ACF OPRE: Child Welfare Systems' Responses to Children of Color

Database of Published Research

View citation for:
Albert, 1994	Eckenrode et al., 1988	National Black Child Development Institute, 1989
Ards, et al., 2001	Fanshel and Shinn, 1978	Needell and Barth, 1998
Ards, et al., 1998	Fein et al., 1990	Nelson et al., 1993
Ards and Harrell, 1993	Fein et al., 1983	Neuspiel et al., 1993
Barth et al., 2000	Finch et al., 1986	Olsen, 1982
Barth, 1997	Fluke et al., 1999	Plantz et al., 1989
Barth and Courtney, 1994	Garland et al., 1998	Runyan et al., 1982
Barth et al., 1994	Garland and Besinger, 1997	Saunders et al., 1993
Barth et al., 1986	Goerge at al., 1994	Schmidt-Tieszen and McDonald, 1998
Berrick et al., 1994	Goerge and Harden, 1993	Spearly and Lauderdale, 1983
Block and Libowitz, 1983	Goerge, 1990	Stenho, 1982
Cappelleri et al., 1993	Groeneveld and Giovanni, 1977	Terling, 1999
Close, 1983	Gurak et al., 1982	Trocme et al., 1994
Courtney and Wong, 1996	Hampton and Newberger, 1984	Wildfire, 2000
Courtney, 1995	Jones and McCurdy, 1992	Wulczyn, 2000
Courtney, 1994	Katz et al., 1986	Wulczyn et al., 1999
Courtney, 1994	Korbin et al., 1998	Wulczyn et al., 1997
Courtney, 1994	Lauderdale et al., 1980	Wulczyn, 1991
Downs, 1986	McCabe et al., 1999	Zellman, 1992
	McMurtry and Lie, 1992

Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Albert, 1994
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Albert, V. (1994). From child abuse report to child welfare services. In R. P. Barth, M. E. Courtney, J. D. Berrick, and Albert, V. From Child Abuse to Permanency Planning: Child Welfare Services Pathways and Placements (pp. 55-75). New York: Aldine de Gruyter.	Database: Social Service Reporting System (SSRS) which contains child abuse and neglect data from the California counties of San Diego, Santa Clara, and San Mateo. Sample: 26,506 children whose first referral occurred between January 1991 and September 1991 (Caucasian 50%, Hispanic 28%, African-American 14%, Asian >1%).	What are the characteristics of children who are referred for abuse and neglect? What does an exploration of the types and frequency of referrals for these children reveal? What does an examination of the dispositions of these referrals reveal?	Chi-square measure of association	Types of maltreatment and numbers of maltreated children were found to be unequally distributed across racial or ethnic groups. Children of African-American descent were disproportionately referred to the child welfare system. Severe neglect, general neglect and caretaker absence or incapacity were more frequent referral reasons for children of African-American descent. Children of African-American descent or those referred for neglect or caretaker absence or incapacity were found to have relatively more multiple referrals than their counterparts.
Ards et al., 2001
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Ards, S. D., Myers, S. L., Chung, C., Malkis, A., and Hagerty, B. (2001). Racial Reporting Bias and Child Maltreatment. A paper presented at the International Society for the Prevention of Child Abuse and Neglect VIII European Conference on Child Abuse and Neglect. August 24-27, 2001: Istanbul, Turkey.	Database & Sample: The 1981, 1988 and 1996 National Incidence Studies of Child Abuse and Neglect (NIS-1, NIS-2, and NIS-3).	The 1990 National Child Abuse and Neglect Data System found that African Americans were over represented among reported and substantiated abuse and neglect cases, and the National Incidence Study of Child Abuse and Neglect, collected in 1980, 1986 and 1993, showed no apparent overrepresentation of children of color. How can these disparate findings be explained?	Computation of incidence and report rates	There may be an appearance of racial reporting bias disproportionality due to the aggregation (bias) of poor and non-poor children in the NIS study.
Ards et al., 1998
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Ards, Sheila, Chung, Chanjin, and Myers, Samuel L., Jr. (1998). The effects of sample selection bias on racial differences in child abuse reporting. Child Abuse & Neglect, 22, 103-115.	Database: The 1981 National Incidence Study of Child Abuse and Neglect (NIS-1). Sample: 3,000 child abuse cases (2,499 White children and 511 Black children).	Did the NIS-1 design features result in a sample selection bias?	Logistic regression models using maximum likelihood methods. (The models were estimated with and without correction for selection bias using a two- step procedure proposed by Heckman.)	Substantial differences were found in the characteristics of Black and White victims by source of report and by type of maltreatment. White victims of child maltreatment were more likely to be reported to Child Protective Services (CPS) if they were among the lower class, known by a law enforcement or medical agency, or female. White victims suffering from emotional maltreatment were less likely to be known to CPS, while White victims suffering from physical and sexual abuse were more likely to be known. Black maltreated children were more likely to be known by CPS if they lived in rural areas, were older, or if they suffered from physical abuse. Black victims were less likely to be reported to CPS if they suffered from emotional abuse. Conclusion: Sample selection bias caused by the exclusion of family, friends and neighbors in the NIS-1 sample design is a concern for the analysis of Black victims. Without correction for selection bias, Blacks are less likely to be reported than equally situated Whites. With correction for selection bias there is little evidence of racial disparities in reporting, and identical Black and White children are just as likely to be reported to CPS. Conclusions about racial differences in child maltreatment must be reached cautiously, given the NIS-1 study design.
Ards and Harrell, 1993
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Ards, S. and Harrell, A. (1993). Reporting of child maltreatment: A secondary analysis of the National Incidence Surveys. Child Abuse & Neglect 17, 337-344.	Databases: The National Studies on Incidence and Prevalence of Child Abuse and Neglect, 1981 (NIS-1) and 1988 (NIS-2). Sample: 1,799 child abuse cases from 1981 (NIS-1), and 1,487 child abuse cases from 1988 (NIS-2).	Which kinds of child maltreatment cases are underreported? Which kinds are overreported, or not reported at all? Why?	Logistic regression using maximum likelihood procedures Multivariate analysis	Older victims were less likely to be known to Child Protective Services (CPS) than younger victims. There was a hierarchy of type of abuse reported to CPS agencies, with a larger portion of sexual abuse cases than physical and/or emotional abuse cases, and educational neglect cases were least likely to be reported to CPS. The victim's race, sex and income did not play a role in whether or not a case was reported to CPS.
