« Back to Methodology Index

Methodology Child Maltreatment Allegations—Child Count

Report Description

This report provides detail on the number of children with a child maltreatment allegation in the specified year.

Refresh Schedule

This report is refreshed quarterly. For information on publication schedules, please refer to Source Data below.

Time Periods

This report uses rolling year time periods. Available Intervals: Jan-Dec, Apr-Mar, Jul-Jun, & Oct-Sep. The option to display All intervals is also offered. The earliest available time period for this report is Jan 1, 1998 – Dec 31, 1998. The end date of the most recent time period corresponds with the most recent quarterly extract. For Over Time Views you have the option to select the earliest year displayed, which controls the display for tables and graphs.

Report Data

This report uses the Allegation File. For information on this file, please refer to Source Data below.

Notes on Source Data

The main sources of data for this site are the University of California, Berkeley quarterly extracts from the California Department of Social Services (CDSS) Child Welfare Services/Case Management System (CWS/CMS). These extracts are pulled approximately one month after each quarter ends, and the data are fully refreshed each quarter. Due to the time it takes to process, run, and validate the data, information on the website and in the California CWS Outcomes System reports is between three and six months old.

Refresh Schedule:

Extract name Data cutoff Website refreshed by
Quarter 1 April 1 July 1
Quarter 2 July 1 October 1
Quarter 3 October 1 January 1 (following year)
Quarter 4 January 1 April 1

For example, the CWS/CMS 2025 Quarter 3 Extract (Q3 25) reported on events up to the data cutoff of October 1, 2025. Reports based on the Q3 25 extract were available on the website by January 1, 2025.

For reporting purposes, we use the CWS/CMS extracts to construct three primary longitudinal analysis files types (allegation, foster care, and case) which track children’s histories in the Child Welfare System. In addition to specifying supervising county and capturing child-level demographics, these files include additional information:

Allegation File

– This file includes one row for each allegation for each child, including allegation type, reporter type, responses such as investigation, as well as allegation disposition (e.g., substantiated, inconclusive, and unfounded).

More information on the allegation file

Foster Care Files

– These files include one row for each placement home for each child, including dates of entries and exits to foster care, placement type, and length of stay.

More information on the foster care files

Case Files

– The primary file includes one row for each case service component for each child. Secondary files serve to locate cases within caseloads, offices, and counties. The files include information about children’s case services, including case openings and closings, and service components.

More information on the case files

Population Data Files

– Includes California Department of Finance (DOF) annual child population counts and Poverty Population Estimates derived from the US Census Bureau’s American Community Survey (ACS).

More information on the population data files

Quarterly Wage File

– Quarterly Wage data from 1993 forward for all children age 16 and older with a record of an out-of-home foster care placement in the Child Welfare Services / Case Management System (CWS/CMS) since January 1, 1998.

More information on the quarterly wage file

Postsecondary Enrollment Data

– Postsecondary Enrollment data for children age 16 and older with a record of an out-of-home foster care placement in the Child Welfare Services / Case Management System (CWS/CMS).

More information on the postsecondary enrollment file

Methods

This report counts each child receiving a child maltreatment allegation once for each analysis year. If a child has more than one allegation in a given year, they are counted in the cell considered to represent the most severe occurrence. In order to determine the appropriate cell, children are grouped in a two-step process:

  1. Disposition Outcomes: Children with substantiated reports are counted first, followed by those with inconclusive reports, unfounded reports, children who had allegations but did not have a disposition (called assessment only/evaluated out), and finally, children for whom a disposition had not yet been determined (called not yet determined).
  2. Allegation Type: Children are then grouped according to the allegation type severity hierarchy established by CWS/CMS (see additional details in the “Allegation Type” link provided below).

This differs from other allegation reports and for this reason the reports are not directly comparable. See Allegation Report Differences for details.

In the county-level tables, we have again counted each child once per year, but for each county in which a referral was made. Thus, a child who has received referrals in multiple counties will appear in the table of each county where referred. As a result, the sum of the children across county tables may not equal the total in the statewide tables.

