NEWS MEDIA, MENTAL ILLNESS AND HOMELESSNESS IN …summit.sfu.ca/system/files/iritems1/11437/etd6235_AMoshrefzadeh.pdf · defense, Gratien Prefontaine, for his kindness and participation.

NEWSMEDIA,MENTALILLNESSANDHOMELESSNESSINCANADA:HASDEPICTIONOFMENTALILLNESSANDHOMELESSNESSCHANGEDINCANADIANNATIONALNEWSPAPERSSINCETHERELEASEOF‘OUTOFTHE

SHADOWSATLAST’?by

ArezuMoshrefzadehBachelorofArts,UniversityofBritishColumbia,2006

THESISSUBMITTEDINPARTIALFULFILLMENTOFTHEREQUIREMENTSFORTHEDEGREEOF

MASTEROFPUBLICHEALTH

IntheFacultyofHealthSciences

©ArezuMoshrefzadeh2010SIMONFRASERUNIVERSITY

Fall2010

Allrightsreserved.However,inaccordancewiththeCopyrightActofCanada,thisworkmaybereproduced,withoutauthorization,undertheconditionsforFairDealing.Therefore,limitedreproductionofthisworkforthepurposesofprivatestudy,research,criticism,reviewandnewsreportingislikelytobeinaccordance

withthelaw,particularlyifcitedappropriately.

ii

APPROVAL

Name: ArezuMoshrefzadeh

Degree: MasterofPublicHealth

TitleofThesis: NewsMedia,MentalIllnessandHomelessnessinCanada:HastheDepictionofMentalIllnessandHomelessnessChangedinCanadianNationalNewspaperssincethereleaseof‘OutoftheShadowsatLast’?

ExaminingCommittee:

Dr.GratienPrefontaineChairofDefenseAssistantProfessor,FacultyofHealthSciences

Dr.JulianSomersSeniorSupervisorAssociateProfessor,FacultyofHealthSciences

Dr.MichaelHayesSupervisorAdjunctProfessor,FacultyofHealthSciences

Dr.JimFrankishExternalExaminerAdjunctProfessor,FacultyofHealthSciences

DateDefended/Approved: September15,2010

Last revision: Spring 09

Declaration of Partial Copyright Licence The author, whose copyright is declared on the title page of this work, has granted to Simon Fraser University the right to lend this thesis, project or extended essay to users of the Simon Fraser University Library, and to make partial or single copies only for such users or in response to a request from the library of any other university, or other educational institution, on its own behalf or for one of its users.

The author has further granted permission to Simon Fraser University to keep or make a digital copy for use in its circulating collection (currently available to the public at the “Institutional Repository” link of the SFU Library website <www.lib.sfu.ca> at: <http://ir.lib.sfu.ca/handle/1892/112>) and, without changing the content, to translate the thesis/project or extended essays, if technically possible, to any medium or format for the purpose of preservation of the digital work.

The author has further agreed that permission for multiple copying of this work for scholarly purposes may be granted by either the author or the Dean of Graduate Studies.

It is understood that copying or publication of this work for financial gain shall not be allowed without the author’s written permission.

Permission for public performance, or limited permission for private scholarly use, of any multimedia materials forming part of this work, may have been granted by the author. This information may be found on the separately catalogued multimedia material and in the signed Partial Copyright Licence.

While licensing SFU to permit the above uses, the author retains copyright in the thesis, project or extended essays, including the right to change the work for subsequent purposes, including editing and publishing the work in whole or in part, and licensing other parties, as the author may desire.

The original Partial Copyright Licence attesting to these terms, and signed by this author, may be found in the original bound copy of this work, retained in the Simon Fraser University Archive.

Simon Fraser University Library Burnaby, BC, Canada

iii

ABSTRACT

OnMay9,2006,aSenateCommitteereportentitled"OutoftheShadowsat

Last"waspublished,highlightingthecrisisinthementalhealthsysteminCanada.It

stressedthecriticalneedtodevelopthementalhealthsystemandtochangepublic

attitudestowardsmentalillness.Usingagendasettingandframingtheories,the

currentstudyexploreswhetherthedepictionofmentalillnessandhomelessness

changedinCanadianNationalnewspapercoveragesincethereleaseofthisreport.

Relevantarticlesfroma2003‐2009werecodedusingacategoricalcodesheet.The

resultsshowasignificantandlastingincreaseintheagendasettingpotentialof

CanadianNationalnewspapercoverageregardingmentalillnessandhomelessness

sincethereleaseofthereport.Theevidencesuggeststhatthereportappearsto

haveplayedacatalyticroleinincreasingtheoverallfrequencyofreportingona

numberofprominentthemesconcerningmentalillnessandhomelessness.

Keywords:newsmedia;agendasetting;framing;mentalillness;homelessness;SenateReport

iv

DEDICATION

tomyparents,

whoneverletmeforgettheimportanceofeducation

andalwaysencouragedmeinitspursuit

v

ACKNOWLEDGEMENTS

Thankyoutomythesissupervisors,JulianSomersandMichaelHayes,for

theirinsightandguidancethroughouttheresearchprocess.Thankyoutomy

externalexaminer,JimFrankish,forhiscontributions;andtothechairofmy

defense,GratienPrefontaine,forhiskindnessandparticipation.ThankyoutoJen

VanRasselforallherhelp,thecountlessquestionsshehashadtoendureandthe

positivitywithwhichshedoesso.AndthankyoutoMichelJoffres,withoutwhom

thiswouldnothavebeenpossible.

Aspecialthankyoutomyfriends,wholistened,understoodandpushedme

toovercomethechallengesandfinallycompletethisthesis.Thankyoutomyfamily

fortheirconstantsupportandencouragement.Tomyhusband,whosewellof

patienceprovedtobetrulybottomless,thankyouforsharinginallthemini

victoriesalongtheway.Thisprocesswouldhavebeeninfinitelymoredifficult

withoutyou.

vi

TABLEOFCONTENTS

Approval............................................................................................................................. iiAbstract ............................................................................................................................. iiiDedication ......................................................................................................................... ivAcknowledgements.............................................................................................................vTableofContents ...............................................................................................................viListofFigures .................................................................................................................. viiiListofTables...................................................................................................................... ix

1:Introduction ................................................................................................................. 11.1 ResearchContext ....................................................................................................... 11.2 StudyPurpose ........................................................................................................... 2

2:Background .................................................................................................................. 42.1 UnderstandingthePopulation ................................................................................... 4

2.1.1 PrevalenceofMentalIllnessAmongtheHomeless .......................................... 42.1.2 ChallengesinEstablishingPrevalence............................................................. 52.1.3 InteractionBetweenMentalIllnessandHomelessness.................................... 72.1.4 HealthChallenges ........................................................................................... 92.1.5 TheRoleofHousing ..................................................................................... 11

2.2 ‘OutoftheShadowsAtLast’ ..................................................................................... 122.2.1 MentalIllnessandHomelessnessintheSenateReport.................................. 132.2.2 PrimaryPrinciplesandGoals ........................................................................ 152.2.3 TheMentalHealthCommissionofCanada .................................................... 17

2.3 NewsMedia ............................................................................................................. 172.3.1 NewsMediaCoverageofMentalIllness ........................................................ 172.3.2 NewsMediaCoverageofHomelessness ........................................................ 192.3.3 MentalIllness,HomelessnessandNewsMedia ............................................. 202.3.4 NewsMediaandAttitudesofthePublic ........................................................ 20

2.4 AgendaSettingandFraming .................................................................................... 22

3:Methods...................................................................................................................... 253.1 NewspaperSelectionandDateRange....................................................................... 253.2 EstablishingSearchTerms ....................................................................................... 253.3 ArticleSelection....................................................................................................... 263.4 CodingProtocol ....................................................................................................... 27

3.4.1 TheoreticalOrientation ................................................................................ 283.4.2 ArticleCharacteristics .................................................................................. 283.4.3 ArticleThemes ............................................................................................. 293.4.4 ArticleFraming ............................................................................................ 30

vii

3.5 CodingMethods ....................................................................................................... 313.6 StatisticalAnalysis ................................................................................................... 32

4:Results ........................................................................................................................ 334.1 Inter‐coderReliability. ............................................................................................. 334.2 GeneralOverview .................................................................................................... 334.3 ArticleCharacteristics.............................................................................................. 34

4.3.1 QuantityofCoverage .................................................................................... 344.3.2 ProminenceofCoverage ............................................................................... 36

4.4 ArticleThemes(AgendaSetting) .............................................................................. 394.5 Framing................................................................................................................... 45

5:Discussion .................................................................................................................. 595.1 KeyFindings ............................................................................................................ 59

5.1.1 AgendaSettingEffects .................................................................................. 595.1.2 PotentialFramingEffects.............................................................................. 62

5.2 Limitations .............................................................................................................. 675.3 FutureDirections..................................................................................................... 70

6:Conclusions ................................................................................................................ 74

Appendices..................................................................................................................... 77AppendixA ...................................................................................................................... 78AppendixB ...................................................................................................................... 80AppendixC ...................................................................................................................... 87AppendixD ...................................................................................................................... 90AppendixE ...................................................................................................................... 91

References...................................................................................................................... 92

viii

LISTOFFIGURES

Figure1: QuantityofCoverageBeforeandAfterthereleaseofTheSenateReport........... 35Figure2: QuantityofCoveragebyyear ........................................................................... 36Figure3: AverageWordCountbyyear ........................................................................... 37Figure4: ArticlePlacementPre‐SenateReport(No.ofarticles) ....................................... 38Figure5: ArticlePlacementPost‐SenateReport(No.ofarticles) ..................................... 38Figure6: Frequencyofarticlesmentioning‘Recovery’ .................................................... 40Figure7: Frequencyofarticlesmentioning‘PublicSafety’............................................... 40Figure8: Frequencyofarticlesmentioning‘Inequity’ ..................................................... 41Figure9: Frequencyofarticlesmentioning‘AffordableHousing’..................................... 42Figure10:Frequencyofarticlesmentioning‘IntegrationofServices’................................ 43Figure11:FrequencyofarticlesmentioningStigma ......................................................... 44Figure12:FrequencyofArticlesmentioningthemes ........................................................ 45Figure13:EpisodicandThematicArticles(No.ofArticles) ............................................... 45Figure14:ArticleApproachPre‐SenateReport(No.ofArticles) ....................................... 46Figure15:ArticleApproachPost‐SenateReport(No.ofArticles) ...................................... 46Figure16:QuantityofEpisodicandThematicArticlesbyYear.......................................... 47Figure17:ArticlesPrimaryEvent/Story(No/Otherremoved).......................................... 49Figure18:PrimaryEvent/Story:OtherandNo/Unclear ................................................... 51Figure19:PrimaryEvent/Story:PersonalStory ............................................................... 52Figure20:PrimaryEvent/Story:ViolenceAgainstaHomelessand/orMentallyIll

Individual........................................................................................................ 53Figure21:PrimaryEvent/Story:ViolenceByaHomelessand/orMentallyIll

Individual........................................................................................................ 53Figure22:AttributionsofResponsibility .......................................................................... 55Figure23:AttributionsofResponsibility(EpisodicandThematic).................................... 55Figure24:BenefitsoftheSolution.................................................................................... 56Figure25:BenefitsoftheSolution(EpisodicandThematic) ............................................. 57

ix

LISTOFTABLES

Table1: Summaryofarticleexclusion(notmutuallyexclusive) ..................................... 27Table2: Variablesusedtoassesspotentialmediaeffects ............................................... 28Table3: ArticlesPrimaryEvent/Story ........................................................................... 48Table4: ArticlesPrimaryEvent/StoryoverTime .......................................................... 50Table5: SummaryofVoiceinallarticles ....................................................................... 58

1

1: INTRODUCTION

1.1 ResearchContext

Massmediaservesasanimportantsourceofhealthinformationforthe

generalpublic(Martinson&Hindman,2005)andmorespecifically,isthemost

commonsourceofinformationregardingmentalillness(Wahl,Wood&Richards,

2002).Assuch,researchershaveraisedconcernsoverthecontentandaccuracyof

informationpresentedinthemedia,andnewsmediainparticular.Newscoverage

ofmentalillnesshasbeenfoundtoreinforcenegativeandstigmatizingviews(Wahl

etal.,2002).Thesenegativedepictionscanhavesignificanteffectsonthepublic’s

perceptionandtreatmentofthosewithmentalillness(Francis,Pirkis,Dunt&Blood,

2001;Wahl,1992).This,inturn,holdswiderimplicationsforpolicydecisions

regardingindividualslivingwithmentalillnessascommunitiespressfordecisions

tobecenteredonpublicsafetyconcerns(Paterson,2006;Takahashi&Dear,1997).

TheStandingSenateCommitteeonSocialAffairs,ScienceandTechnology

chairedbytheHonourableMichaelJ.L.Kirbyconductedatwo‐yearstudyacross

Canadaintothestateofmentalhealth,mentalillnessandaddictionsservicesin

Canada.TheyreleasedtheirfinalreportonMay9,2006,whichhighlightedthe

crisisinthementalhealthsystem,andtheneedtoaddressit(TheSenate,2006).

Thereportwasaptlynamed‘OutoftheShadowsatLast’,asitwasthefirstnational

documenttobereleasedconcerningmentalhealthandmentalillnessinCanada.

2

Amongothergoals,thereportaimedtoinitiateandadvancechangesindeep‐rooted

publicattitudestowardspeoplelivingwithmentalillness.Itputahumanfaceon

mentalillnessthroughrecountingstoriesanditexpressedahopethatthereport

wouldaffectitsreaders.Thereportstressedpriorityareasintransformingthe

landscapeofmentalillness,includingtheneedtoaddressthecrisisofhomelessness,

stating,“Itwouldbehardtooverestimatetheimportanceofadequatehousingfor

peoplelivingwithmentalillness”(TheSenate,2006;p118).Whilethisreportheld

obviousimplicationsforpolicydevelopmentintheareaofmentalhealth,little

researchhasbeendoneonitseffectontheperceptionofthegeneralpublicandthe

channelsthatinformthem,includingthemedia.

1.2 StudyPurpose

Thisthesiswillexplorethevariationinmediacoverageofmentalillnessand

homelessnessinCanadainrelationtotheSenatereport‘OutoftheShadowsatLast’.

Theprimaryresearchquestionis‘Hasthedepictionofmentalillnessand

homelessnesschangedinCanadianNationalnewspapercoveragesincetherelease

of‘OutoftheShadowsatLast’?Thisthesisaimstodeterminewhetherthegoalof

theSenateReporttocreateachangeinpublicattitudestowardsmentalillnessand

toincreaseawarenessoftheprimaryprinciplesoutlinedwithinithavebeen

advancedinrelationtohomelessnessinnationalnewsmedia.Thenullhypothesis

isthatthereisnochangeinthepotentialfortheseeffectsinthenationalnews

coverageofmentalillnessandhomelessness.Inaddressingtheresearchquestion

andtestingthishypothesis,thisthesiswillspecificallyexamine1)thequantityof

3

articlespresentedaboutmentalillnessandhomelessnessand2)thevariationin

howthesearticleshavebeenpresented.Thiswillapproachthearticlesconsidering

bothagendasettingandframingtheories,andaimtoexplorethefollowingcentral

questions:

1) Howfrequentiscoverageregardingthehomelessmentallyillovertime?

2) Hastheprominenceofreportingonthisissuechangedovertime?

