All differential item functioning (DIF) methods require at least a moderate sample size for effective DIF detection. Samples that are less than 200 pose a challenge for DIF analysis. Smoothing can improve upon the estimation of the population distribution by preserving major features of an observed frequency distribution while eliminating the…

The impact of quality assurance procedures on the correct award of Basic Educational Opportunity Grants (BEOGs) for 1979-1980 was assessed, and a model for detecting error-prone applications early in processing was developed. The Bureau of Student Financial Aid introduced new comments into the edit system in 1979 and expanded the pre-established…

An evaluation of existing and proposed mechanisms to ensure data accuracy for the Pell Grant program is reported, and recommendations for efficient detection of fraud and error in the program are offered. One study objective was to examine the existing system of pre-established criteria (PEC), which are validation criteria that select students on…

The development of the error prone model (EPM) for the 1984-1985 student financial aid validation criteria for Pell Grant recipient selection is discussed, based on a comparison of the 1983-1984 EPM criteria and a newly estimated EPM. Procedures/assumptions on which the new EPM was based include: a sample of 1982-1983 Pell Grant recipients…

The development of a number of error-prone models to select Pell Grant recipients for validation is discussed. The 1983-1984 Pell Grant validation strategy consists of a two-stage approach: selection using Pre-Established Criteria (PEC) followed by selection using Error Prone Modeling (EPM). The database used for model development consists of a…