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, ...)a character or corpus object containing the texts
character vector defining the readability measure to calculate. Matches are case-insensitive. See other valid measures under Details.
if TRUE, treat constituent words in hyphenated as
separate terms, for purposes of computing word lengths, e.g.
"decision-making" as two terms of lengths 8 and 6 characters respectively,
rather than as a single word of 15 characters
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 corpus_trim.
if TRUE, include intermediate quantities in the output
not used
textstat_readability returns a data.frame of documents and
their readability scores.
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. $$0.886593 - 0.03640 \times AWL + 0.161911 \times AFW - 0.21401 \times ASL - 0.000577 \times ASL^2 - 0.000005 \times ASL^3$$
"Bormuth.GP":Bormuth's (1969) Grade Placement score.
$$4.275 + 12.881M - 34.934M^2 + 20.388 M^3 + 26.194 CCS -
2.046 CCS^2 - 11.767 CCS^3 - 42.285(M \times CCS) + 97.620(M \times CCS)^2 -
59.538(M \times CCS)^2$$
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. $$1.29 \times \frac{100 \times n_{wsy=1}}{n_{w}} - 38.45$$
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. $$1.16 \times \frac{100 \times n_{wsy=1}}{ Nw + 1.48 \times \frac{100 \times n_{st}}{n_{w}} - 37.95}$$
"Coleman.Liau.ECP":Coleman-Liau Estimated Cloze Percent (ECP) (Coleman and Liau 1975). $$141.8401 - 0.214590 \times 100 \times AWL + 1.079812 \times \frac{n_{st} \times 100}{n_{w}}$$
"Coleman.Liau.Grade":Coleman-Liau Grade Level (Coleman and Liau 1975). $$-27.4004 \times \mathtt{Coleman.Liau.ECP} \times 100 + 23.06395$$
"Coleman.Liau.Short":Coleman-Liau Index (Coleman and Liau 1975). $$5.88 \times AWL + 29.6 \times \frac{n_{st}}{n_{w}} - 15.8$$
"Dale.Chall":The New Dale-Chall Readability formula (Chall and Dale 1995). $$64 - (0.95 \times 100 \times \frac{n_{wd}}{n_{w}}) - (0.69 \times ASL)$$
"Dale.Chall.Old":The original Dale-Chall Readability formula (Dale and Chall (1948). $$0.1579 \times 100 \times \frac{n_{wd}}{n_{w}} + 0.0496 \times ASL [+ 3.6365]$$
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). $$0.1155 \times 100 \frac{n_{wd}}{n_{w}}) + (0.0596 \times ASL) + 3.2672 $$
"Danielson.Bryan":Danielson-Bryan's (1963) Readability Measure 1. $$ (1.0364 \times \frac{n_{c}}{n_{blank}}) + (0.0194 \times \frac{n_{c}}{n_{st}}) - 0.6059$$
where \(n_{blank}\) = Nblank = the number of blanks.
"Danielson.Bryan2":Danielson-Bryan's (1963) Readability Measure 2. $$ 131.059- (10.364 \times \frac{n_{c}}{n_{blank}}) + (0.0194 \times \frac{n_{c}}{n_{st}})$$
where \(n_{blank}\) = Nblank = the number of blanks.
"Dickes.Steiwer":Dickes-Steiwer Index (Dicks and Steiwer 1977). $$ 235.95993 - (7.3021 \times AWL) - (12.56438 \times ASL) - (50.03293 \times TTR)$$
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): $$\frac{n_{wsy>=2}}{n_{st}}$$
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). $$ -31.517 - (1.015 \times ASL) + (1.599 \times \frac{n_{wsy=1}}{n_{w}}$$
where \(n_{wsy=1}\) = Nwsy1 = the number of one-syllable words.
"Flesch":Flesch's Reading Ease Score (Flesch 1948). $$206.835 - (1.015 \times ASL) - (84.6 \times \frac{n_{sy}}{n_{w}})$$
"Flesch.PSK":The Powers-Sumner-Kearl's Variation of Flesch Reading Ease Score (Powers, Sumner and Kearl, 1958). $$ (0.0778 \times ASL) + (4.55 \times \frac{n_{sy}}{n_{w}}) - 2.2029$$
"Flesch.Kincaid":Flesch-Kincaid Readability Score (Flesch and Kincaid 1975). $$ 0.39 \times ASL + 11.8 \times \frac{n_{sy}}{n_{w}} - 15.59$$
"FOG":Gunning's Fog Index (Gunning 1952). $$0.4 \times (ASL + 100 \times \frac{n_{wsy>=3}}{n_{w}})$$
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). $$3.0680 \times (0.0877 \times ASL) +(0.0984 \times 100 \times \frac{n_{wsy>=3}}{n_{w}})$$
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). $$(\frac{(n_{wsy<3} + 3 \times n_{wsy=3})}{(100 \times \frac{N_{st}}{N_{w}})} - 3) / 2 $$
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). $$ 20 - \frac{n_{wsy=1} \times 150)}{(n_{w} \times 10)}$$
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). $$20.43 - 0.11 \times \frac{n_{wsy=1} \times 150)}{(n_{w} \times 10)}$$
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). $$\frac{[(100 - (\frac{100 \times n_{wsy<3}}{n_{w}})) + (3 \times \frac{100 \times n_{wsy>=3}}{n_{w}})]}{(100 \times \frac{n_{st}}{n_{w}})}$$
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":Bj<U+00F6>rnsson's (1968) L<U+00E4>sbarhetsindex (For Swedish Texts). $$ASL + \frac{100 \times n_{wsy>=7}}{n_{w}}$$
where \(n_{wsy>=7}\) = Nwmin7sy = the number of words with 7-syllables or more. The scaling by 100 arises because the L<U+00E4>sbarhetsindex index is based on just a sample of 100 words)
"nWS":Neue Wiener Sachtextformeln 1 (Bamberger and Vanecek 1984). $$19.35 \times \frac{n_{wsy>=3}}{n_{w}} + 0.1672 \times ASL + 12.97 \times \frac{b_{wchar>=6}}{n_{w}} - 3.27 \times \frac{n_{wsy=1}}{n_{w}} - 0.875$$
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). $$20.07 \times \frac{n_{wsy>=3}}{n_{w}} + 0.1682 \times ASL + 13.73 \times \frac{n_{wchar>=6}}{n_{w}} - 2.779$$
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). $$29.63 \times \frac{n_{wsy>=3}}{n_{w}} + 0.1905 \times ASL - 1.1144$$
where \(n_{wsy>=3}\) = Nwmin3sy = the number of words with 3 syllables or more.