3) Arethemesidentifiedasprioritiesin‘OutoftheShadowsatLast’

representedinthenews,anddoestheircoveragechangeovertime?

4) Whatapproachdothearticlestake(i.e.episodicorthematic)?

5) Aresolutionsclearlyidentifiedinthearticles,andhowaretheyframed?

6) Whosevoicesarequoted?

Researchfindingsmaysuggestwhetherornottheimportanceofaddressing

thehomelesslivingwithmentalillness,asstressedin‘OutoftheShadowsatLast’,is

reflectedinnewspapercoveragesurroundingthispopulation.Findingsmayalso

indicatewhethertherewerechangesinthewayinwhichhomelessnessandmental

illnesswereframedinnationalnewspapercoverageandtheimplicationsthese

changesmayhold.

4

2: BACKGROUND

2.1 UnderstandingthePopulation

2.1.1 PrevalenceofMentalIllnessAmongtheHomeless

Thetransientnatureofthispopulationmakesdatacollectionachallenge,

andestimatesoftheprevalenceofmetalillnessamongthehomelessvaryby

locationandstudy.InCanada,theexactnumberofindividualsconsidered

homelessisunknown(Echenberg&Jensen,2008)andresearchtoassessthe

prevalenceofmentalillnessandaddictionamongthehomelessisnotconductedon

anongoingbasis(TheSenate,2006).Canada’sNationalSecretariaton

HomelessnesshasestimatedthenumberofhomelessinCanadatobearound

150,000people(Echenberg&Jensen,2008)butotherreportsvaryfrom10,000

homelessindividualsonanygivennight(CIHI,2007)to300,000(Laird,2007).It’s

estimatedthatapproximately60%ofindividualswhoarehomelesshavealifetime

diagnosisofmentalillness,whichistwotothreetimestheratefoundinthegeneral

population(Riordan,2004).FurtherresearchinCanadaestimatesthatbetween

30%and40%ofhomelessindividualsarelivingwithamentalillness(TheSenate,

2006).Otherestimatessuggestthatthisiscloserto80%(Howlett,2008).

Estimatesfromotherpartsoftheworldpresentasimilarpicture,butaswith

muchoftheCanadiandata,varyfromstudytostudy.Scott(1993)reportsthatin

GreatBritainbetween30%and50%ofhomelessindividualshaveamentalillness.

5

StudiesintheUnitedStatesofAmericaestimateprevalenceratesformentalillness

amongthehomelesspopulationtobebetween10%and45%dependingonthe

methodologyused(Susseretal.,1993).Examplesofthisshowarangefromaone‐

quarter(Smithetal.,1992),toone‐third(Bachrach,1992;Minkoff&Drake,1992),

toone‐half(Adams,Pantelis,Duke&Barnes,1996).Whilethereisalackofprecise

dataregardingproportionofindividualswithmentalillnessamongthehomeless,

evidencesupportsthatthisproportionisasignificantone(Levine&Rog,1990).

2.1.2 ChallengesinEstablishingPrevalence

Underlyingtheuncertaintysurroundingthequestionoftheprevalenceof

mentalillnessamongthehomelessareanumberofmethodologicalchallenges.One

isthelackofastandarddefinitionofhomelessness(Bachrach,1992;Patterson,

Somers,McIntosh,Sheill&Frankish,2008;Echenberg&Jensen,2008).Studies

rangeinscopefromincludingonlythehomelesslivingonthestreets,todefining

homelessnessasanyformofresidentialdislocation(Susseretal.,1993).InCanada,

notonlydodefinitionsofhomelessnessvaryacrossprovinces,butbetween

jurisdictionsaswell(OAGBC,2009).

Accuratediagnosisofmentalillnessinhomelesspopulationsalsoposesa

challenge(Bachrach,1992;Mojtabai,2005;Susser,Conover&Struening,1989).

Individualsoftendonotwanttobeapproachedorquestioned(Bachrach,1992).As

aresult,somestudiesresorttoself‐reportsorrelyonstaff,oftenunqualified,to

makeassessmentsbasedonsurveysconductedinshelters(Susseretal.,1989).Itis

furtherassertedthatitisnearimpossibletoestablishmentalillnessinanindividual

6

sufferingphysicaldeprivationasthedeprivationconfoundsanyresultingdiagnosis

(Bachrach,1992).

Homelessindividualsandgroupsareofteninvisibletoresearchersfora

numberofreasons(Mosher‐Ashley,Henrikson&French,2000;Ropers,1988).

Beingaverymobilegroup(Bachrach,1992),thereisasignificantchallengeto

locatingthispopulation.Researchisoftenconductedinshelters,butmanyhomeless

peopledonotusethesheltersystemoritsassociatedservices(Ropers,1988).

Sheltersthemselvesoftenhaveadmissioncriteriathatexcludeparticulargroups,

suchasmenorthosewithaddictions(Jacobs,Little&Almeida,1993).Further

complicatingthis,thementallyillhomelesspopulationoverlapswithother

subgroupsofmentallyillpeople,makingitdifficulttodistinguishthemasaseparate

group(Bachrach,1992).Thisisparticularlytrueofinvolvementwiththecriminal

justicesystem,asbothhomelessnessandmentalillnessarestrongpredictorsof

interactionwiththissystem(Riordan,2004).

Goingbeyondthechallengesoflocatingthispopulationarethe

methodologicalissuessurroundinghomelesscounts.Mostinformationregarding

theextentofhomelessnessinCanadacomesfromhomelesscountsconductedby

municipalitiesand/orvolunteerorganizations(OAGBC,2009).Asnotall

municipalitiesparticipatemanygounrepresented,especiallyruralcommunities

(OAGBC,2009).Thedatathatiscollectedisgenerallynotacquiredwiththesame

methodologyacrossthecountrymakingitdifficulttoformcomparisons(OAGBC,

2009).Additionally,thereislittleconsistencyinthesamplingmethodsandthe

timingofthecounts(CIHI,2007).

7

2.1.3 InteractionBetweenMentalIllnessandHomelessness

Thedebateoverwhetherhomelessnessinfluencesthedevelopmentof

mentalillness,orifmentalillnessisapathwaytohomelessnessisacomplexone

(Riordan,2004).Thereisresearchtosupportthathomelessnessisaplausible

causeofmentalillness(Susseratal.,1993).Itissuggestedthatthepotential

emotionaltraumaofbecominghomeless,thephysicaldeprivationandhardshipof

livingonthestreets,andentryintoasubculturewherevictimizationandaddictions

arecommon,havethepotentialtotriggermentalillnessorworsenpre‐existing

disorders(Susseretal.,1993;Pattersonetal.,2008).Conversely,numerousstudies

supportthatmentalillnessisapotentialcauseofhomelessness(Folsom&Jests,

2002;OAGBC,2009,Susseretal.,1993).Twowavesofresearchonpsychiatricand

substanceusedisordersamongthehomelessinthe1980’sidentifiedschizophrenia,

bipolardisorder,depression,alcoholanddrugabuseasriskfactorsfor

homelessness(Susseretal.,1993).Symptomsofmentalillness,suchasanxiety,

paranoiaanddepression,mayleadtohomelessnessbyimpedingindividuals’ability

tomaintainemployment,paydebtsandsustainsocialrelationships(CMHS,2003).

Areviewofresearchmethodologyacknowledgesthedifficultyindeterminingan

exacttimefortheonsetofmentalillnessandestablishingitsrelationtotheonsetof

homelessness(Susseretal.,1993).Whilenodefinitiveconclusionscanbemade

aboutthedirectionalityinwhichmentalillnessandhomelessnessrelatetoone

another,itisapparentthattherelationshipisacomplexonethatlikelyhas

reciprocaleffects(Pattersonetal.,2008).

8

Theadditionaleffectsofmentalillnessandhomelessnessoneachother

increasethevulnerabilityofthispopulationcomparedtohomelesspeoplewithout

mentalillness.Individualswithmentalillnessremainhomelessforlongerperiods

oftime(CIHI,2007;Sullivan,Burnam,Koegal&Hollenburg,2000)andshow

decreasedabilitytocopewithadversecircumstances(Pattersonetal.,2008).There

aresignificantlyhigherratesofmentalillnessamongthechronicallyhomeless

comparedtotherestoftheshelteredhomelesspopulation(Metraux,Marcus&

Culhane,2003).Thehomelesslivingwithmentalillnessaremorelikelytohave

repeatedepisodesofhomelessnesscomparedtootherhomelessindividuals

(Sullivanetal.,2000),andhavegreaterdifficultyexitinghomelessnesscompletely

(CMHS,2003).

Thisincreaseinvulnerabilitypairedwithareductioninsocialassistanceand

socialhousingprograms,hasmadeitmoredifficultforhomelesspeopletoacquire

housingandaccessmentalhealthservices(Riordan,2004).Thisisaparticular

challengeforthosewithseverementalillnessandaddictions(Pattersonetal.,

2008).Thosewithconcurrentdisordersarealsodisadvantagedduetothelimited

availabilityofhousingwithintegratedmentalhealthandsubstanceusetreatment

(CMHS,2003).Thecomplexityoftheapplicationprocessforsocialassistanceanda

reducedcapacitytonavigateitmeanthatmanyhomelesspeoplewithmentalillness

donotreceivebenefitstheymaybeeligiblefor(CMHS,2003).Inaddition,service

providershavegreaterdifficultyengagingmentallyillpeoplewholiveonthestreets

(Tsemberis&Eisenberg,2000).Withfewersocialsupports,decreasedaccessto

benefitsandlowinvolvementwithsupportiveservices,itisunsurprisingthat

9

homelessindividualslivingwithmentalillnessareatanincreasedriskofbeing

arrested,incarceratedorotherwiseinvolvedwiththecriminaljusticesystem

(CMHS,2003;Riordan,2004).

2.1.4 HealthChallenges

Homelessnessandhousinginstabilityhavelongbeenacknowledgedas

publichealthconcerns(Krieger&Higgins,2002;Wolitski,Kidder&Fenton,2007).

Thisislargelyduetotheincreasedriskofmajorhealthproblemsassociatedwith

unstablehousingandhomelessness(Krieger&Higgins,2002).Morbidityand

mortalityratesamongthehomelessarehighlyelevatedcomparedtothegeneral

population,withthehomelesspopulationbeingatmuchgreaterriskofpremature

death(Hwang,2001;Golden,Currie,Greaves&Latimer,1999).Dependingonthe

typeofstudyandgroupcharacteristics(suchasgenderandtypeofhomelessness),

mortalityrateswerefoundtobethreeto31timeshigherthanaverage(Hwang,

2001).Thisisrelatedtothehighrateofunintentionalinjuriesamongstthis

population,primarilyaresultoffallsandpedestrianinjuriesfromcaraccidents

(Hwang,2001).However,deathsresultingfromdrugand/oralcoholoverdoseare

alsocommon(Hwang,2001).

Substanceabuseratesareparticularlyhighamongstthehomelesspopulation

andcoexistenceofmentalillnessandaddictionsseemtobethenorm(Homan,Flick,

Heaton&Mayer,1993;North,Smith,Pollio&Spitznagel,1996;Pattersonetal.,

2008;Riordan,2004;Smith,North&Spitznagel,1992).Prevalenceratesofsevere

mentalillnessarealsoreportedlyhigheramongthehomeless(CIHI,2007)with

10

schizophreniaandbipolardisorderbeingassociatedwithahighriskfor

homelessness(Susser,Moore&Link,1993).Studiesamongshelterusersindicate

higherratesofdepression,anti‐socialpersonalitydisorderandsubstanceabuse

comparedtothegeneralpopulation(CIHI,2007).

Highratesofinfectiousdiseaseshavealsobeenacauseforconcern(Hwang,

2001;Goldenetal.,1999;Wolitzkietal.,2007).Additionally,researchindicatesthat

homelesspopulationshavehigherratesofchronicpulmonarydiseases,respiratory

tractinfections,musculoskeletalconditions,skinandfootproblems,seizuresand

unintentionalinjury(Hwang,2001;CIHI,2007;Krieger&Higgins,2002).While

thesehealthconcernsarenotspecifictohomelessindividualslivingwithmental

illness,thesocialdeterminantsthatleadtoincreasedexposureandrisk(Krieger&

Higgins,2002)encompasstheentirehomelesspopulation.

Healthchallengesfacedbythehomelesspopulationareofevengreater

concerninrelationtothosewithmentalillnessastheinteractionofmentalillness

andhomelessnessleadstopoorerhealthpracticesandoutcomes(Mosher‐Ashleyet

al.,2000;Pattersonetal.,2008;Sullivanetal.,2000).Whileresearchinthisareais

sparse,itindicatesthathomelessindividualslivingwithmentalillnesshavehigher

prevalenceofbothacuteandchronicconditionscomparedtothegeneralhomeless

population(Sullivanetal.,2000).Thisisfurtherexacerbatedbyagreaterdegreeof

stigmatization(Mosher‐Ashleyetal.,2000)andanincreaseddifficultyinaccessing

healthcare(Sullivanetal.,2000).Theyareatanincreasedriskofvictimization,

exposuretocommunicablediseases,suicide,andloweroverallsubjectivequalityof

life(Drake,Bebout&Becker,1999;Sullivanetal.,2000).Theyhavelowerlevelsof

11

subsistenceneedsmet(Drakeetal.,1999)andassucharemorelikelytoengagein

negativehealthbehaviours.Forexample,homelessindividualslivingwithmental

illnesshavebeenfoundtobemorelikelytogettheirfoodfromadumpster(Gelberg

&Linn,1988).Ithasalsobeenreportedthatwhensubsistenceneedsarenotmet,

thosewithmentalillnessrequiredevotingtheirfullattentiontomeetingthematthe

expenseofaddressingotherneedssuchasmentalhealthservices(Sullivanetal.,

2000).

2.1.5 TheRoleofHousing

Adequatehousingplaysacriticalroleintherecoveryofmentalillnessandin

preventinghomelessness.Researchindicatesthatthoseinsupportivehousing

showsignificantreductionsinshelteruse,hospitalizations,lengthofhospitalstays

andtimeincarcerated(Culhane,Metraux&Hadley,2002;Metrauxetal.,2003).

Supportivehousingforthoselivingwithmentalillnessamongthehomelessshow

highretentionrates(Culhaneetal.,2002)andisfoundtoincreasethenumberof

dayshousedsubstantiallycomparedtointensivecasemanagementalone

(Rosenheck,Kasprow,Frisman&Liu‐Mares,2003).Emphasizingtheprovisionand

maintenanceofsupportivehousinghasbeencitedasthemostsuccessfulstrategyin

improvingthequalityoflifeandgeneralwellbeingofthispopulation(Sullivanetal.,

2000;Davis&O’Neill,2005).

Despitetheimportanceofsupportivehousing,servicedeliveryplansoften

facestrongcommunityoppositionthatleadstochangesinzoningby‐lawsthat

discriminateagainstsupportivehousing(Dear&Gleeson,1991;Takahashi&Dear,

12

1997).ThiscommunityoppositioniscommonlyreferredtoastheNIMBY(Notin

MyBackYard)syndromeandiscitedasoneofsociety’sdefiningissues(Takahashi

&Dear,1997).ResearchoftheNIMBYprocessidentifiedspecificgroupsthat

communitiesconsidermostobjectionable,andthementallydisabledandhomeless

wereoftentheleastacceptablegroup(Takahashi&Dear,1997).Studiesidentify

thatgroupsperceivedtobetheleastproductiveandmostpotentiallydangerousare

leastlikelytobeconsideredacceptable(Takahashi&Dear,1997),suggestingthat

theportrayalofindividualswhoarehomelessandlivingwithmentalillnesscan

stronglyinfluencecommunityacceptance,andconsequently,communityopposition

tosupportivehousing.

