Calculate readability
Calculate the readability of text(s) using one of a variety of computed indexes.
textstat_readability( x, measure = "Flesch", remove_hyphens = TRUE, min_sentence_length = 1, max_sentence_length = 10000, intermediate = FALSE, ... )
x |
a character or corpus object containing the texts |
measure |
character vector defining the readability measure to calculate. Matches are case-insensitive. See other valid measures under Details. |
remove_hyphens |
if |
min_sentence_length, max_sentence_length |
set the minimum and maximum sentence lengths (in tokens, excluding punctuation) to include in the computation of readability. This makes it easy to exclude "sentences" that may not really be sentences, such as section titles, table elements, and other cruft that might be in the texts following conversion. For finer-grained control, consider filtering sentences prior first,
including through pattern-matching, using |
intermediate |
if |
... |
not used |
The following readability formulas have been implemented, where
Nw = n_{w} = number of words
Nc = n_{c} = number of characters
Nst = n_{st} = number of sentences
Nsy = n_{sy} = number of syllables
Nwf = n_{wf} = number of words matching the Dale-Chall List of 3000 "familiar words"
ASL = Average Sentence Length: number of words / number of sentences
AWL = Average Word Length: number of characters / number of words
AFW = Average Familiar Words: count of words matching the Dale-Chall list of 3000 "familiar words" / number of all words
Nwd = n_{wd} = number of "difficult" words not matching the Dale-Chall list of "familiar" words
"ARI":Automated Readability Index (Senter and Smith 1967)
0.5 ASL + 4.71 AWL - 21.34
"ARI.Simple":A simplified version of Senter and Smith's (1967) Automated Readability Index.
ASL + 9 AWL
"Bormuth.MC":Bormuth's (1969) Mean Cloze Formula.
0.886593 - 0.03640 * AWL + 0.161911 * AFW - 0.21401 * ASL - 0.000577 * ASL^2 - 0.000005 * ASL^3
"Bormuth.GP":Bormuth's (1969) Grade Placement score.
4.275 + 12.881M - 34.934M^2 + 20.388 M^3 + 26.194 CCS - 2.046 CCS^2 - 11.767 CCS^3 - 42.285(M * CCS) + 97.620(M * CCS)^2 - 59.538(M * CCS)^2
where M is the Bormuth Mean Cloze Formula as in
"Bormuth" above, and CCS is the Cloze Criterion Score (Bormuth,
1968).
"Coleman":Coleman's (1971) Readability Formula 1.
1.29 * (100 * Nwsy1 / Nw) - 38.45
where n_{wsy=1} = Nwsy1 = the number of one-syllable words. The scaling by 100 in this and the other Coleman-derived measures arises because the Coleman measures are calculated on a per 100 words basis.
"Coleman.C2":Coleman's (1971) Readability Formula 2.
1.16 * (100 * Nwsy1 / Nw) + 1.48 * (100 * Nst / Nw) - 37.95
"Coleman.Liau.ECP":Coleman-Liau Estimated Cloze Percent (ECP) (Coleman and Liau 1975).
141.8401 - (0.214590 * 100 * AWL) + (1.079812 * Nst * 100 / Nw)
"Coleman.Liau.grade":Coleman-Liau Grade Level (Coleman and Liau 1975).
-27.4004 * Coleman.Liau.ECP / 100 + 23.06395
"Coleman.Liau.short":Coleman-Liau Index (Coleman and Liau 1975).
5.88 * AWL + (0.296 * Nst / Nw) - 15.8
"Dale.Chall":The New Dale-Chall Readability formula (Chall and Dale 1995).
64 - (0.95 * 100 * Nwd / Nw) - (0.69 * ASL)
"Dale.Chall.Old":The original Dale-Chall Readability formula (Dale and Chall (1948).
0.1579 * 100 * Nwd / Nw + 0.0496 * ASL [+ 3.6365]
The additional constant 3.6365 is only added if (Nwd / Nw) > 0.05.
"Dale.Chall.PSK":The Powers-Sumner-Kearl Variation of the Dale and Chall Readability formula (Powers, Sumner and Kearl, 1958).
(0.1155 * 100 * Nwd / Nw) + (0.0596 * ASL) + 3.2672
"Danielson.Bryan":Danielson-Bryan's (1963) Readability Measure 1.
(1.0364 * Nc / Nblank) + (0.0194 * Nc / Nst) - 0.6059
where n_{blank} = Nblank = the number of blanks.
"Danielson.Bryan2":Danielson-Bryan's (1963) Readability Measure 2.
131.059 - (10.364 * Nc / Nblank) + (0.0194 * Nc / Nst)
where n_{blank} = Nblank = the number of blanks.
