(PS11-C60) Effective Classroom Management Strategies: Exploring the Correspondence Between Teacher Self-report and Observation

Teachers effective use of classroom management strategies is important for children with social, emotional, and behavioral challenges, as it is associated with positive academic and social outcomes. Self-report of strategy use is often used to determine if teachers require implementation support. However, there are concerns about this method, as there is low correspondence between self-report and observation data. Most research examining this correspondence relies on linear relationships and uses measures with vague anchors (e.g., 0 = not at all; 3 = some; 5 = a lot). In this study we examined correspondence when teachers were given an anchor with a concrete mid-point (half of the time). We also explored possible non-linear relationships between response anchors and observation data across multiple strategies (e.g., effective response to rule violations, praise, effective commands).

Participants were 30 teachers, observed for two 30-minute periods using a modified version of the Student Behavior – Teacher Response coding system. We calculated both rates per hour and percentages of strategy use for each variable. Data were averaged across observations. Teachers reported their strategy use through a modified version of the Teacher Strategies Questionnaire with the anchors rarely/never, sometimes, half of the time, often, and very often.

Correlations between self-reported and observed strategy use were low to moderate (rs ranged from .11 to .34). When we grouped the self-report responses by conceptual anchors (rarely/never or sometimes = 1; half of the time = 2; often or very often = 3), one-way ANOVAs indicated that the three response groups did not differ significantly in strategy use. However, some non-linear patterns emerged. For response to rule violations (RVs), teachers who reported responding to RVs rarely/never or sometimes were observed to have responded to 23.36% (SD=11.59) of RVs. However, teachers in both the medium condition (M=37.35%; SD = 10.08) and high condition (M=36.56%; SD=15.07) were observed to respond to RVs as similar rates. In essence, the teachers who had low rates of strategies were accurately aware of their skills whereas teachers with medium to high did not demonstrate such awareness. For praise, the pattern differed because no teacher used the anchors of never/rarely or sometimes. As such, there was no low condition to compare with the medium (M=25.68; SD=17.63) and high (M=31.15; SD=13.95) groups.

Implications involve new ways of interpreting teacher self-report. Since teacher report in the low group were representative of actual strategy use, self-report may be relied upon in this condition to indicate teacher’s need for supports. However, the medium and high levels self-report were not representative of actual strategy use; thus, self-report is not a good indication for these teachers. Additional research should continue to examine meaningful conceptual anchors that may enhance correspondence, as well as potential cut-points in teacher self-report that reliably predict those lower in skills and in need of implementation supports.