"nWS.4":Neue Wiener Sachtextformeln 4 (Bamberger and Vanecek 1984). $$27.44 \times \frac{n_{wsy>=3}}{n_{w}} + 0.2656 \times ASL - 1.693$$
where \(n_{wsy>=3}\) = Nwmin3sy = the number of words with 3 syllables or more.
"RIX":Anderson's (1983) Readability Index. $$ \frac{n_{wsy>=7}}{n_{st}}$$
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). $$ 1.043 \times \sqrt{n_{wsy>=3}} \times \frac{30}{n_{st}} + 3.1291$$
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) $$0.9986 \times \sqrt{Nwmin3sy \times \frac{30}{n_{st}} + 5} + 2.8795$$
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. $$ \sqrt{Nwmin3sy \times \frac{30}{n_{st}}} + 3$$
"SMOG.de":Adaptation of McLaughlin's (1969) SMOG Measure for German Texts. $$ \sqrt{Nwmin3sy \times \frac{30}{n_{st}}-2}$$
"Spache":Spache's (1952) Readability Measure. $$ 0.121 \times ASL + 0.082 \times \frac{n_{wnotinspache}}{n_{w}} + 0.659$$
where \(n_{wnotinspache}\) = Nwnotinspache = number of unique words not in the Spache word list.
"Spache.old":Spache's (1952) Readability Measure (Old). $$0.141 \times ASL + 0.086 \times \frac{n_{wnotinspache}}{n_{w}} + 0.839$$
where \(n_{wnotinspache}\) = Nwnotinspache = number of unique words not in the Spache word list.
"Strain":Strain Index (Solomon 2006). $$n_{sy} / \frac{n_{st}}{3} /10$$
The scaling by 3 arises because the original Strain index is based on just the first 3 sentences.
"Traenkle.Bailer":Tr<U+00E4>nkle & Bailer's (1984) Readability Measure 1. $$224.6814 - (79.8304 \times AWL) - (12.24032 \times ASL) - (1.292857 \times 100 \times \frac{n_{prep}}{n_{w}}$$
where \(n_{prep}\) = Nprep = the number of prepositions. The scaling by 100 arises because the original Tr<U+00E4>nkle & Bailer index is based on just a sample of 100 words.
"Traenkle.Bailer2":Tr<U+00E4>nkle & Bailer's (1984) Readability Measure 2. $$Tr<U+00E4>nkle.Bailer2 = 234.1063 - (96.11069 \times AWL ) - (2.05444 \times 100 \times \frac{n_{prep}}{n_{w}}) - (1.02805 \times 100 \times \frac{n_{conj}}{n_{w}}$$
where \(n_{prep}\) = Nprep = the number of prepositions, \(n_{conj}\) = Nconj = the number of conjunctions, The scaling by 100 arises because the original Tr<U+00E4>nkle & Bailer index is based on just a sample of 100 words)
"Wheeler.Smith":Wheeler & Smith's (1954) Readability Measure. $$ ASL \times 10 \times \frac{n_{wsy>=2}}{n_{words}}$$
where \(n_{wsy>=2}\) = Nwmin2sy = the number of words with 2 syllables or more.
"meanSentenceLength":Average Sentence Length (ASL). $$\frac{n_{w}}{n_{st}}$$
"meanWordSyllables":Average Word Syllables (AWL). $$\frac{n_{sy}}{n_{w}}$$
Anderson, J. (1983). Lix and rix: Variations on a little-known readability index. Journal of Reading, 26(6), 490--496.
Bamberger, R. & Vanecek, E. (1984). Lesen-Verstehen-Lernen-Schreiben. Wien: Jugend und Volk.
Bj<U+00F6>rnsson, C. H. (1968). L<U+00E4>sbarhet. 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.
Caylor, J.S. (1973). Methodologies for Determining Reading Requirements of Military Occupational Specialities.
Caylor, J.S. & Sticht, T.G. (1973). Development of a Simple Readability Index for Job Reading Material.
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.
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 f<U+00FC>r die Deutsche Sprache. Zeitschrift f<U+00FC>r Entwicklungspsychologie und P<U+00E4>dagogische 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.
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.
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.
Tr<U+00E4>nkle, U. & Bailer, H. (1984). Kreuzvalidierung und Neuberechnung von Lesbarkeitsformeln f<U+00FC>r die deutsche Sprache. Zeitschrift f<U+00FC>r Entwicklungspsychologie und P<U+00E4>dagogische 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.
*Nimaldasan is the pen name of N. Watson Solomon, Assistant Professor of Journalism, School of Media Studies, SRM University, India.
# NOT RUN {
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(data_corpus_inaugural[48:58],
measure = c("Flesch.Kincaid", "Dale.Chall.old"))
# }
Run the code above in your browser using DataLab