Pilot CWS-R-SAP – v. 1 Current CWS, 11.5. Child Abuse Victims by Race

 

Pilot Child Welfare System Redesign

Strategic Action Plan

 

These webpages are a work in progress. They are open to the public to encourage comments, ideas, and improvements.

 

Previous Page                                                                                Next Page

 

11.5. Child Abuse Victims by Race

 

The following chart has been created from data in the annual Oregon Child Welfare Data Books for the years 2009 through 2019.

Child Abuse Victims by Race
Child Abuse Victims by Race

 

For this chart, the definition of “Victims” is taken from the annual Oregon Child Welfare Data Books for the years 2009 through 2019: “Once there is a founded disposition, the children for whom there is reasonable cause to believe they were abused are considered victims of child abuse.”

 

It is critical that the Oregon Child Welfare System provide a fair and equitable representation of all races, to avoid discriminatory treatment of some children over others. Oregon’s Child Welfare System was aware of this possibility of unfair treatment from the earliest Child Welfare Data Book from 2009: “Eliminating disparate treatment and overrepresentation of children of color.” The chart on the previous page addresses the representation of races as victims of child abuse or neglect.

 

“Unable to Determine Victims”

First, discussion is focused on the top set of bars, “Unable to Determine Victims,” which represents the percentage of children identified as being victims of child abuse for which their race or ethnicity were unable to be determined. This is a bothersome category for two reasons:

  • Why were CPS Caseworkers “unable to determine” the race or ethnicity of so many of the child abuse victims (climbing toward 20% or one-fifth of all child abuse victims) when they had to have seen and talked with the child abuse victims several times during their assessments that led to the “founded” for child abuse disposition? and

 

  • Why are the percentages increasing each year?

 

“Unable to Determine” Data from 2009 Through 2019

Although data from 2016 through 2019 represented the only complete data for the remainder of the elements in this chart, information on the “Unable to Determine” category is available from 2009 through 2019:

2009=14.8%

2010=13.9%

2011=14.5%

2012=15.1%

2013=14.3%

2014=14.4%

2015=14.0%

2016=13.8%

2017=14.4%

2018=18.3%

2019=19.1%

 

Data Patterns

The data shows a consistent pattern of about 14% “Unable to Determine” from 2009 through 2017. However, 2018 “Unable to Determine” increased 4% above the average and 2019 was 5% above the average. The “Unable to Determine” data from 2009 through 2017 shows a pattern of behavior or training.

The number of children being abused varies by year, with each set of children providing its unique array of races, ethnic groups, and children who cannot be identified as to race or ethnic group. A constant average of about 14% “Unable to Determine” 9 years in a row is not random. It can be thought of as intentional.

Why would CPS Caseworkers categorize so many children as “Unable to Determine” every year for 9 years? And why would the percentage increase so significantly in 2018 and 2019?

 

Possible Reason for the Significant Increase in 2018 and 2019

One possible reason for 19% of the child victims being classed as “Unable to Determine” is to mask the overrepresentation of Children of Color in the group of child abuse victims. Analyzing the data from 2019, with 19.1% (nearly one-fifth) of the child victims not categorized, it is difficult to form any strong conclusions about the data presented in the preceding chart for that year.

  • For example, if all 2,616 of the “Unable to Determine” child victims were White, 2,616 is added to 7,933, resulting in 10,549 White child victims, or 77.1% of all child abuse victims, which has a disproportionality index of 1.16. This change in disproportionality index would not be considered significant, so there would be no reason to “hide” White child victims in the “Unable to Determine” category.

 

  • However, if all 2,616 of the “Unable to Determine” child victims were Black or African American, 2,616 is added to 613, resulting in 3,229 Black or African American child victims, or 6% of all child abuse victims, which has a disproportionality index of 6.2 – which means that the percent of Black or African American children that were victims of child abuse would be 6.2 times higher than the percent of Black or African American children in Oregon’s population.

