Readability: Readability Measures

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

Readability Measures

Description

automated_readability_index - Apply Automated Readability Index to transcript(s) by zero or more grouping variable(s).

coleman_liau - Apply Coleman Liau Index to transcript(s) by zero or more grouping variable(s).

SMOG - Apply SMOG Readability to transcript(s) by zero or more grouping variable(s).

flesch_kincaid - Flesch-Kincaid Readability to transcript(s) by zero or more grouping variable(s).

fry - Apply Fry Readability to transcript(s) by zero or more grouping variable(s).

linsear_write - Apply Linsear Write Readability to transcript(s) by zero or more grouping variable(s).

Usage

automated_readability_index(
  text.var,
  grouping.var = NULL,
  rm.incomplete = FALSE,
  ...
)

coleman_liau(text.var, grouping.var = NULL, rm.incomplete = FALSE, ...)

SMOG(
  text.var,
  grouping.var = NULL,
  output = "valid",
  rm.incomplete = FALSE,
  ...
)

flesch_kincaid(text.var, grouping.var = NULL, rm.incomplete = FALSE, ...)

fry(
  text.var,
  grouping.var = NULL,
  rm.incomplete = FALSE,
  auto.label = TRUE,
  grid = FALSE,
  div.col = "grey85",
  plot = TRUE,
  ...
)

linsear_write(text.var, grouping.var = NULL, rm.incomplete = FALSE, ...)

Arguments

text.var	The text variable.
grouping.var	The grouping variables. Default NULL generates one output for all text. Also takes a single grouping variable or a list of 1 or more grouping variables.
rm.incomplete	logical. If TRUE removes incomplete sentences from the analysis.
output	A character vector character string indicating output type. One of "valid" (default and congruent with McLaughlin's intent) or "all".
auto.label	logical. If TRUE labels automatically added. If FALSE the user clicks interactively.
grid	logical. If TRUE a micro grid is displayed, similar to Fry's original depiction, though this may make visualizing more difficult.
div.col	The color of the grade level division lines.
plot	logical. If TRUE a graph is plotted corresponding to Fry's graphic representation.
...	Other arguments passed to end_inc.

Value

Returns a list of 2 dataframes: (1) Counts and (2) Readability. Counts are the raw scores used to calculate readability score and can be accessed via counts. Readability is the dataframe with the selected readability statistic by grouping variable(s) and can be access via scores. The fry function returns a graphic representation of the readability as the scores returns the information for graphing but not a readability score.

Warning

Many of the indices (e.g., Automated Readability Index) are derived from word difficulty (letters per word) and sentence difficulty (words per sentence). If you have not run the sentSplit function on your data the results may not be accurate.

Fry

The fry function is based on Fry's formula that randomly samples 3 100 word length passages. If a group(s) in does not contain 300+ words they will not be included in the output.

References

Coleman, M., & Liau, T. L. (1975). A computer readability formula designed for machine scoring. Journal of Applied Psychology, Vol. 60, pp. 283-284.

Fry, E. B. (1968). A readability formula that saves time. Journal of Reading, 11(7), 513-516, 575-578.

Fry, E. B. (1969). The readability graph validated at primary levels. The Reading Teacher, 22(6), 534-538.

Flesch R. (1948). A new readability yardstick. Journal of Applied Psychology. Vol. 32(3), pp. 221-233. doi: 10.1037/h0057532.

Gunning, T. G. (2003). Building Literacy in the Content Areas. Boston: Allyn & Bacon.

McLaughlin, G. H. (1969). SMOG Grading: A New Readability Formula. Journal of Reading, Vol. 12(8), pp. 639-646.

Smith, E. A. & Senter, R. J. (1967) Automated readability index. Technical Report AMRLTR-66-220, University of Cincinnati, Cincinnati, Ohio.

Examples

## Not run: 
AR1 <- with(rajSPLIT, automated_readability_index(dialogue, list(person, act)))
ltruncdf(AR1,, 15)
scores(AR1)
counts(AR1)
plot(AR1)
plot(counts(AR1))

AR2 <- with(rajSPLIT, automated_readability_index(dialogue, list(sex, fam.aff)))
ltruncdf(AR2,, 15)
scores(AR2)
counts(AR2)
plot(AR2)
plot(counts(AR2))

AR3 <- with(rajSPLIT, automated_readability_index(dialogue, person))
ltruncdf(AR3,, 15)
scores(AR3)
head(counts(AR3))
plot(AR3)
plot(counts(AR3))

CL1 <- with(rajSPLIT, coleman_liau(dialogue, list(person, act)))
ltruncdf(CL1, 20)
head(counts(CL1))
plot(CL1)

CL2 <- with(rajSPLIT, coleman_liau(dialogue, list(sex, fam.aff)))
ltruncdf(CL2)
plot(counts(CL2))

(SM1 <- with(rajSPLIT, SMOG(dialogue, list(person, act))))
plot(counts(SM1))
plot(SM1)

(SM2 <- with(rajSPLIT, SMOG(dialogue, list(sex, fam.aff))))

(FL1 <- with(rajSPLIT, flesch_kincaid(dialogue, list(person, act))))
plot(scores(FL1))
plot(counts(FL1))

(FL2 <-  with(rajSPLIT, flesch_kincaid(dialogue, list(sex, fam.aff))))
plot(scores(FL2))
plot(counts(FL2))

FR1 <- with(rajSPLIT, fry(dialogue, list(sex, fam.aff)))
scores(FR1)
plot(scores(FR1))
counts(FR1)
plot(counts(FR1))

FR2 <- with(rajSPLIT, fry(dialogue, person))
scores(FR2)
plot(scores(FR2))
counts(FR2)
plot(counts(FR2))

FR3 <- with(pres_debates2012, fry(dialogue, list(time, person)))
colsplit2df(scores(FR3))
plot(scores(FR3), auto.label = FALSE)
counts(FR3)
plot(counts(FR3))

library(ggplot2)
ggplot(colsplit2df(counts(FR3)), aes(sent.per.100.wrds, 
    syllables.per.100.wrds)) +
    geom_point(aes(fill=person), shape=21, size=3) +
    facet_grid(person~time)
    
LW1 <- with(rajSPLIT, linsear_write(dialogue, list(person, act)))
plot(scores(LW1))
plot(counts(LW1))

LW2 <- with(rajSPLIT, linsear_write(dialogue, list(sex, fam.aff)))
plot(scores(LW2), method="lm")
plot(counts(LW2))

## End(Not run)

qdap documentation built on May 31, 2023, 5:20 p.m.