Open AccessTake Me Back: Validating the Wayback Machine
Author(s) -
Murphy Jamie,
Hashim Noor Hazarina,
O’Connor Peter
Publication year - 2007
Publication title -
journal of computer‐mediated communication
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 4.15
H-Index - 119
ISSN - 1083-6101
DOI - 10.1111/j.1083-6101.2007.00386.x
Subject(s) - nomological network , face validity , predictive validity , convergent validity , incremental validity , external validity , psychology , computer science , content validity , criterion validity , test validity , social psychology , data science , knowledge management , psychometrics , machine learning , construct validity , developmental psychology , structural equation modeling , internal consistency
Although fields such as e‐commerce, information systems, and computer‐mediated communication (CMC) acknowledge the importance of validity, validating research tools or measures in these domains seems the exception rather than the rule. This article extends the concept of validation to one of an emerging genre of web‐based tools that provide new measures, the Wayback Machine (WM). Drawing in part on social science tests of validity, the study progresses from testing for and demonstrating the weakest form of validity, face validity, to the more demanding tests for content, predictive, and convergent validity. Finally, the study tests and shows nomological validity, using the diffusion of innovations theory. In line with prior diffusion research, the results of tests for predictive and nomological validity showed significant relationships with organizational characteristics and two WM measures: website age and number of updates. The results help validate these measures and demonstrate the utility of the WM for studying evolving website use.