2.2 ‘OutoftheShadowsAtLast’

InFebruary2003,TheStandingSenateCommitteeonSocialAffairs,Science

andTechnology(hereafterreferredtoastheCommittee)beganastudyofmental

health,mentalillnessandaddictioninCanada,undertheleadershipofSenator

MichaelKirby(MHCC,n.d.).Researchwasconductedoveratwoandahalfyear

periodinwhichtheCommitteereceivedstoriesthroughtheirwebsiteandtravelled

toeveryprovinceandterritoryinCanadatoconductpublichearings(MHCC,n.d.,

TheSenate,2006).TheresultsofthisstudywereformallypublishedonMay9,

2006inthefinalreportoftheCommitteeentitled“OutoftheShadowsatLast–

TransformingMentalHealth,MentalIllnessandAddictionServicesinCanada”

(hereafterreferredtoastheSenateReport).Thisstudyisparticularlysignificantas

itwasthefirststudy(andto‐date,onlypublishedstudy)concerningmentalhealth

13

tobeconductedonanationallevel(MHCC,n.d.).Assuch,althoughtherehavebeen

over300nationalreportsonthephysicalhealthsystem,theSenateReportwasthe

firstnationalreportonmentalhealth(Kirby,2008).

2.2.1 MentalIllnessandHomelessnessintheSenateReport

TheSenateReportaddressesnumerousconcernsrelatedtopolicies,

programs,andthelivedexperienceofthoselivingwithmentalillnessinCanada,and

contains118recommendationsforaddressingtheseissues(TheSenate,2006).

WhilethescopeoftopicsaddressedintheSenateReportrelatingtomentalillnessis

vast,sixthemesthataddressissuesrelatedtohomelessnessandweremostrelevant

tothepopulationbeingdiscussedinthisthesis(thosewhoarehomelessandliving

withmentalillness)wereidentifiedandaresummarizedbelow.

HousingasRecovery

TheCommitteeemphasizesthroughouttheirreportthat“Itwouldbehardto

overestimatetheimportanceofadequatehousingforpeoplelivingwithmental

illness”(p118).Theneedforhousingasaformofrecoverytothoselivingwith

mentalillnessisidentifiedasanessentialneed,andissuggestedtobeaformof

protectionfrommentalillness.TheSenateReportspecificallystressesthestrength

ofsupportivehousingasasuccessfulaspectofrecoveryinitspromotionof

independence,safety,andcommunityintegration.

14

Inequity

Whilehousingisidentifiedasafundamentalaspectofwellbeing,theSenate

Reportdrawsattentiontoincreaseddisadvantagethatthoselivingwithmental

illnesshaveinfindingsafeandadequatehousing.Inadditionofthedifficulties

facedbythosewhoarehomeless,thoselivingwithmentalillnessmayfaceunique

challenges.AsconveyedbyanindividualquotedintheSenateReport:

“We must remember that people also have to face the task ofrecovering from the effects of external and internalized stigma,learnedhelplessness, institutionalization,poverty,homelessnessandthewoundsofabrokenspirit”(p43).

Individualswithmentalillnessmayrequireadditionalsupportiveservicestoenable

themtoliveinthecommunity,increasingthedifficultyoffindingappropriate

housing.Inadditiontothis,theadministrativesysteminplacetoapplyforbenefits,

assistanceandservicesischallengingtonavigatewiththeaddeddifficultiesofliving

withamentalillnessoraddiction,creatingconfusionaboutthesupportservices

availableandhowtoaccessthem.

AffordabilityofHousing

TheSenateReportshedslightonthelargenumberofpeopleoftenlivingin

verysmallhomesinneighbourhoodswithhighcrimeanddrugrates,particularly

thoseonsocialassistance,andstronglystressestheneedforanincreasein

affordablehousing.TheCommitteecitesfindingsoftheCanadianMentalHealth

Organizationthatshowthatthenumberofaffordablehousingunitscreatedbythe

GovernmentofCanadadroppedfrom24,000to940between1980and2000(The

Senate,2006;p119).TheSenateReportassertsthatnotonlyaremorehousing

15

unitsrequired,butthereisaneedformoreGovernmentalassistancesothatpeople

canaffordtorentexistingapartmentsatmarketrates.

IntegrationofServices

TheSenateReporthighlightstheimportanceofaddressingthesocial

determinantsofhealthinpreventingandtreatingmentalillness.TheCommittee

emphasizesthenecessityofbetterintegratingmentalhealthserviceswithabroader

rangeofservicesandsupportsthatarerequiredbythosewithmentalillness,

includinghousingservices.Inparticular,theSenateReportencourages

collaborativecareinitiativestofostercommunity‐basedcare.

StigmaandPublicSafety

AlargeportionoftheSenateReportisdedicatedtoacknowledgingthe

painfulanddamagingeffectsofstigmaonthoselivingwithmentalillness.The

Committeeemphasizesthepressingneedtoaddressstigmarelatedtomentalillness

incommunitiesacrossthecountry.Additionally,theSenateReportpointsto

negativemediadepictionsofmentalillnessthatincreasestigmaagainstthosewith

mentalillness,particularlyportrayalsthatincreasethefearofviolence.

2.2.2 PrimaryPrinciplesandGoals

ThesixthemesidentifiedabovereflectmanyofthePrimaryPrinciplesofthe

SenateReportasstatedbytheCommittee.ThesePrimaryPrinciplesinclude:

• tocreatethebestpossibleconditionsforrecovery;

• toaddresssocialdeterminantsofhealthasarequirementtopromoting

mentalhealthandrecovery;

16

• tointegratementalhealthservicesandsupports;

• toaddressmentalillnessandthementalhealthsystemwithmorerespect

andseriousness

Asmentionedpreviously,inadditiontohighlightingtheneedtoaddress

mentalhealth,mentalillnessandaddictioninCanada,theSenateReportcontains

over100recommendations.TheSenateReportemphasizesthatallits

recommendationsareultimatelytohelpthoselivingwithmentalillness“livethe

bestpossiblelifetheycan”(p479).Themajorityoftheserecommendationsaddress

theorganizationanddeliveryofmentalhealthservices,aimedatimprovingthe

structuralframeworkofthementalhealthsystem.Thereis,however,asocietalgoal

expressedintheSenateReportaswell.Thiscomesfromtherecognitionthat

“profoundchangeisessentialifpersonslivingwithmentalillnessaretoreceivethe

helptheyneedandtowhichtheyareentitled”(TheSenate,2006,p2).The

Committeeexpressesitshopethatreadersofthereportwillbeaffectedbyits

content,andthatthiswillchangedeep‐rootedpublicattitudestowardsmental

illness,subsequentlyreducingthestigmaanddiscriminationfacedbythoseliving

withmentalillness.Theanalysisinthisthesisaimstoaddressthisgoalofthe

SenateReportinrelationtothehomelesspopulationbydeterminingwhetherthe

reportincreasedthepotentialforachangeinpublicattitudesregardingmental

illnessandhomelessness.Inrelationtothis,thepotentialforincreasedpublic

awarenessofthePrimaryPrinciplesoutlinedabovewillalsobeexplored.

17

2.2.3 TheMentalHealthCommissionofCanada

TheSenateReportincludedarecommendationforthecreationofthe

CanadianMentalHealthCommissionwhowould,amongotherresponsibilities,

worktoeliminatediscriminationintheprovisionofaffordablehousingto

individualswithmentalillness(TheSenate,2006).In2007,theMentalHealth

CommissionofCanadaformallylauncheditsactivities,andcontinuedtodraw

attentionthetopicsofhomelessnessandmentalillness.Oneofthefourmajor

initiativesthattheCommissionlaunchedinitsfirstyearwastoaddress

homelessnessrelatedtothoselivingwithmentalillnessthrougharesearchproject

fundedbytheFederalGovernment(Howlett,2008).Thisongoingprojectaimsto

studytheservicesandsystemsforthosewhoarehomelessandlivingwithmental

illnessinfivecitiesacrossCanadawithparticipationfrom2285homelesspeople

livingwithamentalillness(MHCC,n.d.).Itsscopeasthelargeststudyofitskind

currentlyintheworld(MHCC,n.d.)isanindicationoftheimportancetheMental

HealthCommissionhasplacedoninvestigatingandaddressinghomelessness

amongthoselivingwithmentalillness.

2.3 NewsMedia

2.3.1 NewsMediaCoverageofMentalIllness

Themediaisthepublic’smostcommonsourceofinformationregarding

mentalillness(Wahl,Wood&Richards,2002).However,newscoverageofmental

illnesshasshownalong‐standingpatternofnegativeportrayal(Wahl,Wood&

Richards,2002).Studieshavefoundthatnewsarticlespredominantlyportray

18

mentalillnessinanegativetoneandconveyastrongassociationbetweenmental

illnessandviolence(Philo,McLaughlin,&Henderson,1996;Ward,1997;Allen&

Nairn,1997;Nairn,1999;Hazelton,1997).Individualswithmentalillnessareoften

depictedasathreattothecommunity(Allen&Nairn,1997).Storiesassociating

mentalillnesswithviolenceandcrimearegivengreaterprominencewhen

comparedtopositivearticles(Francisetal.,2001,Philoetal.,1996,Day&Page,

1986).Positivestatementsregardingmentalillnessaremostlikelytobefoundin

editorialarticles(Philoetal.,1996).Fewerreferencestoviolencearefoundin

articlesthatcentertheirdiscussionaroundthestatementsofmentalhealth

professionals(Allen&Nairn,1997).Whiletheuseofhealthprofessionalsas

sourcesofinformationmayimprovetheportrayalofthosewithmentalillness,this

issubjecttothejournalistsframingoftheissue(Nairn,1999).

Newscoverageofmentalillnesswasfoundtoreinforcestereotypicaland

stigmatizingviewsofmentalillness(Wahletal.,2002;Philoetal.,1996).Itis

suggestedthatthemediaperpetuateastereotypicalimageofasingleunemployed

malewithschizophrenia,whileothergroupsareunrepresented,suchaswomen,the

elderlyandthosewithmentalillnessesotherthanschizophrenia(Philoetal.,1996).

However,otherstudiessuggestthatcoverageofschizophreniaisuncommon(Wahl,

1996;Wahl,Borostovik&Rieppi,1995).

InCanadaspecifically,thereweretwomajorstudiesregardingmedia

coverageofmentalillness.Matas,el‐Guebaly,Harper,Green,&Peterkin(1986)

studiedthechangeincoverageofmentalillnessinCanadabetween1961and1981

intwonewspapers.Theirresultsshowfewsignificantdifferencesoverthetime

19

periodstudied.Incontrasttootherstudies,theysuggestthatarticleswritten

regardingcriminalincidentsweregenerallypositive.However,theyalsofoundthat

articlesgivenmostprominence(i.e.frontpageofthenewspaper)weremorelikely

toportraymentalillnessasdangerous.DayandPage(1986)didacontentanalysis

ofnewspapersfrom1977‐1984andsimilarlyfoundthatnegativeitemsweregiven

greaterprominencethanpositiveones.Additionally,theirresultsshowthatthe

overalltoneofthearticlesanalyzedwasnegative,withfew(18%)positive

statements.

2.3.2 NewsMediaCoverageofHomelessness

Therehasbeenlittleresearchconductedregardingthedepictionof

homelessnessinnewsmediaandtheextenttowhichthismayimpactpublicopinion

(Min,1999;Lee,Link&Toro,1991).Thefewexistingstudiesfocusprimarilyonthe

UnitedStatesandconsidermostlytelevisionnews(Min,1999).Thisresearchfound

thatnewsstoriesregardinghomelessnessshowanincreasingtrendintheUnited

States(Min,1999)andthatthepublicregardhomelessnessasaseriousissue(Lee

etal.,1991).Whilesomestudiesreportthatthenewscoveragefocusonstoriesof

homelessindividuals,blamingthemfortheircondition(Min,1999),othersreport

thathomelessnessisperceivedtobeaproblemwithstructuralroots(Leeetal.,

1991).Whilethereisnodefinitivepictureonhowhomelessnessisportrayedin

newsmedia,Leeetal.(1991)suggestthatthereisaplausiblelinkbetweenthe

volumeandcontentofmediacoverage(bothprintandbroadcast)andthepublic’s

perceptionofhomelessness.

20

2.3.3 MentalIllness,HomelessnessandNewsMedia

Tomyknowledge,therehasonlybeenonepublishedjournalarticletodate

thatexaminesthemedia,mentalillnessandhomelessnesscollectively.Thisarticle

isentitled“TheMediaandHomelessMentallyIllPersons”andwrittenbyLeonaL.

Bachrachin1990.Thisarticleemphasizesthatwhileprofessionalarticlesfocus

theirattentiononthehomelessandmentallyillindividualswhoareeasiestto

accessandstudy(suchasshelterusers),themediafrequentlyreportsonmore

unusualsituations,givinginsightintolessaccessiblemembersofthispopulation.

Assuch,Bachrach(1990)summarizesthatnewsmediadocumentationofthe

homelessmentallyillshouldbeconsideredanimportantsupplementtoscientific

literatureregardingthispopulation.

2.3.4 NewsMediaandAttitudesofthePublic

Themediahasbeenwidelycitedasthemostimportantsourceofinformation

regardingmentalillness(Granello,Pauley&Carmichael,1999;Lopez,1991;

Benkert,Graf‐Morgenstern,Hillert,Sandmann,Ehmig,Weissbecker,Kleppinger&

Sobota,1997,Philo,1996).InCanada,theSenateCommitteeidentifiesmediaasthe

mosteffectivemeansofspreadinginformationaboutmentalillness(TheSenate,

2006).Howeverresearchfocusedspecificallyontheimpactofportrayalsofmental

illnessisrare(Wahl,1992)andassuch,empiricalevidenceofthisimpactonpublic

attitudesislimited(Dietrich,Heider,Matschinger&Angermeyer,2006).As

mentionedpreviously,thisisevenmorescantwithregardstohomelessness.

21

Whiletheamountofresearchislimited,thelittlethathasbeendonehas

beenconfidentinstatingthatnegativedepictionscanhavesignificanteffectson

people’sattitudesandtreatmentofthosewithmentalillness(Francisetal.,2001;

Wahl,1992).Negativeportrayalsofmentalillnesshavebeenassociatedwith

negativeattitudes(Francisetal.,2001).Dietrichetal.,(2006)showthat

adolescentswhoreadanarticlethatlinkedmentalillnesswithviolentcrimeare

morelikelytodescribeanindividualwithmentalillnessasviolentanddangerous.

Furtherstudieshavefoundthatnegativeattitudesassociatedwithmediadepictions

ofmentalillnessaredifficulttoalteroncepresent.Thepresentationofpositive

imagesofmentalillnessdoesnotappeartochangenegativeattitudes(Francisetal.,

2001).Positivepersonalexperiencescanalsobeoverwhelmedbynegativemedia

images(Philo,1996).

