Big Data Analytics Capability and Organizational Agility: An Empirical Research Model

Abstract

Big data analytics capability (BDAC) is widely claimed to improve organisational agility, but the mechanism through which it does so, the conditions under which the effect is stronger, and the sample sizes needed to test these claims are seldom specified before data are collected. This paper develops an empirical research model in which BDAC affects agility both directly and indirectly through data-driven decision-making capability (DMC), with environmental dynamism moderating the direct path. The model is grounded in the resource-based and dynamic-capabilities perspectives and in prior empirical work on information-technology capability and agility. Rather than reporting field estimates, the paper validates the model’s testability through a simulation harness: survey-type data are generated from the hypothesised structural relationships with indicator reliability of 0.80, and the mediation and moderation analyses are run on samples of 100 to 500 firms. The indirect effect is detected with bootstrap confidence intervals in 98% of samples at N = 100 and in every sample at N ≥ 200, while a worked example at N = 400 recovers the direct, indirect and interaction effects with the expected attenuation from measurement error (observed coefficients of 0.36, 0.46 × 0.27 and 0.10 against design values of 0.30, 0.50 × 0.40 and 0.20). The paper specifies measurement, sampling, analysis and robustness procedures for a field study and argues that simulation-based pre-validation should become standard practice for empirical models in information-systems and strategy research.