Evaluating Quality of Official Statistics

• Activity Number: 541
• Type: Invited
• Date/Time: Wednesday, August 6, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Government Statistics Section
• Author(s): Ron S. Kenett*+
• Companies: KPA, Israel and University of Turin
• Keywords: Information quality ; InfoQ ; integrated data analysis ; student reports ; actionable outcomes

Abstract

In a JRSS(A) paper, Kenett and Shmueli define the concept of Information Quality (InfoQ) as the potential of a dataset to achieve a specific goal using a given empirical analysis method. InfoQ has eight dimensions: Data Resolution, Data Structure, Data Integration, Temporal Relevance, Generalizability, Chronology of Data and Goal, Operationalization and Communication. The talk will present InfoQ (what it is and how it is achieved) through several case studies, including student test reports produced by school systems in the US. Specifically, the National Education Goals Panel (NEGP) encouraged states to determine, from student reports, what types of skills or knowledge a student has, to compare him to other students in the school, district, state, or nation, and identify ways to help the child improve. In fact, NEGP is asking test reports to provide high InfoQ to parents and educators. The talk will compare and contrast the InfoQ dimensions with those typically used by statistical agencies to assess the quality of their products, and discuss how the InfoQ approach can be used to optimize integrated data analysis combining official data with organizational data.