Barth et al, 2000
Barth, Richard P., Miller, Julie M., Green, Rebecca L., and Baumgartner, Joy N. (2000). Children of Color in the Child Welfare System: Toward Explaining Their Disproportionate Involvement in Comparison to Their Numbers in the General Population. Chapel Hill and Research Triangle Park, North Carolina: University of North Carolina School of Social Work, Jordan Institute for Families and Research Triangle Institute.	Database: National Child Abuse and Neglect Data System (NCANDS) data Sample: 1997 data on child maltreatment from 16 states (CO, CT, FL, IL, KY, LA, MA, NJ, OK, PA, RI, SC, UT, VT, WA, and WY); 1996 data from DE, NC, and TX; and 1995 data from MO.	In a study of child abuse and racial disproportionality, how can investigations and placements of children be explained using ecological data? How can service decisions (from investigation to substantiation) for individual children be explained using ecological data?	Multivariate analysis Logistic regression models using SUDAAN	A substantial amount of the variance in county child welfare characteristics was explained by state variance, although the median family income and the Black mortality rate made significant contributions to explaining the total substantiation. African-American children were slightly more likely than Caucasian children to be substantiated for abuse and neglect and to be placed into foster care. African-American adolescents were significantly less likely than Caucasian teenagers to be substantiated or placed into foster care. The odds of adoption for African-American children were found to be significantly lower than they were for other children. Most of the discrepancy observed in the 1-day counts of children in foster care was attributable to different patterns of foster care (i.e., more kinship care) and longer lengths of stay in foster care for African-American children. "Children of Color" is not a useful concept for analyzing child welfare services dynamics. Explaining disproportionate involvement of African-American children in child welfare services requires a consideration of contextual factors related to risks of harm, in addition to child welfare services, practices and policies.
Barth, 1997
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Barth, R.P. (1997). Effects of age and race on the odds of adoption versus remaining in out-of-home care. Child Welfare 76(2), 285-308.	Database: A UC-Berkeley relational database of information on all children who have entered out-of-home care in California since 1988. Sample: A cohort of 3,873 children who were under six years of age when they entered out-of-home care.	Were children under six years of age when they entered out-of-home care reunified with their biological families, did they remain in out-of-home care, or did they experience another outcome?	Multivariate analysis Logit analysis	Age at the time of placement and race/ethnicity were found to have substantial direct effects on outcomes after four years, but there were no significant age by race interactions: African-American children from non-kinship foster care were far less likely than Caucasian children to be reunified with their families. African-American children were found to be more than twice as likely to remain in care as to be adopted. Caucasian children were about twice as likely to be adopted as to remain in care. Latino children were equally likely to remain in out-of-home care as to be adopted. African-American, and to a lesser but statistically significant extent, Latino children, were found to have dramatically diminished opportunities to be placed in permanent adoptive homes, compared to Caucasian children.
Barth and Courtney, 1994
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Barth, Richard P., and Courtney, Mark E. (1994). Timing is everything: An analysis of the time to adoption and legalization. Social Work Research, 18, 139-148.	Database: In a longitudinal study, data was collected on 1,396 adopted California children between mid-1988 and mid-1989. Sample: 496 children who had been in foster care before adoption.	What is the effect of child and foster care characteristics on the timeliness of adoption? What is the effect of child characteristics: - On the odds that a child remains in foster care for a long time before an adoption agreement is signed? - On the length of time for an adoption agreement to be legalized in court? - On major determinants of timeliness in adoption transitions?	Logit regression model Survival analysis Nonparametric methods derived from Kaplan-Meier estimation	Ethnicity is a major determinant in the timeliness of adoption transitions. White children's adoptions are finalized more quickly than those of non-White children. Adoptions of children who were adopted by White parents were legalized more quickly than children adopted by non-White parents.
Barth et al., 1994
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Barth, Richard P.; Courtney, Mark; Needell, Barbara; and Jonson-Reid, Melissa (1994). Performance Indicators for Child Welfare Services in California. Berkeley: Child Welfare Research Center.	Database: Three data sources were used - 1.) The SOC 291 Report on Preplacement Preventive Services 2.) The Foster Care Information System (reconfigured for longitudinal analysis) for children with at least one placement in foster care 3.) The U.S. Census for per capita child population counts Sample: Three types of samples were used - 1.) Random samples of children entering foster care for the first time either between July 1989 and December 1992 or between January 1988 and December 1992 2.) A cohort sample of children entering foster care during the first half of 1988 and followed for at least four years 3.) A cohort sample of children exiting care to reunification with birth families in the last half of 1989	What are the results of an analysis of the possible role of outcome and performance indicators for child welfare services funding and management?	Event history analysis	Reunification: Generally, African-American children were reunified with their birth parents at a far slower rate than children belonging to other ethnic groups. Number of Placements: The mean number of placements during the first spell in care did not differ significantly by age group or ethnicity. Reentry to Foster Care: African-American children appeared to reenter care at a slightly higher rate than children of other ethnicities. Outcomes After Four Years: African-American infants and children were more likely to still be in care and less likely to be adopted or reunified with their families than children of other ethnicities. For children placed between the ages of 1 and 3, Caucasian children were the most likely to be reunified and the least likely to still be in care, followed by Hispanic children and then African-American children. Young Caucasian children were far more likely to be adopted then either African-American or Hispanic children. When positive exits (reunification, adoption, guardianship, or placement with relatives) were compared to still being in care after four years, African-American children were less likely to have a positive exit than children from other ethnic groups. This difference existed for children first placed in kinship care and those first placed with non-kin. Performance indicators clearly vary depending on case characteristics, and meaningful analyses of performance must take into account many factors, including ethnicities. As such, analysis of performance indicators requires stratification by such factors.
Barth et al., 1986
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Barth, R.P., Snowden, L.R., Broeck, E.T., Clancy, T., Jordan, C., Barusch, A.S. (1986). Contributors to reunification or permanent out-of-home care for physically abused children. Journal of Social Service Research 9(2/3), 31-45.	Database: Data were collected from case record reviews of closed cases designated in the agency court log as physical abuse cases under California's Welfare and Institutions code. Sample: A purposive sample was selected of 107 physical abuse cases opened in 1980 or 1981 and closed between 1980 and 1984 (the sample included only one child randomly selected from each family). A second purposive sample was then drawn to increase the number of reunified children to 80 and children in permanent out-of-home placements to 21.	What are the client and service characteristics that contribute to legal reunification of the family and child or permanent out-of-home placement for physically abused children served under permanency planning statutes?	Discriminate function analysis	In this sample, the families with the least likelihood of having their children reunified were those who had abused their children most severely, had children with school problems, and had the fewest socioeconomic resources, in that order of importance. Children who were Black and older were, to a lesser extent, more likely to have the court order that their residence be outside of their birth homes.
