Predictors of Successful Arcade Machines

Abstract

Amusement Arcade Machines in a commercial location were measured for various attributes by the author using semantic differential and objective scales. The ability of each scale to predict the actual revenue taken in by each machine was assessed using a series of univariate regression analyses. The resulting profile of predictive capabilities can be used to aid selection of critical design criteria. A multiple linear regression was used to generate recommendations for the selection of arcade machines for purchase. Over 50% of the variability in machine success was predicted by scales measuring the apparent age of the machine, the vividness of the fantasy involved in the game, amount of movement on the display, and the extent to which the player was in control of the game. The implications of the findings for machine design and selection are discussed. A Factor Analysis of the rating scales yielded five underlying dimensions which were interpreted as Complexity, Type of Fantasy, User Involvement/Control, Novelty, and Loudness.