Asnewsmediafocusesontheassociationbetweenmentalillnessand

violence,andthepublicbecomeincreasinglyconcerned,thismayleadtopolicy

decisionsregardingindividualswithmentalillnessthatgiveprioritytopublicsafety

(Paterson,2006).Aspreviouslydiscussed,thisisofparticularimportancein

relationtohousingasproposalsforsupportivehousinggenerallyencounter

neighbourhoodresistance,themajorityofwhichisbasedonfearsregarding

increasedcrimeandreducedpropertyvaluesasaconsequence(Takahashi&Dear,

1997).

22

2.4 AgendaSettingandFraming

AgendaSettingandFramingaretheoriesconcernedwithwhatthemedia

choosetofocusonandhownewsispresented.Agendasettingreferstotheprocess

bywhichproblemsbecomerecognizedasissuesthatmeritpublicattention(Cook,

Tyler,Goetz,Gordon,Protess,Leff&Molotch,1983),andisoneofthemost

commonlycitedinfluencesofnewsmedia(Collins,Abelson,Pyman&Lavis,2006).

ThecentralclaimofAgenda‐SettingTheorywasfirstsummarizedbyCohen(1963):

“Thepressmaynotbesuccessfulmuchofthetimeintellingpeoplewhattothink,

butitisstunninglysuccessfulintellingitsreaderswhattothinkabout.”(ascitedin

Weaver,1996,p37).AgendaSettingtheoryassertsthatwhilemediamaynot

necessarilytellthepublichowtothinkaboutissues,theyshapepublic

consciousnessthroughtellingthepublicwhatissuesareimportant(Brosuis&

Kepplinger,1990).Thisresultsinastrongcorrelationbetweentheimportance

mediaplacesonparticularissuesandtheimportancethepublicattributetothose

issues(Mitchell,2007;McCombs&Shaw,1972).Studieshavesuggestedthistobea

dose‐responserelationshipwiththelevelofimportancegiventoanissuebythe

publicproportionaltotheamountofcoverageitreceivesinthemedia(Collinsetal.,

2006;Funkhouser,1973;Glynn,Herbst,O’Keefe&Shapiro,1999).

Framinggoeshandinhandwithagendasettingtheory.Whileagendasetting

analyzeswhichissuesaregivenimportancethroughbeingselectedasnews,framing

focusesonthewayinwhichissuesarepresentedorinterpretedinthenews

(Entman,1993).Amediaframeisaparticularperspectiveonanewsstory,which

23

canshapetheinterpretationandperspectiveofthepublic,usuallywithouttheir

awareness(Dorfman,2003).

Asframinghaswideapplications,moredetaileddefinitionsofvariousmedia

frameshavebeenwidelydiscussed(Scheufele,1999;Scheufele,2006;Weaver,

2006;McCombs,Shaw&Weaver,1997;Iyengar&Simon,1993).Fortheanalysisin

thisthesis,theframingcomponentwillbediscussedasthenewspaperarticles

approachtothestory.Iyengar(1990a,1991,1996,1997)differentiatesbetween

twotypesofapproach:EpisodicandThematicnewsframing.Episodicframing

“depictspublicissuesintermsofconcreteinstancesorspecificevents”(Iyengar,

1990b,p.7),whilethematicframing“placesthepublicissuesinsomegeneralor

abstractcontext”(Iyengar,1990b,p.7).Episodicframescentertheircoverageon

personalexperiencesandevents(Iyengar,1990b).Theyfocusonthedetails

surroundingtheeventitself,payinglittleattentiontocauses,trendsorbroader

implicationsofevents(Dorfman,Woodruff,Chavez&Wallack,1997).Thistendsto

generatelargenumbersofarticlesoverashortperiodoftime(Collinsetal.,2006).

Incontrast,theobjectofcoverageinthematicframesisgenerallyimpersonal

(Iyengar,1990b)andcoverageofaspecifiedissueismorecontinual(Collinsetal.,

2006).Thematicarticlesfocusonrootcausesandtrends(Dorfmanetal.,1997),

placingissuesinanappropriatecontext(Iyengar,1990b).Whilemuchofframing

researchisconductedwithtelevisionnews,studieshavefoundsimilarframing

effectsandpatternsareexhibitedinnewspapers(Collinsetal.,2006;Dorfman,

2003;McManus&Dorfman,2002).Whilenotallarticlesarepurelyepisodicor

24

thematic,oneframecangenerallybeidentifiedasthepredominantone(Iyengar,

1996;Scheufele,1999).

Thedifferenceintheseframingstylesbecomesparticularlymeaningfulwhen

consideringtheimpliedattributionofresponsibilitytheyconvey.Evidenceshows

thatwhatpeopleidentifyascausesandsolutionstoanissuedependsignificantlyon

thewayinwhichtheissueisframed(Iyengar,1990a).Resultssuggestthatepisodic

framingengendersindividualisticattributionsofresponsibility,(Iyengar,1990a).

Thepersonalizationfoundinepisodicstoriesdrawsattentionawayfromstructural

causes(Dorfman,2003),focusingthepubliconparticularindividualsorgroups

insteadofpolitical,historicalorstructuralissues(Iyengar,1996).Asaresult,

characterdeficiencies(suchaslaziness,apathy,etc.)areblamedforcreating

adversecircumstances(Iyengar,1990b),whichinturnshieldssocietyand

governmentfromresponsibility(Iyengar,1996).Incontrast,thematicframing

assignscausationtosocietalfactors(suchaseconomicconditions,inadequate

governmentalefforts,etc.)andassuch,engendersstrongsocietalandstructural

attributionsofresponsibility(Iyengar,1996).

25

3: METHODS

3.1 NewspaperSelectionandDateRange

NewspaperswerechosenonthecriteriaofbeingaCanadiannational

newspaper,ofwhichtherearetwo:‘TheGlobeandMail’and‘NationalPost’.Both

arebroadsheetEnglishlanguagenewspaperswithadailypublicationschedulefrom

MondaytoSaturday.‘TheGlobeandMail’isconsideredcentristinideology,while

theNationalPostisconsideredconservative/right(Mitchell,2007),however

neithernewspaperisexplicitlyassociatedwithaspecificpoliticalparty(Collinset

al.,2006).Theirweeklyreadership,asreportedIn2007,was2768100(Globeand

Mail)and1570200(NationalPost)(NADBANK,2007).

TheCanadianNewsstandDatabasewasusedtoselectarticlesfromMay9,

2003toMay9,2009(inclusive).Thisrangecoveredthreeyearspriorto,andthree

yearsfollowingthereleaseoftheSenatereport.

3.2 EstablishingSearchTerms

Initialsearchtermswereselectedtotargetarticlesrelevanttotheresearch

question:

‐mentalillness,mentaldisorder,ormentallyill;and

‐homeless,homelessnessorhousing.

Thesearchterm‘mentalhealth’wasconsidered,butasithadmuchwider

implicationsandlessdirectassociationwithmentalillnesswhencomparedtothe

26

otherinitialsearchterms,itresultedinmanyirrelevantarticles.Asampleof

articlesusing‘mentalhealth’asasearchterm(andexcludingtheotherinitialsearch

termsrelatedtomentalillness)werestudiedforanyfurtherphrasesthatmaybe

usedtopointmoredirectlytomentalillness.Itwasfoundthatmanyoftherelevant

articlesusedtheterms‘mentalhealthissue/s’and‘mentalhealthproblem/s’to

refertomentalillness,andthesephraseswereincorporatedintotheinitialsearch.

‘Mentalhealthsystems’wasalsoaddedtothesearchterms.Oftheremaining120

articlesassociatedwiththe‘mentalhealth’searchterm,arandomsampleof40%of

thearticleswasselectedforaninitialanalysis,whichshowedthat90%ofthese

articleswerenotrelevanttohomelessnessormentalillness.Thesearticleswould

ultimatelybeexcludedfromthedataanalysisatalaterpointforbeingirrelevant.

Giventhelowproportionofrelevantarticlesgeneratedfromthe‘mentalhealth’

searchterm,itwasnotusedasasearchtermforthisanalysis.

Thefinalsearchtermsweretherefore:

• mentalillness,mentaldisorder,mentallyill,mentalhealthsystem,mental

healthissue/sormentalhealthproblem/s;and

• homelessorhomelessnessorhousing.

3.3 ArticleSelection

Oncethesearchtermswereestablished,allarticleswithinthespecifiedtime

periodcontainingthesearchtermswereselected,resultingin414articles.

Obituaries,letterstotheeditor,articlesunder150words,andentertainment/art

reviewswerethenexcluded.Articlesthatappearedmorethanoncewerealso

excluded.Thesearticleswereonesthatappearedinvariousregionaleditionsofthe

27

newspaperandwereidenticalincontent.Atotalof81articleswereexcludedas

summarizedinTable1,leaving333articlesforanalysis.

Table1: Summaryofarticleexclusion(notmutuallyexclusive)

Reasonforexclusion Numberofarticles

Entertainment/ArtReview 23

LettertotheEditor 22

Articlerepetition 22

Wordcountunder150 14

Obituary 8

IndexArticle 2

3.4 CodingProtocol

Acodesheetwasdevelopedtogatherdataregardingpublicationdetailsand

potentialagendasettingandframingeffects.Althoughmediacontentanalysesoften

usequalitativecodingmethods(Collinsetal.,2006),thecodesheetwasdesignedto

consistprimarilyofcategoricalvariables.Thiswastoreducethepossibilityofcoder

subjectivityandtolaterallowforbothadescriptiveandquantitativeapproachto

thedataanalysis.Pilottestingwascarriedoutamongundergraduatevolunteers

andtheresultsofthisidentifiedvariablesthatneededtobeaddedorclarifiedto

improvereliabilityacrosscoders.Thefinalcodesheet(AppendixA)contained23

items.

28

3.4.1 TheoreticalOrientation

Thecodesheetwasdesignedtoallowanalysisofpotentialagendasettingand

framingeffectsassummarizedinTable2.Potentialagendasettingeffectswere

examinedbycodingforthequantityofcoverageofmentalillnessandhomelessness

inthenewspapers,theprominenceofthearticles,andthequantityofcoverageof

selectedthemesidentifiedintheSenateReport.Thepotentialforframingeffects

wasexploredinthecomparisonoftheamountofepisodicandthematiccoverage

(articleapproach),theassociationofthearticleapproachtothesolutionsidentified

inthearticles,andthevoicethatwaspresentedthroughthearticles.

Table2: Variablesusedtoassesspotentialmediaeffects

PotentialEffectsExplored Variable

AgendaSetting QuantityofCoverage

ArticleProminence

Articlethemes

Framing ArticleApproach

Voice

Responsibilityandbenefitsattributed

tosolutions

3.4.2 ArticleCharacteristics

Eacharticlewasassignedanidentificationnumber(ID#)andcodedforbasic

articlecharacteristics–newspapername,articledate,typeofarticle(newsor

29

editorial/column),authorandarticleheadline.Thesewereprimarilyfor

identificationofthearticlesandtheirplacingalongatimeline.Thearticle’sword

count,whetheraphotographaccompaniedit,anditsplacementwithinthe

newspaperwascodedtoassessarticleprominencetoseewhetherthischangedover

time.Acodeforarticlerelevancewasaddedafterthepilottestsshowedthat

althoughsearchtermswerepresentwithinthearticle,theyweresometimes

misidentified(e.g.thedatabaseidentified‘elementalproblem’asthesearchterm

‘mentalproblem’)ormentionedonlyinpassing(e.g.inthejobtitleofaquoted

individual),makingfurthercodingnotpossible.Thesearticleswerecodedas

keywordonlyandlaterremovedfromtheanalysis.Remainingarticleswerealso

codedforrelevanceusingaroughapproximationofhowmuchofthearticlewas

relatedtomentalillnessand/orhomelessness.Astheoverlapofthehomelessand

thosewithmentalillnesswasoftennotclearlydefinedduetothenatureofthis

population,articleswerecodedforbeingaboutmentalillnessorhomelessnessin

general,notnecessarilythehomelessmentallyillspecificallybecauseofthislackof

clarity.

3.4.3 ArticleThemes

TheAgendaSettingcomponentofthecodingwasprimarilyinthedetection

ofarticlethemes.Sixmajorthemesrelatedtomentalillnessandhomelessnesswere

identifiedin‘OutoftheShadowsatLast’,aspreviouslydiscussedintheBackground

sectionofthisthesis:

• Recovery:Identificationofhousingasaformofrecoveryforthoseliving

withmentalillness;

30

• PublicSafety:Concernsrelatingtothesafetyofthepublicatlarge;

• Inequity:Issuesrelatedtotheuniquechallengesfacedbythispopulation

ortheirincreasedriskofbeinghomeless/mentallyill;

• AffordableHousing:Problemsrelatingtothecostofhousing(e.g.lackof

affordablehousing,needforallocationoffundingtowardscheaper

housing,etc);

• IntegrationofServices:Indicationoftheneedforservicesoragencies

(e.g.health,housing,police,healthauthorities,etc)toworktogether,or

shareinformation;

• Stigma:Specificmentionofstigma,discriminationorNIMBYism.

AsAgendaSettingtheoryfocusesonthepresenceofparticularcontent,the

themeswerecodedasbeingeither‘mentioned’or‘notmentioned’inanarticle,

regardlessofthecontextorlengthofdiscussionaroundit.

3.4.4 ArticleFraming

Theframingcomponentofthecodingwasinarticleapproachandvoice.

Articleswerecodedforwhethertheywereepisodicorthematicintheirapproach

andwhatevent/storytheywerewritteninresponseto.Episodicarticleswere

thoseinwhichtheprimaryfocusofthearticlewasaspecificeventorstory,with

littleattentiontothebroadercontextoftheissue.Thematicarticleswereidentified

asthoseinwhichthemajorityofthestoryfocusedonthecontext,trendsorcauses

ofaparticularissueinsteadofaspecificeventitself.Althoughmanyarticles

containedelementsofbothapproaches,thedominantapproachwasidentifiedand

coded.Articleswerealsocodedforwhethertheystatedasolutiontoanyissues

raisedinrelationtohousingandmentalillness.Pilotstestsofthecodesheetshowed

thatcodingforsolutionsthatwereimpliedinthearticleledtoverylowcorrelation

31

betweencoders,signallinganeedforamoreobjectivecode.Assuch,thecodesheet

wasalteredtocodeonlyforclearlystatedsolutionsmadebythearticle.The

solutionswerethencodedforwheretheresponsibilitylayintheir

creation/implementationandwhowouldbenefitasaresult.Thefinalaspectthat

wascodedwasvoice,whichwasdefinedasanyonedirectlyquotedinthearticle,to

seewhatperspectivesandgroupswererepresentedwithadirectvoice,and

whetherthischangedovertime.Somearticleshadquotationswithnopersona

attachedtothem(innameorposition),andthesecouldnotbecoded.

