Skip to content

Instantly share code, notes, and snippets.

@vpnagraj

vpnagraj/answer.R

Last active August 12, 2024 22:08
Show Gist options
  • Save vpnagraj/ad5d9e90769300d3e446d45e5bef026e to your computer and use it in GitHub Desktop.
Save vpnagraj/ad5d9e90769300d3e446d45e5bef026e to your computer and use it in GitHub Desktop.
Download ZIP
Example of reprex in R for asking *and* answering questions. The question.R script generates the content for question.md. The answer.R script generates the content for answer.md. The proof.R script uses the answer with data in example.csv.
Raw
answer.md 
library(tidyverse)

## define the example tibble
dat <-
  tribble(
  ~individual, ~Q1_A, ~Q1_B, ~Q1_C, ~Q1_D,   ~Q2_A, ~Q2_B, ~Q2_C,   ~Q2_D,
  "alice",    NA, "cat",    NA,    NA, "tacos",    NA,    NA,      NA,
  "bob", "dog",    NA,    NA,    NA,      NA,    NA,    NA, "pizza"
) 


## first ... use pivot_longer to get all of the "question columns" (starting with Q)
## this will split the question columns into two columns (question and choice) by the "_"
## the values will be the answer
## the values_drop_na=TRUE will exclude the responses that have NA (i.e., the "non choices")
dat %>%
  pivot_longer(cols = starts_with("Q"), names_prefix = "", names_sep = "_", names_to = c("question","choice"), values_to = c("answer"), values_drop_na = TRUE)
#> # A tibble: 4 × 4
#>   individual question choice answer
#>   <chr>      <chr>    <chr>  <chr> 
#> 1 alice      Q1       B      cat   
#> 2 alice      Q2       A      tacos 
#> 3 bob        Q1       A      dog   
#> 4 bob        Q2       D      pizza

Created on 2024-08-12 with reprex v2.0.2

Raw
answer.R
reprex::reprex({
library(tidyverse)
## define the example tibble
dat <-
tribble(
~individual, ~Q1_A, ~Q1_B, ~Q1_C, ~Q1_D, ~Q2_A, ~Q2_B, ~Q2_C, ~Q2_D,
"alice", NA, "cat", NA, NA, "tacos", NA, NA, NA,
"bob", "dog", NA, NA, NA, NA, NA, NA, "pizza"
)
## first ... use pivot_longer to get all of the "question columns" (starting with Q)
## this will split the question columns into two columns (question and choice) by the "_"
## the values will be the answer
## the values_drop_na=TRUE will exclude the responses that have NA (i.e., the "non choices")
dat %>%
pivot_longer(cols = starts_with("Q"), names_prefix = "", names_sep = "_", names_to = c("question","choice"), values_to = c("answer"), values_drop_na = TRUE)
})
Raw
example.csv
Name Income_A Income_B Income_C Income_D Age_A Age_B Age_C Age_D Alcohol_A Alcohol_B Alcohol_C Alcohol_D Exercise_A Exercise_B Exercise_C Exercise_D Politics_A Politics_B Politics_C Politics_D
Alice Jones 100-200k >65 1 drink daily >3 times weekly Republican
Bob Williams 50-100k 18-25 >3 drinks daily 2-3 times weekly Democrat
Carmen Smith >200k 0 40-64 None 1 time weekly Prefer not to say
Raw
proof.R
library(tidyverse)
## read in the example dataset
dat <- read_csv("example.csv")
## take a look at how the data is formatted
dat
## tailor the approach in answer.R to the dataset in question
dat %>%
pivot_longer(cols = c(starts_with("Inc"),starts_with("Age"),starts_with("Alc"),starts_with("Exe"),starts_with("Pol")), names_prefix = "", names_sep = "_", names_to = c("question","choice"), values_to = c("answer"), values_drop_na = TRUE)
Raw
question.md

i am trying to format answers to multiple choice questions in a data frame

i would like one row per respondee with question, choice, and answer in separate columns

library(tidyverse)

## here is what i have
tribble(
  ~individual, ~Q1_A, ~Q1_B, ~Q1_C, ~Q1_D,   ~Q2_A, ~Q2_B, ~Q2_C,   ~Q2_D,
  "alice",    NA, "cat",    NA,    NA, "tacos",    NA,    NA,      NA,
  "bob", "dog",    NA,    NA,    NA,      NA,    NA,    NA, "pizza"
)
#> # A tibble: 2 × 9
#>   individual Q1_A  Q1_B  Q1_C  Q1_D  Q2_A  Q2_B  Q2_C  Q2_D 
#>   <chr>      <chr> <chr> <lgl> <lgl> <chr> <lgl> <lgl> <chr>
#> 1 alice      <NA>  cat   NA    NA    tacos NA    NA    <NA> 
#> 2 bob        dog   <NA>  NA    NA    <NA>  NA    NA    pizza


## here is what i want
tribble(
  ~individual, ~question, ~choice, ~answer,
  "alice",      "Q1",     "B",   "cat",
  "alice",      "Q2",     "A", "tacos",
  "bob",      "Q1",     "A",   "dog",
  "bob",      "Q2",     "D", "pizza"
)
#> # A tibble: 4 × 4
#>   individual question choice answer
#>   <chr>      <chr>    <chr>  <chr> 
#> 1 alice      Q1       B      cat   
#> 2 alice      Q2       A      tacos 
#> 3 bob        Q1       A      dog   
#> 4 bob        Q2       D      pizza

Created on 2024-08-12 with reprex v2.0.2

Raw
question.R
## i am trying to format answers to multiple choice questions in a data frame
## i would like one row per respondee with question, choice, and answer in separate columns
reprex::reprex({
library(tidyverse)
## here is what i have
tribble(
~individual, ~Q1_A, ~Q1_B, ~Q1_C, ~Q1_D, ~Q2_A, ~Q2_B, ~Q2_C, ~Q2_D,
"alice", NA, "cat", NA, NA, "tacos", NA, NA, NA,
"bob", "dog", NA, NA, NA, NA, NA, NA, "pizza"
)
## here is what i want
tribble(
~individual, ~question, ~choice, ~answer,
"alice", "Q1", "B", "cat",
"alice", "Q2", "A", "tacos",
"bob", "Q1", "A", "dog",
"bob", "Q2", "D", "pizza"
)
})
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment