Management Fashion: Lifecycles, Triggers, and Collective Learning Processes

Administrative Science Quarterly, Dec, 1999 by Eric Abrahamson, Gregory Fairchild

Four major employee-management fashion waves rose and fell in this period: job enrichment, quality circles, total quality management, and business process reengineering (Cole, 1979, 1989). We focused on the second major employee-management fashion from this period, quality circles, so that we could examine whether its lifecycle was influenced by the previous employee-management fashion, job enrichment. Practical considerations also drove us to focus on quality circles, because we could obtain data on a moderately sized, full wave of discourse, in computer-readable format, making it possible both to read all this discourse and to use computer assisted text analysis to analyze its evolution. This would have been impossible had we focused on earlier fashions, such as job enrichment, that antedate the computerization of articles' texts, or later fashions, such as total quality management or business process reengineering, that are too big and have not yet undergone a full fashion cycle.

Data and Analysis

Lifecycles. In question #1, we asked about the lifecycle of management discourse about a management fashion. In terms of quality circles, we ask how and why the volume of discourse that management-knowledge entrepreneurs used to promote quality circles rose and fell following a particular pattern. To measure the volume of discourse, we counted the number of articles listed under the quality circle subject heading in ABI Inform from 1977, when the first quality circle article appears in this database, to 1995 (see Abrahamson, 1989, 1997; Barley and Kunda, 1992; Shenhav, 1995 for similar approaches). We used subject headings for two reasons. First, we compared this search strategy to a second strategy by which we searched electronically the titles and abstracts of all ABI Inform articles for word strings denoting quality circles: "QC(s)," "QCC(s)," "quality control circle(s)," and "quality circle(s)." These searches yielded a second set of articles. There was an 88.7 percent overlap between this set and the set obtained with our first search strategy. The remaining 11.3 percent of articles usually pertained to different subjects. Second, we learned from conversations with several ABI Inform content analysts that ABI Inform's assignment of articles to subject headings, like the "quality circle" heading, is accomplished and checked by trained and experienced content-analysts, working in industry- and function-specific teams.

The total number of articles indexed in ABI Inform increased markedly from 1974 to 1995. To adjust for this growth, we multiplied the number of quality circle articles in any one year by the ratio between the total number of articles indexed in 1984 and the total number of articles indexed that year. This adjustment technique is analogous to the technique used by economists to transform nominal into real currency amounts, thereby factoring out the effect of inflation. We used a similar adjustment for every other article count in this study.