3.5 CodingMethods

Codingofthearticleswasdoneinthreestages.Theinitialstageinvolved

usingaPerlscriptthatwasdesignedforthisspecificsetofarticles(SeeAppendixB

forthefullscript).Thisscriptscannedalltheselectedarticlesandidentifiedthe

newspapername,identificationnumber,articledate,articletype,headline,author,

wordcount,pagenumberanddirectlinktoeacharticle.Itthenenteredthisdata

intoaspreadsheetwherearandomsamplewascheckedforaccuracybyacoder.As

thesedetailswerenotsubjecttointerpretation,anautomatedformofcodingwas

thoughttodecreasehumanerrorandprovidegreateraccuracyofresults.Inthe

secondstage,allofthearticleswerecodedfortheremainingitemsonthecodesheet

byonecoder.ThecoderwasblindtoallidentifyinginformationexceptforID#,such

asarticledate,headlineandauthornametodecreaseanybiasincoding.Toreduce

errorsinrecordingandimproveinternalvalidity,thiscodingwasconducted

throughtheuseofaformthatidentifiedvariablesforselectionandautomatically

32

transferredthedataintoaspreadsheet(SeeAppendixCfortheform).Inthefinal

stage,asampleof20%ofthearticles(n=68)wasrandomlyselectedandcodedby

anotherindividualtoassessintercoderreliability.

3.6 StatisticalAnalysis

CodedarticleswereenteredintoanExcelSpreadsheet(Excel2008,Mac

version)formanagement.Basicdescriptivestatisticssuchasfrequency

distributionswereconductedusingthissoftware.Datawasalsoimportedintoan

SPSSDatabase,version17.0,forcross‐tabulationtests.Thesetestscalculatedthe

Pearsonexactchi‐squaretest(usingthe95%confidencelevel)fortestsof

significantdifferencesintheappearanceofarticlethemesanddifferencesin

episodicandthematiccoverage.Forsmallsamples,Fischersexacttestwasused.

33

4: RESULTS

Asstatedpreviously,articleswereselectedfromMay9,2003–May,9,2009

(inclusive)forcodingananalysis.Assuch,whenresultsarereferredtobyyearin

theproceedinganalysis,theyearisconsideredtobeginonMay9(i.e.2003‐2004

referstoMay9,2003–May8,2004inclusive).Theresultsalsoreferencethethree

yearspriortoandthethreeyearsfollowingthereleaseoftheSenateReport,which

willbereferredtoaspre‐SenateReportandpost‐Senatereportrespectively.

4.1 Inter­coderReliability.

Aspreviouslymentioned,arandomsampleof20%ofthearticles(n=68)

werecodedbyasecondcoder.Inter‐coderagreementwasassessedseparatelyfor

specificvariables,namelydepthofdiscussion,eacharticletheme,episodic/thematic

framingandpresenceofasolution.Kappascoresforthesevariablesrangedfrom

0.731–1.00(p=<0.001forallscores)showingsubstantialtoalmostperfect

agreement(Viera&Garrett,2005),indicatingahighlevelofinter‐coderreliability.

TheKappascoreforeachassessedvariablecanbefoundinAppendixD.

4.2 GeneralOverview

Atotalof333newspaperarticleswerecoded.Articlesinwhichthedepthof

discussionregardingthehomelessandmentallyillwereconfinedto‘keywordonly’

wereremoved(n=102),leaving231articlesforanalysis.Thedepthofdiscussionof

theremainingarticleswasprimarily‘majority/entirearticle’(n=179).Withinthe

34

sampleddatarange,articleswerefoundfrom13‐August‐2003to7‐May‐2009.Of

these,55articleswerefromtheNationalPostnewspaper,and176werefromThe

GlobeandMailnewspaper.Thenumberofauthors(bothjournalistsandpublic

figures)representedintheanalyzedsamplewas127,withthemaximumnumberof

articleswrittenbyasinglejournalistbeing12.

4.3 ArticleCharacteristics

4.3.1 QuantityofCoverage

QuantityofCoveragewasusedtoinvestigateagendasettingeffectsandwas

countedasthetotalnumberofarticlesinagivenperiodoftime.Therewasa

greaternumberofarticlesinthethreeyearsfollowingthereleaseoftheSenate

Report(n=169)thanthethreeyearsprior(n=62),ascanbeseeninFigure1.

35

Figure1: QuantityofCoverageBeforeandAfterthereleaseofTheSenateReport

Toseehowthisrelatesproportionallytohowmanyarticleswerepublished

ingeneralinthoseyearsandtoestablishwhetherthisincreasewasduetoageneral

increaseinnewspaperarticlepublication,asearchwasconductedtoestablishthe

totalnumberofarticlespublishedbyTheGlobeandMailandNationalPost

newspapersinthesixyearsbeingstudied.Theresultsfound434,880articlesinthe

threeyearspre‐SenateReport,and374,270articlespost‐SenateReport.These

figuresrepresenttotalnumberofarticles,includingtypesofarticlesthatwere

excludedfromthestudysamplesuchasletterstotheeditorandarticlesfewerthan

150words.Assuch,thesefiguresarenotmeanttoconveyanaccurate

representationofnewsarticlesprintedinthattime.Theyarehowever,importantin

36

exploringwhethertheincreaseinthenumberofarticlesrelatedtomentalillness

andhomelessnessismeaningful.Consideringthis,therewasasignificantincrease

inthenumberofarticlespost‐SenateReport(p<0.0001at99%confidenceinterval).

ThetrendsinthequantityofcoveragebyyearareshowninFigure2.The

numberofarticlesincreasedeveryyear,with15articlesin2003‐2004,17articlesin

2004‐2005,29articlesin2005‐2006,46articlesin2006‐2007,61articlesin2007‐

2008and63articlesin2008‐2009.

Figure2: QuantityofCoveragebyyear

4.3.2 ProminenceofCoverage

Articleprominencewasexploredbyanalyzingthechangeintheaverage

articlewordcount,theplacementofthearticlewithinthenewspaperandthe

presenceofartworkorphotographsaccompanyingthearticle.Themeanword

countofthearticlesoverthesixyearsstudiedwas973words/article,withasmall

increasebetweenthemeanwordcountpre‐SenateReport(909words/article)and

37

post‐SenateReport(996words/article).Therewaslittlevariationinthemean

wordcountbyyearasshowninFigure3,withthegreatestvariationbeingseen

between2007‐2008and2008‐2009.

Figure3: AverageWordCountbyyear

ForArticlePlacement,inthethreeyearsprecedingthereleaseoftheSenate

Report,therewerethreearticlesthatappearedonthefrontpageofthenewspaper,

10thatappearedonthesectionfrontpage,and49articlesthatappearedinother

pagesofthenewspaper(seeFigure4).Comparatively,inthesubsequentthree

years,sevenarticlesappearedonthefrontpageofthenewspaper,33onthesection

frontpage,and129inotherpartsofthenewspaper(seeFigure5).When

comparingFigures4and5,itisimportanttonotethatinadditionanincreaseinthe

totalnumberofarticlesappearinginthepaperfrontpageandsectionfrontpagesof

38

thenewspapers,thereisanincreaseintheproportionofarticlesinprominent

positionsofthenewspaperpost‐SenateReport.

Figure4: ArticlePlacementPre­SenateReport(No.ofarticles)

Figure5: ArticlePlacementPost­SenateReport(No.ofarticles)

Therewerefewarticleswithanaccompanyingphotograph(n=33)inthe

analyzedsample.InthethreeyearspriortothereleaseoftheSenateReport,10%

39

ofthearticleshadaphotograph(n=6),whileinthethreeyearsfollowing,16%ofthe

articleshadaphotograph(n=27).

4.4 ArticleThemes(AgendaSetting)

Potentialagendasettingeffectswerefurtherexploredthroughthe

appearanceofthemesderivedfrom‘OutoftheShadowsatLast’.Nostatistically

significantdifferenceswereobservedpreandpost‐Senateinthementionsofanyof

thethemes(p‐valuesforeachthemecanbefoundinAppendixE).However,there

wasanincreaseinthenumberofarticlesmentioningeachtheme,asoutlinedbelow.

Recovery

Thenumberofarticlesthatmentionedthethemeof‘Recovery’showedan

increasefromthetimepre‐SenateReport(n=6)toPost‐SenateReport(n=34).

Figure6showsthenumberofarticlesmentioningthethemeofrecoverybyyear,

showingveryfewmentionsinbetween2003and2006,whichweremorethan

doubledin2006andfurtherincreasedin2008‐2009.

40

Figure6: Frequencyofarticlesmentioning‘Recovery’

PublicSafety

Thenumberofarticlesthatmentionedthethemeof‘PublicSafety’increased

inthethreeyearsfollowingthereleaseoftheSenateReport(fromn=10ton=27).

Whilethenumberofarticleswiththisthemewaslowbetween2003and2005(only

onearticleineachyear),itincreasedtoeightarticlesin2005‐2006,andremained

betweeneightandtenarticlesinthethreeyearsfollowing,andshowninFigure7.

Figure7: Frequencyofarticlesmentioning‘PublicSafety’

41

Inequity

Thefrequencyofarticlesmentioningthethemeof‘Inequity’showedan

increaseinthepost‐SenateReportyears.Thefrequencyofarticlesmentioningthis

themeshowedanincreasefromatotalof12articlesinthethreeyearspriortothe

releaseoftheSenateReporttoatotalof38articlesinthethreeyearsfollowing.The

trendintheincreaseofthenumberofarticlesisshownbyyearinFigure8.

Figure8: Frequencyofarticlesmentioning‘Inequity’

AffordableHousing

Thetheme‘AffordableHousing’wasthemostfrequentlymentionedtheme

amongallthearticles,beingmentionedinatotalof63articlesoverthesixyears

analyzed.Thenumberofarticlesthatmentionedthisthemeincreasedfromatotal

of16articlespre‐SenateReporttoatotalof47articlesinthethreeyearsfollowing

thereleasethereport.Therewasamarkedincreaseinthenumberofarticles

mentioningthisthemein2006‐2007.Thisdecreasedin2007‐2008,butthenumber

42

ofarticlesinthisyearstillremainedhigherthaninanyofthethreeyearspre‐Senate

Report.Thebreakdownofthenumberofarticlesmentioningthisthemebyyearis

showninFigure9.

Figure9: Frequencyofarticlesmentioning‘AffordableHousing’

IntegrationofServices

Thetheme‘IntegrationofServices’increasedfromafrequencyofthree

articlesintheyearspriortothereleaseoftheSenateReportto20articlesinthe

threeyearsfollowing.Thequantityofarticlesmentioningthisthemeshoweda

generalincreaseovertime,asseeninFigure10,exceptfor2004‐2005inwhichno

mentionsofthisthemewerefound.

43

Figure10: Frequencyofarticlesmentioning‘IntegrationofServices’

Stigma

Thenumberofarticlesmentioningstigmaincreasedpost‐SenateReport.The

frequencyofarticlesincreasedfromatotaloffivearticlesmentioningstigmainthe

threeyearsprecedingthereleaseoftheSenateReport,toatotalof30articlesinthe

followingthreeyears.Figure11showsthefrequencyofarticlesmentioningthis

themebyyear.Ascanbeseen,therewasamarkedincreaseinthenumberof

articlesthatmentionedthethemeofstigmain2006‐2007.Whilethefrequencyof

articlesmentioningthisthemedecreasedin2007‐2008,itagainshowedanincrease

in2008‐2009.

44

Figure11: FrequencyofarticlesmentioningStigma

Summaryofthemes

Asummaryofthefrequencyoftheappearanceofspecificthemes,bothpre

andpost‐SenateReport,isshowninFigure12.Asdemonstratedinthisfigure,all

themesshowedanincreaseinthefrequencyofarticlesinwhichtheywere

mentionedPost‐SenateReport.ThethemesofStigma,RecoveryandIntegrationof

Servicesshowedamoremarkedincreaseincomparisontootherthemes,increasing

byafactorof6,5.7and6.7respectively.Incomparison,thethemesofPublicSafety,

InequityandAffordabilityofHousingincreasedbyafactorof2.7,3.2and2.9

respectively.

45

Figure12: FrequencyofArticlesmentioningthemes

4.5 Framing

Potentialframingeffectswereexploredthroughtheanalysisofarticle

approach,namelyepisodicorthematic.Ofallthearticlesusedforanalysis(n=231),

themajoritywereepisodic(n=200).ThiscanbeseeninFigure13.

Figure13: EpisodicandThematicArticles(No.ofArticles)

46

Figures14and15showtheamountofepisodicandthematiccoveragepre‐

SenateReportandpost‐SenateReport.Whiletheamountofepisodiccoverageand

thematiccoveragebothincreasepost‐SenateReport,Figure15showsanincreasein

theproportionofthematicarticlesPost‐SenateReport.

Figure14: ArticleApproachPre­SenateReport(No.ofArticles)

Figure15: ArticleApproachPost­SenateReport(No.ofArticles)

47

Figure16showstheamountofepisodicandthematiccoverageineachother

analyzedyears.Bothepisodicandthematiccoverageshowanincreaseeachyear,

exceptfor2008‐2009inwhichthenumberofepisodicarticlesdropped.The

numberofthematicarticleswaslowinthefirstfewyearsofanalysisandincreased

intheyearsfollowingthereleaseoftheSenatereport(2006‐2009).Aswiththe

coverageofmanyofthethemes,thenumberofthematicarticlesshowedamore

pronouncedincreasetheyearthattheSenateReportwasreleased,followedbya

decreaseinthefollowingyear.Thiswasthenfollowedbyanotherpronounced

increasein2008‐2009.

Figure16: QuantityofEpisodicandThematicArticlesbyYear

PrimaryEvent/Story

Articleswerecodedfortheirprimaryevent/story,thesummaryofwhichis

presentedinTable3.

48

Table3: ArticlesPrimaryEvent/Story

MainStory/EventEpisodic(n)

Thematic(n)

Total(n)

Other 356 41

No/unclear 25 16 41

Personalstory 23 1 24

Newbuilding/zoningallocation/opening 21 1 22

Politicalevent/election 14 1 15

Releaseofreport 11 4 15Violence/crimeagainstanMI&Hindividual 15 0 15

Violence/crimebyanMI&Hindividual 13 0 13

Fundingallocation 10 1 11

Newpolicy/strategy 8 0 8

Deathofahomelessindividual 7 0 7

Healthpolicy/administrativechange 5 0 5

Homelesscount 5 0 5

Policeaction 4 1 5

2010WinterOlympicGames 4 0 4

ThemostfrequentlycodedcategorieswereNo/Unclear(n=41)orOther

(n=41)accountingfor35%ofthetotalsampleofarticles.Thesewerearticlesthat

eitherdidnothaveacleareventorstory,orthoseinwhichthestorycouldnotbe

categorizedinameaningfulwaywithothers.Examplesofthisincludereportsofthe

creationofahomelesschoir,thelossofahomelessman’sdogandacommentaryon

theuseofasafeinjectionsite.Forvisualclarity,thesetwovariableswereremoved

forFigure17,tomoreeasilyshowthefrequencyoftheremainingvariables.Stories

49

thatfocusedonpersonalstories(n=24)andNewbuilding/zoning/allocation/

opening(n=22)werethenextmostfrequentstories.ThiswasfollowedbyPolitical

events(suchaselections),thereleaseofareport,andviolence/crimeagainstan

individualidentifiedashomelessandmentallyill(n=15).

