Linguistic inquiry and word count: LIWC 2001

Tags: LIWC2007 Operators Manual, category, punctuation, Operators Manual, numbers, Linguistic Inquiry and Word Count, deutschen Version, Pennebaker, SelectAll, German translation, The University of Texas at Austin, University of Auckland, James W. Pennebaker, Operator's Manual, Roger J Booth, Written Text, Language Analysis, Austin, Texas
Content: Operator's Manual
Linguistic Inquiry and word count: LIWC2007 James W. Pennebaker, Roger J Booth, and Martha E. Francis The University of Texas at Austin and The University of Auckland, New Zealand The LIWC2007 software, Operator's Manual, and the LIWC2007 Language Analysis Manual are published by LIWC.net, Austin, Texas 78703 USA.
LIWC2007 Operators Manual
Page 2
Contents Getting Started ................................................................................................................................ 1 Running LIWC2007 on a PC....................................................................................................... 2 Running LIWC2007 on a Macintosh........................................................................................... 2 Reading and Analyzing LIWC2007 Output ................................................................................ 3 Customizing LIWC2007 Output ................................................................................................. 3 Analyzing Text in Segments........................................................................................................ 3 Handling numerals, abbreviations and emoticons....................................................................... 3 Handling punctuation .................................................................Error! Bookmark not defined. Creating and Using Custom Dictionaries .................................................................................... 4 Conditional Categories ................................................................................................................ 6 Preparing written text For LIWC2007 Analysis........................................................................... 6 1. text file organization. .................................................................................................... 6 2. Text file computer entry................................................................................................. 6 3. Cleaning the text files..................................................................................................... 6 Naming Text Files ....................................................................................................................... 7 Typing Conventions: Writing and Interview Samples ............................................................... 7 1. Spelling, abbreviations, contractions. ............................................................................ 7 2. End of sentence markers and hyphens. .......................................................................... 7 3. Other common problems:............................................................................................... 8 Transcribing Oral Transcripts: Special Problems....................................................................... 8 1. Nonfluencies................................................................................................................... 8 2. Fillers.............................................................................................................................. 9 3. Transcribers' comments. ................................................................................................ 9 technical support ........................................................................................................................... 9 Getting Some Practice: Running the Samples ............................................................................. 10
LIWC2007 Operators Manual
Page 1
Getting Started
TheLIWC2007programcomeswiththefollowingfiles:

LIWC2007 theactualapplicationfile(LIWC2007.EXEforWindowsand
LIWC2007forMacintosh)incorporatingtheLIWC2007and
LIWC2001masterdictionaries.NotethattheLIWCStudentVersions
onlyincludetheinternalLIWC2007dictionaryandnootherancillary
files.

SAMPLES adirectoryofsampletextfiles,includinginaugurationspeechesby
Lincoln,FranklinRoosevelt,andClinton
(Lincoln.txt,FDR.txt,Clinton.txt)


2poemsbySylviaPlathandAnneSexton
(Plath.txt,Sexton.txt)


2talkshowsegments:HowardStern(radio),DonnaShelala(TV)
(Radio.txt,talkshow.txt)


2filesofapassagefromHuckleberryFinn--oneoriginal,one"cleaned"
(Huckraw.txt,Huckcln.txt)


2psychologyjournalabstracts
(Abstr1.txt,Abstr2.txt)

DICTIONARIES Thedictionariesincluded:


LIWC2007.dicisacopyoftheinternaldefaultdictionary.Notethatthisis
nottheactualinternaldictionarythattheLIWC2007programruns.
Anychangestothisdictionarywillonlytakeeffectifthisdictionaryis
loadedaspartofthe"LoadNewDictionary"commandinthe
"Dictionary"menu.


LIWC2001.dicisacopyoftheinternaldefaultdictionaryusedontheoriginal
LIWC2001program
Spanish2001.dic is a Spanish translation of the LIWC2001 dictionary developed by Ramirez-Esparza, N., Pennebaker, J.W., Garcia, F.A., & Suria, R. (2007). La psychologнa del uso de las Palabras: Un programa de comutadora que analiza textos en Espaсol (The psychology of word use: A computer program that analyzes texts in Spanish). Revista Mexicana de Psicologнa, 24, 85-99.
German2001.dic is a German translation of the LIWC2001 dictionary developed by Wolf, M., Horn, A., Mehl, M., Haug, S., Pennebaker, J. W., & Kordy, H. (2008, in press). Computergestьtzte quantitative Textanalyse: Дquivalenz und Robustheit der deutschen Version des Linguistic Inquiry and Word Count [Computer-aided quantitative Text analysis: Equivalence and robustness of the German adaption of the Linguistic Inquiry and Word Count]. Diagnostica.
LIWC2007 Operators Manual
Page 2


Pronoun.dicisashortsampledictionaryofpronouns.Itisincludedasa
simpleexampleofthedictionarysystem.


LIWC2007wordcategoryfile(LIWC2007dictionaryposter.xls)isanExcel
filethatlistsallthewordsthatareintheLIWC2007dictionaryby
category.
Running LIWC2007 on a PC
Toruntheapplication,doubleclickontheLIWC2007iconorLIWC2007.EXEfile.Oncethe LIWC2007applicationlaunches,explorethevariousoptions.
Toanalyzewhatevertextfilesyouspecify,gointothe"File"menuandselect"Process Text..."(orclickonthe`AskLIWC2007toprocessafile(s)'icon).Multiplefilescanbe processedinoneoftwoways.Eithershift-clickonallthefilenamesyourrequire,or alternatively,youcananalyseallfilesinaparticulardirectorybyclickingthe"SelectAll" button.TheSelectAlloptionwillanalyzeallfilesinthatparticulardirectorythataretext (.txt)orWorddocument(.doc)files.Ifyouhavedirectorieswithinthecurrentdirectory, youcanhaveLIWC2007processallthetextfileswithintheseaswellbycheckingthe" Includefilesinencloseddirectories"checkboxbeforeclickingthe"SelectAll"button".
Tip:Ifyouhavealargenumberoftextfilestoprocess,itisgenerallymostefficienttoput themallinonedirectory(ordirectories)andthenusethe"SelectAll"buttononthat directory.Youcanalsoselectmultiplefileswithinadirectorybyusingshift-clickingor control-clicking.OnaPCcomputer,pointtoaparticularfileandholddowntheshiftkey beforeclicking.Youcanthenpointtoalaterfileinthesamedirectoryandagaindepress theshiftkeybeforeclicking.Allfilesbetweenthetwoclicked-onfileswillnowbeselected. Control-clickingsimplyrequiresthatyouholddownthecontrolkeyandclickonthe individualfilesyouwishtoselect.
Youthengettheopportunitytospecifyanameandlocationforyouroutputfilebefore LIWC2007beginsprocessing.LIWC2007processesthefilessequentially,showingyouits progress,storingtheoutputinthefileyouspecifiedandthendisplayingresultsina windowonthescreen.Theoutputfileissavedintab-delimitedtextthatincludesthe variablenamesonthefirstline.Thisallowsittobereaddirectlyintoprogramssuchas Excel,SPSS,orSAS.