Berrick et al., 1994
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Berrick, J.D., Barth, R.P., and Needell, B. (1994). A comparison of kinship foster homes and foster family homes: Implications for kinship foster care as family preservation. Children and Youth Service Review 16(1/2), pp. 33-63	Database: The University of California at Berkeley Foster Care Database (UCB-FCD) containing information on all children in foster care in California from January, 1988 through 1994. Sample: 246 kin providers and 354 foster care providers	What are the differences in providers to and services received by kinship foster care providers and family foster care providers? What are their demographic characteristics? What children are served in care? What issues of visitation with birth parents exist?	Comparative analysis	African-American foster parents reported fewer hours of social worker contact with their children than other ethnic groups. Caucasian foster parents were provided with more services by their agencies than other ethnic groups. Children of color, especially Hispanic children, were less likely to be placed in ethnically similar homes than were Caucasian children. Kinship foster parents were largely represented by women of color, many of whom were single parents.
Block and Libowitz, 1983
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Block, N. M. and Libowitz, A. S. (1983). Recidivism in Foster Care. New York: Child Welfare League of America.	Database: Records from the Jewish Child Care Association (JCCA) on all children under age 18 who were discharged from the JCCA during the years 1978 and 1979 to parents, relatives, friends or adoptive parents. Sample: 335 children	What is the nature and extent of recidivism? What are the causes of recidivism in the sample? What are the predictors of recidivism? What services should be provided to reduce recidivism?	Descriptive analysis Statistical analysis using chi-square tests	Race was found to be associated with recidivism. Black and Hispanic children recidivated more often than White children.
Cappelleri et al., 1993
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Cappelleri, J. C., Eckenrode, J., and Powers, J. L. (1993). The epidemiology of child abuse: Findings from the Second National Incidence and Prevalence Study of Child Abuse and Neglect. American Journal of Public Health 83(11), 1622-1624.	Database: The Second National Incidence and Prevalence Study of Child Abuse and Neglect (NIS-2). Sample: Child abuse and neglect cases from 28 counties in 19 states between September 7 and December 6, 1986.	What is the impact of five key factors - age at discovery, gender, ethnicity, income, and county-metro status--on sexual abuse and physical abuse?	Multivariate logistic regression analysis	Sexual abuse was more likely to be found among White children than Black children. Black children were physically abused at a significantly higher rate than White children and children of "other" ethnic groups.
Close 1983
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Close, M. M. (1983). Child welfare and people of color: Denial of equal access. Social Work Research & Abstracts 19(4), 13-20.	Database: The National Study of Social Services to Children and their Families Sample: 1,530 child welfare cases reported between January 1, 1977 and March 31, 1977 (1,046 White, 384 Black, and 100 Hispanic).	Are the objectives of the child welfare system, as reflected in practice, more limited for children of color than for White children? Do service plans and the range of services provided reflect differential levels of comprehensiveness? Do types of services recommended and provided seem to reflect the differential assessments, expectations, and unequal access to certain services? In this connection, does the system discriminate against families of color by providing them with fewer supportive-supplemental services and by responding more slowly to the crises they experience?	Two-way contingency analysis One-way analysis of variance	It was found that children of color did not have equal access to preferred services in the child welfare system. Children of color and their families received fewer services overall and had considerably less contact with child welfare staff. The child welfare system responded more slowly to crises that developed in families of color. Black and Hispanic children seem to have been denied equal access to emergency services and younger Hispanic children were grossly underrepresented in the groups of children for whom day care was recommended or provided.
Courtney and Wong, 1996
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Courtney, M. E. and Wong, Y. L. I. (1996) Comparing the timing of exits from substitute care. Children and Youth Services Review 18(4/5), 307-334	Database: Administrative data recorded by the child welfare authorities in the state of California containing records of the foster care histories of all children who entered or exited substitute care since January, 1988. Sample: 8,625 children who entered a first episode in the foster care system in California between Jan. 1 and June 30, 1988 and whose foster care status was monitored through the end of December, 1992.	How do selected child, family, and child welfare service system variables contribute to the timing of the three most common forms of exit from substitute care (discharge to parent, relative, or guardian; adoption; running away from care)?	Competing-risk analysis using Cox proportional-hazards model	Race and ethnicity played a role in both of the preferred modes of exit from out-of-home care (discharge to parent, relative, or guardian or adoption). Relative to other groups, being African-American was associated with a significant decrease in the probability of both discharge to family or guardian and adoption. African-American children had lower hazards of favorable discharge, indicating that many of them would likely remain in care indefinitely. Latinos were somewhat less likely to exit care to adoption than Caucasians or children of "other" backgrounds, but more likely to do so than African Americans.
Courtney, 1995
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Courtney, Mark E. (1995). Reentry of Children to Foster Care. Social Service Review, 69, 226-241.	Database: Administrative data system used in California to track longitudinal records of foster care histories of all children who entered or exited care after January 1, 1988. Sample: 6831 California children in foster care who were discharged from their first episode between January 1 and June 30, 1988 and whose foster care status was monitored through June of 1991.	What are the characteristics of those children who return to foster care after they are reunited with their families? What are the effects of reentry on children and families in the foster care system?	Event history analysis using the Cox Proportional Hazard Model	A child's race was found to be associated with the hazard rate of reentry to the foster care system after being reunited with his or her family. The hazard rate for African Americans was higher than all other groups, which had hazard rates which did not significantly differ from one another. Reentry rates for African-American children from AFDC-eligible families were found to be twice as high as those for Caucasian children from families not eligible for AFDC.
