Data Reliability Audit Process for Auditing States Less Frequently than Annually
DEAR COLLEAGUE LETTER
DATE: January 21, 2004
TO: ALL IV-D DIRECTORS
RE: Data Reliability Audit Process for Auditing States Less Frequently than Annually
As you may know, OCSE will, for the first time since the passage of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), implement provisions in the statute and regulations that permit auditing states’ data reliability on a cycle less frequently than annually. For the 2003 audit period, 12 states have been identified as not needing a Data Reliability Audit (DRA). Three of the 12 states will be excused from a DRA audit for two years, provided they do not fail a performance standard in the interim. The primary reason we are taking these steps is that these states have demonstrated the ability to report consistently reliable data over a period of at least two years and have passed all performance standards.
The criteria used in making these determinations are complex and detailed, because the problems still existing among the states with unreliable data are varied, as are the requirements in the audit regulations. In general, however, states that have passed three consecutive DRA audits and met all performance standards will be audited once every three years. States that have passed two consecutive DRA audits and met all performance standards will be audited in two years. If these states subsequently pass the next DRA audit two years hence and meet all performance standards, they will qualify for the once-every-three-years audit cycle. Credit is given for marginal performance but only if the marginal performance occurred in the first year of the two or three-year period used to exclude a state from the annual audit cycle. Marginal performance is performance that has not met the DRA 95 percent standard for reliable data, but failure to achieve this threshold cannot be proven with statistical precision. States that are exempt from annual audits will go back into the annual audit pool during the next audit cycle if they fail a performance standard or if, for their next scheduled audit period, they fail to report reliable data or achieve only marginal performance on any line which is evaluated for data reliability.
States that are excused from an audit are still required to provide OCSE’s Office of Audit with audit trails to support the data being reported for incentive purposes as well as a management assertion that the system used to report the data has not undergone significant change since the last audit. These requirements are necessary to help ensure and protect the integrity and reliability of the data and the incentive payment system. The audit staff will perform a limited Data Reliability Review (DRR) or desk audit of the audit trails, primarily to ensure that the audit trails match the data being reported. The work does not entail on-site review work or a review of cases. States will receive a letter summarizing the results of the DRR if there are no problems. A state exempt from annual audit could be subject to the DRA in the same year as the DRR if the DRR discloses serious problems between the audit trails and the data or there is some other irregularity about the data being reported, such as a sudden, major change that cannot be explained.
Our goal in establishing the above framework for excluding states from annual audits is first and foremost to preserve and protect the data reliability and incentive payment system, which is gaining recognition as a model for measuring program results in the Federal government. We also want to reward the hard work and success states have achieved in establishing systems that both achieve performance levels and report reliable data on a consistent basis. Consistency is the key. Passing the DRA in one year but achieving only marginal or unreliable performance in the next is not a hallmark of a reliable data reporting system, which is why no credit is given for passing only one year.
Since we started with this process of rewarding performance based on reliable data, we have seen tremendous progress in the quality of the data and data reporting systems. The number of states failing the DRA for at least one measure decreased by half, from 25 in FY 2001 to 12 in the last audit. I encourage states that have reliable systems no longer subject to an annual audit to continue to closely monitor results. It is important that you maintain and continue to improve your data and procedures to ensure that you continue to meet program goals and enjoy the audit exemption. I encourage the remaining states to work even harder on their data reporting process to establish a consistent pattern of reporting reliable data. Our goal is to reach a point at which all states are on a three-year annual cycle. Achieving this goal is important to us in OCSE as well, because it means that we have reliable data nationwide. It also frees up audit resources for other important responsibilities mandated by law, such as administrative cost audits.
Many of you have already heard from your local Area Audit Office regarding your audit cycle. If not, the auditors will soon be contacting you to explain this process. Please do not hesitate to call them if you have any questions or concerns.
I am proud of the data reliability and incentive payment system, and of our accomplishments during the last audit period. I hope we will see even greater results with more states earning an exemption from the annual audit cycle during the current audit period.
Sherri Z. Heller, Ed.D.
Office of Child Support Enforcement
cc: ACF Regional Administrators
ACF/CSE Program Managers