A Discrepancy-Based Measurement Approach for Data Integration

Abstract

This paper presents a measurement instrument for diagnosing the level of data integrationat the organization and organization subunit levels. Incorporating consideration forspecific theoretical properties related to data integration into its design, the measurementinstrument relies on a discrepancy (or gap) measure to adequately account for theseproperties. Data from forty-eight organizations of the Group Insurance industry are usedto test the reliability and validity of the measurement instrument. The results suggest thatthe discrepancy measure, based on the difference between respondents' ideal (i.e.normative) and actual estimates of data integration, appears to be a valid indicator of dataintegration. However the efficacy of using a discrepancy measure, over the simpler useof its parts, must be considered in light of practical and theoretical considerations.