The purpose of this study was to test the validity of the games for understanding model by comparing it to a technique approach to instruction and a control group. The technique method focused primarily on skill instruction where the skill taught initially was incorporated into a game at the end of each lesson. The games for understanding approach emphasized developing tactical awareness and decision making in small game situations. Two physical education specialists taught field hockey using these approaches for 15 lessons (45 min each). The control group did not receive any field hockey instruction. Data were collected from 71 middle school children. Pretests and posttests were administered for hockey knowledge, skill, and game performance. Separate analyses of variance or analyses of covariance were conducted to examine group differences for cognitive and skill outcomes. The games for understanding group scored significantly higher on passing decision making than the technique and control groups during posttest game play and significantly higher than the control group for declarative and procedural knowledge. The games for understanding group scored significantly higher on control and passing execution than the other groups during posttest game play. For hockey skill, there were no significant differences among the treatment groups for accuracy, but the technique group recorded faster times than the control group on the posttest.