Running LIWC2007 on a Macintosh
Toruntheapplication,doubleclickontheLIWC2007icon.OncetheLIWC2007application launches,explorethevariousoptions.
Toanalyzewhatevertextfilesyouspecify,gointothe"File"menuandselect"Process Text...".Selectthefilesthatyouwithtoanalyzebyshift-clickingorcommand-clickingthem inthedialogbox.
Youthengettheopportunitytospecifyanameandlocationforyouroutputfilebefore LIWC2007beginsprocessing.LIWC2007processesthefilessequentially,showingyouits progress,storingtheoutputinthefileyouspecifiedandthendisplayingresultsina windowonthescreen.Theoutputfileissavedintab-delimitedtextthatincludesthe
LIWC2007 Operators Manual
Page 3
variablenamesonthefirstline.ThisallowsittobereaddirectlyintoSPSSorExcel programs.
Reading and Analyzing LIWC2007 Output LIWC2007storestheoutputinthefileyouspecifiedandthendisplaysresultsinawindow onthescreen.Bydefault,allLIWC2007outputvariablesarelistedconsecutivelyinthe outputfile.Theoutputfileissavedintab-delimitedtextthatincludesthevariablenames onthefirstline.ThisallowsittobereaddirectlyintoSPSSorExcelprograms. ToviewanyLIWC2007outputfile,choosethe"Open"commandwithinthe"File"Menu(or clickonthe`Openanexistingdocument'iconinWindows)andspecifyanoutputfilename. Alternatively,theoutputfilecanbeopenedwithanywordprocessingprogram(e.g.,Word, WordPerfect).Forthebestviewoftheoutputfile,however,aspreadsheetprogram,such asExcel,isrecommended.
Customizing LIWC2007 Output Insomecases,youmayprefertoanalyzeonlyasubsetoflanguagedimensionsratherthan thefullsetofvariables.Todothis,openthe"Categories"menu.Withineachoption(e.g., standardinformation,linguisticdimensionsetc.),checkboxesareavailableforeach LIWC2007dimension.Byclickingoneachdimensionandremovingthecheckmark,the outputcategorycanbeomittedfromtheanalyses.Notethatthecategorypreferenceswill remainineffectuntiltheyarere-checkedandwillbesavedwhentheapplicationisquit.To usealldimensions,choose"Useallcategories"fromthe"Categories"menu.
Analyzing Text in Segments EachtextfileanalyzedbyLIWC2007canbetreatedasawholeorbrokenintosegmentsin oneofthreeways.Thisiscontrolledbythe"Analyzeinsegments..."commandonthe "Options"menu.Youhavefourchoiceshere:(1)Notextsegmentation;(2)Definenumberof segments,inwhichcaseyoucanchoosehowmanysegmentsyouwishtodivideyourtext filesinto;(3)Definewordspersegment,inwhichcaseyouchoosehowmanywordsineach segment;and(4)Definesegmentdelimiter,inwhichcaseyouwillhavesegmentsofyour textseparatedbyanumberofblanklinesandLIWC2007willusethesetobreakyourtext intosegments.TheactivesegmentationisdisplayedintheWindowsversionofLIWC2007 onthestatuslineatthebottomoftheapplicationwindow,andthetheMacintoshversionin afloatingwindowcalled"AnalysisStatus".NotethatintheLIWCoutputfile,thesecond columnreferstotheactualsegmentsequence.
Handling numerals and punctuation The"Extras..."itemofthe"Categories"menuallowsyoutodeterminehowLIWC2007 handlesnumerals(e.g.12,38,156).IneachcaseyoucanhaveLIWC2007ignorethemby
LIWC2007 Operators Manual
Page 4
clickingonthe"Ignorethem"radiobuttonorhavethemlistedasseparatecategoriesby clickingonthe"Listthemseparately"radiobutton.
Numeralscanalsobeaddedtothe"numbers"category.Thedefault"numbers"category looksonlyforwords(e.g.,seven,thousands).Byclickingonthe"Addto`numbers' category"button,numeralsequencesareconsideredwordunitsandarecountedinthe samewayasnumberwords.
The"Punctuation..."itemofthe"Categories"menuallowsyoutodeterminehowLIWC2007 countsandreportspunctuationcharacters.Byclickingontheiteminthedialogboxyou canswitchonoroffthefollowingpunctuationcharactersandLIWC2007willcountthem andreportthemasapercentageoftotalwords:
Period .
Comma ,
Colon
:
SemiC ;(semi-colon)
QMark ?(questionmark)
Exclam !(exclamationmark)
Dash
-
Quote "(quotationmark)
Apostro `(apostrophe)
Parenth ()or[]or{}(LIWC2007willcounteachpairofparentheses)
OtherP
(otherpunctuationincludesallASCIIcharactersfrom33-47,58-64,91- 96,123-126notincludedinthelistabovei.e.allnon-alphanumericand non-controlcharacters)
AllPct Allpunctuation

Creating and Using Custom Dictionaries
Aswellascontainingadictionaryandcategorysetintegraltotheapplication,LIWC2007 permitsuseofdictionariesandcategoriescreatedbytheuser.Thisisdonebyselectingthe "Loadnewdictionary..."optionfromthe"Dictionary"menu.Externaldictionariesmustbe plaintextfilessetoutinthefollowingformat.Forexample,onecouldcreateaspecificself- referencingdictionary:

%
1
I
2
me
3
my
4
we
5
us
6
our
LIWC2007 Operators Manual
Page 5
7
singular
8
plural
9
possessive
%
me
27
mine
379
my
379
myself
27
our
689
us
58
we'*
48

