Resumen
The literature on research evaluation has noted important differences in citation time patterns between disciplines, high and low ranked journals and types of publications. Delays in the receipt of citations suggest that the diffusion of knowledge following discovery is slower and given the passage of time the research contribution may be less valuable. This paper provides a framework for the comparison of different citation time patterns. Using principles drawn from the literature on stochastic dominance we show that comparisons of time patterns can be based on very general characteristics of cost of delay functions. When a particular function is used to represent the cost of delay, the magnitude of the impact of differences in citation time patterns can be assessed using simple exponential discounting. We demonstrate the application of this framework in assessing different citation time patterns by applying it to comparisons of 10-year citation records for: leading journals in economics, different business subject areas, journals in economics compared with those in neuroscience and the research output of individual economists.