The current study analyses the classroom social context; the rate of teacher task talk with respect to social consequences (praise and reprimand), in relation to the rate of on-task and unwanted behaviour of a target student in the classroom. Teacher behaviour talk (talk related to student conduct), social talk (social, non-academic talk) and proximity to the target student are also assayed. Teacher verbal behaviour was recorded towards whom it was addressed: to the whole class, to a target student, to other students, and analysed severally and combined (summed) in relation to student on-task and unwanted behaviour. Data were obtained by continuous recording of classroom behaviour and collated in to 60 second intervals. All correlations were calculated on seconds of occurrence of that behaviour per minute. This allowed for lagging the independent variable relative to dependent variables to better reflect the subsequent nature of the independent variable, such as praise and reprimand, rather than assaying contiguous relationships alone. Results indicated that teacher verbal behaviour directed toward the target student did not relate significantly with student on-task behaviour or student unwanted behaviour. The teacher verbal behaviour that related most significantly with both student on-task behaviour and student unwanted behaviour was teacher task talk to the whole class, followed by teacher task talk to the whole class, to the target student and to other students combined. Teacher reprimand and behaviour talk directed toward the target student did not relate significantly with student on-task or unwanted behaviour. Teacher reprimand and behaviour talk to the whole class, to the target student and to other students combined was significantly related to the decrease in student unwanted behaviour and increase in on-task behaviour given lag analysis for those teachers maintaining a high (greater than 50% of available time) rate of task talk. For teachers that did not do so, teacher verbal behaviour did not maintain significant relationships with student on-task or unwanted behaviour. Results for this lower rate of teacher task talk were characterized by variability both in student behaviour and teacher verbal behaviour. A case study was conducted increasing the general rate of teacher task talk which supported the above findings. These results indicate that there is substantial commonality between student behaviour within the classroom and that teacher talk to the whole class and combined (i.e. the sum of addressing the whole class, the target student and other students) is significantly more predictive of individual student behaviour than is teacher talk directed to a target student alone. Teacher social talk was significantly related to teacher reprimand and behaviour talk. Teacher proximity to the target student was not significantly related to student on-task or unwanted behaviour. The findings are discussed in relation to previous research findings and teaching practice.