Ifyourexternaldictionaryincludescategorydefinitions,theymustprecedethedictionary
andbeenclosedbetween%delimitersasshownabove.Ifyourdictionarydoesn'tinclude
categorydefinitions,LIWC2007willusethedefaultinternalcategories.Theremustonlybe
onecategorydefinitionsperlinebeginningwiththecategorynumberandfollowedbythe
categorynameseparatedfromthenumberbyspace(s)and/ortab(s).LIWC2007will
acceptupto999categories.
Eachnumberreferstothecategorytowhicheachwordisassigned.Hence,theword"me" isassociatedwithcategory2(thesoloworddictionaryof"me")andcategory7(1stperson singular),theword"our"isassociatedwithcategories7(our),8(1stpersonplural),and9 (possessive).
Thedictionarylistcomprisesonewordorword-stemperlinefollowedbyalistofcategory numberswithwhichthewordisassociated.Againallelementsinthelineareseparatedby space(s)and/ortab(s).Intheexampleabove,theword"us"isassociatedwithcategories5 and8.Word-stemsarepartialwordsterminatedbyanasterisk.Thus,inthedictionary,use ofanasterisk(*)attheendofthewordsignalsLIWC2007toignoreallsubsequentletters. Consequently,"we'*"willcountthewordswe're,we'll,we'd,etc.incategories4and8.
Helpfultipsfromusers:*
· Thefirstlineofthedictionarymustbethe%symbol,followedbytheCategory numbersandnames.AfterthelastCategoryname,asecond%symbolmustbeinsertedto signalthebeginningoftheWordentriesandnumbers.
· Itismoreefficienttocreateyourowndictionarythantotrytointegrateyour dictionaryintotheLIWC2007defaultdictionary.
· CategorynamesshouldbeONEwordonlywithnopunctuation.Categorynamesdo notneedtobeinalphabeticalorder.
· Wordsshouldbelistedinalphabeticalorder.SinglewordsonlybutNOnumbersor punctuation(exceptionsincludeapostropheandhyphen).

OK:
Love
500



Love-sick 500 503
NOT OK Lov4sale 521
Love sick 500 503
LIWC2007 Operators Manual
Page 6
· Wordentriesshouldonlyappearonce.Linkingwordstomultiplecategoriesisfine.

OK:
Love 500 504

NOTOK: Love 500



Love 504
· Becarefulintheuseofasterisksandavoiddoublecountingwords.Forexample,the followingcaseisproblematic:

NOTOK: Thank*
27



Thanksgiving27 94
In this case, the word "Thanksgiving" would be counted twice in category 27.
*With Special thanks to Nancy Collins at UC-Santa Barbara.
Phrases rather than words
LIWCwasoriginallycreatedtoexaminewordsorwordstems.LIWC2007cannowsearch forphrases.ThisoptioniscurrentlyavailableontheMacversionbutnotyetonthePC version.

Preparing Written Text For LIWC2007 Analysis

TheaccuracyofLIWC2007outputdataisdeterminedbythequalityofthetextfilesthatare analyzed.Inordertoinsurebestresults,itisnecessarytoproperlypreparetextessaysfor LIWC2007analysis.Theessentialstepsforessaytextorganization,entry,andeditingare asfollows:
1. Textfileorganization.

Eachlanguagesampleshouldbeputinitsownfileandnamedinasystematicand
meaningfulway.Forexample,datafromastudywithtwoconditionsandthreedays
ofwritingmightbesavedinfilesusingthisnamingstrategy:

[PARTICIPANT#][DAY#].[CONDITION]--4568day1E.txt,4568day2E.txt,and
4568day3E.txt
2. Textfilecomputerentry.

EssaysshouldbeenteredintothecomputerusingMicrosoftWorddocumentsoras
standardtextorASCIIfiles.Filespreparedwithotherwordprocessingprograms
(e.g.,WordPerfect)willnotwork.Also,filesinformatssuchaspdf,html,jpg,etc
willnotprovidevalidoutput.Agoodruleofthumbisthatfilesendingin.txtor.doc
willprobablyprovideaccurateresults;otherfiletypesprobablywon't.
3. Cleaningthetextfiles.

