both of these test models, the CSBI and CAP Inventory have been found to be effective for both studies in determining child sexual behaviors related to children that have been sexually abused (Friedrich, 1992) and child abuse and its relation to child born medical conditions (Milner, 1991).
In this paper I will determine the validity of these tests by looking at these criteria the tests history and development and theories behind them the administrating procedures and their effectiveness and the designated population and sampling information.
I will then provide my own critique of these tests in regards to their usefulness and what can be accomplished with them in the future. I will conclude by briefly comparing and contrasting the two tests.
Child Sexual Behavior Inventory (CSBI)
Test History and Development
The particular version of the Child Sexual Behavior Inventory used in this study was developed by William N. Friedrich and two others in 1986 (Purcell, Beilke, Friedrich, 1986). It was a test that was drawn from an earlier model entitled the Child Behavior Checklist (CBCL) authored by T. Achenbach and C. Edelbrock in 1983. The CBCL itself has since expanded to include children up to age 18 (CBCL4-18) with a separate model used for preschool aged children (CBCL2-3). CBCL4-18 is intended to be one component of a broad, multi-informant assessment of a childs behavioral adjustment (Doll).
Friedmans model originally contained 40 measurable items, however, after doing interviews with caregivers of sexually abused children, Friedrich decided to expand it to 48 items featuring additional factors such as the childs exposure to nudity in the family setting as well as bathroom habits in the household (Friedrich, 1992) the result was an instrument that yielded higher levels of both specificity and sensitivity and a higher overall accuracy in classifying abused versus nonabused cases (Bernt, 2001). It has been an ever-evolving model determined to produce more accurate and efficient results.
Test Administration Procedures
The CSBI can be administered with paper-and-pencil format or by interview. Friedrich suggests that it be done through the paper-and-pencil format so that a written evaluation is completed (Bernt, 2001). The test is written at a 5th grade level and takes less than 10 minutes to complete. The content is explicit given the subject matter and requires no official training to complete (Bernt, 2001). The test does, however, require that the evaluator have graduate training in the area of psychology particularly in the area of child sexual abuse (Bernt, 2001). Scoring is done in less than 5 minutes and is done by hand (Bernt, 2001).
Designated Population and SamplingNorming Information
The CSBI is geared toward three separate age groups (2-5, 6-9, and 10-12 years old) and both genders. Norms are derived from interview results from mothers or primary female caregivers (Bernt, 2001). Other caregivers in the equation, such as fathers, teachers, daycare providers and the like, were addressed, however, the main focus of the norm was on the female caregivers (Bernt, 2001).
It was found that the parent of the normative group, (versus the sampling group), were more educated and better off financially as well as the child was more apt to be living in a 2-parent household (Friedrich, 1992). Also it was found that the sample group was more ethnically diverse but that the non-White children in the sample did not differ from the White children as far as scores go (Friedrich, 1992).
Reliability and Validity
The resulting coefficient alpha of the 35-item CSBI was .82 for the normative sample and .93 for the clinical sample. These coefficients indicate appropriate reliability (Friedrich, 1992). Bernt found similar results with the test in his review. He found that the test reliability over a 2-week period was .91 and that over a 4-week period dropped to .85 (Bernt, 2001). He also determined that the CSBI
has been found to discriminate between sexually abused and nonabused preschoolers between sexually abused school-age children and children treated for emotionalpsychiatric problems with no such history between sexually aggressive (with no history of sexual abuse) and nonaggressive sexually abused children and between sexually aggressive and physically and nonaggressive children (Bernt, 2001).
Critique of Test, Its Usefulness and Future Recommendations
The CSBI is a very useful tool in determining whether child sexual abuse has occurred and is, in actuality, the best scale to use in that it utilizes supplemental information such as projective tests, child interviews and physical examination but it also avoids the pitfalls that go along with child interviews with the use of the written test (Bernt, 2001). The findings also provide support for the fact that sexual abuse rarely occurs in isolation but rather is associated with greater levels of familial distress and fewer educational and financial resources in the family (Friedrich, 1988).
The biggest issue with the validity of the test is that samples are drawn only from Minnesota and California. More sampling needs to be done around the country in order to get a truer representation of the general population.
Also there could be enormous consequences to follow the event of false positives leading to child sexual abuse (Bernt, 2001). Great care needs to be taken by evaluators and care givers to ensure that this does not take place.
The Child Abuse Potential (CAP) Inventory
Test History and Development
The CAP Inventory was created in 1986 by Joel S. Milner and was designed to screen individuals reported for physical child abuse in protective services settings, such as departments of social services, where abuse rates in reported cases range from 35 to 50 (Milner, 1991). It has been used in hospital settings by Suspected Child Abuse and Neglect (SCAN) teams to look at patients that are suspected of abuse and for also those that are in the at-risk category (Milner, 1991).