Notes

County is assigned in this analysis using the county specific code from the assignment table corresponding with the allegation received date during the specified time period.

Age in this analysis is based on the first allegation received date during the specified time period.

Cells containing a period (“.”) represent a value of zero. In cells representing quotients, a period may also indicate the indeterminate form 0/0.

The following methodology may include references to report features not available on the public site version. On the public site, masking is performed to protect the privacy of individuals served by CDSS and comply with CDSS data de-identification guidelines. Values of 1 to 10 and calculations based on values of 1 to 10 are masked (‘M’ or ‘*’). In stratified views of the data, additional values (the lowest available) are masked to prevent calculation of values 1 to 10.

Secure Site Features

This report may also be run with filters in place to restrict the data to various subgroups of other variables (e.g., Black subgroup of the Ethnicity variable, <1 year old subgroup of the Age variable, etc.). Please see Report Dimensions and Filters for details.

Report Dimensions and Filters

Child age is organized by the following intervals and derived from a child’s date of birth as captured in the CWS/CMS variable birth_dt:

  1. < 1 yr*
  2. 1-2 yrs
  3. 3-5 yrs
  4. 6-10 yrs
  5. 11-15 yrs
  6. 16-17 yrs
  7. 18-20 yrs OR 18-21 yrs**

The calculation of age is based on a date appropriate for the particular report. For example, the ‘Entries to Foster Care’ report calculates the age of children and youth on the day they enter foster care. ‘Exits from Foster Care’, on the other hand, calculates age at exit.

When Age Group is not selected as a row or column dimension, users may filter reports to include children of specific ages (e.g., only 3 year olds) or to create a different age category than those shown above (e.g., a group of 2-6 year olds). This is done by checking those age boxes for which report data are to be included.

* In some reports, this category is separated into infants with ages of less than one month (‘<1 mo’) and others less than one year old (i.e., ‘1-11 mo’).

** The available age range varies by report.

  • The Child and Family Services Review, round 3 (CFSR3) reports, for example, are restricted to children and youth less than 18 years old. The age groupings and available age filters reflect these ranges.
  • In some reports, ‘18-21 yrs’ is separated into ‘18 + 60 days’, ‘18 + > 60 days’, and ‘19-21 yrs’:
    • 18 + 60 days: Youth who are in the first 60 days of their 18th year
    • 18 + > 60 days: Other 18-year-old youth (i.e., older than 18 plus 60 days)
    • 19 to 21 year olds.

    These categories are intended to support analysis of the effects of the implementation of Assembly Bill 12 (AB12). AB12 allows youth to voluntarily remain in or reenter foster care. Youth exiting as part of the ‘18 + 60 days’ group are more likely to have experienced a conventional exit; those in the ‘18 + > 60 days’ and ‘19-21 yrs’ group are more likely to have made use of the AB12 changes.

  • In Transition-Age Youth Research & Evaluation Hub (TAY-Hub) reports, the age range is limited to youth in care age 18-20 or 18-21 years. In these cases, rather than presenting grouped ages, the ages 18, 19, 20, and 21 (if applicable), are presented as separate rows or columns. This approach is intended to allow a focus on older youth and to provide a detailed stratification by age.

Child ethnicity is collapsed into six groups based on 31 codes from the CWS/CMS variables p_ethnctyc and hisp_cd. Ethnic groups and codes are listed below:

  1. Black
    • Black (823)
    • Ethiopian (826)
  2. White
    • White (839)
    • White-Armenian (840)
    • White-Central American (841)
    • White-European (842)
    • White-Middle Eastern (843)
    • White-Romanian (844)
  3. Latino
    • Those children coded as Hispanic using the Hispanic Origin indicator (hisp_cd = ‘Y’), regardless of primary ethnicity (p_ethnctyc) selection
    • Hispanic (830)
    • Mexican (3164)
    • South American (3165)
    • Caribbean (3162)
    • Central American (3163)
  4. Asian/PI
    • Asian Indian (822)
    • Cambodian (824)
    • Chinese (825)
    • Filipino (827)
    • Guamanian (828)
    • Hawaiian (829)
    • Japanese (831)
    • Korean (832)
    • Laotian (833)
    • Other Asian/Pacific Islander (834)*
    • Other Asian (5922)
    • Other Pacific Islander (5923)
    • Hmong (835)
    • Polynesian (836)
    • Samoan (837)
    • Vietnamese (838)
  5. Native American
    • Alaskan Native (820)
    • American Indian (821)

*Code is inactive.