Eachfiletobeanalyzedshouldbeexaminedandadjustedformisspellingsand
inappropriateworduse(e.g.,"its"ratherthan"it's").Itisalwayswisetorunallfiles
LIWC2007 Operators Manual
Page 7
throughstandardspell-checkprograms.BecauseLIWC2007convertsalltextfilesto lowercasebeforeprocessingthem,grammar,capitalization,andsentencestructure donotneedtobecorrected.
Naming Files Becausethefilenamesarepartoftheoutputfile,certainconventionsshouldbeadoptedin thepreparationofthefilesandfilenames: 1. Separatefilesforseparatetextsamples.LIWC2007analyzesdataonefileata time.Ifparticipantswriteresponsestotwoquestionsorperhapswriteontwo separatedays,eachquestionordayshouldbeaseparatefile.Ifresponsestoboth questions(orbothdays'writing)arewithinthesamefile,LIWC2007willanalyze themasasinglewritingsample. 2. Thefilenameshouldbedescriptive,includingIDnumber,condition,andquestion ordaynumber. 3. FilesmustbeinTEXTorWordDocumentformat.LIWC2007cannotread WordPerfectorotherwordprocessingfiles.Notethatvirtuallyallwordprocessing programsallowyoutoconvertyourfilesintoASCII,TEXT,orWordformat.
Typing Conventions: Writing and Interview Samples Inmakingcorrectionsorcleaningtextfiles,keepinmindwhatyourgoalsareinanalyzing thedata.LIWC2007doesnotdiscriminatebetweenupper-andlower-caseletters.Itcan onlycountwordsthatareinitsdictionaries.Misspellings,colloquialisms,foreignwords, andabbreviationsareusuallynotinthedictionaries.Thefollowingitemsshouldbe checkedbeforeanyfilesareanalyzed:
1. Spelling,abbreviations,contractions.

Correctallspellingerrors.ItisbesttousestandardUnitedStatesspelling(although
thestandarddefaultdictionaryalsocontainsmostBritishEnglishspellingsaswell).

Meaningfulabbreviationsshouldbespelledout."Jan"shouldbeJanuary.More
obscureabbreviationsoracronyms,suchas"AT&T",canremainassuchunlessyou
havereasontowantthetermtobeexpandedandcountedasfourseparatewords:
"AmericanTelephoneandTelegraph".

Commonverbcontractionsareinthedictionaryanddonotneedtobechanged.
Theseinclude:don't,won't,isn't,shouldn't,can't,couldn't,I'm,I'll,I'd,we're,we'd,
you're,he's,it's,etc.Mostotherswillbesimplycountedaspossessivenouns:
"Sally'sshoes"willbecountedthesamewayas"Sally'sgoingtothestore."Inthe
secondcase,change"Sally's"to"Sallyis."
2. Endofsentencemarkersandhyphens.

TheWordspersentence(WPS)categoryisbasedonthenumberoftimesthatend-
of-sentencemarkersaredetected.Theseincludeallperiods(.),questionmarks,and
LIWC2007 Operators Manual
Page 8
exclamationpoints.Onepotentialproblemisthatcommonabbreviations(suchas "Dr.","Ms.","U.S.A.","D.O.A.")willbecountedasmultiplesentencesunlessthe periodsareremoved.Becarefulthattheremovaloftheperiodsdoesn'tmakeanew word.Forexample,theUnitedStates,or"U.S.",becomes"US"(1stpersonplural pronoun)whentheperiodsareremoved.Inthiscase,changeitto"USA".

Timemarkers(e.g.,6a.m.or7:30p.m.)canalsobeaproblem.Because"a.m."
withouttheperiodsisaverb,"am",changetimeto6amor7:30pm.

Whenwordsstartorendwithhyphens,theyarereadbyLIWC2007aspartofthe
word.LIWC2007,forexample,lists"self-esteem"asameaningfulwordinoneofits
dictionaries.Incasesofhyphenatedphrasessuchas"this-or-that"LIWC2007will
searchforasinglewordandwon'tfindit.Tocorrect,change"this-or-that"to"this-
or-that".

Watchoutforhyphensbetweenphrases,asin"wewenttothestore-Idon'tknow
why."LIWC2007willthinkthat"store-I"isoneword.Insertblanksoneithersideof
thehyphenssothatbothwordswillbecounted.

3. Othercommonproblems:
Typedentry Changeto:
w/
with
b/
between
&
and

`cause
because
gotta
gotto
lotta
lotof
and/or
and-or
`anor`n and
mos
months
sec
second
@
at

Transcribing Oral Transcripts: Special Problems
Althoughnotdesignedforspokenlanguage,wehavefoundLIWC2007tobeusefulin analyzingconversationsandinterviews.Toaccommodatecertaindimensionsofspoken language,wehaveadoptedthefollowingconventions:
1. Nonfluencies.