Because it was created in a hospital setting, it seems only fitting that it be tested there as well (Milner, Gold, Ayoub, Jacewitz, 1984 Milner, Gold Wimberly, 1986 Milner Wimberly, 1980).
The notion was also considered that those that scoring high on the CAP could be non abusing parents that experience a lot of stress in their lives (Grisso, 1986).
Test Administration Procedures
Over a period of 5 years, 1,151 parents were tested through a nonrandom selection process. Conveniently, they were all parents of children with medical problems or who had their own personal medical problems (Milner, 1991).
The actual CAP Inventory itself is a 160-question test given in an agree-disagree format (Milner, 1991) that contains six descriptive factor scales Distress, Rigidity, Unhappiness, Problems With Child and Self, Problems With Family, and Problems From Others (Milner, 1991). The current version (as of 1991) contains three validity scales a Lie scale, a Random Response scale, and an Inconsistency (Milner, 1991). These scales are then used to form three validity indexes the faking-good index, the faking-bad index, and the random-response index (Milner, 1991).
Those in the test are originally approached by hospital staff and asked if they want to fill out a questionnaire. If they are going to participate they are given an informed-consent form to fill out. The test is completed in the hospital room or waiting room and upon completion, hospital staff is available for debriefing (Milner, 1991).
Designated Population And SamplingNorming Information
The CAP Inventory is aimed at parents of children with medical conditions or parents who themselves have a medical condition. 1,151 parents at Hillcrest Medical Center in Tulsa, Oklahoma answered the questionnaire (Milner, 1991). The study contained 190 males and 961 females, all parents that either had a medical condition or had children with medical conditions (Milner, 1991). All subjects participated on a voluntary, anonymous basis (Milner, 1991).
Reliability and Validity
It was found that, ultimately, Abuse scale scores were a result of parent and child relations and not directly derived from medical conditions with the child or parent (Talbott, 1985). However, Milner in his findings found that medical problems could increase the stressors between parent and child and therefore, would have an effect on the relationship (Milner, 1991).
The questionnaires were evaluated by hospital staff, data was collected and scores and percentages were given.
One fallacy of the study was that it only looked at a limited range of factors. There are numerous other medical and non-medical conditions that could create stress and therefore skew the results of the CAP Inventory(Milner, 1991). Also, other combinations of stressors and stress-related conditions need to be considered (Bhagat, McQuaid, Lindholm, Segovis, 1985).
There also wasnt a normative group in the test. This would help to gain a better comparison versus focusing solely on those with existing medical conditions (Milner, 1991).
Critique of Test, Its Usefulness and Future Recommendations
The CAP Inventory was used to try to determine if a child or parents medical problems increased at-home stressors that would ultimately lead to child abuse andor neglect. The findings of the test revealed that there are many variables to consider. There was a limited scope in the range of potential medical stressors and abuse scores (Milner, 1991). The sample sizes were probably too small and there wasnt a normative in the study (Milner, 1991). Nothing of real true value was determined from this test as there was just too large of range of specificity rates (66.7 to 100) on the Abuse scale (Milner, 1991).
So the question of whether or not medical problems with either child or parent lead to higher rates of child abuse still remains. Future tests need to add a normative group and widen the range of factors to get a better idea of specificity on the Abuse scale. At this level, I just dont see the test as being very useful.
Comparisons Between the CSBI and CAP Inventory
The CSBI was clearly more valid in my eyes for the simple fact that there was a sample group and a normative group. The CSBI utilized information from Minnesota and California, whereas the CAP Inventory only looked at patients from one hospital. They both used questionnaires utilizing the paper-and-pencil method versus the interview method, which proves to be more effective in gathering data. Both tests are evaluated by professional personnel, however those evaluating the CSBI had some form of graduate training where those evaluating the CAP Inventory were hospital staff, including doctors, nurses and psychiatric staff.
Conclusion
Both of these tests could be useful tools in determining child sexual abuse and child abuse amongst those with medical conditions, however, I feel that the CSBI is far and away above the CAP Inventory in being able to derive information for their respective causes. The CSBI found more hard data and is actually a useful tool used to determine child sexual abuse. The CAP Inventory contains too many variables and not enough of a sample. It also needs to have a normative grouping. A lot of work still needs to be done in order for it to become a valid and useful tool in determining child abuse amongst those with medical conditions.
Categories:
0 comments:
Post a Comment