question_type: Count of Question Type

Description

Transcript apply question counts.

Usage

question_type(text.var, grouping.var = NULL, neg.cont = FALSE,
  percent = TRUE, zero.replace = 0, digits = 2,
  contraction = qdapDictionaries::contractions, bracket = "all",
  amplifiers = qdapDictionaries::amplification.words, ...)

Arguments

text.var

The text variable

grouping.var

The grouping variables. Default NULL generates one word list for all text. Also takes a single grouping variable or a list of 1 or more grouping variables.

neg.cont

logical. If TRUE provides separate counts for the negative contraction forms of the interrogative words.

percent

logical. If TRUE output given as percent. If FALSE the output is proportion.

zero.replace

Value to replace 0 values with.

digits

Integer; number of decimal places to round when printing.

contraction

A two column key of contractions (column 1) and expanded form replacements (column 2) or a vector of contractions. Default is to use qdapDictionaries's contractions data se

bracket

The type of bracket (and encased text) to remove. This is one or more of the strings "curly", "square", "round", "angle" and "all". These strings correspond to: {, [, (, < or a

amplifiers

A character vector of terms that increase the intensity of a positive or negative word. Default is to use qdapDictionaries's amplification.words data set.

...

Other arguments passed to bracketX.

Value

Returns a list of:
rawA dataframe of the questions used in the transcript and their type.
countA dataframe of total questions (tot.quest) and counts of question types (initial interrogative word) by grouping variable(s).
rnpDataframe of the frequency and proportions of question types by grouping variable.
indsThe indices of the original text variable that contain questions.
missingThe row numbers of the missing data (excluded from analysis).
percentThe value of percent used for plotting purposes.
zero.replaceThe value of zero.replace used for plotting purposes.

Details

The algorithm searches for the following interrogative words (and optionally, their negative contraction form as well): 1) whose 2) whom 3) who 4) where 5) what 6) which 7) why 8) when 9) were* 10) was* 11) does* 12) did* 13) do* 14) is 15) are* 16) will* 17) how 18) should 19) could 20) would* 21) shall 22) may 23) might* 24) must* 25) can* 26) has 27) have* 28) had* 29) ok 30) right 31) correct 32) implied do/does/did The interrogative word that is found first (with the exception of "ok", "right"/"alright", and "correct") in the question determines the sentence type. "ok", "right"/"alright", and "correct" sentence types are determined if the sentence is a question with no other interrogative words found and "ok", "right"/"alright", or "correct" is the last word of the sentence. Those interrogative sentences beginning with the word "you", "wanna", or "want" are categorized as implying do/does/did question type, though the use of do/does/did is not explicit. Those sentence beginning with "you" followed by a select interrogative word (and or their negative counter parts) above (marked with *) or 1-2 amplifier(s) followed by the select interrogative word are categorized by the select word rather than an implied do/does/did question type. A sentence that is marked "ok" over rides an implied do/does/did label. Those with undetermined sentence type are labeled unknown.

Examples

Run this code

## Inspect the algorithm classification
x <- c("Kate's got no appetite doesn't she?",
    "Wanna tell Daddy what you did today?",
    "You helped getting out a book?", "umm hum?",
    "Do you know what it is?", "What do you want?",
    "Who's there?", "Whose?", "Why do you want it?",
    "Want some?", "Where did it go?", "Was it fun?")

left_just(preprocessed(question_type(x))[, c(2, 6)])

## Transcript/dialogue examples
(x <- question_type(DATA.SPLIT$state, DATA.SPLIT$person))

## methods
scores(x)
plot(scores(x))
counts(x)
plot(counts(x))
proportions(x)
plot(proportions(x))
truncdf(preprocessed(x), 15)
plot(preprocessed(x))

plot(x)
plot(x, label = TRUE)
plot(x, label = TRUE, text.color = "red")
question_type(DATA.SPLIT$state, DATA.SPLIT$person, percent = FALSE)
DATA[8, 4] <- "Won't I distrust you?"
question_type(DATA.SPLIT$state, DATA.SPLIT$person)
DATA <- qdap::DATA
with(DATA.SPLIT, question_type(state, list(sex, adult)))

out1 <- with(mraja1spl, question_type(dialogue, person))
## out1
out2 <- with(mraja1spl, question_type(dialogue, list(sex, fam.aff)))
## out2
out3 <- with(mraja1spl, question_type(dialogue, list(sex, fam.aff),
   percent = FALSE))
plot(out3, label = TRUE, lab.digits = 3)

Run the code above in your browser using DataLab

Data Engineering and BI courses are free this week!