"Dickes.Steiwer":Dickes-Steiwer Index (Dicks and Steiwer 1977).
235.95993 - (73.021 * AWL) - (12.56438 * ASL) - (50.03293 * TTR)
where TTR is the Type-Token Ratio (see textstat_lexdiv())
"DRP":Degrees of Reading Power.
(1 - Bormuth.MC) * 100
where Bormuth.MC refers to Bormuth's (1969) Mean Cloze Formula (documented above)
"ELF":Easy Listening Formula (Fang 1966):
(Nwmin2sy / Nst)
where n_{wsy>=2} = Nwmin2sy = the number of words with 2 syllables or more.
"Farr.Jenkins.Paterson":Farr-Jenkins-Paterson's Simplification of Flesch's Reading Ease Score (Farr, Jenkins and Paterson 1951).
-31.517 - (1.015 * ASL) + (1.599 * Nwsy1 / Nw)
where n_{wsy=1} = Nwsy1 = the number of one-syllable words.
"Flesch":Flesch's Reading Ease Score (Flesch 1948).
206.835 - (1.015 * ASL) - (84.6 * (Nsy / Nw))
"Flesch.PSK":The Powers-Sumner-Kearl's Variation of Flesch Reading Ease Score (Powers, Sumner and Kearl, 1958).
(0.0078 * ASL) + (4.55 * Nsy / Nw) - 2.2029
"Flesch.Kincaid":Flesch-Kincaid Readability Score (Flesch and Kincaid 1975).
0.39 * ASL + 11.8 * (NSy /Nw) - 15.59
"FOG":Gunning's Fog Index (Gunning 1952).
0.4 * (ASL + 100 * (Nwmin3sy / Nw)
where n_{wsy>=3} = Nwmin3sy = the number of words with 3-syllables or more. The scaling by 100 arises because the original FOG index is based on just a sample of 100 words)
"FOG.PSK":The Powers-Sumner-Kearl Variation of Gunning's Fog Index (Powers, Sumner and Kearl, 1958).
3.0680 * (0.0877 * ASL) +(0.0984 * 100 * (Nwmin3sy / Nw)
where n_{wsy>=3} = Nwmin3sy = the number of words with 3-syllables or more. The scaling by 100 arises because the original FOG index is based on just a sample of 100 words)
"FOG.NRI":The Navy's Adaptation of Gunning's Fog Index (Kincaid, Fishburne, Rogers and Chissom 1975).
(((Nwless3sy + 3 * Nw3sy) / (100 * Nst / Nw))-3) / 2
where n_{wsy<3} = Nwless3sy = the number of words with less than 3 syllables, and n_{wsy=3} = Nw3sy = the number of 3-syllable words. The scaling by 100 arises because the original FOG index is based on just a sample of 100 words)
"FORCAST":FORCAST (Simplified Version of FORCAST.RGL) (Caylor and Sticht 1973).
20 - (Nwsy1 * 150) / (Nw * 10)
where n_{wsy=1} = Nwsy1 = the number of one-syllable words. The scaling by 150 arises because the original FORCAST index is based on just a sample of 150 words.
"FORCAST.RGL":FORCAST.RGL (Caylor and Sticht 1973).
20.43 - 0.11 * (Nwsy1 * 150) / (Nw * 10)
where n_{wsy=1} = Nwsy1 = the number of one-syllable words. The scaling by 150 arises because the original FORCAST index is based on just a sample of 150 words.
"Fucks":Fucks' (1955) Stilcharakteristik (Style Characteristic).
AWL * ASL
"Linsear.Write":Linsear Write (Klare 1975).
[(100 - (100 * Nwless3sy / Nw)) + (3 * 100 * Nwmin3sy / Nw)] / (100 * Nst / Nw)
where n_{wsy<3} = Nwless3sy = the number of words with less than 3 syllables, and n_{wsy>=3} = Nwmin3sy = the number of words with 3-syllables or more. The scaling by 100 arises because the original Linsear.Write measure is based on just a sample of 100 words)
"LIW":Björnsson's (1968) Läsbarhetsindex (For Swedish Texts).
ASL + (100 * Nwmin7sy / Nw)
where n_{wsy>=7} = Nwmin7sy = the number of words with 7-syllables or more. The scaling by 100 arises because the Läsbarhetsindex index is based on just a sample of 100 words)
"nWS":Neue Wiener Sachtextformeln 1 (Bamberger and Vanecek 1984).
(19.35 * Nwmin3sy / Nw) + (0.1672 * ASL) + (12.97 * Nwmin6char / Nw) - (3.27 * Nw1sy / Nw)- 0.875
where n_{wsy>=3} = Nwmin3sy = the number of words with 3 syllables or more, n_{wchar>=6} = Nwmin6char = the number of words with 6 characters or more, and n_{wsy=1} = Nwsy1 = the number of one-syllable words.
