Big data and its epistemology

Abstract

The article considers whether Big Data, in the form of data‐driven science, will enable the discovery, or appraisal, of universal scientific theories, instrumentalist tools, or inductive inferences. It points out, initially, that such aspirations are similar to the now‐discredited inductivist approach to science. On the positive side, Big Data may permit larger sample sizes, cheaper and more extensive testing of theories, and the continuous assessment of theories. On the negative side, data‐driven science encourages passive data collection, as opposed to experimentation and testing, and hornswoggling (“unsound statistical fiddling”). The roles of theory and data in inductive algorithms, statistical modeling, and scientific discoveries are analyzed, and it is argued that theory is needed at every turn. Data‐driven science is a chimera.

Big Data in Organizations and the Role of Human Resource Management

“…Knowledge is generated by means of big data in various ways and there may be an end to a certain type of theory, but big data calls for entirely new types of theories (Boellstorff 2015, Tokhi & Rauh 2015. Generating hypotheses and then collecting data may no longer be feasible, since the required data are already available (Frické 2015). In the conclusion of his paper, Frické counters Andersons's claim, demanding focus on thorough and precise scientific work rather than deleting theory and using only big data:…”

“…This is, however, not the case, since "there is no automatic technique for turning correlation into causation" (Spiegelhalter 2014: 264). Statistics are, generally speaking, a process that takes place post hoc (Frické 2015). There are several statistical biases that potentially distort any data set in one direction or another.…”

“…There is a chance that a hypothesis could be falsely rejected (type I error) or falsely not rejected (type II error), as shown in Table 5. (Sheskin 2004: 54) Although the threshold value of .05 seems rigorous, there is a chance that out of 100 independent tests, approximately five tests are either false positives or false negatives (Frické 2015). Such an example simplifies the problem too drastically.…”

The Analysis of Agent’s Knowledge Based on Game Behavior

“…Data is an unprocessed material which can show the objective existence of things and be perceived by agent, such as graphic symbol, number, letter, sound and so on [12,13].…”

A Framework for the Assessment of Research and Its Impacts

“…Are we moving toward a data-driven science (Kitchin, 2014), supporting "the end of theory" (Anderson, 2008), or will theory-driven scientific discoveries remain unavoidable (Frické, 2015)? There is little agreement in the literature.…”