In-School and Out-of-School Factors That Build Student Achievement

Endnotes

¹In the Nashville study, high achievers were students whose NCE scores in reading total were at or above the 50th percentile. Lower-achieving students had scores below the 50th percentile.

²Reviews of this paper have generated discussion about this question and particularly the term biological capability. It may be that some interpret the statement as referring to students' inherent or genetic ability. While this is a bristle term, to be sure, it remains true that teachers' ratings of students varied slightly when asked what percent of their lower-achieving students have the biological capability to one day attend and complete college. Lower-achieving students were slightly less likely (26 percent vs. 31 percent) than high achievers to have a teacher who believed that more than one-half of her poor-reader students had the biological capability to one day attend and complete college. For decades, dating back to the "Pygmalion in the Classroom" research of Rosenthal and Jacobson (1968), the expectations of teachers toward their students have been shown to powerfully affect student performance. The questions and results of this research align with this prior research.

³These activities were identified by (1) documenting responses on students' weekly time-use protocols, (2) classifying all responses into four categories of typical human activity (high-yield learning, labor, leisure, and health maintenance) that were conceptualized by the author, and (3) performing a quantitative content analysis of the responses in each conceptual category. The author is unaware of any previous conceptualization of students' out-of-school activities that includes the specific, comprehensive range of activities discussed here.

⁴Two trained research assistants served as coders. Their classifications of parents' responses had a consistency rate of 95 percent. A detailed protocol of typical parent responses for each coding category is available upon request.

⁵Data for coding behaviors were available for 97 percent of the coding instances. Data were missing for one high-achiever writing activity and talk activity, for three low-achiever hobby activities, and for one low-achiever writing activity.