The Construction of Data Quality Measurement Scale ─Semiconductor-Related Industry Data Consumer’s Viewpoint

• Main Authors: Ching-Yu Lu, 呂靜喻
• Other Authors: Bertrand Miao-Tsong Lin
• Format: Others
• Language: zh-TW
• Published: 2000
• Online Access: http://ndltd.ncl.edu.tw/handle/84181761792443732186

碩士 === 銘傳大學 === 資訊管理研究所 === 88 === Although the quality of data is recognized as crucial in the information age and the topic of data quality is gaining more and more research efforts, most of previous studies focus on how to ensure the quality of the data of concern. To the best of our knowledge, none work has been conducted from the data consumer’s viewpoint. As a consequence, in this study we construct a data quality measurement scale based on consumer’s viewpoint. As the semiconductor-related industry is one of the most successful and promising economic activities in Taiwan, we confine our study to data consumers in this area. In this study we first congregate some data quality attributes that have been proposed in the literature, and then develop a data quality scale. Through an empirical survey, purification of the first stage scale, and verification of reliability and validity of the first stage scale, a scale for the second stage is produced. The stage-two scale is further processed by purification and verification. Then, we develop a more reliable and parsimonious data quality scale. Through first stage purification and verification, we identify 39 items related to seven dimensions, namely presentation and flexibility, reliability, manipulation, completeness, cost effectiveness and security, format, and component. During the second stage purification and verification, we identify 20 items related to three dimensions, namely reliability, usability, and completeness. With the scale attained as for reference basis, system analysts may develop information systems conforming to data consumer’s needs. On the other hand, data consumers can use this scale to measure and evaluate the information systems they are working with. Aligning data quality with user and/or organization’s needs will strengthen the competing edge.