This study examined the features and correlations between teaching behaviors and learning activities in urban high school physical education (PE) class settings using direct instruction model. Participants were sixteen urban high school PE teachers and their students. Results indicated that the teachers spent their class times on the major teaching behaviors were: Informing 28.8%, Structuring 25.4%, Observing 17.0%, and Feedback 9.2%. The students spent their class times on the major learning activities were: Motor Engaged 56.2%, Cognitive Engaged 20.7%, and Waiting for a Turn 9.7%. Correlation analyses revealed that Informing, Questioning, and Feedback teaching behaviors were positively associated with Motor Engaged and Cognitive Engaged learning activities. When the teachers exhibited the behavior category of None of the Above, their students showed no motor and cognitive learning activities engagements. Findings suggested that, PE teachers should develop and employ teaching behaviors that promote and demonstrate physical skills and fitness because those teaching behaviors are positively associated with students' physical activity levels.
Research on teaching in physical education (RT-PE) has accomplished remarkably with respect to how teaching behaviors are related to learning outcomes (Brophy & Good, 1986; Graham & Heimerer, 1981; Keating, Kulinna & Silverman, 1999; Martin & Kulinna, 2004; Silverman, Tyson & Morford, 1988). Silverman (1991) defined RT-PE as "research on what teachers and students do and how this affects or relates to leaming and to the social dynamics of the class." (p. 352). This definition is broad and tells us that, when a study is conducted on teaching in physical education, both teacher's teaching behaviors and student's learning activities should be included in the parameters of a study.
Recently, research studies have provided sufficient information regarding the characteristics of effective teaching (Pangrazi, 2007; Kulinna, Cothran & Regualos, 2006; Siedentop & Tannehill, 2000). Based on the work done by previous researchers, the following characteristics were described as effective teaching/learning environments: (a) clear objectives and content covered; (b) well-organized and appropriate expectations; (c) meaningful task and high rates of success; (d) smooth transition and low rates of management time; (e) appropriate guidance and active supervision; (f) high rates in student-engaged time and low rates in student waiting time; and (g) teacher's enthusiasm and equitable support. The characteristics stated above have become important guidelines for training novice teachers to grow to be an effective teacher (Pangrazi, 2007; Kulinna, Cothran & Regualos, 2006; Siedentop & Tannehill, 2000).
A number of previous studies in RT-PE focused on how teacher and student behaviors were associated with direct instruction, and eventually the direct instruction model in physical education was developed (Anderson, Evertson, & Brophy, 1979; Graber, 2001; Rosenshine, 1979; Rosenshine & Stevens, 1986; Sweeting & Rink, 1999). Rosenshine (1979) illustrated that direct instruction model in physical education possesses clear learning goals, adequate time for instruction and practice which is characterized by appropriate subject matters for students' abilities, lowlevel cognitive questions but meaningful task and high success by monitoring student performance and providing immediate and specific feedback.
Silverman, Tyson, and Morford (1988) found that students spent their class time on motor skill practice with teacher's feedback was positively associated with their motor learning achievement. Furthermore, regarding how time is spent in physical education classes, Silverman (1991) summarized that most students spent less than 30% of the class time on motor activities but the majority of …