Doughty, Chase and O'Shields (2004) reviewed the Precision Teaching literature which they point out claimed that Precision Teaching is superior in producing learning outcomes and they posed concerns about the lack of well designed research to support these claims. One issue of concern was in much of the research Precision Teaching involved more practice than the teaching method to which it was compared. This present research aimed to compare the effects of fast practice and slow practice on learning three sets of 20 statistics definitions by tertiary students using a within-subject design. Following Doughty et al.'s (2004) suggestions, the amount of practice of the three sets of definitions was matched. All three sets were learned to some degree of accuracy before practice started. One set was practiced fast and another set was practices slowly and accurately over the same time period and until the fast practice set met a rate aim of 30 correct per min, the third set was practiced slowly and accurately for the same number of trials after this. There were non-timed tests of all sets of definitions prior to the start of any practice, after the rate aim was achieved, after the second slow practice finished and four weeks later. The amount of feedback for fast practice and the first slow and accurate practice was matched, but less feedback was provided for the second slow practice set. Three of the initial eight participants completed all experimental conditions. Results showed that response latencies to complete a response became shorted with both types of practice over practice but were shorter for the fast practiced items than for the slowly practiced items during the practice periods. The tests showed that accuracy was high after the extended practice and after four weeks no practice, regardless the method of practiced. The types and amount of feedback did not appear to have any effect on the learning outcome. Limitations that prevented a firm conclusion were discussed.