ERIC Number: ED199290
Record Type: RIE
Publication Date: 1981-Apr
Reference Count: 0
A Technique for Relating School Variables to Individual Student Achievement.
Johnstone, Whitcomb G.
When the school or district is the unit of analysis in school effects research, the outcome of interest is mean achievement, and the variance explained by school or district level predictors is between schools or between districts. When the individual is the unit of analysis, the outcome of interest is individual achievement and the variance explained by school or district level predictors is between individuals. A technique called situational regression (SR) is described which provides a link between these levels of analysis. SR involves regressing a criterion variable measure at the individual level on one or more predictor variables measured at the group level. Predictors are measured characteristics of the groups e.g., schools, to which individuals belong and may include socioeconomic status (SES), percent daily attendance at the school (ATTEN) and the number of resource teachers assigned to the school (REST). Data illustrating the SR technique were taken from composite student achievement scores of all grade 5 students in one district and regressed on SES, ATTEN, and REST. These three school level variables together accounted for almost 17 percent of the individual variance in achievement scores and represented nearly 70 percent of the between school variation in achievement. (RL)
Publication Type: Speeches/Meeting Papers; Reports - Research
Education Level: N/A
Authoring Institution: N/A
Identifiers: Criterion Variables; Situational Regression Technique
Note: Paper presented at the Annual Meeting of the American Educational Research Association (65th, Los Angeles, CA, April 13-17, 1981).