Statistical power in information systems research: Application and impact on the discipline

Scholars agree that Information Systems (IS) is a young but maturing discipline. To be a mature discipline, a portion of the research community believes it is essential to use a higher ratio of empirical studies that use inferential statistics. When hypothesis testing is used, statistical power is a measure of the ability of a statistical test to correctly detect significant effects among research variables. The current study examined 167 statistical tests for 65 information systems (IS) studies published in MLS Quarterly from 1990-1997 to determine effect sizes and corresponding power levels. Previous studies of information systems research had determined that power levels used in the IS discipline lagged behind its referent disciplines. This study of current IS research confirmed an increase in power levels in the last decade, implying that IS scholars are using inferential statistics more effectively, enhancing the discipline's maturity.