This manual describes procedures for collecting data about two classroom elements: (1) teacher-student interactions; and (2) amount and type of student engagement in academic work. Although the system was originally developed from a study of influences on sex-related differences in mathematics, it is designed to be adapted as needed for investigations of other classroom behaviors. Part 1 of the manual contains explanations of the use of a modified version of a coding system developed by Brophy and Good (1970) to measure teacher-student interactions. These interactions are classified as public and nonpublic, and examples of measuring the incidence and qualities of behaviors are given. In a section reviewing general coding conventions, the validity of the procedure is interpreted, and observation techniques are suggested. In Part 2, the method used to study the proportion of time students are engaged in mathematical activities, and characteristics of those activities are introduced, based on a system designed by Romberg, Small, Carnahan, and Cookson (1978). Coding categories and data collection procedures are defined. Data collection worksheets are appended. (FG)