Secondary Data Analyses From Published Descriptive Statistics

Abstract

The Problem

The need for human resource development (HRD) to develop theory that will lead to improved practice was recognized in published literature in 2002. Two years after Holton’s initial mandate for theory in HRD, Torraco identified the need for researchers to “interpret old data in new ways” as a core role for theory development in HRD research. Yet over a decade later, few scholarly articles published in journals dedicated to HRD report on secondary data analyses (i.e., interpret old data in new ways). Even fewer have considered how quantitative analysis of secondary data can be used to inform theory.

The Solution

A three-pronged solution is provided to address the void of secondary data analyses reported in HRD literature. First, this article demonstrates how secondary data analyses can be conducted using only published descriptive statistics as data. Second, it presents user-friendly software to aid scholars and scholar-practitioners in their pursuit of such analyses. Third, it reviews the concept of the general linear model (GLM) as the technique presented is limited to supporting quantitative statistical analyses that fall within the GLM.

The Stakeholders

Stakeholders include scholars and scholar-practitioners who are interested in conducting quantitative data analyses to inform HRD theory. In addition, scholars and scholar-practitioners who conduct scholarly reviews of academic literature will see why it is important for authors to comply with reporting guidelines that call for relevant descriptive statistics such as means, standard deviations, and correlation matrices to be included in quantitative research reports.