Hm,hmm,uh,uhh,uhm,um,umm,anderarepartofthenonfluencydictionary.
Otherformswillnotbecaught(e.g.,oooohshouldbechangedtoumifusedasa
nonfluency).

Stutteringcanbeaccommodatedbyalteringthestutteringpartofaphrasetoa
nonfluencymarker.Forexample,"The,thebo-,theboatwentintothewater"could
bechangedto"Uh,theboatwentintothewater."Thetranscriberwillhaveto
decidehowmanyuh'swouldbeappropriate.
LIWC2007 Operators Manual
Page 9

Uh-uhanduh-huhshouldbechangedto"no"and"yes".Huh?shouldbechangedto
"what?"Or,ifyouarevery,veryproper,to"Excusememadam,Ididn'tquitecatch
whatyousaid."
2. Fillers.

Everydayspeechislitteredwith"meaningless"fillers.Unfortunately,thesefillers
usesomeofthemostimportantwordsinourdictionaries.Watchoutforthe
following:

Youknow.Asin,"wewent,youknow,tothestoreand,youknow,boughtgum."
Changetooneword:youknow."Wewent,youknow,tothestore..."

Imean.Asin,"wewent,Imean,tothestore..."Changetooneword:Imean.

Idon'tknow.Asin,"wewent,Idon'tknow,tothestore..."Changeto:Idontknow.

Like."Wewent,like,tolikethestoreandlikewelikeboughtlikegum."Becareful
withlikebecausesometimesitisusedappropriately.Asanonfluency,changeitto:
rrlike.Notethatallwordsstartingwith"rr"willbecodedasanonfluency.Hence,if
youaretranscribingaudiotapesmadeinthe1950's,theword"well"wouldlikelybe
usedtheway"like"istoday.Hence,youwouldenteritas"rrwell."
3. Transcribers'comments.

LIWC2007isdesignedonlyforspokenlanguage.Transcribersofteninsertremarks,
suchas[subjectlaughs],[shakyvoice],[whispers].Werecommendremovingthese.

Occasionally,thetranscribercannotunderstandawordorpassage.Ratherthan
writing[can'tunderstandword]or[?],thetranscribershouldputanonsenseword,
suchas"xxxx"initsplace.LIWC2007willcountthexxxxasaspokenwordbutnot
assignittoadictionary.Forentirepassages,don'tinsertanything.


Technical Support
Technicalsupportforset-upandhardware/softwarecompatibilitycanbeobtainedby [email protected]romtheFirst Author,JamesW.Pennebaker,DepartmentofPsychology,TheUniversityofTexas,Austin, Texas78712([email protected])orfromthesecondauthor,RogerBooth, SchoolofMedicalScience,TheUniversityofAuckland,Auckland,NewZealand ([email protected]).Moreextendedconsultationisavailableonafeebasis.

LIWC2007 Operators Manual
Page 10
Getting Some Practice: Running the Samples

IncludedwiththeLIWC2007programisasubdirectorycalledSAMPLES.Itiscomposedof 11textfilesofvaryinglengths.Theseinclude:
InauguraladdressesofLincoln,FranklinRoosevelt,andClintonatthebeginningof thefirsttermofoffice:

LINCOLN.TXT

FDR.TXT

CLINTON.TXT
TwopoemsfromAnneSextonandSylviaPlath:

SEXTON.TXT

PLATH.TXT
Tworatherdryabstractsfromesotericsocialpsychologyjournalsbyesteemed socialpsychologists:

ABSTR1.TXT

ABSTR2.TXT
Twotranscriptsfromthemedia-onefromtheHowardSternShow;theotherfroma morningprograminterviewwithDonnaShelala:

RADIO.TXT

TALKSHOW.TXT
ApassagefromMarkTwain'sHuckleberryFinnwhichispresentedinitsoriginal, uneditedformaswellasinaformtranslatedinto"proper"AmericanEnglish.The purposeofthesetwoformsistogivetheresearcherasenseofhowextensive editingcanchangetheoutput(notasmuchasyoumightthink):

HUCKRAW.TXT

HUCKCLN.TXT
ThisgroupoffilesisintendedtogivetheLIWC2007userasenseofthediversityoftext samplesthatcanbeanalyzedandthesimilaritiesanddifferencesamongthem.To appreciatethatnatureofthesamples,simplyopenanyoftheminWordPad,Word, WordPerfect,oreventheLIWC2007"openfile"menu.

