Homework 3

Habits of Mind Problem 1: Come up with a testable research question. Carry out the experiment and do the test. What do you find?

Habits of Mind Problem 2: Value-added modeling is a hot topic in education. Various value-added models are used to analyze student achievement data. Often, these analyses produce estimates of “teacher effects.” Corcoran (2010) discusses value-added measures of teacher effectiveness in the context of two different school systems: NYC and Houston.

Corcoran, S. P. (2010). Can teachers be evaluated by their students’ test scores? Should they be? The use of value-added measures of teacher effectiveness in policy and practice. Providence, RI: Annenberg Institute for School Reform at Brown University. Retrieved from http://www.annenberginstitute.org/pdf/valueAddedReport.pdf.

Please feel free to read the entire article, but focus on two main sections: “What Is a Teacher’s Value-Added?” (pp. 4-5) and “Are value-added scores precise enough to be useful?” (pp. 21-25). Appendices B and C provide two sample teacher data reports. Use the two confidence intervals reported on the data report in Appendix B to comment on the following questions:

·  What is the parameter of interest? Describe what this parameter is measuring. How is it different from other parameters we have discussed in class?

·  Why does the report provide a confidence interval for each result, rather than merely reporting the estimates?

·  What are common sources of variability affecting the margin of error?

·  Corcoran (2010) writes, "Value-added estimates become more precise with additional years of data..." (p. 22). Describe what kind of additional data you think he is referring to.

·  Compare and contrast the two intervals on Mr. Jones’s report. Why is the “Last 4 Years” interval considered more precise than the “2008-2009” interval? Does this seem reasonable? Why or why not?

·  In practice, there are various value-added models used to estimate teacher effect estimates. In fact, each can lead to very different teacher effect estimates. What do you think should be considered when determining whether one estimated teacher effect is more credible than another? What do you think are appropriate uses for these kind of estimates?

De Veaux et al. (2006) Book Problems:

1.  Problem 18, p. 445

2.  Problem 14, p. 468

3.  Problem 12, p. 541

4.  Problem 33, p. 543