The efficacy of matching information systems development methodologies with application characteristics - an empirical study

An experimental study was conducted to determine whether a data-centered sy stem development methodology produced system designs superior to those from a process-centered methodology when applied to a data-intensive system pro blem. Experimental subjects (N= 30) were advanced undergraduate systems ana lysis students who were given standardization training in system developmen t techniques and then randomly divided into two groups. One group was taugh t to use a data-centered methodology and the other a process-centered metho dology. Both groups then "solved" a system design problem that was intentio nally designed to be data-intensive. The resulting designs were systematica lly scored for quality using a panel of system development experts. Results showed that the system designs produced by the data-centered group were no t significantly better (P < 0.05) than those produced by the process-center ed group. The quality of the system designs by both of the methodology assi sted groups were found to be significantly superior (p < 0.01) to those pro duced by the group using no methodology whatsoever. This suggests that the development of system development methodologies that are geared toward spec ific application development technologies (hypertext, object-oriented, or r apid prototyping, for example) may not have merit and that further investme nt in "strongly typed" development methodologies may not be productive. (C) 1999 Elsevier Science Inc. All rights reserved.