survey_report.R in R

Clean a survey data frame and report grouped results in base R.

#!/usr/bin/env Rscript
# Clean a small survey data frame and report grouped results in base R.

survey <- data.frame(
  respondent = 1:8,
  region = c("north", "south", "north", "east",
             "south", "east", "north", NA),
  rating = c(4, 5, NA, 3, 5, 2, 4, 4),
  minutes = c(12, 25, 18, 7, 31, 9, 22, 14),
  stringsAsFactors = FALSE
)

clean <- survey[!is.na(survey$region) & !is.na(survey$rating), ]
clean$region <- factor(clean$region,
                       levels = c("north", "south", "east"))
clean$engaged <- clean$minutes > 15

cat("responses:", nrow(survey), "->", nrow(clean), "after cleaning\n\n")

by_region <- aggregate(cbind(rating, minutes) ~ region,
                       data = clean, FUN = mean)
by_region$rating <- round(by_region$rating, 2)
by_region$minutes <- round(by_region$minutes, 1)
print(by_region)

cat("\nengagement by region:\n")
print(table(clean$region, clean$engaged))

cat("\noverall mean rating:", round(mean(clean$rating), 2), "\n")
cat("correlation rating~minutes:",
    round(cor(clean$rating, clean$minutes), 3), "\n")

How it works

1. Rows with missing region or rating are dropped.
2. factor orders regions; a derived column flags engagement.
3. aggregate and table produce the grouped report.

Keywords and builtins used here

• aggregate
• c
• cat
• cbind
• cor
• factor
• mean
• nrow
• print
• round
• table

The run, in numbers

Lines

31

Characters to type

994

Tokens

272

Three-star pace

105 tpm

At the three-star pace of 105 tokens a minute, this run takes about 155 seconds.

Type this snippet

Step 1 of 1 in Encore, step 8 of 8 in Data frames in base R.

← Previous