Courtney, 1994
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Courtney, Mark E. (1994). Factors associated with the reunification of foster children with their families. Social Service Review, 68, 81-108.	Database: Administrative data from California's Foster Care Information System on all children who entered foster care for the first time between January 1988 and May 1991. Sample: 8,748 children	What are the effects of child, family and system variables on the hazard rate for family reunification?	Event history analysis using a Cox proportional-hazards model	Ethnicity interacts with both age and placement jurisdiction to affect the hazard rate for return home. African-American children generally go home at about half the rate of Caucasian children regardless of age-group. They also go home more slowly than Latino children. A child's ethnicity and the region from which the child was placed interact to help explain transition home. Being of Latino or "other" ethnic heritage is associated with an increase in the hazard rate for reunification relative to other effects in the model when a child is from Los Angeles. Kinship care is more likely to be provided to non-White and poor children by non-White and poor kin caregivers.
Courtney, 1994
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Courtney, M. E. (1994). Reunification from kinship and nonkinship foster care In R. P. Barth, M. E. Courtney, J. D. Berrick, and Albert, V. From Child Abuse to Permanency Planning: Child Welfare Services Pathways and Placements (pp. 105-134). New York: Aldine de Gruyter.	Database: Administrative data from the California Foster Care Information System Sample: 8,748 children who entered foster care for the first time between January 1988 and May 1991.	What are the child, family, and foster care system variables that are associated with the timing of family reunification?	Event history analysis using a proportional-hazards regression analysis	Ethnicity and region of placements appeared to interact with respect to the rate at which foster children returned home. African-American children in particular tended to remain in foster care longer than other children.
Courtney, 1994
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Courtney, M. E. (1994). Time to adoption. In R. P. Barth, M. E. Courtney, J. D. Berrick, and Albert, V. From Child Abuse to Permanency Planning: Child Welfare Services Pathways and Placements (pp. 153-176). New York: Aldine de Gruyter.	Database: The University of California at Berkeley Foster Care database (UCB-FC) containing current histories and placement histories of more than 80,000 children who entered foster care in California between January 1988 and June 1991. Sample: All children who entered foster care between January 1988 and May 1989 and who were adopted by May of 1991 (n=864), were compared to a random sample of children who entered at the same time and did not return home or get adopted (n=1,754).	What are the effects of child, family, and foster care characteristics on the likelihood that foster children will be adopted and of the timeliness of transitions that take place in the movement toward adoption? What are the factors affecting the likelihood of being adopted? What is the effect of child characteristics on the odds that a child will remain in foster care for a long time before an adoption agreement is signed between the adoptive parent(s) and the adoption agency? How long does it take for an adoption agreement to be legalized by the superior court?	Logistic regression analysis using maximum likelihood methods	African-American children were found to be five times less likely to be adopted in the first three and one-half years of care than Caucasian children. Overall, children who entered kinship care were one-half as likely to be adopted during their first three and one-half years of care than other children. This effect was found to be far greater for Hispanic children and considerably less for Caucasian children. Children entering group home care as their initial placement, especially Caucasian children, were less likely to be adopted within the first three and one-half years of placement. Adoptions by parents of color took longer to be legalized than adoptions by Caucasian parents.
Downs, 1986
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Downs, S. W. (1986). Black foster parents and agencies: Results of an eight-state study. Children and Youth Services Review 8, 201-218.	Database: The 1980 Survey of Foster Parents in Eight States (Alabama, Arkansas, North Dakota, Rhode Island, Texas, Utah, Virginia, and Wisconsin) Sample: 1,279 foster parents (32 percent of whom were Black and 68 percent of whom were White)	How do White and Black foster parents differ in adopting foster kids? How do Blacks perceive their relationship to child welfare agencies? Are agencies supportive of Black foster parent adoption? What is the foster parent's perception of agency support and communication?	Bivariate and univariate analyses	Black foster parents were found to have somewhat more problematic relationships with child welfare and collateral agencies, and Black foster parents were somewhat more likely to care for children of relatives.
Eckenrode et al., 1988
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Eckenrode, John; Powers, Jane; Doris, John; Munsch, Joyce; and Bolger, Niall (1988). Substantiation of child abuse and neglect reports. Journal of Consulting and Clinical Psychology, 56(1), 9-16.	Database: Reports received by the New York State Child Abuse and Maltreatment Register between April 1, 1985 and August 31, 1985. Sample: 198 physical abuse reports, 796 sexual abuse reports and 880 child neglect reports from the state central registry.	What are the case factors that predict the substantiation of reports following an investigation by child services?	Multiple regression analysis	A child's ethnicity was found to have an impact on the outcome of the investigation and varied as a function of the type of maltreatment. Allegations of physical abuse were more likely to be substantiated in the cases of Black and Hispanic children than in those of White children. In the case of Black children, this higher substantiation rate is due to the higher likelihood of cases involving Black children going to court.
Fanshel and Shinn, 1978
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Fanshel, D. and Shinn, E. B. (1978). Children in Foster Care: A Longitudinal Investigation. New York: Columbia University Press.	Database: Data collected by the Columbia University School of Social Work on all children (ages 0-12) who entered foster care in New York City in 1966 and were then followed until 1971 in order to study their adjustment over a five year time span. Sample: 624 children (representing 467 family groups) who had been in care for at least 90 days and had never been in care before. 132 of the children were White Catholic or Protestant, 31 were Jewish, 259 were Black Catholic or Protestant, and 202 were Puerto Rican. (In families with three or nine children in foster care, subjects were selected randomly and no family could be represented by more than 2 children.)	How do children separated from their parents fare over time with extended tenure in foster care?	Factor analysis Multiple regression and correlational analysis	Parental Visiting of Children: Ethnicity was found to be one of the stronger predictors of parental visiting. Black children experienced the least amount of parental visiting - 50 percent of Black children were either unvisited or visited minimally, compared to 67 percent of White children, 70 percent of Puerto Rican children and more than 90 percent of Jewish children who were visited frequently. Discharge and Other Outcomes: More Black children entered care because of abuse or neglect than did White or Puerto Rican children. Of those children who entered foster care because of physical illness of their parents, Black children were discharged significantly less often than White or Puerto Rican children. Circumstances at Discharge: Sixty-two percent of Jewish children returned to a home where the parents lived together. This was true of 30 percent of the White children and 20 percent of the Puerto Rican children. Black children were the least likely to return to an intact family situation, although relatives were an important resource of Black children for discharge placement.
