Citation Context

...esting, so these results should cause us some alarm, but they do seem to indicate that something real is occurring. Stability across grade and subject: Mandeville and Anderson (1987) and others (e.g. =-=Rockoff, 2004-=-; Newton, et al., 2010) found that stability fluctuated across grade and subject matter. Though limited, stability was greater for mathematics courses than for reading courses, raising issues of fairn...

Citation Context

...the relatively low levels of stability of the teacher effect sizes. Stability across statistical models: Linear composites in general tend to perform similarly regardless of how one gets the weights (=-=Dawes, 1979-=-). Tekwe, Carter, Ma, Algina, Lucas, & Roth (2004) compared four regression models and found that unless the models involve different variables, the results tend to be quite similar. Three of the mode...

Citation Context

... (HLM) to control the effects of prior achievement, attendance, and school-level variables and to measure the conditional growth in student performance. A second effort grew out of work in Tennessee (=-=Sanders & Horn, 1994-=-, 1998). This value-added approach, as it is sometimes called, was a great deal more statistically sophisticated. It involved a layered multiple regression model (TVAAS), that looked for the effects o...

Citation Context

...rally weaker than we expect for use of assessments for students (coefficients of .9 for high-stakes interpretations, or .7 when combined with other information in classroom decisions, about students; =-=Nitko, 2001-=-, pp. 76-7).Practical Assessment, Research & Evaluation, Vol 17, No 17 Page 12 Schafer, Lissitz, Zhu, Zhang, Hou & Li, Student Growth Models Table 9. Correlation between 2008-2009 and 2009-2010 Chang...

Citation Context

...e results should cause us some alarm, but they do seem to indicate that something real is occurring. Stability across grade and subject: Mandeville and Anderson (1987) and others (e.g. Rockoff, 2004; =-=Newton, et al., 2010-=-) found that stability fluctuated across grade and subject matter. Though limited, stability was greater for mathematics courses than for reading courses, raising issues of fairness and comparability ...

Citation Context

... eight models more completely. The acronyms we used are bolded within the text. A complete list of the major acronyms along with brief characterizations is given later in Table 5. Betebenner’s model (=-=Betebenner 2008-=-, 2012) is quite popular, currently being used in Colorado, for example. It uses quantile regression to estimate the conditional percentile of each student’s performance in the second year compared to...

Citation Context

...ge 5), for example, has calculated that 69% of its teachers are teaching non-tested subjects and grades. In Memphis, Tennessee the current testing program does not apply to about 70% of the teachers (=-=Lipscomb, Teh, Gill, Chiang, & Owens, 2010-=-). This is a problem that is quite common today, although it is not the only methodological problem. For example, most teachers do not actually work alone with students. They have other teachers, othe...

Citation Context

...ulting quasi-vertical scales, constructed without using common items, are linear transforms of what has been called a growth scale that for some purposes actually may be superior to a vertical scale (=-=Schafer, 2006-=-). Once we had consistently-scaled scores, we subtracted the growth (spline) score at the earlier grade (pretest) from the growth (spline) score at the later grade (posttest), as though they were from...

Citation Context

...rowth Models but this result was described in a later paper as little better than a coin flip between teachers with National Board for Professional Teaching Standards certification and those without (=-=Sanders and Wright, 2008-=-). Goldhaber and Hanson (2010) found with North Carolina data that VAM estimates seem to provide better measures of teacher impact on student test scores than do measures obtained at the time a teache...

Citation Context

...udy the behavior of outcomes similar to those that might result from a vertical scale. The growth scores were based on look-up tables derived from an earlier study using statewide data for each test (=-=Schafer, Hou, & Lissitz, 2009-=-). Each table was developed as a spline function created to give moderated (consistent) meaning to various points along the performance continuum across grades and contents, scaled using 2008 data. Th...

Citation Context

... No 17 Page 2 Schafer, Lissitz, Zhu, Zhang, Hou & Li, Student Growth Models an effort in Dallas, Texas. The original version of the Dallas system went into effect in 1994 and examined school effects (=-=Webster & Mendro, 1994-=-); this was expanded to include teacher effects in the 1995-1996 school year (Webster, Mendro, Bembry, & Orsak, 1995). The model was composed of two stages. The first stage used multiple regression to...

Citation Context

...ed with more nuances. The transition table described in Table 4 (TProg) was developed to reflect an assumption that greater change is required to maintain a higher achievement level than a lower one (=-=Schafer, 2007-=-, 2008). Thus, the values assigned to cells on the diagonal progress as achievement levels increase. Above and below that diagonal, all degrees of measured differences are valued differentially. This ...

Citation Context

...d content combination. As a consequence, there are striking differences within almost every state in the degrees of idealism/realism expressed in their resulting cut scores across years and contents (=-=Schafer, Liu, & Wang, 2007-=-). A danger in using those cut scores to measure change, as do the transition procedures, is that they may be unfair to teachers in different grades (or contents). We studied that by correlating grade...