Figure17: ArticlesPrimaryEvent/Story(No/Otherremoved)

Astherewerefewthematicarticles,thestoriesrepresentedinepisodic

articlesshowedasimilartrendasthoseinthefullgroupofarticles(asseeninTable

3).Thethematicarticleshowever,showedadifferentpattern.WhileNo/unclear

!" #" $!" $#" %!" %#"

!"#"$%&'()*$+,-./&0$12.)3$

4)2,(5$/6,&0-728.&'&3(*2(&9)$052':)$

46.),)33$06;'($

<6,&0)$20(&6'$

=)2(5$6>$2$56.),)33$&'8&9&8;2,$

?)@$/6,&0-73(*2():-$

A;'8&':$2,,602(&6'$

B&6,)'0)70*&.)$C-$2$.)'(2,,-$&,,756.),)33$&'8&9&8;2,$

<6,&(&02,$)9)'(7),)0(&6'$

D),)23)$6>$*)/6*($

B&6,)'0)70*&.)$2:2&'3($2$.)'(2,,-$&,,756.),)33$&'8&9&8;2,$

?)@$C;,8&':7E6'&':$2,,602(&6'76/)'&':$

<)*36'2,$3(6*-$

!"#$"%$&'()*+,-$

.'()*+,-$/')0&'1$23,4(56("'1$7!"58(9,'$',0"3,:;$

50

andOtherwerestillthemostfrequentstory,thenextmostfrequentstorywasthe

releaseofareport.

ThePrimaryEvent/Storywasthenstratifiedbypublicationyear,theresults

ofwhicharesummarizedinTable4.

Table4: ArticlesPrimaryEvent/StoryoverTime

PrimaryEvent/Story2002­2003

2003­2004

2005­2006

2006­2007

2007­2008

2008­2009

Other 2 4 5 9 10 11

No/unclear 1 3 7 10 9 11

Personalstory 3 2 0 5 5 9

Newbuilding/zoningallocation/opening 1 0 2 4 10 5

Politicalevent/election 7 0 1 1 1 5

Fundingallocation 0 0 0 4 2 5

Violence/crimebyanindividualwhowashomelessand/ormentallyill 0 2 2 3 3 3

Releaseofreport 0 0 3 5 6 1

Healthpolicy/administrativechange 0 0 0 0 2 3

Policeaction 0 0 0 2 0 3

Deathofahomelessindividual 1 1 1 0 2 2

Violence/crimeagainstanindividualwhowashomelessand/ormentallyill 0 1 5 1 7 1

Newpolicy/strategy 0 2 0 1 4 1

Homelesscount 0 2 2 1 0 0

2010WinterOlympicGames 0 0 1 0 0 3

51

Oftheseresults,thereareafewnotabletrends.Articlescategorizedas

No/unclearandOthershowedasimilaryearlyincreasingtrendtothetotal

frequencyofallarticles,asdemonstratedinFigure18.

Figure18: PrimaryEvent/Story:OtherandNo/Unclear

Articleswithafocusonpersonalstories(seeFigure19)showedadecline

fromthreearticlestononeinthethreeyearspre‐SenateReport.Intheyear

followingthereleaseoftheSenateReporttherewasanincreaseinthenumberof

articles,andafurtherincreasecanbeseenin2008‐2009.

!"

#"

$"

%"

&"

'!"

'#"

#!!()#!!$" #!!$)#!!*" #!!*)#!!%" #!!%)#!!+" #!!+)#!!&" #!!&)#!!,"

!"#$"%$&'()*+,-$

.,/'$

0')1/'2$34,5(67("'28$9(:,'$/5;$!"6<5*+,/'$

!"#$%&

'()*+,-$.%&

52

Figure19: PrimaryEvent/Story:PersonalStory

Articleswrittenaboutapoliticalevent/electionshowedgenerallylow

coveragewithpeaksofhighercoverageinspecificyears.Thiscanbeexpectedas

mostofthesearticlesreferredtopoliticiansandtheirelectionplatformregarding

homelessness(andsometimesmentalillness).Assuch,itcouldbeexpectedthat

therewouldbeincreasedcoverageregardingthistopicinyearswithanelectionor

politicalcontraversy.Articlesregardingthereleaseofareportwereonlypresent

after2005,andwhiletheyshowedincreasingcoverageuntil2008,thenumberof

articlesonthistopicdeclinedin2008‐2009.Articlesregardingfundingallocations

wereonlypresentpost‐Senatereport.Articlesinwhichtheprimaryevent/story

wasviolenceagainstanindividualwhowashomelessand/orlivingwithmental

illnessshowedgenerallylowyearlycoveragewithpeaksofhighercoverageintwo

specificyears,asshowninFigure20.

!"

#"

$"

%"

&"

'"

("

)"

*"

+"

#!"

$!!%,$!!&" $!!&,$!!'" $!!',$!!(" $!!(,$!!)" $!!),$!!*" $!!*,$!!+"

!"#$"%$&'()*+,-$

.,/'$

0')1/'2$34,5(67("'28$0,'-"5/+$7("'2$

53

Figure20: PrimaryEvent/Story:ViolenceAgainstaHomelessand/orMentallyIllIndividual

Incomparison,articlesinwhichtheprimaryevent/storywasviolencebyan

individualwhowashomelessand/orlivingwithmentalillnessshowedmore

generallysteadyreportingwithalowoverallyearlyfrequencyofarticlesregarding

thistopic.ThiscanbeseeninFigure21.

Figure21: PrimaryEvent/Story:ViolenceByaHomelessand/orMentallyIllIndividual

!"

#"

$"

%"

&"

'"

("

)"

*"

$!!%+$!!&" $!!&+$!!'" $!!'+$!!(" $!!(+$!!)" $!!)+$!!*" $!!*+$!!,"

!"#$"%$&'()*+,-$

.,/'$

0')1/'2$34,5(67("'28$9)"+,5*,$&:/)5-($/$;"1,+,--$/5<6"'$

=,5(/++2$>++$>5<)4)<?/+$

!"

#"

$"

%"

&"

'"

("

)"

*"

$!!%+$!!&" $!!&+$!!'" $!!'+$!!(" $!!(+$!!)" $!!)+$!!*" $!!*+$!!,"

!"#$"%$&'()*+,-$

.,/'$

0')1/'2$34,5(67("'28$9)"+,5*,$:2$/$;"1,+,--$/5<6"'$

=,5(/++2$>++$>5<)4)<?/+$

54

Articleswrittenaboutthedeathofahomelessindividualwereinfrequent

anddidnotmarkedlychangeinthetimeperiodstudied.

Solutions–ResponsibilityandBenefits

Articleswereanalyzedforwhethertheyprovidedaclearsolutiontoany

problemsidentifiedregardingthehomelessorthosewithmentalillness.Ofallthe

articlesanalyzed,49articles(21%)offeredaclearsolution.Ofthese,33werecoded

asepisodicarticles,and16werethematic.Whilethereweremoreepisodicarticles

amongthosewithsolutions,itisimportanttonotethatthe33articlesaccountfor

onlyfor16.5%ofallepisodicarticles,whileincontrast,51.6%ofthethematic

articlesofferedclearsolutions.Thematicarticlesofferedasolutionsignificantly

morethanepisodicarticles(p=<0.0001).

Articlesthatcontainedaclearsolutionwerethenanalyzedtoseewhether

theresponsibilityforimplementingthesolutionlaywithgovernmentalagencies,or

withtheindividualsaffectedbymentalillnessand/orhomelessnessthemselves.

TheresultsaresummarizedinFigure22.

55

Figure22: AttributionsofResponsibility

Asshowninthisfigure,themajorityofthesolutionsattributedthe

responsibilityofthesolutiontoGovernmentalagencies(92%).Whenstratifiedby

articleapproach(Figure23),itcanbeseenthatwhilebothepisodicandthematic

articlesmostfrequentlyattributedresponsibilityforsolutionstoGovernmental

agencies,nothematicarticlessuggestedthattheindividualsaffectedshouldbe

responsible.

Figure23: AttributionsofResponsibility(EpisodicandThematic)

!"#

$#$#

%&&'()*&(+,-#+.#/0-1+,-()(2(&3#

!"#$%&'$&()*+

,&-.#.-/)*0+)11$2($-+

3&2*$)%4&"(+$56*.2.(+

56

Articlesthatpresentedsolutionswerethencodedtoseewhichgroupswere

identifiedasrecipientsofthesolution,meaningthosewhowouldbenefitfromthe

solutionifitweretobeimplemented.TheresultsareshowninFigure24,showing

thatthemajorityofthesolutionsofferedwereidentifiedasbeingofbenefittothose

whowerehomelessand/ormentallyill(n=22,45%).

Figure24: BenefitsoftheSolution

Whenstratifiedbyarticleapproach(Figure25),resultsshowedthatepisodic

articlesthatofferedsolutionsprimarilysuggestedthattheindividualsaffected

wouldbenefit,whilethematicarticlesweremainlyunclearornotexplicit.

!"

##"

$%"

&"

'()(*+,-"./",0("1.23,+.)"

!"##$%&'()*')+*,-.)

/%0&1&0$*+2)*33.4'.0)

5%4+.*,6%"').78+&4&')

9"':)

57

Figure25: BenefitsoftheSolution(EpisodicandThematic)

Authors

Authorsofthenewspaperarticlesweregroupedbythenumberofarticles

writtenbyeachindividualauthorinthesample.Thishelpedtodistinguishbetween

regularjournalistsandarticleswrittenbyone‐timeauthors(oftenpoliticalorother

publicfigures).Thematicarticleswerefoundtobesignificantlymorelikelytobe

writtenbyjournalistswhohadwritten4ormorearticlesrelatedtomentalillness

andhomelessness(p=0.0298).

Voice

ThevoiceofanyoneindirectquotationswasnotedandsummarizedinTable

5below.Inarticlesfrombothpreandpost‐SenateReport,individualsidentifiedas

homelessand/orlivingwithmentalillnessweremostfrequentlyquoted.

58

Table5: SummaryofVoiceinallarticles

Whoisquoted(Individualidentifiedas) Pre­Senate

Report(No.

ofarticles)

Post­Senate

Report(No.

ofarticles)

Homeless/MentallyIll/HomelessandMentallyIll 20 42

Politician 16 32

HealthProfessional,Researcher/Expert 10 32

Organization/ProfessionalSociety 9 34

Shelter/Housingrepresentative 7 1

Policeofficer/representative 6 19

Ministry/HealthAuthority 6 15

FamilyMember 6 10

Judge 3 9

Advocate 3 18

Governmentrepresentative 2 16

Businessperson 0 6

Articlesisaninterview 1 6

None 10 30

Other 12 16

59

5: DISCUSSION

Overall,theresultsshowthattherehasbeenachangeinthedepictionof

mentalillnessandhomelessnessinCanadiannationalnewspapercoveragesincethe

releaseoftheSenateReport.TheresultssuggestthattheSenatereporthada

lastingandmeaningfuleffectonthepublicagendaadvancedbythenational

newspapercoverage.Theresultsfurthersupportthatnewspapercoverage

regardinghomelessnessandmentalillnessshowsanincreasingtrendinthe

presentationofthemesidentifiedasprioritiesintheSenateReport.Additionally,

whilethemajorityofcoverageisfocusedonepisodicreporting,thereisanincrease

inthenumberofarticlesprovidingcontextualinformationandsolutionsregarding

thispopulation.

Thischapterdiscussesthesekeyfindingsinthecontextofagendasettingand

framingtheories,followedbyadiscussionofthelimitationsofthisanalysisand

directionsforfutureresearch.

5.1 KeyFindings

5.1.1 AgendaSettingEffects

Therewasasignificantincreaseinthenumberofarticleswritteninthethree

yearsfollowingthereleaseof‘OutoftheShadowsAtLast’comparedtothethree

yearspriortoitsrelease.Thequantityofcoverageincreasedeachyearinthetime

periodstudied.Theaveragelengthofthenewspaperarticlesshowedlittlechange.

60

Howevertherewasanincreaseinthenumberofarticlesappearingonthefront

pageofthenewspaperandfrontpageofsectionsofthenewspapersaswellasan

increaseinthenumberofarticlesaccompaniedbyaphotograph.Thisindicatesan

increaseintheprominenceofarticlesrelatedtohomelessnessandmentalillness.

Theincreasedquantityofcoverageandthegreaterprominenceofrelevantarticles

implythatthepotentialforexposuretoarticlesregardingmentalillnessand

homelessnessincreasedpost‐SenateReport.Thisinturnsignifiesthatthepotential

foragendasettingeffectsincreasedfollowingthereleaseoftheSenateReport.

Theanalysisalsoindicatesthatinthethreeyearsprecedingthereleaseofthe

SenateReport,therewereanincreasingnumberofarticleswritteneachyear.A

likelyexplanationforthisisfoundintheSenateReportitself.Inthereport,the

Committeerecognizesthatduringthetwoandahalfyearsinwhichtheywere

conductingtheirresearch,themomentumtoaddressandchangethementalhealth

systemwasalreadybuilding.

Whiletheamountofcoverageregardingmentalillnessandhomelessness

increasedintheyeartheSenateReportwaspublished,thereisfurtherevidenceof

thisincreaseinthesecondandthirdyearfollowingthereleaseoftheSenateReport.

Aseffectsofreportsaretypicallyexpectedtobeshorter,thisislikelyfueledbyother

influences.Itispossiblethatpastnewspapercoverageitselfgeneratesadditional

coverageasreportersaremoreassuredofliteracyaroundthesetopics,andassuch

canensurereadershipandinterestoffurthercoverage.ThecreationoftheMental

HealthCommissionofCanadaisalsoalikelyinfluenceonthecontinuedincreaseof

coverage.TheCommitteeemphasizedthatwhiletheSenateReportwoulddraw

61

attentiontomentalhealthissues,nationalfocusontheseissuescouldonlybe

maintainedthroughthecreationoftheMentalHealthCommission.Assuch,it

possiblethatboththecreationoftheMentalHealthCommissionandtheir

subsequentactivitieshavecontinuedtokeepmentalillnessandhomelessness

salienttopics.Similarly,itislikelythatpublicfocusandattentionbroughttoissues

concerningmentalillnessandhomelessnessaffordedothergroupsanopportunity

tocommentonthesetopicsinapublicforum.Thecontinuedincreaseincoverageis

suggestivethattheSenateReportappearstohaveplayedacatalyticrolein

increasingtheoverallfrequencyofreportingonmentalillnessandhomelessness.

Themes

Whilenostatisticallysignificantchangeswereobservedintheappearanceof

theidentifiedthemes,therewasanincreaseinthefrequencyofallthethemespost‐

SenateReport.Thistrendismeaningfulasitimpliesthattheagendasetting

potentialofnationalnewscoveragewithregardstothesethemesincreasedpost‐

SenateReport.Thisismorepronouncedinthepresentationofthethemesof

Recovery,StigmaandIntegrationofServices,whichshowedamoremarked

increase.Whilethereisnoclearindicationofwhytheincreaseinthecoverageof

somethemeswasmorepronouncedthanothers,thedifferencesarelikelyfueledby

otherensuingevents,reportsandfindings.Aswithoverallcoverageofmental

illnessandhomelessness,itispossiblethattheSenateReportactedasacatalystin

enablingtheseissuestobefurtherpubliclydiscussed.

62

ThispotentialcatalyticroleoftheSenateReportisfurthersupportedin

notingthepatternsofthereportingofthemesovertime.Fiveofthesixselected

themes(allexceptIntegrationofServices)showedapatterninwhichthenumberof

articlesmentioningthethemeincreasedintheyeartheSenateReportwasreleased,

declinedsomewhatinvaryingdegreesinthefollowingyear,andincreasedagainin

2008‐2009,inmostcasesabovethepreviouspeaknumberofarticlesin2006‐2007.