Fein et al., 1990
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Fein, Edith; Maluccio, A.; and Kluger, M. (1990). No More Partings: An Examination of Long-Term Foster Family Care. Washington, DC: Child Welfare League of America.	Database: The Connecticut State Computerized Management System. Sample: 779 youths in foster family care two years or more on January 1, 1985.	What are the characteristics of children in both long- and short-term care? What issues do foster youths and families face as the youths approach emancipation? What is the functioning of children in foster family care? What obstacles do foster youths face when it comes to permanency planning?	Descriptive analysis Multiple regression analysis	White foster care children and families receive more services and supports than minority children and families. African-American foster children and youths differ from youths of other races in the following ways: - More enter care because of neglect - They have more stable placements - They are placed in care at an earlier age and remain in care longer - Their biological parents are more likely to need treatment for substance abuse - Fewer enter care in adolescence (raising questions about where black adolescents are living) - They are the least likely to receive services. Hispanic foster children and youths differ from youths of other races in the following ways: - They are younger - More enter care because of abuse. White foster children and youths differ from youths of other races in the following ways: - They are older - The appear to have a more complex and extensive range of problems and needs - They are more likely to be receiving help for behavioral, emotional and other problems - Their foster families are more likely to have used services such as mental health clinics and special ed programs.
Fein et al., 1983
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Fein, E., Maluccio, A. N., Hamilton, V. J., Ward, D. E. (1983). After Foster Care: Outcomes of Permanency Planning for Children. Child Welfare, 62(6), 485-562	Database: Survey data collected by the Research Department of Child and Family Services in Connecticut on all children ages 0-14 who'd been in care for at least 30 days. Sample: 187 children who were discharged to a permanent home from temporary foster care.	Which children leave foster care and where do they go? Do these children remain in their permanent homes? How well are the children functioning? What services do the children need and use, and how are these related to stressful life events? What are the characteristics and needs of different types of permanent homes? What are the characteristics, histories, and situations of those children whose placements disrupted? How does a caseworker's permanency planning affect outcome?	Descriptive analysis Multiple regression analysis	48 percent of White children went to their biological homes, and 29 percent went to adoptive homes, whereas 66 percent of Black children went to their biological homes, 19 percent were placed with relatives, and 9 percent were adopted. 86 percent of Hispanic children were returned to their biological homes. Children were found to be doing moderately well in terms of Family Adjustment, and less well but adequate in terms of Emotional and Developmental Functioning. Although Black children and families had such characteristics as fewer previous returns to biological homes (during previous placements) and lower incomes, Black children were found to be doing better in some areas of functioning than other children.
Finch et al., 1986
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Finch, Stephen J., Fanshel, David, and Grundy, John F. (1986). Factors associated with the discharge of children from foster care. Social Work Research & Abstracts, 22, 10-18.	Database: The New York City-based Child Welfare Information Services (CWIS). Sample: 20,066 children who were active in foster care on December 31, 1974, and information about the status of children for successive periods through December 31, 1976 in order to create a continuous time-series file.	What are the variables that have a significant association with whether a foster child is discharged?	Multiple regression analysis	The probability of being adopted during the study period was higher for a White child than for a Black, Hispanic, or other child of the same age, with the same number of years in care, and who was free for adoption. Hispanic children who had adoption as a discharge goal had a greater probability of being returned to their parents or relatives than non-Hispanic children. The variables associated with adoption were different from the variables associated with discharge to parents and relatives.
Fluke et al., 1999
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Fluke, J. D., Yuan, Y. Y. T., and Edwards, M. (1999). Recurrence of maltreatment: An application of the National Child Abuse and Neglect Data System (NCANDS). Child Abuse & Neglect 23(7), 633-650.	Database: The National Child Abuse and Neglect Data System Detailed Case Data Component (NCANDS DCDC), which was derived from state-level child protection and child welfare information systems. Sample: Maltreatment recurrence cases for the years 1994 and 1995 from 10 states: 99,288 cases from Illinois, 22,572 from Louisiana, 55,546 from Massachusetts, 33,587 from Missouri, 51,590 from North Carolina, 99,475 from New Jersey, 12,632 from Pennsylvania, 98,301 from Texas, 2,419 from Vermont, and 80,814 from Washington.	What are the aspects of maltreatment recurrence in - patterns across states? - patterns in the literature? - unreported patterns of recurrence?	Event history analysis	For all states except Vermont and North Carolina, Asian/Pacific Islanders had the lowest rates of recurrence. For some states, including Louisiana, North Carolina, and Vermont, survival distributions were statistically different for African Americans and Whites. However, the pattern of differences across states between African Americans and Whites was not consistent. In some states, time to recurrence for African Americans was shorter than for Whites, while in other states the reverse was true. In Illinois, Massachusetts, Missouri, New Jersey, Texas and Washington, survival distributions for White and African-American children were not significantly different.
Garland et al., 1998
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Garland, Ann F.; Ellis-MacLeod, Elissa; Landsverk, John A.; Ganger, William; and Johnson, Ivory (1998). American Journal of Orthopsychiatry, 68(1), 142-146.	Database: A longitudinal study of children placed in out-of-home care between May of 1990 and October of 1991. Sample: 1,332 subjects representing 834 families.	What is the outcome of a test of the "visibility hypothesis" (which suggests that there is a higher probability for minority children to be placed in foster care when living in geographic locations where their proportions in the population are relatively low, compared to areas where their proportions are high) in a large, ethnically diverse county in California?	Linear analysis of placement ratios	The patterns of minority representation in the foster-care cohort support the visibility hypothesis (e.g., the more "visible" a child is in the community, the more likely that the child would be placed in foster care). The "visibility" pattern found was specific to African-American children, who represented the greatest proportion of minority children in the sample. The pattern was not present for either Hispanic or Asian-American children.
Garland and Besinger, 1997
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Garland, Ann F. and Besinger, Bridgett A. (1997). Racial/Ethnic differences in court referred pathways to mental health services for children in foster care. Children and Youth Services Review, 19, 651-666.	Database: The Foster Care Mental Health (FCMH) research project, which included data on 1037 children ages 0-17 years who were placed in foster care from May 1990 to October 1991 and who remained in out-of-home care at least 5 months. Sample: 184 cases, chosen to achieve roughly equal representation of all three major racial/ethnic groups (African-American, Caucasian and Latino).	What is the role of the court process in referring children in foster care to mental health services? What are the racial/ethnic differences in the patterns of referral and service use?	Multivariate logistic regression analysis	The court process was found to play a significant role in referring children to services. Significant differences by race and ethnicity were found in mental health service utilization prior to the child's protective placement, as well as service orders and post-placement service use. Caucasian youth were more likely to receive orders for psychotherapy and to have documented use of psychotherapy than were African-American and Hispanic youth, even when the possible confounding effects of age and type of maltreatment were controlled. Caucasian youth were more likely to enter the system with a history of counseling and to receive counseling during the first eight months of out-of-home care. Although no statistically significant racial/ethnic differences on frequencies of other types of service use were found, there were higher rates of use by Caucasians on almost every type of service use prior to and post-removal from the home.