Note: In late 2017, CDSS provided instruction to counties to ask clients which of the federally recognized races (those marked with an asterisk in CWS/CMS) they identify with. Once a federally recognized race is selected, a secondary “Hispanic” ethnicity can be selected in Other Ethnicity. If the client does not identify with any federally recognized race, then workers were asked to select “Declines to State” as the primary race. However, the Hispanic Origin indicator should always be marked as “Yes” for children with Latino backgrounds in order to avoid accidental categorization into Missing.

When ethnicity is not selected as a dimension on the website, the default filter includes Missing values and all ethnicities. Users can also filter *DYNAMIC* reports to include only children of specific ethnic groups (e.g., only Native American and White children) by checking those ethnicities for which report data are to be included.

Census-based Ethnicity’ option (available on some reports) labels will read ‘Hispanic’ rather than ‘Latino’ for consistency with Census Bureau usage.

Ethnic groups for population data and rates reports are based on the California Department of Finance annual population projections.

  1. Black
  2. White
  3. Latino
  4. Asian/PI
  5. Native American
  6. Multi-Race

In the rates reports and disparity indices, the denominators–child population based on California Department of Finance data–may include children/youth in the multi-race category. However, the numerators–children with allegations, children with investigations, etc., based on CWS/CMS–always have null values for multi-race, since we do not construct a multi-race category from CWS/CMS data. Conversely, ethnicity may be missing in CWS/CMS and, in those cases, rates numerators are categorized as missing. Department of Finance population data does not include a missing category and, therefore, that row is always null for the denominators of the rates reports. Given these differences between the data sources and the resulting null values, no rates are calculated for the multi-race or missing rows.

See Population methodology for important details.

Note: For Transition Age Youth (TAY) reports, the category Missing is suppressed, therefore the total for this dimension will differ from the total for other dimensions in the same report.

The Census collects and reports data on race (Asian/Pacific Islander, Black, White, and Native American) separately from data on Hispanic status. One may report as being of more than one race. A person of any race can report as being Hispanic or not. The following table compares the standard CWS/CMS (Primary Ethnicity) method of reporting race and ethnicity with the Census method of reporting race and ethnicity. For our purposes, the important differences between these two methods are:

Primary Ethnicity Method Census Method
Data Source Race/Ethnicity is based on the CWS/CMS primary race/ethnicity variable and the CWS/CMS Hispanic status variable. CWS/CMS Secondary Ethnicity data is not utilized. In addition to the data used in the Primary Ethnicity Method, the Census Method uses CWS/CMS Secondary Race/Ethnicity variable.
Hispanic status Hispanic is treated as a race – One can be Asian/Pacific Islander, Black, White, Hispanic, Native American. If a person’s Hispanic status variable is yes, that person is reported as Hispanic irrespective of the race/ethnicity reported in the primary race/ethnicity variable. Hispanic status is reported separately from race. A Hispanic, or non-Hispanic person, is asked to report their race, i.e., Asian, Black, White, etc. Because most Hispanics do not report another race on CWS/CMS, the race of most Hispanics is missing.
Multiple Races Only the primary race/ethnicity (adjusted for Hispanic status) is reported, so nobody is reported as being of mixed race. Both primary and secondary races are considered. Persons reporting two or more races (e.g., Black primary race/ethnicity and White secondary race/ethnicity) are reported as being of mixed race.
Native Americans Only people who report a primary race/ethnicity of Native American and who do not identify as Hispanic are reported as Native Americans. A person is reported as Native American irrespective of Hispanic status. Native Americans of mixed race (e.g., White and Native American) are reported as being of mixed race.