 

  • Similarly, if all 2,616 of the “Unable to Determine” child victims were American Indian or Alaska Native, 2,616 is added to 398, resulting in 3,014 American Indian or Alaska Native child victims, or 22% of all child abuse victims, which has a disproportionality index of 13.75 – which means that the percent of American Indian or Alaska Native children that were victims of child abuse would be 13.75 times higher than the percent of American Indian or Alaska Native children in Oregon’s population.

 

Hopefully, not all of the “Unable to Determine” child victims will be placed on a single row in the chart; the extremes are shown to demonstrate that the data as it appears on the chart cannot be confidently used when comparing the disproportionality of one race or ethnic group over another among child abuse victims.

 

It also points to the need for investigation into the victims designated as “Unable to Determine” to remove – or confirm – the suspicion of “hiding” overrepresentation of Children of Color.

 

Black or African American Child Victims

With the caveat above taken into account, the data as displayed in the bottom set of bars in the chart, representing the percentage of Black or African American children in Oregon, depicts a 0.1% increase in percentage each year from 2016 through 2019. To be equitable, the percentage of Black or African American victims should also be in the corresponding 3.5%-3.8% range.

However, the next set of bars above, as displayed in the table, representing the percentage of Black or African American child victims, averages 4.5% for 3 of the years and 4.8% for 2018. This indicates that the percentage of Black or African American child victims is overrepresented, possibly indicating a slight bias toward, or targeting of, Black or African American children and their families during CPS assessment; presuming cultural traditions are not factored in.

 

Asian/Pacific Islander Child Victims

With the same caveat, the data as displayed in the third set of bars from the bottom in the chart, representing the percentage of Asian/Pacific Islander children in Oregon, depicts a 0.2% increase in percentage from 2016 through 2019. To be equitable, the percentage of Asian/Pacific Islander victims should also be in the corresponding 5.4%-5.6% range.

However, the next set of bars above as displayed in the chart, representing the percentage of Asian/Pacific Islander child victims, averages 1.4% to 1.6% for the years between 2016 and 2019. This indicates that the percentage of Asian/Pacific Islander child victims is underrepresented, possibly indicating a slight bias away from, or avoidance of, Asian/Pacific Islander children and their families during CPS assessment; presuming cultural traditions are not factored in.

 

White Child Victims

With the same caveat, the data as displayed in the fifth set of bars from the bottom in the chart, representing the percentage of White children in Oregon, depicts a reduction in percentage each year from 2016 through 2019, ranging from a high of 67.6% to a low of 66.7%. To be equitable, the percentage of White victims should also be in the corresponding 66.7%-67.6% range.

However, the next set of bars above as displayed in the chart, representing the percentage of White child victims, demonstrates a steady and significant decline in the percentages from 64.2% in 2016 to 58.0% in 2019. This indicates that the percentage of White child victims is historically underrepresented, and continually further underrepresented, possibly indicating a slight bias away from, or avoidance of, White children and their families during CPS assessment; presuming cultural traditions are not factored in.

 

Example of Implicit Bias that Leads to Misdiagnosis

“I think we need to look at abusive head trauma and why it is according to much of the research that abusive head trauma cases are misdiagnosed for white kids. I think that suggests that we really need to go back in and look at that data and it is possible that implicit bias could be contributing to that misdiagnosis of abusive head trauma with regard to white kids and that might give us information that will allow us to move forward.” – Dr. Rita Cameron Wedding CECANF Testimony – New York[1] [bolding added for emphasis]

 

“It could be that disproportionality is a double-edged sword that directly disparately treats African Americans while inadvertently depriving Whites of proper assessments and diagnoses.” – The Honorable Judge Patricia M. Martin[2] [bolding added for emphasis]

 

Hispanic (Any Race) Child Victims

With the same caveat, the data as displayed in the fifth set of bars from the top in the chart, representing the percentage of Hispanic (any race) children in Oregon, depicts an increase in percentage from 21.9% to 22.3% during 2016 through 2019. To be equitable, the percentage of Hispanic (any race) victims should also be in the corresponding 21.9%-22.3% range.