Goerge et al., 1994
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Goerge, R. M., Wulczyn, F. H. and Harden, A. W. (1994) Foster Care Dynamics in California, Illinois, Michigan, New York, and Texas, 1983-1992: A Report from the Multi-State Foster Care Data Archive. Chicago: Chapin Hall Center for Children.	Database: A multistate foster care data archive containing foster care histories for all children placed in a state-supervised substitute living arrangement from 1983-1992 in Illinois, Michigan and New York, from 1988-1992 in California, and from 1985-1992 in Texas. The database was constructed from information drawn directly from the administrative databases that the separate state agencies use to manage and operate their own child welfare caseloads. Sample: 440,374 cases	How many children are in foster care? Is the size of this population changing? What forces are involved in these changes? Are they persistent? Do they occur evenly over time? Have the shifting patterns of admissions and discharges had different effects on the foster care population? Has the composition of foster care caseloads, as defined by characteristics of the population of children in care, changed in recent years? What is the risk of any child in the general population being placed in foster care? For what groups of children are these risks increased? Once a child is removed from home, how long should we expect him or her to remain in foster care? What factors are related to the duration of stay in care?	Event history analysis using a proportional hazards model	Across all five states, White children accounted for about 33 percent of first admissions to foster care, African Americans accounted for 40 percent, and Hispanics for under 20 percent. In Illinois, the percentage of children placed who were White decreased by one-third between 1988 and 1992. During this period, the percentage of African-American children placed increased by one-quarter. In California, the proportion of Hispanic children increased from 23 percent in 1988 to 29 percent in 1992 while the proportion of both African-American and White children declined. In New York, the percentage of White children placed increased. African-American children stayed in care longer than other racial or ethnic groups. In California and New York this disparity was large and in Illinois it was even larger.
Goerge and Harden, 1993
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Goerge, R. M. and Harden, A. W. (1993). The Impact of Substance-Affected Infants on Child Protection and Substitute Care Caseloads: 1985-1992. Chicago: Chapin Hall Center for Children.	Database: The Child Abuse and Neglect Tracking System and The Child and Youth Centered Information System of the Department of Children and Family Services (DCFS) in Illinois. Sample: 12,087 separate substance misuse allegations for children under the age of one from 1985 to 1992.	How did Substance-Affected Infants (SAI) experience the substitute care system between 1985 and 1992? How has this population affected, and how will it continue to affect, the aggregate caseload of DCFS?	Aggregate analysis Survival analysis methods from the Kaplan-Meier technique	The proportions of Substance-Affected Infants (SAI) placed in substitute care were found to be: White: 11.5% Black: 83.9% Hispanic: 4.0% Other: 0.6% First placement spell duration was found to be significantly longer for SAI than for non-SAI children. There was no variation by race for most children, although Black children in relatives' homes in Cook County and non-Black children in relatives' homes outside of Cook County exhibited some variation.
Goerge, 1990
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Goerge, R. M. (1990). The reunification process in substitute care. Social Service Review, 64(3), 422-457.	Database: Data recorded by the Illinois Department of Children and Family Services (DCFS) on all children who entered foster care between July 1, 1976 and May 31, 1984. Sample: 1,196 children.	Does the probability of reunification decrease as a child spends more time in substitute care? Do certain types of children, classified by reason for placement, type of placement, age at placement, race or ethnicity, and region of residence, exhibit a stronger decreasing probability of reunification than others? What types of children are more likely to remain in foster care throughout their childhoods?	Event history analysis	The race of the child and region of placement were found to combine to explain large differences in reunification. The results of placement-level analysis show that foster children in Cook County stayed in their placements longer than children outside of Cook County. Race had no effect on the duration of these placements. Black children outside of Cook County were found to have a different experience than any Cook County children or non-Black children outside of Cook County. They had the highest probability of staying in care; nearly 4 percent in the first placement, 8 percent in the second placement, and 5 percent in the third placement.
Groeneveld and Giovanni, 1977
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Groeneveld, L. P. and Giovannoni, J. M. (1977). Disposition of child abuse and neglect cases. Social Work Research & Abstracts 13(2) 24-30.	Database: The National Clearinghouse on Child Abuse and Neglect Sample: The first 2,400 cases in the database, reported between January and August 1974. Cases were reported by Arizona, Montana, North Carolina, Rhode Island, and Texas.	What are the factors that affect the progress of child abuse cases at the various stages of case management?	Multivariate regression analysis	Ethnicity was found to be one of three variables that had a significant effect on child abuse cases at any point in the system. It was found to have an effect on removal in neglect cases; non-Caucasian children were more likely to be removed from their families than Caucasian children.
Gurak et al., 1982
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Gurak, Douglas T., Smith, David Arrender, and Goldson, Mary F. (1982). The Minority Foster Child: A Comparative Study of Hispanic, Black and White Children. New York: Fordham University Hispanic Research Center, Monograph 9.	Database: The "Under-Care" Module of the Child Welfare Information System (CWIS) of New York City. For the comparative analysis, data were used from the records of children in care in 1979 in seven New Jersey District Offices (counties) in New Jersey's Division of Youth and Family Services. Sample: 953 New York children in foster care in 1979. For the comparative analysis: 149 children with cases active between October 1, 1977 and October 1, 1978 and 500 children whose cases were active beginning June 1979 and ending June 1978.	Do ethnic differentials in outcomes exist? Can any such differences be attributed to ethnic differentials in entry-level characteristics? Can any such differences which may remain subsequent to controlling for entry-level characteristics be accounted for by process-level characteristics of the foster care system?	Multivariate regression analysis	Ethnic differentials in outcomes do exist and these outcomes cannot be explained by group differences in entry-level characteristics. Several process-level factors clearly contribute to these ethnic differentials. Blacks and Hispanics remain in care longer than Whites. Blacks and Hispanics experience lower progress towards exit rates than Whites. Blacks and Hispanics are placed in less efficient agencies than Whites.