"nWS.2":Neue Wiener Sachtextformeln 2 (Bamberger and Vanecek 1984).
(20.07 * Nwmin3sy / Nw) + (0.1682 * ASL) + (13.73 * Nwmin6char / Nw) - 2.779
where n_{wsy>=3} = Nwmin3sy = the number of words with 3 syllables or more, and n_{wchar>=6} = Nwmin6char = the number of words with 6 characters or more.
"nWS.3":Neue Wiener Sachtextformeln 3 (Bamberger and Vanecek 1984).
(29.63 * Nwmin3sy / Nw) + (0.1905 * ASL) - 1.1144
where n_{wsy>=3} = Nwmin3sy = the number of words with 3 syllables or more.
"nWS.4":Neue Wiener Sachtextformeln 4 (Bamberger and Vanecek 1984).
(27.44 * Nwmin3sy / Nw) + (0.2656 * ASL) - 1.693
where n_{wsy>=3} = Nwmin3sy = the number of words with 3 syllables or more.
"RIX":Anderson's (1983) Readability Index.
Nwmin7sy / Nst
where n_{wsy>=7} = Nwmin7sy = the number of words with 7-syllables or more.
"Scrabble":Scrabble Measure.
Mean Scrabble Letter Values of All Words
. Scrabble values are for English. There is no reference for this, as we created it experimentally. It's not part of any accepted readability index!
"SMOG":Simple Measure of Gobbledygook (SMOG) (McLaughlin 1969).
1.043 * sqrt(Nwmin3sy * 30 / Nst) + 3.1291
where n_{wsy>=3} = Nwmin3sy = the number of words with 3 syllables or more. This measure is regression equation D in McLaughlin's original paper.
"SMOG.C":SMOG (Regression Equation C) (McLaughlin's 1969)
0.9986 * sqrt(Nwmin3sy * (30 / Nst) + 5) + 2.8795
where n_{wsy>=3} = Nwmin3sy = the number of words with 3 syllables or more. This measure is regression equation C in McLaughlin's original paper.
"SMOG.simple":Simplified Version of McLaughlin's (1969) SMOG Measure.
sqrt(Nwmin3sy * 30 / Nst) + 3
"SMOG.de":Adaptation of McLaughlin's (1969) SMOG Measure for German Texts.
sqrt(Nwmin3sy * 30 / Nst) - 2
"Spache":Spache's (1952) Readability Measure.
0.121 * ASL + 0.082 * Nwnotinspache / Nw) + 0.659
where n_{wnotinspache} = Nwnotinspache = number of unique words not in the Spache word list.
"Spache.old":Spache's (1952) Readability Measure (Old).
0.141 * ASL + 0.086 * (Nwnotinspache/ Nw) + 0.839
where n_{wnotinspache} = Nwnotinspache = number of unique words not in the Spache word list.
"Strain":Strain Index (Solomon 2006).
Nsy / (Nst / 3) / 10
The scaling by 3 arises because the original Strain index is based on just the first 3 sentences.
"Traenkle.Bailer":Tränkle & Bailer's (1984) Readability Measure 1.
224.6814 - (79.8304 * AWL) + (12.24032 * ASL) - (1.292857 * 100 * Nprep / Nw)
where n_{prep} = Nprep = the number of prepositions. The scaling by 100 arises because the original Tränkle & Bailer index is based on just a sample of 100 words.
"Traenkle.Bailer2":Tränkle & Bailer's (1984) Readability Measure 2.
234.1063 - 96.11069 * AWL - 2.05444 * 100 * (Nprep / Nw) - 1.02805 * 100 * (Nconj / Nw).
where n_{prep} = Nprep = the number of prepositions, n_{conj} = Nconj = the number of conjunctions, The scaling by 100 arises because the original Tränkle & Bailer index is based on just a sample of 100 words)
"Wheeler.Smith":Wheeler & Smith's (1954) Readability Measure.
ASL * 10 * (Nwmin2sy / Nw)
where n_{wsy>=2} = Nwmin2sy = the number of words with 2 syllables or more.
"meanSentenceLength":Average Sentence Length (ASL).
Nw / Nst
"meanWordSyllables":Average Word Syllables (AWL).
Nsy / Nw
textstat_readability returns a data.frame of documents and
their readability scores.
Kenneth Benoit, re-engineered from Meik Michalke's koRpus package.
Anderson, J. (1983). Lix and rix: Variations on a little-known readability
index. Journal of Reading, 26(6),
490–496. https://www.jstor.org/stable/40031755
Bamberger, R. & Vanecek, E. (1984). Lesen-Verstehen-Lernen-Schreiben. Wien: Jugend und Volk.