Hereisastep-by-stepprocedureforLIWCingthe11files: 1. StarttheLIWC2007applicationbyclickingontheLIWC2007iconorLIWC2007.EXE.
2. WithintheLIWC2007application,gointothe"File"menuandpress"Processtext...".
3. Navigateyourwaytothedirectorycontainingthesamplefiles.
LIWC2007 Operators Manual
Page 11
4. Clickthe"SelectAll"button.
5. LIWC2007willdisplayastandarddialogboxwithadefaultname(LIWCresults.dat) andlocationforthefiletocontaintheresults.Youcanchangetheseifyouwish.
6. Pressthe"Save"button.Voila!
Oncompletion,LIWC2007willsavetheresultsinthespecifiedfileandalsoopenitand displaythedatainawindowforyoutosee.Beautiful,isn'tit?Youcanscrollthefiletothe rightandseethatall74variablesarethereasarethefilenames.
Toseethedatamorecompletely,however,useeitherExcelorSPSStoopenLIWC results.datfile.IfyouuseSPSS,openthefileasatab-delimitedfileandbesuretocheckthe box"Readvariablenames."Thefirstpartoftheoutputfileshouldlooksomethinglikethis inExcel:

Filename Lincoln.txt FDR.txt Clinton.txt Huckcln.txt Huckraw.txt Plath.txt Sexton.txt radio.txt talkshow.txt Abstr1.txt Abstr2.txt
WC 3639 1881 1584 603 654 100 237 272 621 107 196
WPS 28.43 22.13 17.22 21.54 21.80 33.33 14.81 5.44 24.84 17.83 24.50
Sixltr 23.36 23.34 20.71 8.79 8.10 26.00 12.24 7.72 18.20 45.79 36.73
Dic 82.77 85.06 86.36 88.06 76.76 74.00 89.45 93.75 93.24 76.64 66.84
funct 61.14 56.14 57.20 64.84 55.50 39.00 60.34 58.46 57.33 34.58 39.29
pronoun 10.52 12.39 14.77 18.24 15.90 5.00 23.21 21.32 15.62 0.93 0.51
i 1.59 1.65 0.88 2.49 2.75 0 13.50 6.99 0.64 0 0
we 0.63 3.35 7.83 1.16 1.07 0 0 1.10 2.42 0 0
posemo 3.24 4.36 4.67 2.32 2.29 4.00 7.59 6.25 2.74 7.48 1.53
negemo 1.70 2.45 1.64 1.00 0.92 5.00 4.22 0.74 1.13 1.87 0
OK,yourfiledoesn'tlookexactlylikethis.Andyes,thereareanother65variablesinyour outputfile.However,eventhissmallsampleofverbalmaterialyieldssomeintriguing findings.

Important:Allvariables(exceptrawwordcount[WC]andwordspersentence[WPS]) reflectpercentageoftotalwords.So,forexample,1.6%ofLincoln'sinauguraladdresswas comprisedof1stpersonsingular"I"words(I,me,my)comparedwith7%ofthespeech samplefromHowardStern.Clinton,morethananypresident,usedatremendouslyhigh rateof1stpersonpluralwords(e.g.,weus)inhisspeech(7.8%).Naturalspokentext generallyhasalowerpercentageoflongwords(i.e.,wordsgreaterthansixletters[sixltrs]) thanformaltext.Otherstrikingdifferences(e.g.,useofemotionwords)canbeseeninthe actualLIWC2007analysis.
Bylookingatthistable,itiseasytoseehowlanguageusecandifferfrompersontoperson andfromcontexttocontext.Obviously,whenattemptingtogetareliablepictureof languageusewithinagivenpersonorsituation,themoreandlengthierthetextsamples, thebetter.

File: linguistic-inquiry-and-word-count-liwc-2001.pdf
Title: Microsoft Word - LIWC2007_OperatorManual.doc
Author: pennebaker
Published: Wed Feb 10 22:34:10 2010
Pages: 13
File size: 0.22 Mb


, pages, 0 Mb

William L. Ditto, 33 pages, 0.48 Mb

, pages, 0 Mb

All my relations, 7 pages, 1.43 Mb

, pages, 0 Mb

The great plan of happiness, 5 pages, 0.04 Mb
Copyright © 2018 doc.uments.com