Incomparison,thetotalnumberofarticlesinthesamplesteadilyincreasedinthis

periodoftime,signifyingthatthispatternwasnotduetoageneraltrendinthe

numberofarticleswritten.Hadthecoverageofthemesbeeninfluencedbythe

SenateReportalone,itwouldbeexpectedthatafterapeakofcoverageintheyearin

whichthereportwasreleased,adeclineinthecoverageofthesethemeswouldbe

observedinthefollowingyears.However,theobservedtrendreinforcesthatother

influenceslikelycausedcoverageofthesethemestonotonlycontinuetobe

presentedinnewspaperreportsatanelevatedlevelcomparedtoprioryears,butto

furtherincreasein2008‐2009.

5.1.2 PotentialFramingEffects

Theanalysisofpotentialframingeffectswascenteredondifferencesin

articleapproach.Overthesixyeartimeperiodanalyzed,alargeportionof

newspapercoverageusedepisodicframing.Therewasnostatisticallysignificant

differenceinarticleapproachwhencomparingpreandpostSenateReport,andboth

episodicandthematicarticlesincreasedpost‐SenateReport.However,the

proportionofcoverageusingathematicapproachincreasedpost‐SenateReport,

63

andthemajorityofthematicarticleswereinthethreeyearsafterthereleaseofthe

SenateReport.Althoughepisodiccoverageshowedamarkedincreaseoverthe

studyperiod,thenumberofepisodicarticlesdecreasedinthelastyearofanalysis

despiteitbeingtheyearwithmostcoverageregardingmentalillnessand

homelessness.Overall,thoughthemajorityofarticlesremainepisodicallyframed,

articleapproachshowedanincreaseintheamountofthematicreportingsincethe

releaseoftheSenatereport.

BasedonIyengar’s(1991)theoryofframingeffects,thiswouldsuggestthat

duetotheepisodicframingofthemajorityofnewscoverage,responsibilityfor

addressingconcernsrelatedtomentalillnessandhomelessnesswouldlikelybe

attributedtoaffectedindividualsthemselves.However,theanalysisofthesolutions

presentedinthearticlesindicatedthatthismightnotbethecaseforallepisodic

articlesinthisstudy.Theresponsibilityforimplementingsolutionsinbothepisodic

andthematicarticleswasalmostentirelyattributedtogovernmentalagencies.This

maysupportCollinsetal,’s(2006)suggestionsthatIyengar’s(1991)theoryof

framingeffectsmayapplytospecificissuessuchascrimeandunemploymentbut

notuniversallytoallpublicpolicyissues.Itisalsopossiblethatthedissimilarityis

duetodifferencesstudydesign,asIyengar’sresearchfocusedonreceptionstudies

andimpliedattributionsofresponsibility,whiletheresearchinthisthesisfocuses

onlyonclearlypresentedsolutions.Itisalsoimportanttonotethatwhilethe

episodicarticlesthatidentifiedsolutionsshowedstructuralattributionsof

responsibility,theyareasmallproportionofallepisodicnewspapercoverage.

64

Theincreaseinthematicreportingislikelytohaveapositiveeffectonthe

potentialforchangingpublicattitudes.Thematicarticlesplaceindividualsand

eventsinthecontextofbroadersocialandstructuralsystemsandassuch,allow

readerstomoreeasilyidentifythatifaproblemistobesolved;thecontextmustbe

addressed(Dorfman&Wallack,2008).Furthermore,thematicarticleswere

significantlymorelikelytoofferclearsolutions.Definingclearsolutionstosocial

issueshasbeenidentifiedasanimportantfactorinadvocatingforchangethrough

themedia(Dorfman&Wallack,2008).Theincreaseinthematicarticlespost‐Senate

Reportraisedthepotentialofthenewsmediatoraiseattitudesofchangeand

advancetheunderstandingofthepublicregardingthebroadercontextofmental

illnessandhomelessness.

Additionally,thematicarticlesweresignificantlymorelikelytobewrittenby

anauthorwithfourormorearticlesinthesamplestudied.Thisimpliesthat

journalists,asopposedtopoliticiansorotherpublicfigureswhowouldauthoronly

onearticleinthesamplestudied,wrotethematicarticles.Italsosuggeststhatmost

thematicarticlesarewrittenbyjournalistswhohaveaninterestinand/or

knowledgeofmentalillness,homelessnessorboth,andareprolificintheir

reportingoftheseissues.

ArticlesPrimaryEvent/Story

Theprimaryevent/storyatthecentreofeachnewspaperarticlewas

explored,andtheresultsofthisanalysisofferedadditionaldetailsregardingthe

depictionofmentalillnessandhomelessnessinnationalnewscoverage.More

65

specifically,theanalysisshowedthatwhiletherewasadiverserangeof

events/storieswritteninrelationtomentalillnessandhomelessness,coverage

centredonviolencewasnotamongthemostfrequentlyreportedevent/story.

Additionally,therewasanincreaseinthecoverageofpersonalstoriespost‐Senate

Report,suggestiveofahigherpotentialforpublicsympathytowardsthehomeless

andmentallyill.

Forbothepisodicandthematiccoverage,themostfrequentlycoded

categorieswere‘Other’and‘No/unclear’.Thisindicatesthatalargeportionof

articlesregardlessoftheirapproach,werewrittenaboutanisolated

event/occurrenceanddidnotfocusonanyparticularissueregardingthis

population.Overthesix‐yeartimeperiodstudied,thenumberofarticlesregarding

thesetopicsshowedasimilarpatterntotheoverallincreaseintheamountofnews

coverage.Thismaybeindicativeofthenatureofnewsreporting,whichrequires

findingnewandunusualoccurrencestoreport(Bachrach,1990),resultinginwide

diversityofstoriesbeingreportedregardingmentalillnessandhomelessness.

Coverageofpersonalstorieshowever,showedadifferentpattern.The

numberofarticlescentredonpersonalstoriesincreasedPost‐SenateReportand

showedfurtherincreaseintheyearsthatfollowed.TheSenateReportitself

containedmanypersonalanecdotesandputahumanfaceonmentalillness.It

appearsthatsincethereleaseoftheSenateReport,thisisincreasinglyoccurringin

nationalnewsmediaaswell.Thisfindingisparticularlynoteworthyasresearch

suggeststhatpersonalstoriescanbemoreemotionallyengagingforthepublicand

havehigherpotentialtochangepublicattitudesthroughelicitingsympathy(Gross,

66

2008).Whileithasbeensuggestedthatepisodicreportscanminimizeattitudesof

changebyfocusingontheindividualratherthanonstructuralforces(Iyengar,1991;

Iyengar,1990a,Dorfman&Wallack,2008),researchalsoindicatesthatcompelling

storiescanleadtostrongfeelingsofsympathytowardsthesubjectofthestoryand

inspireaneagernesstochangethestructurethatcreatedthatindividual’ssituation

(Gross,2008).

Theothernoteworthypatternsarethoserelatedtostoriesofviolenceand

crime.Asdiscussedinpreviouschapters,newscoverageofmentalillnesshasbeen

showntoperpetuateanassociationbetweenmentalillnessandviolence(Philo,

McLaughlin,&Henderson,1996;Ward,1997;Allen&Nairn,1997;Nairn,1999;

Hazelton,1997).Thisstudyfoundthatrelativelyfewarticleswerewritten

regardingmentalillnessandhomelessnessthatcentredonviolenceorcrimeand

thisshowednomarkedchangeovertimeshowinglowcoverageofspecific

incidents.Incontrast,newsreportsthatcentredonviolenceagainstanindividual

whowashomelessand/orlivingwithmentalillnessshowedpeaksofhigher

coverageinparticularyears,suggestingaproliferationofarticlessurrounding

particularincidencesofviolence/crime.Thisdifferenceincoverageislikelydueto

thenationalscaleofthenewspapersanalyzedandthedifferenceinthetypesof

crimereported.Whilestoriesofviolence/crimebyanindividualidentifiedas

homeless/mentallyillweregenerallyaboutaggressivepanhandlingincidents,

storiesofviolence/crimeagainstanindividualidentifiedashomeless/mentallyill

wereconcernedmoreviciousattacks.Assuch,nationalnewscoverageofmental

illnessandhomelessnessdoesnotappeartoperpetuatetheassociationofthe

67

homelessand/orthoselivingwithmentalillnesswithviolencethroughincreased

coverageofviolentincidents,exceptpossiblyasvictimsofit.

Voice

Ingeneral,thearticlesquotedawiderangeofpeople,bothbeforeandafter

thereleaseoftheSenateReportandshowednomeaningfulchangeovertime.

Therewerenomajordifferencesobservedinwhowasquotedinthearticleswhen

comparingpreandpostSenateReport.Thegroupthatwasmostfrequentlyquoted

inbothtimeperiodswasindividualsidentifiedashomeless,mentallyill,orboth

homelessandmentallyill,signifyingthatthepopulationbeingdiscussedwasgiven

thegreatestvoice.Thegroupsthatsawthegreatestincreasesinthenumberof

articlesquotedwerethoseidentifiedashomeless/mentallyill/homelessand

mentallyill,healthprofessionals/researcher/expert,andrepresentativesfroman

organizationorprofessionalsociety.Theonlygroupthatshowedadecreasewas

Shelter/Housingrepresentatives.Thisdifferencehowever,wassmall.

5.2 Limitations

Althoughthesearchtermsusedtoselectthenewspaperarticlesforanalysis

providedarelativelycompletesetofarticlesregardingmentalillnessand

homelessnessinthechosennewspapers,thesearchtermsdonotaccountforevery

possiblearticlerelatedtomentalillnessandhomelessnessanditislikelythatsome

articlesweremissed.AsmentionedintheMethodssection,thetermmentalhealth

wasnotused,whichmayhaveleftsomerelevantarticlesoutofthesample.Names

68

ofspecificmentaldisorderswerenotusedinthesearchterms,theadditionofthese

mayhavegeneratedmorearticlesforanalysisaswell.

Thesearchtermsusedalsomisstheinclusionofarticlesrelatedtochildren.

Childrenarenotlegallyconsideredhomelessbut‘wardsofstate’.Whileitisnot

anticipatedthattherewouldbemanyarticleswrittenaboutissuesrelatedtomental

illnessandhomelessnessandchildren,thesearticlescouldbeparticularlyimportant

asthepublicisgenerallydrawntoandsympathizeswithstoriesaboutchildren.Itis

alsomorelikelythatthepublicwouldattributechallengesrelatedtochildrenon

structural/societaldeficitsratherthanpersonalcharacteristics.

Incodingthearticles,sixthemesfrom‘OutoftheShadowsatLast’were

identifiedandcoded.Thesethemeswereidentifiedbymyselfalone,andwereones

Ithoughtaddressedmentalillnessandhomelessnessmostdirectlyinthereportand

highlightedtheprimaryprinciplesitcontained.Thereareotherthemesand

vulnerablegroupsstronglyemphasizedintheSenateReportthatwerenotincluded

inthecodesheetastheydidnotdirectlyaddresshomelessness.Thesetopics

howevercouldbeapplicableandtheidentifiedvulnerablegroupscouldoverlap

withthehomelessaswell.Examplesofthisincludementalillnessinseniors,

children/youthandwomen.Assuch,thesixthemesthatwereidentifiedand

analyzedarenotacompletereflectionoftheissuesraisedanddiscussedregarding

thispopulationintheSenateReport.

Aswithothercodingresearch,thecodingprocessissubjectiveinnature.To

monitorandminimizethesubjectivityofcoding,asecondcoderwasusedona

69

selectionofarticles,andhadtheintercoderreliabilityshownlowcorrelation,the

codesheetwouldhavebeenrevisedtoimprovereliability.Duringthedevelopment

ofthecodesheetandcodingprocess,categoriesweredevelopedforcodingto

increaseobjectivity,asthemajorityofthecodingwasconductedbyasinglecoder

(exceptforthesampleselectedforintercoderreliability).Whilethesecategories

aidedinidentifyingcodesobjectively,whatwaslostwastheabilitytocodecontent

thatwasimpliedandnotclearlystated.Thisisparticularlyrelevantinthecodingof

solutionsandtheresponsibilityforimplementingsolutions.Whilesomearticles

impliedwaysinwhichhomelessnesscouldbeovercome,theydidnotstatetheseas

aclearsolution,andassuchcouldnotbecoded.Forexample,inthedescriptionof

anissue,numerousarticleswouldassignblameonaparticularindividualorgroup

forthecreationoforlackofactionregardingthatparticularissue.Whilethis

pointingofblamestronglysuggestedwheretheresponsibilityforthesolutionlay,it

wasnotpossibletocodethis,astheauthordidnotpresentasolutiontotheissue

directly.

Withregardstotheanalysisofmediaeffects,whilethisanalysisexplores

agendasettingandframingeffects,findingscanonlyhaveimplicationsregarding

thepotentialeffectsofthenewspapercoverage;itisnotpossibletoknowtheactual

effectsonpublicperception.Itispossibletospeculatethattheremaybeeffects

basedonpreviousresearchconductedonpublicopinion.However,without

conductingreceptionstudiesexaminingpublicawarenessandopinioninthis

particularstudy,dataanalysiscanonlyraiseinferencesaboutpotentialeffects.

70

Whileitisnotpossibletosaythatanydifferencesobservedbetweenthe

threeyearspriortoandthethreeyearsfollowingthereleaseoftheSenateReport

werecauseddirectlybythereport,itislikelytobeaconsiderablefactorfora

numberofreasons.Asmentionedpreviously,thiswasthefirstnationalstudyinto

thementalhealthsystem,mentalillnessandaddictionsinCanada(MHCC,n.d.).As

such,theSenateReportwasthefirstnational‐levelreportaddressingmentalillness

tobepublishedinCanada.Thereportcontains118recommendationsandassuch,

isareportthatholdsimplicationsforactionthatmaybecarriedoutonvarious

levels.Asaresult,itislikelytohavebeenconsideredameaningfulandanticipated

documentbyindividualsandgroupsinthecommunityandhealthrelatedfields.It

wasalsotheonlynationalreporttoaddressissuesrelatedtomentalillnessor

homelessnesstobepublishedinthatparticularyear.

5.3 FutureDirections

Whiletheresultsofthisanalysisaddtopreviousresearchregarding

potentialmediaeffectsofnewsreportingonhealth‐relatedtopics,theyalsoraise

numerousquestionsthatcanbeexploredinfurtherresearch.Aswithmost

previousresearchconductedregardingmediaportrayalsofmentalillness,this

studywasconductedontheassumptionthatmediaportrayaldoesinfluencepublic

opinion.Whilethereisresearchtosupportthatthisisindeedthecase(Franciset

al.,2001;Wahl,1992;Dietrichetal.,2006),therehavebeencomparativelyfew

studiesthatexaminethisspecificallyandassesscommunityattitudestowards

mentalillness(Francisetal.,2001).Amorecomplexanalysisexaminingchangesin

71

publicawarenessandattitudestowardsmentalillnessandhomelessnesswouldbe

usefulinestablishingactualeffectsofnewspapercoverage.Thiswouldenablean

examinationoftheimpliedattributionsofresponsibilityraisedinthenews

coverage,providingamoreaccuratestudyofthelinkbetweenarticleapproachand

itsassociatedsolutions.Thiswouldalsoallowamoreaccuratecomparisonwith

Iyengar’s(1991)research,whichinturnwouldleadtostrongerconclusionson

whetherhistheoryofframingeffectswassupportedbythisresearch.