Björnsson, C. H. (1968). Läsbarhet. Stockholm: Liber.
Bormuth, J.R. (1969). Development of Readability Analysis.
Bormuth, J.R. (1968). Cloze test readability: Criterion reference
scores. Journal of educational
measurement, 5(3), 189–196. https://www.jstor.org/stable/1433978
Caylor, J.S. (1973). Methodologies for Determining Reading Requirements of
Military Occupational Specialities. https://eric.ed.gov/?id=ED074343
Caylor, J.S. & Sticht, T.G. (1973). Development of a Simple Readability
Index for Job Reading Material
https://archive.org/details/ERIC_ED076707
Coleman, E.B. (1971). Developing a technology of written instruction: Some determiners of the complexity of prose. Verbal learning research and the technology of written instruction, 155–204.
Coleman, M. & Liau, T.L. (1975). A Computer Readability Formula Designed for Machine Scoring. Journal of Applied Psychology, 60(2), 283.
Dale, E. and Chall, J.S. (1948). A Formula for Predicting Readability:
Instructions. Educational Research
Bulletin, 37-54. https://www.jstor.org/stable/1473169
Chall, J.S. and Dale, E. (1995). Readability Revisited: The New Dale-Chall Readability Formula. Brookline Books.
Dickes, P. & Steiwer, L. (1977). Ausarbeitung von Lesbarkeitsformeln für die Deutsche Sprache. Zeitschrift für Entwicklungspsychologie und Pädagogische Psychologie 9(1), 20–28.
Danielson, W.A., & Bryan, S.D. (1963). Computer Automation of Two Readability Formulas. Journalism Quarterly, 40(2), 201–206.
DuBay, W.H. (2004). The Principles of Readability.
Fang, I. E. (1966). The "Easy listening formula". Journal of Broadcasting & Electronic Media, 11(1), 63–68.
Farr, J. N., Jenkins, J.J., & Paterson, D.G. (1951). Simplification of Flesch Reading Ease Formula. Journal of Applied Psychology, 35(5): 333.
Flesch, R. (1948). A New Readability Yardstick. Journal of Applied Psychology, 32(3), 221.
Fucks, W. (1955). Der Unterschied des Prosastils von Dichtern und anderen Schriftstellern. Sprachforum, 1, 233-244.
Gunning, R. (1952). The Technique of Clear Writing. New York: McGraw-Hill.
Klare, G.R. (1975). Assessing
Readability. Reading Research
Quarterly, 10(1), 62-102. https://www.jstor.org/stable/747086
Kincaid, J. P., Fishburne Jr, R.P., Rogers, R.L., & Chissom, B.S. (1975). Derivation of New Readability Formulas (Automated Readability Index, FOG count and Flesch Reading Ease Formula) for Navy Enlisted Personnel.
McLaughlin, G.H. (1969). SMOG Grading: A New Readability Formula. Journal of Reading, 12(8), 639-646.
Michalke, M. (2014). koRpus: An R Package for Text Analysis (Version 0.05-4). Available from https://reaktanz.de/?c=hacking&s=koRpus.
Powers, R.D., Sumner, W.A., and Kearl, B.E. (1958). A Recalculation of Four Adult Readability Formulas.. Journal of Educational Psychology, 49(2), 99.
Senter, R. J., & Smith, E. A. (1967). Automated readability index. Wright-Patterson Air Force Base. Report No. AMRL-TR-6620.
*Solomon, N. W. (2006). Qualitative Analysis of Media Language. India.
Spache, G. (1953). "A new readability formula for primary-grade reading
materials." The Elementary School Journal, 53, 410–413.
https://www.jstor.org/stable/998915
Tränkle, U. & Bailer, H. (1984). Kreuzvalidierung und Neuberechnung von Lesbarkeitsformeln für die deutsche Sprache. Zeitschrift für Entwicklungspsychologie und Pädagogische Psychologie, 16(3), 231–244.
Wheeler, L.R. & Smith, E.H. (1954). A Practical Readability Formula for the
Classroom Teacher in the Primary Grades. Elementary English, 31,
397–399. https://www.jstor.org/stable/41384251
*Nimaldasan is the pen name of N. Watson Solomon, Assistant Professor of Journalism, School of Media Studies, SRM University, India.
txt <- c(doc1 = "Readability zero one. Ten, Eleven.",
doc2 = "The cat in a dilapidated tophat.")
textstat_readability(txt, measure = "Flesch")
textstat_readability(txt, measure = c("FOG", "FOG.PSK", "FOG.NRI"))
textstat_readability(quanteda::data_corpus_inaugural[48:58],
measure = c("Flesch.Kincaid", "Dale.Chall.old"))Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.