“…In other studies showing differences between observations and teacher self-reports on surveys or interviews, the length or frequency of observations is unclear (Squire & Applebee, 1968; Steele, House, & Kerins, 1971; Walberg & Thomas, 1972), or there were only one or two observations for periods as short as 20 minutes (Beam & Horvath, 1975; Hardebeck, Ashbaugh, & McIntyre, 1974; Squire & Applebee, 1968). Subsequent research indicates that to elicit reliable and valid measures of teachers’ overall instruction, three observations are required for one stable observation, and at least three stable observations over an extended period of time are required (Taylor, Pearson, Clark, & Walpole, 1999).…”