Given a dataframe with percentile ranks by year, calculates
year-over-year and cumulative change. This enables the "39th to 78th
percentile" style analysis from MarGrady Research.
Usage
percentile_rank_trend(
df,
percentile_col,
year_col = "end_year",
entity_cols = c("district_id")
)
Arguments
- df
Dataframe with a percentile column and year column
- percentile_col
Character. Name of the percentile column to track.
- year_col
Character. Name of year column. Default "end_year".
- entity_cols
Character vector. Columns identifying entities to track
over time (e.g., c("district_id", "subgroup")).
Value
df with added columns:
{percentile_col}_yoy_change: Year-over-year change
{percentile_col}_cumulative_change: Change from first year
{percentile_col}_baseline: Value in the first year
Examples
if (FALSE) { # \dontrun{
# Track Newark's percentile rank over time
grate_ranked %>%
filter(district_id == "3570") %>%
percentile_rank_trend(
percentile_col = "grad_rate_percentile",
entity_cols = c("district_id", "subgroup")
)
} # }