Skip to contents

Summarize how many rows have N columns at or above (or below) various thresholds?

Source: R/colcounter_xyz.R
colcounter_summary.Rd

Like colcounter() or cols.above.count() but will handle multiple thresholds to compare to each indicator, etc.

Table of counts, percents, cumulative counts, cumulative percents of places with N, or at least N, of the indicators at or above the benchmark(s)

Usage

colcounter_summary(
  x,
  thresholdlist,
  or.tied = TRUE,
  na.rm = TRUE,
  below = FALSE,
  one.cut.per.col = FALSE
)

Arguments

x

Data.frame or matrix of numbers to be compared to threshold value, like percentiles for example.

thresholdlist

vector of numeric threshold values to compare to

or.tied

if TRUE, include ties (value in x equals threshold)

na.rm

if TRUE, used by colcounter() to count only the non-NA columns in given row

below

if TRUE, count x below threshold not above threshold

one.cut.per.col

if FALSE, compare each threshold to all of x. If TRUE, specify one threshold to use for each column.

Value

A table of frequency counts

See also

colcounter_summary_all() colcounter_summary() colcounter_summary_cum() colcounter_summary_pct() colcounter_summary_cum_pct() tablefixed()

Examples

 pdata <- data.frame(a=rep(80,4),b=rep(93,4), col3=c(49,98,100,100))
  ### pdata <- EJAM::blockgroupstats[ , names_e_pctile]
 pcuts <-  5 * (0:20)  
colcounter_summary(        pdata, pcuts)
colcounter_summary_pct(    pdata, pcuts)
colcounter_summary_cum(    pdata, pcuts)
colcounter_summary_cum_pct(pdata, pcuts)
colcounter_summary_cum_pct(pdata, 5 * (10:20))
a3 <- colcounter_summary_all(    pdata, pcuts)

x80 <- colcounter(pdata, threshold = 80, or.tied = T)
x95 <- colcounter(pdata, threshold = 95, or.tied = T)
table(x95)
tablefixed(x95, NCOL(pdata))
cbind(at80=tablefixed(x80, NCOL(pdata)), at95=tablefixed(x95, NCOL(pdata)))