1
Abstract
Has NCLB Improved Teacher and Teaching Quality for Disadvantaged Students?
Laura M. Desimone, VanderbiltUniversity
Thomas M. Smith, VanderbiltUniversity
David Frisvold, University of Michigan
Correspondence about this paper should be directed to Laura M. Desimone, , 615 400 2276 (phone).
Purpose/Importance
Recent studies offer compelling evidence that teacher and teaching quality are critical factors in student success. One of the key underlying rationales for the NCLB teacher quality provisions is that improving teaching quality will address the wide variation in student achievement gains that currently exists between advantaged and disadvantaged students.
Research Questions
To examine the extent to which NCLB has improved teacher and teaching quality for economically disadvantaged students, we ask (1) how large were the teacher quality gaps between advantaged and disadvantaged students at the onset of NCLB? (2) has teacher quality for disadvantaged students improved? And (3) can any improvements that have occurred be associated with NCLB-related policy changes?
Methods
Design. Our analyses track change in state policy and teacher characteristics between 2000 and 2003. We estimate the relation between state-level implementation of NCLB-related policies and subsequent improvements in teacher quality as measured by several key indicators. We focus on a single subject (mathematics) in a single grade (eighth).Given our design and data, our conclusions must necessarily fall short of direct attribution to NCLB; instead we identify trends experienced by disadvantaged students in states where considerable NCLB-related implementation was occurring.
Sample/Data. We use three sources of data for this study: (1) a database of state policies related to NCLB implementation, which we constructed from existing national data sources, (2) data from student and teacher surveys from the eighth grade national 2000 NAEP and (3) the 2000 and 2003 state NAEP.
Analyses. Our analyses are a combination of weighted mean comparisons, cross-sectional three-level hierarchical linear models (teachers within schools within states) and a state-specific fixed effects model that determines the relation between state policies and teacher quality from variation over time within states.
Measures. Grounded in research and NCLB provisions, the measures of teacher quality that we use are experience, certification, and degree in mathematics. Our measures of teaching quality are inquiry-oriented/conceptual and procedural teaching. The policy variables we use are grounded in a theory that posits five attributes that contribute to successful reform implementation: (1) consistency (alignment), (2) specificity (detail and clarity of guidance), (3) authority (buy-in and support), (4) power (rewards and sanctions), and (5) stability (of policies and administrators, teachers, and students).
Results
Our analyses show that on three key indicators—having a teacher with certification, a degree in mathematics, and more than two years of experience—disadvantaged students were worse off than advantaged students in 2000, and this did not change much by 2003. However, the differences were quite small.We found that in states implementing NCLB-related policies, if there was any positive movement in teacher quality it was small, and in a few statesteacher quality decreased.
Conclusions
This analysis suggests that most states are not on target for making the kinds of teacher quality improvements required by NCLB. Our analysis provides limited evidence that particular state policies may eventually move states in the right direction, but raises concern that results may not be substantial enough, or fast enough, to satisfy the legislation or common ideals about equality in teacher quality.