dispersion_plot: Lexical Dispersion Plot

In trinker/qdap: Bridging the Gap Between Qualitative Data and Quantitative Analysis

Description

Generate a lexical dispersion plot of terms.

Usage

dispersion_plot(
  text.var,
  match.terms,
  grouping.var = NULL,
  rm.vars = NULL,
  color = "blue",
  bg.color = "grey90",
  horiz.color = "grey85",
  total.color = "black",
  symbol = "|",
  title = "Lexical Dispersion Plot",
  rev.factor = TRUE,
  wrap = "'",
  xlab = "Dialogue (Words)",
  ylab = NULL,
  size = 4,
  plot = TRUE,
  char2space = "~~",
  apostrophe.remove = FALSE,
  scales = "free",
  space = "free",
  ...
)

Arguments

text.var	The text variable.
match.terms	A vector of quoted terms or a named list of quoted terms. If the latter terms will be combined into a single unified theme named according to the list names. Note that terms within the vectors of the list cannot be duplicated.
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.
rm.vars	The repeated measures variables. Default NULL generates one facet for all text. Also takes a single repeated measures variable or a list of 1 or more grouping variables.
color	The color of the word symbols.
bg.color	The background color.
horiz.color	The color of the horizontal tracking stripe. Use horiz.color = bg.color to eliminate.
total.color	The color to use for summary 'all' group. If NULL totals are dropped.
symbol	The word symbol. Default is "|".
title	Title of the plot
rev.factor	logical. If TRUE reverses the plot order of the factors.
wrap	a character to wrap around the words (enables the reader to visualize spaces). Default is "'", use "" to remove.
xlab	The x label.
ylab	The y label.
size	The size of the plotting symbol.
plot	logical. If TRUE the plot will automatically plot. The user may wish to set to FALSE for use in knitr, sweave, etc. to add additional plot layers.
char2space	A vector of characters to be turned into spaces.
apostrophe.remove	logical. If TRUE removes apostrophes from the output.
scales	Should scales be fixed ("fixed", the default), free ("free"), or free in one dimension ("free_x", "free_y")
space	If "fixed", the default, all panels have the same size. If "free_y" their height will be proportional to the length of the y scale; if "free_x" their width will be proportional to the length of the x scale; or if "free" both height and width will vary.
...	Other argument supplied to strip.

Value

Plots a dispersion plot and invisibly returns the ggplot2 object.

Note

The match.terms is character sensitive. Spacing is an important way to grab specific words and requires careful thought. Using "read" will find the words "bread", "read" "reading", and "ready". If you want to search for just the word "read" you'd supply a vector of c(" read ", " reads", " reading", " reader").

See Also

term_match

Examples

## Not run: 
term_match(raj$dialogue, c(" love ", "love", " night ", "night"))
dispersion_plot(raj$dialogue, c(" love ", "love", " night ", "night"))
dispersion_plot(raj$dialogue, c("love", "night"), rm.vars = raj$act)
with(rajSPLIT , dispersion_plot(dialogue, c("love", "night"), 
    grouping.var = list(fam.aff, sex), rm.vars = act))

## With grouping variables
with(rajSPLIT , dispersion_plot(dialogue, c("love", "night"), 
     grouping.var = sex, rm.vars = act))

## Drop total with `total.color = NULL`
with(rajSPLIT , dispersion_plot(dialogue, c("love", "night"), 
     grouping.var = sex, rm.vars = act, total.color = NULL))

## Change color scheme
with(rajSPLIT, dispersion_plot(dialogue, c("love", "night"), 
    bg.color = "black", grouping.var = list(fam.aff, sex), 
    color = "yellow", total.color = "white", horiz.color="grey20"))
    
## Use `word_list`
## Presidential debates by all
wrds <- word_list(pres_debates2012$dialogue, stopwords = Top200Words)
wrds2 <- spaste(wrds[["rfswl"]][["all"]][, "WORD"])
wrds2 <- c(" governor~~romney ", wrds2[-c(3, 12)])
with(pres_debates2012 , dispersion_plot(dialogue, wrds2, rm.vars = time))

## Presidential debates by person
dat <- pres_debates2012
dat <- dat[dat$person %in% qcv(ROMNEY, OBAMA), ]

wordlist <- c(" tax", " health", " rich ", "america", " truth", 
    " money", "cost", " governnor", " president", " we ", 
    " job", " i ", " you ", " because ", " our ", " years ")

with(dat, dispersion_plot(dialogue, wordlist, total.color = NULL, 
    bg.color = "white", grouping.var = person, rm.vars = time,
    color = "black", horiz.color="grey80"))

wordlist2 <- c(" i'd ", " i'll ", " i'm ", " i've ", " i ", 
    " we'd ", " we'll ", " we're ", " we've ", " we ", 
    " you'd ",  " you'll ", " you're ", " you've ", " you ", " your ",
    " he'd ", " he'll ", " he's ", " he ")

with(dat, dispersion_plot(dialogue, wordlist2, 
    bg.color = "black", grouping.var = person, rm.vars = time,
    color = "yellow", total.color = NULL, horiz.color="grey20"))
   
with(dat, dispersion_plot(dialogue, wordlist2, 
    bg.color = "black", grouping.var = person, rm.vars = time,
    color = "red", total.color = "white", horiz.color="grey20"))

## `match.terms` as a named list        
wordlist3 <- list(
    I = c(" i'd ", " i'll ", " i'm ", " i've ", " i "),
    we = c(" we'd ", " we'll ", " we're ", " we've ", " we "),
    you = c(" you'd ",  " you'll ", " you're ", " you've ", " you ", " your "),
    he = c(" he'd ", " he'll ", " he's ", " he ")
)

with(dat, dispersion_plot(dialogue, wordlist3,
    bg.color = "grey60", grouping.var = person, rm.vars = time,
    color = "blue", total.color = "grey40", horiz.color="grey20"))

colsplit2df(scores(with(dat, termco(dialogue, list(time, person), wordlist3))))

## Extras:
## Reverse facets

x <- with(pres_debates2012 , dispersion_plot(dialogue, wrds2, rm.vars = time))

## function to reverse ggplot2 facets
rev_facet <- function(x) {
    names(x$facet)[1:2] <- names(x$facet)[2:1]
    print(x)
}

rev_facet(x)

## Discourse Markers: See...
## Schiffrin, D. (2001). Discourse markers: Language, meaning, and context. 
##    In D. Schiffrin, D. Tannen, & H. E. Hamilton (Eds.), The handbook of 
##    discourse analysis (pp. 54-75). Malden, MA: Blackwell Publishing.

discoure_markers <- list(
    response_cries = c(" oh ", " ah ", " aha ", " ouch ", " yuk "),
    back_channels = c(" uh-huh ", " uhuh ", " yeah "), 
    summons = " hey ", 
    justification = " because "
)

(markers <- with(pres_debates2012, 
    termco(dialogue, list(person, time), discoure_markers)
))
plot(markers, high="red")

with(pres_debates2012, 
    termco(dialogue, list(person, time), discoure_markers, elim.old = FALSE)
)

with(pres_debates2012, 
    dispersion_plot(dialogue, unlist(discoure_markers), person, time)
)

## End(Not run)

trinker/qdap documentation built on Sept. 30, 2020, 6:28 p.m.