Home > Federal & State Reporting Systems > AFCARS Resources > Technical Bulletins > Technical Bulletin #9

IV. Consistency Check Groups - Foster Care

TABLE OF FOSTER CARE GROUP CHECKS:

Consistency checks are done by evaluating the data in a group of related elements, or in the case of race elements, among the race categories of a given race element. If there is an inconsistency among them, it determines which element is the most likely cause of the inconsistency. In general, the one element that could be changed to make the relationship among the elements in the group consistent is determined to be the most likely error. However, this rule has several exceptions based on experience and input from several sources.

Each of the consistency check groups are given below with the principle of the check, the conditions under which the check is not done, and a table of the possible relationships between the elements and which element(s), if any, will be marked as errors. If the element in this last column appears in bold-italics with asterisks around it (e.g., ** #19 **), this means that the data in the element would be deleted from that record by the Federal system if the file were to be submitted to AFCARS.

Note: if the "Element(s) to be marked" column says need more information, the DQU checks to see if any of the elements in the group have already been marked as the most likely error in an earlier check. If this is the case, it is assumed that the element is the likely error for the current check and no further action is taken. However, if none of the elements have failed an earlier check, the DQU cannot determine the most likely error and all elements in the group will be marked.

It is also important to note that while the DQU may continue to use the data in an element even if it has already been marked as incorrect, the DQU never overwrites one error with another. For example, if an element is marked as out of range it may still be used in certain consistency checks. However, if it is determined to be the most likely cause of the inconsistency, it will remain marked as out of range. The same would hold true for an element that is involved in more than one consistency check group. An element identified as the most likely cause of more than one inconsistency will always be marked as the most likely error in the first group it failed.

Foster Care Group #1 (marked as "a" or "A" in the Detailed Data Quality report)

Elements Involved:

Conditions under Which Check is Not Performed:

Consistency checks between two date elements are not performed if either element is blank or out of range.

Principle of Check:

All dates should be logically ordered. For example, the Discharge Date must come after the Latest Removal Date. Also, the length of time between dates should be within a reasonable range. For example, the Child's Birth Year should not be more than 21 years before his or her latest removal date. This check identifies the date(s) that appear to be "out of place."

Special Notes:

Group #1 is performed differently than the other groups. In order to determine which element(s) is out of place, the DQU compares every pair of date elements (there are sixteen foster care date elements, thus there are 120 pairs of elements checked). If a pair of elements is inconsistent according to a given standard, it is difficult to tell which one is wrong. To address this problem, the DQU maintains two tables. One table maintains information on which check failed (i.e., pairs of date elements were inconsistent); the other table maintains information on the number of times each date element was involved in a check that failed.

After all pairs of date elements have been evaluated, the DQU looks at the table that maintains the information on which checks failed. If a check failed between two elements, the DQU looks at the other table to see which of the two elements had the most errors. The one with the most errors is marked with an "a." If both elements had an equivalent number of errors, both of them are marked with an "a."

The tables below indicate every pair of date elements and the relationship that is checked for them. If a relationship exists and the Error Status column says "Error," then the table that maintains the information on which checks failed will be updated for that pair of elements and the table that maintains the number of errors for each element will be increased by one for each element in the "Involved Elements" column.

Table 3. Foster Care Consistency Group #1.2

Table 4. Foster Care Consistency Group #1.5

Table 5. Foster Care Consistency Group #1.6

Table 6. Foster Care Consistency Group #1.18

Table 7. Foster Care Consistency Group #1.20

Table 8. Foster Care Consistency Group #1.21

Table 9. Foster Care Consistency Group #1.22

Table 10. Foster Care Consistency Group #1.23

Table 11. Foster Care Consistency Group #1.45

Table 12. Foster Care Consistency Group #1.46

Table 13. Foster Care Consistency Group #1.47

Table 14. Foster Care Consistency Group #1.48

Table 15. Foster Care Consistency Group #1.50

Table 16. Foster Care Consistency Group #1.51

Table 17. Foster Care Consistency Group #1.56

Foster Care Group #2: (marked as "b" or "B" in the Detailed Data Quality report)

