To evaluate test speededness and to derive implications for test program activity, this study reviewed the literature on speed and power, identifying four major approaches to the assessment of speed: the Gulliksen approach, the Cronbach and Warrington approach, the Stafford approach, and the approach of the Educational Testing Service (ETS) as described by Swineford. Two new models were formulated: the Reilly-Donlon approach and the biserial methods. These new methods provide correlations that are analogous to those of Cronbach and Warrington but derivable from a single, conventional administration of a test. The new methods depend centrally on the assumption of normality in the rates of work. Evidence from ETS program files in connection with this hypothesis was evaluated. Methods for estimating the parameters of the distribution of rates of work by graphic methods and the use of normal probability paper were studied. Extensions of these methods to two practical problems were explored: (1) The estimation of time available for review as an ancillary consequence of test speededness and (2) the estimation of the time demands of individual passages and items. The potential implications of omissions for the evaluation of test speededness were demonstrated. (Author/PN)