COMPARISON OF TWO PREDICTIVE MODELS OF A LAW ENFORCEMENT OFFICER PHYSICAL AGILITY CRITERION TASK TEST

Abstract

PURPOSE The purpose of this study was to compare the relative efficiency of two predictive models of a law enforcement officer physical agility criterion task test. METHOD Data were gathered on 121 male law enforcement officer recruits (mean±SD age = 26.9±5.1 yrs) at the Nebraska Law Enforcement Training Center who participated in mandatory fitness training (1h•d−1, 3 d•wk−1) as part of a 12-week basic law enforcement training program. Multiple regression analyses were conducted to examine the research hypothesis that 1.5 mile run times (CRF), percent body fat (PERFAT), leg press/body weight ratio (LEGRATIO), bench press/body weight ratio (BENCHRATIO), and abdominal muscular endurance (SITUPS) combine to produce a sufficient predictive model on a criterion task test model of law enforcement officer physical agility (LEOPAT), and that upper body 1-RM strength (BENCHPRESS), lower body 1-RM strength (LEGPRESS), sit-reach flexibility (FLEX), and lean body mass (LBM) would not further improve the predictive utility of the model. RESULTS The model including the five variables accounts for a significant proportion of variance (R 2 = .389, F(2,120)=14.649, p = .0001), which is consistent with the hypothesis. When BENCHPRESS, LEGPRESS, FLEX, and LBM were added to the equation, the variance accounted for increased to R 2 = .403 (F(2,120) = 10.461 p = .0001) which was not significantly more than the variance accounted for by the smaller model (F-change(1,114) = 1.205, p = .275). Consistent with the hypothesis, none of the additional four predictors contributed to this larger model. CONCLUSION Together these results provide support for the hypothesis that the 5-variable model used to predict performance on the LEOPAT was more efficient than the 9-variable model. Inclusion of SITUPS, PERFAT, FLEX, and LBM do not increase the predictive utility of the model further.