Data set on 1,965 Ohio elementary school buildings for 2001-2002.
The ‘ohio’ dataset contains 6 objects as follows.
Original data ohioschool.dat from http://www.spatial-econometrics.com/ (J. LeSage and R. Pace 2009).
The data set contains information on, for instance, school building ID, Zip code of
the location of the school, proportion of passing on five subjects, number of teacher,
number of student, etc. The variables are:
col 1: zip code
col 2: lattitude (zip centroid)
col 3: longitude (zip centroid)
col 4: buidling irn
col 5: district irn
col 6: # of teachers (FTE 2001-02)
col 7: teacher attendance rate
col 8: avg years of teaching experience
col 9: avg teacher salary
col 10: Per Pupil Spending on Instruction
col 11: Per Pupil Spending on Building Operations
col 12: Per Pupil Spending on Administration
col 13: Per Pupil Spending on Pupil Support
col 14: Per Pupil Spending on Staff Support
col 15: Total Expenditures Per Pupil
col 16: Per Pupil Spending on Instruction % of Total Spending Per Pupil
col 17: Per Pupil Spending on Building Operations % of Total Spending Per Pupil
col 18: Per Pupil Spending on Administration % of Total Spending Per Pupil
col 19: Per Pupil Spending on Pupil Support % of Total Spending Per Pupil
col 20: Per Pupil Spending on Staff Support % of Total Spending Per Pupil
col 21: irn number
col 22: avg of all 4th grade proficiency scores
col 23: median of 4th grade prof scores
col 24: building enrollment
col 25: short-term students < 6 months
col 26: 4th Grade (or 9th grade) Citizenship % Passed 2001-2002
col 27: 4th Grade (or 9th grade) math % Passed 2001-2002
col 28: 4th Grade (or 9th grade) reading % Passed 2001-2002
col 29: 4th Grade (or 9th grade) writing % Passed 2001-2002
col 30: 4th Grade (or 9th grade) science % Passed 2001-2002
col 31: pincome per capita income in the zip code area
col 32: nonwhite percent of population that is non-white
col 33: poverty percent of population in poverty
col 34: samehouse % percent of population living in same house 5 years ago
col 35: public % of population attending public schools
col 36: highschool graduates, educ attainment for 25 years plus
col 37: associate degrees, educ attainment for 25 years plus
col 38: college, educ attainment for 25 years plus
col 39: graduate, educ attainment for 25 years plus
col 40: professional, educ attainment for 25 years plus
The derived dataset for analyzing the percentage passed based on Zip codes. The variables are: y: the percentage passed (4th or 9th grade) in each school TchExp: average Teacher's experience Subjects: for five study subjects of Citizenship, Maths, Reading, Writing and Science Stu.Tch: student by teacher ratio School: school index Zip: Zip code
The derived dataset for analyzing the median of 4th grade scores based on school districts. The variables are: MedianScore: the median of 4th grade prof scores district: school districts
A SpatialPolygonsDataFrame object (see package sp) containing the map information of ohio school districts.
The spatial distance matrix based on Zip codes. The codes generated this matrix are:
Zsp <- model.matrix(~ factor(Zip) - 1, data = ohioGrades)
uzipC <- matrix(0, nrow = ncol(Zsp), ncol = 2)
Zip <- as.numeric(substr(colnames(Zsp), start = 12, stop = 16))
for (i in 1: ncol(Zsp)) {
Cord <- as.matrix(ohioSchools[(ohioSchools$V1 == Zip[i]), 2:3])
uzipC[i,] <- Cord[1,]
}
Dst <- as.matrix(dist(uzipC))
for(i in 1:nrow(Dst)) {
x <- Dst[i,]
x <- ifelse(x == 0, 0, 1/x)
Dst[i,] <- ifelse(x > 4, 4, x)
}
ohioZipDistMat <- Dst/4
The spatial distance matrix based on school districts. The codes generated this matrix are:
ccNb <- poly2nb(ccShape)
W <- matrix(0, 616, 616)
for (i in 1:nrow(W)) {
tmp <- as.numeric(ccNb[[i]])
for (k in tmp) W[i,k] <- 1
}
W[353,] <- W[,353] <- 0
districtShape <- as.numeric(substr(as.character(ohioShape@data$UNSDIDFP), 3, 7))
dimnames(W) <- list(districtShape, districtShape)
districtSchool <- floor(ohioSchools[,5]/10)
districtSchool <- factor(districtSchool[districtSchool %in% districtShape])
levelsShape <- levels(factor(districtShape))
levelsSchool <- levels(districtSchool)
levels(districtSchool) <- c(levelsSchool, levelsShape[!(levelsShape %in% levelsSchool)])
ohioDistrictDistMat <- W[levels(districtSchool),levels(districtSchool)]
J. LeSage and R. Pace (2009). Introduction to Spatial Econometrics. Chapman \& Hall/CRC, Boca Raton. M. Alam, L. Ronnegard, X. Shen (2014). Fitting spatial models in hglm. Submitted.