However, the next set of bars above as displayed in the chart, representing the percentage of Hispanic (any race) child victims, ranges between 12.1% and 13.9% during 2016 through 2019. This indicates that the percentage of Hispanic (any race) child victims is underrepresented, possibly indicating a slight bias away from, or avoidance of, Hispanic (any race) children and their families during CPS assessment; presuming cultural traditions are not factored in.

 

American Indian or Alaska Native Child Victims

With the same caveat, the data as displayed in the third set of bars from the top in the chart, representing the percentage of American Indian or Alaska Native children in Oregon, depicts a consistent average of 1.6% during 2016 through 2019. To be equitable, the percentage of American Indian or Alaska Native victims should also be in the corresponding 1.6% range.

However, the next set of bars above as displayed in the chart, representing the percentage of American Indian or Alaska Native child victims, has a high of 3.6% in 2017 and a low of 2.9% in 2019. This indicates that the percentage of American Indian or Alaska Native child victims is overrepresented, possibly indicating a strong bias toward, or targeting of, American Indian or Alaska Native children and their families during CPS assessment; presuming cultural traditions are not factored in.

 

Analysis

The data as displayed in the chart above shows twice (2x) to three times (3x) as many American Indian or Alaska Native child victims when compared with the percentage of American Indian or Alaska Native children in Oregon. This is sufficient evidence to warrant further investigation into the apparent disproportionality of American Indian or Alaska Native children as child victims.

In addition, the significantly high number of “Unable to Determine” children in the child victim demographic warrants further investigation to avoid the appearance of hiding greater disproportionality with respect to race among the child victims. Another reason for further investigation is the fact that the 2,616 children identified as “Unable to Determine” in 2019 had been thoroughly assessed – presumably seen by someone from CPS – and their report was determined to be founded. Did no one attempt to inquire as to the race or ethnicity of almost 20% (almost one-fifth) of the children they assessed?

 

Recommendations:

Action 11.5.1: Continue to monitor and eliminate racial and ethnic bias when assessing families for child abuse. Although some progress appears to have been made in reducing the percentage of Black or African American child victims, more progress is needed to align with the actual proportionality of Black or African American children in the population of Oregon. Of even more concern, very little progress has been made to reduce the percentage (2-3 times expected) of American Indian or Alaska Native child victims.

 

Action 11.5.2: Conduct a complete analysis of the American Indian or Alaska Native child victims and their families to determine whether racial profiling was introduced during the assessment. A significant overrepresentation of 2x to 3x is sufficient cause to review ALL cases, at least back to 2016. It is further recommended that one or more representatives from local American Indian tribes be included in the investigation to further ensure that bias was not introduced or continued during the investigation. Based on that investigation, make the necessary modifications to CPS rules, procedures, training, etc. to address the issue and ensure that American Indian and Alaska Native children are treated more equitably in the future.

 

Action 11.5.3: Fully reexamine every case of “Unable to Determine” race or ethnic group from at least the last 4 years, to determine the reason for the high percentage in that population. Determine if racial or ethnic bias is a part of the reason why almost one-fifth of all children assessed as child abuse victims were not able to be identified as belonging to a particular race or ethnic group. Determine the stated reasons why CPS Caseworkers assessing (i.e., seeing and talking with) the child and the family were unable to determine race or ethnic group. If an additional race or ethnic grouping is needed, then add it to the data collection metrics. Take action on any racial bias uncovered during the investigation, to ensure the equitable treatment of all races and ethnic groups by CPS and all of the Child Welfare System, as well as the honest reporting by Child Welfare System staff.

 

[1] The Dissenting Report of The Honorable Judge Patricia M. Martin, CECANF Commissioner, March 14, 2016.

[2] Ibid.

 

Previous Page                                                                                Next Page

 

 

To submit questions or comments, please email Jo@Jo-Calk.com. I welcome all input, ideas, and suggestions. Thank you for caring for children.

Blessings,

Jo Calk