In the Census format, race is missing for most Hispanic children because their only primary or secondary race/ethnicity reported on CWS/CMS is one in the Hispanic group (i.e., Hispanic, Mexican, Central American, etc.). Because secondary race/ethnicity is considered, the count of Hispanic children is somewhat greater in the Census Method than in the Primary Ethnicity Method.

The consideration of reported secondary race/ethnicities in the Census method results in some children being reported as being of mixed race. As a result the count of non-Hispanic Black, White and Asian/Pacific Islander children is lower in the Census Method than in the Primary Ethnicity Method.

Because many people who identify as Native Americans also identify as Hispanic, the Census method reports more Native Americans than the Primary Ethnicity method does. However, this increase is dampened because secondary race/ethnicity is considered and children with a Primary Race of Native American often report other secondary races, e.g., White.

The Census Method first creates two ethnic groups – Hispanic and non-Hispanic. A child is identified as Hispanic if the hisp_cd on the Client table is Yes or the p_ethnctyc on the Client table or the ethnctyc on the Client_SCP_Ethnicity table is one of the following:

  • Hispanic (830)
  • Mexican (3164)
  • South American (3165)
  • Caribbean (3162)
  • Central American (3163)

After Hispanic status has been determined, race is determined by collapsing the non-Hispanic race codes from the CWS/CMS variables p_ethnctyc on the Client table and ethnctyc on the Client_SCP_Ethnicity table into five race groups. Reporting of mixed race is determined as described above. Race/ethnic groups and codes are listed below:

  1. Black
    • Black (823)
    • Ethiopian (826)
  2. White
    • White (839)
    • White-Armenian (840)
    • White-Central American (841)
    • White-European (842)
    • White-Middle Eastern (843)
    • White-Romanian (844)
  3. Asian/PI
    • Asian Indian (822)
    • Cambodian (824)
    • Chinese (825)
    • Filipino (827)
    • Guamanian (828)
    • Hawaiian (829)
    • Japanese (831)
    • Korean (832)
    • Laotian (833)
    • Other Asian (5922)
    • Other Pacific Islander (5923)
    • Hmong (835)
    • Polynesian (836)
    • Samoan (837)
    • Vietnamese (838)
  4. Native American
    • Alaskan Native (820)
    • American Indian (821)
  5. Missing

Child sex at birth is based on the CWS/CMS variable gender_cd and categorized as Female (1), Male (2) or Intersex (3).

Children for whom Sex at Birth is not coded are reported as Missing.

When Sex at Birth is not selected as a dimension, the default filter includes Missing values and Female, Male and Intersex. Users can also filter reports to include only children of a specific gender (e.g., only Male children) by checking the sex at birth for which report data are to be included.

Note: For Transition Age Youth (TAY) reports, the categories Intersex and Missing are suppressed, therefore the total for this dimension will differ from the total for other dimensions in the same report.

Children receiving multiple allegations are categorized according to the severity hierarchy established by CWS/CMS. For example, when allegations of Physical Abuse, General Neglect and Emotional Abuse are entered for a specific child in a single report, only one allegation will be counted and it will fall under Physical Abuse since this type of abuse is the highest in the hierarchy. The severity hierarchy is listed below. CWS/CMS codes are drawn from the variable alg_tpc:

    1. Sexual Abuse (2181)
    2. Physical Abuse (2179)
    3. Severe Neglect (2180)
    4. General Neglect (2178)
    5. Exploitation (2177)
    6. Emotional Abuse (2176)
    7. Caretaker Absence/Incapacity (2169)
    8. At-Risk, Sibling Abused (5001)
    9. Substantial Risk (5624) *Category not currently active in CWS/CMS
    10. Missing

Users can subset referral reports by selecting the disposition type filter and choosing those dispositions to be included. Disposition types and codes are drawn from the CWS/CMS variable ,algdsp_t as listed below

  1. Substantiated (45)
  2. Inconclusive (47)
  3. Unfounded (46)
    Effective Q2 2011, the category ‘Assessment Only’ (children for whom a referral was received, but no disposition entered) is divided into two groups, using the CWS/CMS variable, Disposition Code (REFR_CLT.DISPTN_CD):
  4. Assessment Only/Evaluated Out (DISPTN_CD=’A’)
  5. Not Yet Determined (the remainder, where DISPTN_CD is not equal to ’A’)

We would expect the number of Not Yet Determined for a given interval to decrease over time, as determinations are made and entered into CWS/CMS.

