Testimony Measure: Verification of Performance Data Could Be Improved
OIG-10-1: Nov 18, 2009
Tonya R. Ford
This is a publication by GAO's Inspector General that concerns internal GAO operations. GAO's mission is to support Congress in meeting its constitutional responsibilities and to help improve the performance and ensure the accountability of the federal government for the benefit of the American people. To accomplish its mission, GAO provides objective and reliable information and informed analysis to Congress, federal agencies, and the public. To help monitor its performance in serving Congress, GAO uses the testimony performance measure. On occasion, more than one GAO official may testify at the same hearing. Performance data for this measure comes from GAO's Congressional Hearing System, which is managed by the Office of Congressional Relations. Data is entered into the system by administrative staff from (1) organizational units (generally referred to as mission teams) whose officials are testifying and (2) the Office of Congressional Relations under limited circumstances. The Congressional Hearing System captures information, for example, about the date and topic of the hearing, the name of the committee or subcommittee holding the hearing, and the name of the GAO witness. The Government Performance and Results Act of 1993 (GPRA) requires federal agencies to develop strategic plans with long-term, outcome-oriented goals and objectives; annual goals linked to the long-term goals; and annual reports on the results achieved. As a legislative branch agency, GAO is not required to comply with GPRA but generally does follow the intent of this law. To follow the intent of GPRA, GAO uses a strategic planning and management process based on strategic goals and objectives. To help monitor its performance, the agency has developed various performance measures, including the testimony measure.
We identified a total of eight errors in GAO's fiscal year 2008 testimony performance data. From our random sample of 54 hearings, we found six errors. Five of the six errors resulted from 5 hearings--at which two GAO officials provided separate testimonies--being counted twice (in this case, as 10 hearings) in the Congressional Hearing System. The sixth error from our sample was the result of a planned testimony being issued as a statement for the record and the Congressional Hearing System's data not being updated to reflect this change. Statements for the record are not to be included in the testimony measure. Separate from our sample, we learned about two other errors. The first error involved a January 2008 testimony that was not entered into the agency's hearings database in a timely manner and thus was not included in the total number of hearings reported for fiscal year 2008. The second error was identified by the Office of Congressional Relations' review of fiscal year 2008 testimony performance data. The error involved a January 2008 hearing, where two GAO officials provided separate testimonies, that was inadvertently entered twice into the Congressional Hearing System--and thus was counted as two hearings. Office of Congressional Relations managers state they have corrected the eight errors. Our work shows that several factors contributed to inaccurate fiscal year 2008 performance data for GAO's testimony measure. First, while the Office of Congressional Relations has internal procedures that in part address maintaining the Congressional Hearing System database and producing related hearing reports, these procedures do not specifically mention the testimony measure, its definition, or the performance data. The procedures also do not include specific steps for checking and testing the performance data before it is reported in GAO's annual performance report. Second, while Office of Congressional Relations procedures stress the importance of updating Congressional Hearing System data when changes occur, one of the errors we identified happened because data had not been updated when a planned testimony was issued as a statement for the record. Third, unit administrative staff who enter data in the Congressional Hearing System told us that it is difficult for them to know when more than one GAO testimony will be delivered at a hearing. Finally, a report used to check GAO hearing data was not effective in identifying data errors related to hearings being counted twice. We believe the actions initiated by the Office of Congressional Relations--updating its procedures and developing a new report from the Congressional Hearing System--will help reduce the risk of future errors in GAO testimony performance data. As procedures are updated, we believe consideration should be given to adding specific steps that address the three types of data errors identified in our review: (1) the miscounting of hearings where more than one GAO testimony is presented, (2) not properly updating hearing data when a planned testimony is not presented, and (3) untimely posting of data. In addition, we believe that steps are needed to verify the accuracy of testimony performance data prior to their publication in GAO's annual performance report. Please review the full report for a list of Inspector General recommendations.