Elements Involved:

#10 Has the Child Been Clinically Diagnosed with a Disability
#11 Mental Retardation
#12 Visually/Hearing Impaired
#13 Physically Disabled
#14 Emotionally Disturbed
#15 Other Medically Diagnosed Condition
#33 Condition Associated with Removal - Child Disability

Conditions under Which Check is Not Performed:

1) if #10 is blank or out of range
2) if #11 is out of range
3) if #12 is out of range
4) if #13 is out of range
5) if #14 is out of range
6) if #15 is out of range

Principle of Check:

If one of the conditions associated with the removal of the child was that he/she has a medically diagnosed disability, then the answer to the question "Has the child been clinically diagnosed with a disability?" should be "yes" and at least one of the disabilities (#11-15) should be identified. Likewise, if a child is indicated as having a type of disability (such as visual or hearing impairment) the answer to "Has the child been clinically diagnosed with a disability" should be "yes". However, even though a child has been clinically diagnosed with a disability does not mean that it was a condition associated with removal.

Table 18. Foster Care Consistency Group #2

Foster Care Group #3: (marked as "c" or "C" in the Detailed Data Quality report)

Elements Involved:

#16 Has the Child Been Previously Adopted
#17 Age at that Adoption

Conditions under Which Check is Not Performed:

1) if #16 is out of range
2)if #17 is out of range

Principle of Check:

If a child is known to have been adopted before, then the child's age at that adoption should be reported. Likewise, if a child's age at a previous adoption is known, the answer to the question "Has the child been previously adopted?" should be "yes".

Table 19. Foster Care Consistency Group #3

Foster Care Group #4: (marked as "d" or "D" in the Detailed Data Quality report)

Elements Involved:

#18 Date of First Removal from Home
#19 Total Number of Removals from Home
#20 Date of Discharge from Previous Episode
#21 Date of Latest Removal from Home

Conditions under Which Check is Not Performed:

1) if #19 is blank or out of range
2) if any two of #18, 20, and 21 are blank or out of range

Principle of Check:

If a child is in his or her first episode, the first removal date and the latest removal date should be the same and there should be no previous discharge date. If the child is in his or her second or more episode, the first removal date should be equal or prior to the previous discharge date which, in turn, should be equal or prior to the date of latest removal.

Table 20. Foster Care Consistency Group #4

Foster Care Group #5: (marked as "e" or "E" in the Detailed Data Quality report)

Elements Involved:

#21 Date of Latest Removal from Home
#23 Date of Placement in Current Setting
#24 Total Number of Placements in the Current Episode

Conditions under Which Check is Not Performed:

1) if #21 is either blank or out of range
2) if #23 is either blank or out of range
3) if #24 is either blank or out of range

Principle of Check:

If the child is in his or her first placement, the date of placement will most likely be the same as the date of latest removal. If the child is in his or her second or more placement, the date of placement should most likely be after the date of latest removal.

Table 21. Foster Care Consistency Group #5

Foster Care Group #6: (marked as "f" or "F" in the Detailed Data Quality report)

Elements Involved:

#02 Report Period End Date
#05 Date of Most Recent Periodic Review
#21 Date of Latest Removal from Home
#25 Manner of Removal from Home
#43 Most Recent Case Plan Goal
#56 Date of Discharge

Conditions under Which Check is Not Performed:

1) if #21 is either blank or out of range
2) if #56 < #21

Principle of Check:

If the child has been in care for longer than 30 days or if there is a valid Date of Most Recent Periodic Review then the Manner of Removal cannot be "not yet determined" and the Case Plan Goal cannot be "not yet established."

A child is considered to be in care for longer than 30 days if the Discharge Date is more than 30 days after the Date of Latest Removal. If there is no Discharge Date, then the Report Period End Date is used.

A Review Date is considered to be valid if it occurred on or after the Date of Latest Removal and on or before the Discharge Date, if one is entered.

Table 22. Foster Care Consistency Group #6