Reporters are organized into the groups listed below. The CWS/CMS codes for each group are drawn from the variable col_relc:

  1. Family/Friend
    • Aunt (572)
    • Brother (573)
    • Cousin (577)
    • Friend (580)
    • Godparent (581)
    • Grandparent (583)
    • Great Aunt (584)
    • Great Grandparent (585)
    • Great Uncle (586)
    • Relative (596)
    • Sister (598)
    • Uncle (601)
    • Indian Custodian (6243)
    • Aunt (Maternal) (7282)
    • Aunt (Paternal) (7283)
    • Brother (Half) (7284)
    • Cousin (Maternal) (7285)
    • Cousin (Paternal) (7286)
    • Grandparent (Maternal) (7287)
    • Grandparent (Paternal) (7288)
    • Great Grandparent (Maternal) (7289)
    • Great Grandparent (Paternal) (7290)
    • Great Aunt (Maternal) (7291)
    • Great Aunt (Paternal) (7292)
    • Great Uncle (Maternal) (7293)
    • Great Uncle (Paternal) (7294)
    • Relative (Maternal) (7295)
    • Relative (Paternal) (7296)
    • Sister (Half) (7297)
    • Uncle (Maternal) (7298)
    • Uncle (Paternal) (7299)
  2. Neighbor
    • Neighbor (591)
  3. Law Enforcement/Legal
    • Attorney (571)
    • Law Enforcement (589)
    • Parole Officer (594)
    • Probation Officer (595)
  4. CASA/GAL
    • Child Advocate/Court Appointed Special Advocate (CASA) (574)
    • Guardian Ad Litem (GAL) (587)
  5. Counselor/Therapist (576)
  6. CWS Staff (578)
  7. Day Care/Fost Care
    • Day Care Staff (579)
    • Foster Care Staff (599)
  8. Medical
    • Medical Professional (590)
    • Dental Professional (5924)
  9. Education
    • School Personnel (597)
    • Teacher (600)
  10. Other Professional (593)
  11. Other (in CWS/CMS: all other codes)
  12. Missing

When reporter type is not selected as a dimension, the default filter includes Missing values and all categories. Users can also filter reports to include only certain reporters (e.g., only Medical and Education) by checking those categories for which report data are to be included.

Investigation indicates whether the allegation was part of a referral where investigation was either intended or completed.

The Referral Response Type (RFR_RSPC) indicates whether investigation is intended. Using the Referral Response Type allows identification of investigated referrals prior to the completion of the investigation. The Disposition Type (ALG_DSPC) indicates, at the allegation level, whether an investigation was completed and, if completed, its conclusion.

1. Investigated

Any one of the following qualifies the allegation for the ‘Investigated’ category:
Referral Response Type (RFR_RSPC)

  • 3 Day (1516)
  • 5 Day (1517)
  • 10 Day (1518)
  • Immediate (1520)

Disposition Type (ALG_DSPC)

  • Substantiated (45)
  • Unfounded (46)
  • Inconclusive (47)

That is, to be in the investigated group, either the Referral Response Type must equal 1516, 1517, 1518, or 1520 or the Disposition Type must equal 45, 46, or 47. In most instances, qualifying values for both variables will be present.

2. Not Investigated

The values that do not qualify the allegation for the ‘Investigated’ category are:
Referral Response Type (RFR_RSPC)

  • Evaluate Out (1519)
  • N/A Secondary Report (1521)

Disposition Type (ALG_DSPC)

  • [Not yet determined] (0)
  • Unknown at Conversion (5369)
  • Entered in Error (5918)

« Back to Methodology Index