Fetch Summative Educator Evaluation Rating Distributions
Source:R/fetch_staff_doedata.R
fetch_staff_evaluations.RdDownloads the NJ DOE standalone summative educator-evaluation workbook and returns, for each entity and staff category (teachers vs principals/APs/VPs), how many educators landed in each of the four rating tiers (ineffective, partially effective, effective, highly effective) plus the total evaluated. Rating distributions are a rarely-analyzed window into evaluation-system rigor (e.g. how concentrated ratings are at the top of the scale).
Value
Data frame with end_year, the entity identifiers, the raw
category and normalized staff_category, the five numeric rating
columns (ineffective, partially_effective, effective,
highly_effective, total), and the entity flags
(is_state, is_county, is_district, is_school,
is_charter).
Details
Source. Standalone Excel workbooks under
nj.gov/education/doedata/staff/ – distinct from the SPR-sourced staff
fetchers. Only three years exist: end_year 2014 (SY2013-14),
2015 (SY2014-15), and 2016 (SY2015-16); any other year errors.
Sheets differ by year. The 2014 workbook ships one combined sheet
(school rows plus district-total rows, school_id == "999"); the 2015
and 2016 workbooks split district totals and school totals into two sheets
that are read and stacked.
Staff category. The raw CATEGORY ("TEACHERS" /
"PRIN/AP/VP") is normalized to staff_category
("teachers" / "principals_vps") and the raw label is kept as
category.
Suppression -> NA (never a guessed number). Small cells are masked
with "*"; every rating column maps "*" to NA. A real
published 0 stays 0. A trailing data-certification note row in
each workbook is dropped.
Entity flags. is_school (per-school rows) vs is_district
(district totals, school_id == "999"). A statewide aggregate row
(county "99" / district "9999") is present in 2014 and 2015 and
is flagged is_state; it is returned at level = "district". The
2016 workbook publishes no statewide row.
Examples
if (FALSE) { # \dontrun{
# School-level teacher rating distribution, 2015-16
fetch_staff_evaluations(2016)
# District-total rows, including the statewide aggregate (2014)
library(dplyr)
fetch_staff_evaluations(2014, level = "district") %>%
filter(is_state, staff_category == "teachers")
# Share of teachers rated highly effective, by district (2015)
fetch_staff_evaluations(2015, level = "district") %>%
filter(staff_category == "teachers", !is_state) %>%
mutate(pct_highly = 100 * highly_effective / total) %>%
select(district_name, highly_effective, total, pct_highly)
} # }