"... In PAGE 44: ...Findings Research Question 1: To what extent do school board members perceive that school-linked services are necessary and are available to students in their district and developed in a collaborative manner? The survey contained three questions which asked board members to indicate whether their community supported the development of school-linked services within the school district. Table8 indicates that three fourths or more (75.3% and 84.... In PAGE 69: ...9% and 80.4%) responded affirmatively ( Table8 ). The questionnaire explored what school-linked services were provided and the type of collaborating agency.... ..."

"... In PAGE 44: ...0%) were found in fewer school districts. School districts collaborate with many different agencies and organizations to provide services ( Table9 ). Most of the services are supported by partnerships with agencies at the state and local level and to a smaller degree by federal programs.... ..."

"... In PAGE 10: ... By pairing charter schools with their host (meaning geographically co-located) districts, we attempt to create the local market for educational services in which both the charter schools and the public school districts compete. Table1 provides descriptive statistics from the data set that we have constructed. Except for a few observations that have been deleted because of missing values for key variables, the number of students included in the table is exactly equal to those who took the MEAP test in the districts included in the analyses.... ..."

"... In PAGE 19: ... 17 Table2 gives variance components estimates in a variance components model for total scores of the AT rubrics, the CE rubrics, and separate totals for AR in Reading Comprehension and AR in Mathematics. The variance component for Rater for AT is zero because the raters divided the observational tasks to save time and produced only a single AT rating per rubrics per classroom.... In PAGE 25: ... 23 Table 9 gives a brief variance components analysis for the assignment ratings, totaled within the AR and CE rubrics (AT depends on social interaction, so that is very similar to the variance components analysis of Table2 above). It is interesting to note that Rater has a negligible variance component in this analysis.... ..."

"... In PAGE 18: ...ousing unit, (e.g., value or rent, size, and age). These household-level data are summarized in the first column of Table1 . 27 In general, these conditions require restrictions on preferences that limit the contribution to utility made by preferences for endogenous choice characteristics (such as public school quality or crime rates) relative to the contribution to utility made by exogenous choice characteristics.... In PAGE 19: ... See the Data Appendix for further details. 31 As is apparent by comparing the first and last columns of Table1 , the characteristics of the household-level data match the characteristics of the district-level data quite well. 32 While I would have liked to use only households with children in the sixth grade, the nature of this special tabulation of the Census forces me to include all households with children in grades 5-8.... ..."

"... In PAGE 33: ...5 percent urban. Table5 provides data relating to distribution of responses by a district apos;s student enrollment. In both groups the highest response was provided by school board members from districts who ranged from 1,000 to 4,999 (44.... ..."

"... In PAGE 9: ...the dependent variable as reported by the principal) X7 = Influence of the school council (index of the domains of influence indicated for the dependent variable as reported by the principal) X19 = Interaction term between start-up and the influence of the district School type X8 = Dummy variable for start-up1 charters (0 if it is not a start-up, 1 if it is a start-up) X9 = Dummy variable for conversion2 charters (0 if it is not a conversion, 1 if it is) X10 = Dummy variable for private schools (0 if it is not a private school, 1 if it is) School characteristics X11 = Proportion of Hispanic students X12 = Proportion of Black students X13 = Proportion of American Indian students X14 = Proportion of Asian/Pacific Islander students X15 = Proportion of students served by Title I X16 = Dummy variable for urban setting (0 if it is not urban, 1 if it is) X17 = Dummy variable for suburban setting (0 if it is not suburban, 1 if it is) X18 = Dummy variable for for-profit charters (0 if it is not a for-profit, 1 if it is) X20 = Total ungraded and K-12 enrollment Results On examination, Table2 shows that principals residing in private schools that converted to a charter for the most part have higher levels of influence than their public conversion and start-up charter school counterparts in all areas of autonomy except for school spending. Organizations such as private schools have had high degrees of autonomy to begin with because they are free from district mandates.... ..."