Cookbook Specifications Help Solve Key Data Governance Issues
Even as the amount of student data continues to grow, and the amount of reporting also continues to grow, data governance models are as diverse as the academy itself. However, a few points seem to cause collective concern. For example, the CIO of SUNY at Buffalo, J. Brice Bible, writes in CIO Review, “… the data definitions and nomenclature used for [software and operating systems] and other college or department-based systems are not consistent and therefore difficult to integrate into shared reporting services.” As clients of the Cookbook, however, these challenges are addressed directly; as Bible states, “The process of populating Data Cookbook with agreed-upon definitions has created visibility into the value and on-going need for a strong Data Governance of campus.”
Data governance issues stem from several factors, one being clear communication of terms and their meaning(s) across different contexts. The Cookbook offers a number of features to compare definitions and specifications in the public community, along with features to align reporting needs with clear communicative outputs. The problem of definitional clarity is similarly elaborated in a recent EDUCAUSE white paper, The Compelling Case for Data Governance, which outlines two key areas: official versus ad hoc data definitions and operational uncertainty.
Official versus ad hoc data definitions means that schools should have clear procedures to “establish the mechanisms by which the official definition is derived, communicated, and accessed by all constituents…” Sometimes schools have internal requests for reports that require a customized or irregular analysis of the data, and the way certain terms are defined can become muddled, particularly as the defined points differ from an officially-defined term. The problem compounds when making a distinction between definitions used externally and those used internally. The authors of this white paper support having clear flexibility to allow for maximum use of student data, while maintaining the coherence of term meaning.
Operational uncertainty occurs when “multiple versions of the same data” are reported to the administration and subsequent “lack of trust and confidence” can hinder data-driven decision-making. Furthermore, the problem compounds when external reporting faces the same unreliability.
Both issues deal with the integrity of the data, from how it is defined to when it should be used. Student data, which can help in institutional decision-making, stands to lose some of its power here, which is unfortunate when the data can yield important insights. The Cookbook specification feature can help alleviate the integrity issue. With room to spell out instances when the official definition should be used, and when it is permissible to use an alternative definition, a data specification details the rationale to communicate across various stakeholders. Similarly, a specification can communicate how “multiple versions of the same data” hang together so that when a report needs to be generated, stakeholders know the parameters. In short, the use of specifications address problems of data integrity directly.
James E. Willis, III, Ph.D.
IData Content Curator