Hampton and Newberger, 1984
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Hampton, R. L. and Newberger, E. H. (1984). Child abuse incidence and reporting by hospitals: Significance of severity, class, and race. American Journal of Public Health 75(1), 56-60.	Database & Sample: The 1981 National Study of the Incidence and Severity of Child Abuse and Neglect.	What are the variables associated with the initial diagnosis of child abuse by hospitals and with the proportion of cases subsequently reported to child protective service agencies?	Multivariate analysis	Race was found to be one of the factors that distinguished the reported from the unreported cases of abuse. Hospitals failed to report to child protection agencies almost half of the cases that met the study's definition of abuse, and compared to other agencies in the sample, hospitals identified children who were younger, Black, lived in urban areas, and had more serious injuries. Black and Latino families were more likely to be reported than White families.
Jones and McCurdy, 1992
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Jones, Elizabeth D. and McCurdy, Karen (1992). Child Abuse & Neglect, 16, 201-251.	Database: The 1988 National Incidence Study of Child Abuse and Neglect (NIS-2) on maltreated children. Sample: 2,814 cases.	What is the relative impact of demographic characteristics of the child, family structure, and economic variables on types of child abuse and neglect?	Descriptive analysis Logit analysis	Physical neglect, in comparison with other types of abuse, is the most predictable and distinguishable type. It is most clearly related to economic factors such as low income and Aid to Families with Dependent Children (AFDC) status, regardless of race. Only one comparison of maltreatment, sexual abuse versus neglect, reveals significant differences between Blacks and Whites with Blacks evidencing a higher risk than Whites of being neglected as opposed to sexually abused. However, minority status has little influence on the likelihood of neglect occurring. Rather, neglect appears to be a problem of economics.
Katz et al., 1986
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Katz, Mitchell H., Hampton, Robert L., Newberger, Eli H., Bowles, Roy T., and Snyder, Jane C. (1986). Returning children home: Clinical decision making in cases of child abuse and neglect. American Journal of Orthopsychiatry, 56(2), 253-262.	Database: Records of suspected abused or neglected children seen at Children's Hospital in Boston, Massachusetts. Sample: Hospital records of 185 children from 1978 to 1981.	What are the factors that influence the decision to remove children from their parents' care in cases of abuse and neglect?	Bivariate analysis Multivariate log-linear analysis	Families who were Medicaid-eligible and those with a previous report of suspected child maltreatment were more likely to have their children removed. Minority families were not, however, more likely to lose their children.
Korbin et al., 1998
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Korbin, J. E., Coulton, C. J., Chard, S., Platt-Houston, C., Su, M. (1998) Impoverishment and child maltreatment in African-American and European-American neighborhoods. Development and Psychopathology 10, 215-233	Database & Sample: Phase I) Quantitative Study: All 1991 "substantiated" and "indicated" reports of child maltreatment from the Cuyahoga County Department of Human Services computerized records. Census tracts were identified using 1990 census data: 94 predominantly African-American and 189 predominantly European-American census tracts. Phase II) Qualitative Study: 13 census tracts were selected for ethnographic study to represent neighborhoods with different rates of child maltreatment reports. For the ethnographic analysis, four census tracts were drawn that represented neighborhoods with different rates (highest and lowest quartiles) of child maltreatment reports.	What is the relationship between neighborhood structural factors and child maltreatment report rates in African-American and European-American census tracts?	Aggregate level and regression analysis Analysis of ethnographic observations and resident narratives	High rates of reported maltreatment in low income and minority populations were found in this study. Impoverishment had a significantly weaker effect on maltreatment rates in African American than in European American neighborhoods. While African American groups lived disproportionately in poor, divested areas, when other factors that favored community social organization were in place, their neighborhoods were found to be supportive of children and families. Neighborhood conditions were found to be linked with child maltreatment report rates.
Lauderdale et al., 1980
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Lauderdale, M., Valiunas, A., and Anderson, R. (1980). Race, ethnicity, and child maltreatment: An empirical analysis. Child Abuse & Neglect 4, 163-169.	Database: The Texas Department of Human Resources Central Registry incidence data. Sample: All 36,945 validated cases of abuse and neglect reported between 1975-1977.	What is the empirical relationship of child maltreatment to ethnicity?	Computation of rates of abuse and neglect for each major ethnic group in the Texas population (Anglos, Blacks, Mexican-Americans)	Child maltreatment was found to vary by ethnicity. Anglos had the lowest rate of abuse and neglect, Mexican Americans had a rate 10.5 percent higher, Blacks had a rate 37.3 percent higher than the Anglo rate. Abuse accounted for 33.5 percent of the validated cases of maltreatment among Anglos, 29.4 percent among Blacks and 25.2 percent among Mexican Americans. Neglect accounted for 55.4 percent of maltreatment among Anglos, 61.2 percent among Blacks, and 65.6 percent among Mexican Americans.
McCabe et al., 1999
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
McCabe, Kristen; Yeh, May; Hough, Richard L.; Landsverk, John; Hurlburt, Michael S.; Culver, Shirley Wells; and Reynolds, Beth (1999). Racial/Ethnic representation across five public sectors of care for youth. Journal of Emotional and Behavioral Disorders, 7, 72-82.	Database: The Patterns of Youth Mental Health Care in Public Service Systems Project (POC) enumeration of youth (birth to 18 years) who were active in one or more of five public sectors of care (including child welfare) in San Diego County during a 6-month period in the 1996-1997 fiscal year. Sample: 11,515 youth	What would the results of an investigation of ethnic minority service use be through examination of representation rates across four racial/ethnic groups and five service sectors simultaneously in a large population of service users?	Multivariate analysis Introduction of unique methodology for determining the representative over- and underrepresentation of racial/ethnic groups in each sector by providing comparisons to three different sets of Census data (1996 Full Census data; 1996 Census estimates of youth at or below 200% of poverty level; 1997 San Diego County School Enrollment Census data).	The patterns of representation varied greatly, depending upon racial/ethnic group and comparison group. African Americans were overrepresented in child welfare regardless of the comparison group. Rates of overrepresentation were reduced when socioeconomic status was taken into account by using a 200% Poverty Census comparison group. In each comparison group, Asian/Pacific Islander Americans were underrepresented in the child welfare sector. Patterns among Caucasian Americans differed by comparison group. In comparison with the full Census and School Enrollment data, Caucasians were underrepresented in child welfare. In comparison with the 200% Poverty Census group, however, Caucasians were overrepresented. The comparisons for Latinos in all three groups reflected underrepresentation in child welfare.
