In order to provide accurate estimates of how much teachers affect the achievement of their students, this study used panel data covering over a decade of elementary student test scores and teacher assignment in two contiguous New Jersey school districts. The test score data, which spanned the years 1989-1990 to 2000-2001, came from nationally standardized basic skills reading and math tests. Data were also collected on students' gender, ethnicity, special education classification, and English as a Second Language enrollment, as well as on school, grade, and teacher identifiers. The study estimated teacher fixed effects while controlling for fixed student characteristics and classroom specific variables. Data analysis indicated that there were large and statistically significant differences among teachers. A one standard deviation increase in teacher quality raised students' reading and math test scores by approximately .20 and .24 standard deviations, respectively, on a nationally standardized scale. In addition, teaching experience had statistically significant positive effects on reading test scores, controlling for fixed teacher quality. (Contains 25 references.) (SM) Reproductions supplied by EDRS are the best that can be made from the original document.

Are teachers' impacts on students' test scores ("value-added") How can we measure and improve the quality of teaching in primary schools? One prominent but controversial method is to evaluate teachers based on their impacts on students' test scores, commonly termed the "value-added" (VA) approach.

.

were generous with their time in putting together data on our behalf. Timothy Daly, Michelle Rhee, and Andrew Sokatch of The New Teacher Project and Abigail Smith and Cynthia Skinner of Teach for America were extremely helpful in answering our questions regarding these programs. We thank Caroline Hoxby and Andrew Leigh for providing us with data on college median SAT scores. Dan Obus provided excellent research assistance. We would also like to thank seminar participants at the

We estimate the willingness to pay for reductions in crime risk using the location and move-in dates of sex offenders. We find significant effects of sex offenders' locations that are geographically localized. House prices within 0.1 miles of a sex offender fall by 4 percent on average. We then use this finding to estimate the costs to victims of sexual offenses, and find costs of over $1 million per victim--far greater than previous estimates. However, we cannot reject the alternative hypotheses that individuals overestimate risks posed by offenders or that living near an offender poses significant costs exclusive of crime risk.