McMurtry and Lie, 1992
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
McMurtry, Steven L. and Lie, Gwat-Yong (1992). Differential exit rates of minority children in foster care. Social Work Research and Abstracts, 28, 42-48.	Database: Written case records drawn from the central office of the Arizona Foster Care Review Board. Sample: A study of 775 foster care children in Maricopa County, Arizona, who were originally placed in out-of-home care between January 1, 1979 and December 31, 1984 and whose cases were open for at least six months.	How do children's racial/ethnic characteristics affect their placement outcomes, their exit outcomes and the length of time it takes them to exit foster care?	Event history analysis using a proportional hazards model	Black children were found to be represented more than three times as often among foster children as among residents of the county. Significant differences between ethnic/racial groups were found for the age of the child when initially placed in care. Most notably, White children were an average of more than one and a half years older than non-White children when first placed. The groups also differed significantly on the presence of disabilities, with Black children being least likely and Hispanic children being most likely to have a physical or developmental disability. Finally, a significant effect for groups was found for total duration of care. Overall, Black children spent an average of more than three years in care, White and Hispanic children averaged two and a half years, and other minority children averaged two years. Results show that children did not differ across ethnic groups on rates of exit into adoption or outcomes such as emancipation or placement with relatives. The major difference was among children who returned home, which occurred at only half the rate for Black children as for White children.
National Black Child Development Institute, 1989
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
National Black Child Development Institute. (1989). Who Will Care When Parents Can't? A Study of Black Children in Foster Care. NCBDI: Washington, DC.	Database: Case records from child public welfare agencies in Detroit, Houston, Miami, New York, and Seattle. Sample: 222 Detroit children, 311 Houston children, 98 Miami children, 246 New York children and 126 Seattle children placed in a state supervised living arrangement for at least a 24-hour period during the calendar year of 1986.	What is the profile of black children, their families, and the social context of their lives before and during foster care?	Descriptive analysis	Escalating numbers of Black children were found to be entering foster care at ever younger ages and remaining in care for longer periods of time because of dramatic discrepancies between needs identified and services provided. Black children entering foster care were at a disproportionately high risk of serious health, education, and mental health problems. 46 percent of the foster children examined were discharged during the course of the study period, which was more than two years in length.
Needell and Barth, 1998
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Needell, B. and Barth, R. P. (1998). Infants entering foster care compared to other infants using birth status indicators. Child Abuse & Neglect 22(12), 1179-1187.	Database: Foster care data were taken from the California Children's Services Archive at the University of California at Berkeley. Birth data were taken from the California Birth Statistical Mater File, also housed at the Archive. Sample: 26,460 maltreated infants who entered foster care between 1989 and 1994, and a random sample of 401 infants born in this time period who didn't enter foster care.	How do infants in foster care compare with a random sample of infants not in foster care? What are their similarities and differences?	Probability matching Logit regression	40.5 percent of mothers in the foster care sample were African American as compared to 7.6 percent of the non-foster care sample. Mothers of infants in care were more than twice as likely to be African American than mothers of other infants. Hispanic and other ethnic groups were underrepresented in the group of infants in care. Foreign-born mothers, especially Hispanic women, were much less likely to have infants in care.
Nelson et al., 1993
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Nelson, K. E., Saunders, E. J. and Landsman, M. J. (1993). Chronic child neglect in perspective. Social Work 38(6), 661-671.	Database: Case file data on families referred to the Allegheney County (Pittsburgh, Pennsylvania) Children and Youth Services for child neglect . Sample: 182 families referred between October 1986 and August 1989.	What are the dynamics of child neglect in three groups of families referred to a metropolitan county child protection agency because of child neglect?	Discriminate analysis	Black children were found to be overrepresented in cases of neglect in the child welfare system in Pittsburgh. Although Black families constituted only 31.1 percent of the families with children under age 18 at the time of the study, 45.3 percent of the primary caregivers in the study were minorities, and all but two were Black.
Neuspiel et al., 1993
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Neuspiel, Daniel R., Zingman, Terry M., Templeton, Virginia H., DiStabile, P., and Drucker, E. (1993). Custody of cocaine-exposed newborns: Determinants of discharge decisions. American Journal of Public Health 83(12), 1726-1729.	Database: Maternal and infant medical records at a public hospital in New York City. Sample: All 99 newborn infants testing positive for cocaine or benzoylecgonine from July 1990 through May 1991 (49% Black, 40% Hispanic, 11% other).	What are the factors that predict custody status of infants born to substance-abusing women?	Kruskal-Wallis one-way analysis of variance Multiple logistic regression analysis Chi-square tests T tests	Placement outside the family was greater with Blacks than with other ethnicities and races. Denial of custody to Black mothers was higher. Being Black was found to be predictive of discharge away from the mother and less predictive of non-family discharge custody. Black mothers were found less likely to keep their newborns. No interactions were found between ethnicity and prior child welfare reporting.
Olsen, 1982
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings
Olsen, Lenore (1982). Services for minority children in out-of-home care. Social Service Review, 56, 572-585.	Database: The National Study of Public Social Services to Children and Their Families, which identified 1.7 million children on the case loads of public social service agencies across the nation in 1977, one-third of whom were living apart from their natural families at the time. Sample: 17,000 cases.	How are minority group children served by the child welfare system? What are the characteristics of children and families receiving services, particularly among children of different minority groups? How does age interact with minority group membership to affect the experiences of minority children in the child welfare system?	Bivariate analysis Analysis of covariance	In comparison to other ethnic groups, it was found that there were fewer service plans for Black children and fewer plans for regular contact between the Black child and their principal child-caring person (PCCP). Black children also spent a considerably longer time in care than other children. Hispanic children were found to be less likely to have plans for regular contact with their PCCP and there was a tendency for agencies to be unable to specify why these children were not free for adoption. Hispanic youth were treated as more behaviorally disturbed than other groups of children. Only one-third of the Native American children ages 6-11 had service goals, and virtually none of their families had service recommendations. Asian children were more likely to have service plans, with particular attention given to improving family relationships. Asian youth spent less time in care than other children.
Plantz et al., 1989
Citation (Author(s), Title, Source, Year)	Database/ Sample	Research Question(s)	Method of Analysis	Research Findings