Anexpansionofanumberofmethodologicallimitationsinthisstudywould

giveamorecompletepictureofthepotentialchangesinmediaportrayalofmental

illnessandhomelessness.ThisstudyfocusedonCanada’stwonationalnewspapers.

Examiningothernewspaperswouldprovideinsightintohowreportingmaydiffer

inregional,localandelectronicnewspapers.Forexample,previousresearchhas

foundthattabloidnewspaperstendtoportraymentalillnessmorenegatively

(Francisetal.,2001),andindividualscitingelectronicresourcesholdlesstolerant

attitudestowardsmentalillness(Granelloetal.,1999).Itwouldalsobeusefulto

includeananalysisofcommentsandletterspublishedinthenewspapersasthis

mayprovideinsightintopublicinterestinthesetopics.

ThisstudyfocusesononlysixthemesfromtheSenateReport,butas

previouslydiscussed,otherthemesandrecommendationsarepresentwithinthe

report.Ananalysisincludingmoreofthesethemesandrecommendationswould

furtheraddinsightintowhethertheSenateReportinfluencedthecontentof

newspaperarticlessinceitsrelease.Additionally,researchoveranexpandedtime

periodwouldbeusefulinprovidingapictureofthedepictionofmentalillnessand

72

homelessnessintheyearsbeforeanyresearchorpublicannouncementrelatedto

theSenateReporthadbegun.

Itwouldbeinterestingtoseehowthereportingofotherissuesinfieldof

healthmayhavechangedinthesametimeperiod.Thiswouldhelpestablishhow

theincreaseinarticlesregardingmentalillnessandhomelessnesscomparedto

otherhealthissues,andwhethertheyshowsimilarorconflictingtrends.

Additionally,thiswouldfurthersupporttheunderstandingofhowmuchtheSenate

Reportinfluencedtheincreaseinreportingregardingmentalillnessand

homelessness.

Whilefurtherresearchonothernewsmediachannelswouldbeuseful,itis

alsoimportanttonotetheincreasinguseofsocialmedia,particularlybyyouth,and

thesubsequentneedtofurtherstudytheroleitplaysintheacquisitionofhealth

andsocialinformation.Researchhasbeenbeguntoexaminetherolethatsocial

mediamayplayinhealth,howeveramorecompleteanalysisofthepotentialagenda

settingeffectsofsocialmediawouldgreatlyaddtotheexistingresearchintothese

effectsinothermediaforms.

Thisstudyraisesawarenessontheimportanceofresearchinotheravenues

ofmediastudiesthatwouldaidinthediscussionoftheresults.Anexplorationof

thesourcesusedbyreportersandtheinfluencesthesesourceshaveonhow

newspaperarticlesarewrittenwouldassistinunderstandingwhythedifferences

betweenarticlesandtheirapproacharepresent.Understandingthemotivation

73

behindjournalisticwritingwillalsoaidtheunderstandingofhowpublichealth

professionalscanbestusenewsmediaasaneffectivetool.

Onabroaderlevel,themostimportantandusefulresearchthatthisstudy

hashighlightedaneedforisfurtherstudyintohowthemediaaffectspublicopinion

andpolicy.Understandingwhatthemostinfluentialmediastreamsareforvarious

issues,andwhichformsofmediahavegreaterabilitytoinform,topersuadeandto

settheagenda,willallowformediachannelstobeusedmoreeffectivelywithin

publichealth.

74

6: CONCLUSIONS

SincethereleaseoftheSenateReport‘OutoftheShadowsatLast’,therehas

beenasignificantandlastingincreaseintheagendasettingpotentialofCanadian

Nationalnewspapercoverageregardingmentalillnessandhomelessness.

Furthermore,publicawarenessofspecificconceptsidentifiedintheSenateReport

hasincreasedsinceitsrelease.Thedepictionofmentalillnessandhomelessness

hasshownachangeintheincreasedthematicframingofcoverage,raisingthe

potentialforincreasedpublicunderstandingofthecontextuallandscape

surroundingthispopulation.

Theanalysesshowthatthereweresignificantlymorearticlesregarding

mentalillnessandhomelessnessafterthereleaseoftheSenateReport.The

increaseincoveragecontinuedintheyearsfollowingthepublicationoftheSenate

Reportsuggestingthatthereportplayedanimportantcatalyticroleinincreasing

theoverallfrequencyofreportingregardingmentalillnessandhomelessnessin

nationalnewscoverage.Thishaslikelyallowedawiderdiversityofsourcesto

continuethediscourseregardingthistopiconapublicandnationallevel.

Furthermore,theagendasettingpotentialofnationalnewscoveragewithregardsto

therelevantthemeshasincreased.Thisshowsaheightenedpotentialforan

increaseinpublicawarenessoftheprimaryprinciplesandissuesraisedinthe

SenateReport.

75

Thenationalnewsreportsofmentalillnessandhomelessnesscovereda

widevarietyofeventsandstories.Amongthese,therewasanincreaseintheyears

followingtheSenateReport’spublicationinthenumberofarticlesdescribing

personalstories.Thisissuggestiveofanincreaseinthepotentialforheightened

sympatheticattitudesinthepublictowardsindividualswhoarehomelessand/or

livingwithmentalillness.Itwasalsoobservedthatstoriesthatofferedsolutions

regardingmentalillnessandhomelessnessplacedtheresponsibilityfor

implementingthesesolutionslargelyongovernmentalstructures.Thebenefitsof

thesesolutionsweregenerallyeitherunspecifiedorsuggestedtoonlyimpactthe

individualsaffected.Contrarytopreviousresearch,thisanalysisfurtherfoundthat

thenewspapercoveragedidnotfrequentlyportraymentalillnessasbeing

associatedwithviolence.

Withregardstotheapproachofthearticles,articleswereprimarilyepisodic

intheirframing,withthemajorityofthearticlespayinglittleattentiontothe

contextandrootcausesofmentalillnessandhomelessness.Thisshowedno

statisticallysignificantchangeoverthetimeperiodstudied.However,theevidence

supportsthattheamountofthematiccoverageofmentalillnessandhomelessness

showedanincreaseintheyearsfollowingthereleaseoftheSenateReport,andas

such,thepotentialtoinfluencepublicattitudesregardingthecausesandpossible

solutionsregardingthispopulationincreased.

Ultimately,theresultsofthisanalysisleadtoarejectionofthenull

hypothesisthatstatestherewouldbenochangeinthepotentialforalteringpublic

attitudestowardsmentalillnessandhomelessnessthroughnationalnewspaper

76

coveragesincethereleaseoftheSenateReport.Thisresearchsupportsthatthe

SenateReportverylikelyincreasedthepotentialforachangeinpublicattitudes

relatedtomentalillnessandhomelessaspresentedinnationalnewscoverage.

SincethereleaseoftheSenateReport,nationalnewscoverageregardingmental

illnessandhomelessnessincreasinglymadethepublicawareoftheimportanceof

boththinkingabouttheseissuesandthethemesandprinciplesthatrelatetothem.

Theincreaseinthematiccoveragecoupledwithanincreaseinreportingofpersonal

storiesamongtheepisodicarticleshaveraisedthepotentialforthepublictobe

sympatheticoftheimpactofmentalillnessandhomelessnessonindividuallives

andawareofthebroadercontextualframeworksthatsurroundthem.

ThisthesissuggeststhatreportsliketheSenateReporthavethepotentialto

influencethepublicagendaasadvancedbythemedia.Italsohighlightstheneedfor

increasedresearchontherelationbetweenpolicy,mediaandpublicopinion.More

specifically,itsuggeststhatpublichealthprofessionalsandpolicymakersneedto

furtherexaminetheeffectivenessofmediachannels,andthefactorsthatinfluence

thiseffectiveness.Examiningtheinteractionbetweenpolicydevelopmentand

mediacoverageisofparticularimportanceforpoliciesinwhichactionsandresults

areinfluencedbypublicopinion.Whilethereisstillmuchtolearnregardingmedia

effectsonpublicattitudes,especiallyinrelationtospecifichealthissues,thishas

becomeincreasinglyimportanttoexploreashealthpoliciesencouragecommunity‐

basedcareandrelyonpublicsupport.

77

APPENDICES

78

AppendixA

Section Information Response

Coding details Date of coding dd/mm/yyyy

Coder name Text

Newspaper name Categorical National Post, Globe & Mail

Date dd/mm/yyyy

Type of article Categorical news, edit/column

ID number Numeric 000-414

Article Characteristics

Headline Text

Word count Numeric

Placement Categorical paper front pg, section front page, other Article prominence

Artwork/Photograph Categorical yes, no

Article Relevance Depth of discussion Categorical

keyword only, up to couple sentences, up to one paragraph, ¼ article, ½ article, ¾ article,

majority/entire article

Recovery Categorical mentioned, not mentioned

Public Safety Categorical mentioned, not mentioned

Inequity Categorical mentioned, not mentioned

Affordability Categorical mentioned, not mentioned

Integration of Services Categorical mentioned, not mentioned

Articles Themes

Stigma Categorical mentioned, not mentioned

79

Is the article primarily episodic or thematic?

Categorical episodic, thematic

Is it in response to a particular event or a story?

Categorical personal story, political event/election, new

building/zoning allocation/opening, violence/crime against someone with MI or H,

violence/crime by someone with MI or H, police action, release of a report, other (to be

sorted into categories post-coding)

Is there a clearly stated solution/s?

Categorical Yes, No

Who is responsible for the solution?

Categorical Collective/Governmental, Individuals affected,

Unclear/not explicit

Article Approach

Who does the solution/s benefit?

Categorical Community at large, Individuals affected, Both,

Unclear/not explicit

Voice (direct quotes) Who speaks?

Categorical Government representative, Politician,

Researcher/expert, Health professional, Advocate, Person with mental illness, Person

who is homeless, Person who is both homeless and has mental illness, Business

person, police rep/officer, judge, other (to be sorted into categories post-coding)

80

AppendixB

#!/usr/bin/perl‐w

#Input:Atextfile(CanadianNewsstandDatabasesearchresultemailedvia

# ProQuest)containingarticleswhichareidentifiedbyaheadingin

# theform"Documentxofxx".

#Output:Atabdelimitedfilecontainingasummaryofdataabouteacharticle,

# asoutlinedbelow.

#

#Readininputfile.Parsethefileintoarticlesusingthe"Documentxofxx"

#headingsasdelimiters,thenfurtherparseeacharticle,capturingthe

#following:NewspaperName,Date,Section,DocumentType,ProquestDocumentID,

#ISSN,Headline,Author,Wordcount,andPageNumber.Printtheresultsina

#tab‐delimitedfile.

#

#Author:JoubinMoshrefzadeh(joubin_at_gmail_dot_com)

#Copyright(C)2009JoubinMoshrefzadeh

#

useSwitch;

my$inpFile=shift;

my$outpFile=shift;

die"Usage:".__FILE__."<inputFile.txt><outputFile.txt>\n"

unless(defined($inpFile)&&defined($outpFile));

open(INFILE,"<$inpFile")ordie"can'topen$inpFile:$!";

open(OUTFILE,">$outpFile")ordie"can'topen$outpFile:$!";

#keepstrackofwhichstanza/articlewe'reon,andcorrespondinglytheelement

#inourhashwhichwillcontainthisarticle'sinfo

my$stanza=0;

81

my$firstParagraph=""; #tocontainfirstparagraphofeachstanza

my$line=0; #keeptrackoflinenumberwithinastanza

#aflagfordeterminingwhenwe'refinishedreadingthefirstparagraph

#ofaparticularstanza

my$done=0;

#Forstoringalltheimportantbitsfromeacharticle,

#oncethey'vebeenparsedout;mightaswellinitializeit

#forthefirststanza/article

my%searchResults=(

$stanza=>{

TEXT =>'', #thecompletetextofastanza/article

AUTHORS =>'', #thearticle'sauthors

DOCTYPE =>'', #thearticle'sdoctype

DATELINE =>'', #thearticle'sdateline

SECTION =>'', #thearticle'ssection

PUBLICATION =>'', #thearticle'spublication

SOURCETYPE =>'', #thearticle'ssourcetype

ISSN =>'', #thearticle'sISSN

PROQUESTDOCID =>'', #thearticle'sProquestDocID

WORDCOUNT =>'', #thearticle'swordcount

URL =>'', #thearticle'sURL

PAPERNAME =>'', #thearticle'snewspapername

ARTICLEDATE =>'', #thearticle'sdate

PAGENUMBER =>'', #thearticle'spagenumber

HEADLINE =>'', #thearticle'sheadline

}

);

#beginprocessingfile

while(<INFILE>){

#Eachtime"Documentxofy"isseen,wehaveanewarticle/stanza.

82

#Doinitializationofvariablesforthisstanza.

if($_=~/Document[0‐9]{1,2}of[0‐9]{1,2}/){

$stanza++; #incrementthestanzacounter

$done=0; #resetthisflag

$firstParagraph="";#resetthisvariable

$line=0; #resetthelinecounter

#initializethehashforstoringthedetailsofthisstanza

$searchResults{$stanza}={

ARTICLETITLE =>'',

TEXT =>'',

AUTHORS =>'',

DOCTYPE =>'',

DATELINE =>'',

SECTION =>'',

PUBLICATION =>'',

SOURCETYPE =>'',

ISSN =>'',

PROQUESTDOCID =>'',

WORDCOUNT =>'',

URL =>'',

}

}

#appendeachlinebeingreadtothishashelement(soitwillcontain

#alltextforthisstanza)

$searchResults{$stanza}{TEXT}.=$_;

#Extractfirstparagraph(terminatedbythelinecontaining

#"AbstractSummary")andparseoutthearticleheadline

if($_=~/Abstract\(Summary\)/){

#setthisflagsoweknowforremainderofthisstanza

#thatwe'repastthefirstparagraph

83

$done=1;

#Splitfirstparagraphusing"."asdelimiter

#andassigningresultstoanarray.

my@elements=split/\./,$firstParagraph;

#Dependingonthesizeoftheresultingarray,

#theHeadlinewillbeinelementnumber...

my$numElements+=@elements;

switch($numElements){

case7 {

$searchResults{$stanza}{HEADLINE}=$elements[0];

}

case8 {

$searchResults{$stanza}{HEADLINE}=$elements[2];

}

case9 {

$searchResults{$stanza}{HEADLINE}=$elements[3];

}

else {

#can'tdeterminetheheadline

$searchResults{$stanza}{HEADLINE}=

"COULDN'TGETHEADLINEINFO";

}

}

}

if($done==0){

#We'restillreadingthefirstparagraphso...

$line++;

#Weonlywantthelinesafterthe"Documentxofxx"line...

if(($_!~/Document[0‐9]{1,2}of[0‐9]{1,3}/)&&($line>2)){

84

#cleanupandappendlinestogethertoformonestring

$firstParagraph.=$_;

chomp$firstParagraph;

$firstParagraph.="";

}

}

#Nowwe'rebasicallyfocusingonthelastfewlinesofeachstanza

#whichcontainaheading,thenacolon,thensomevalue.Based

#ontheheading,setthevalueofthecorrespondinghashelementfor

#thisstanza.

chomp;

my($kee,$val)=split/:/,$_,2;

if($kee){

switch($kee){

case/Author/{

$searchResults{$stanza}{AUTHORS}=$val;

}

case/Documenttypes/{

$searchResults{$stanza}{DOCTYPE}=$val;

}

case/Dateline/{

$searchResults{$stanza}{DATELINE}=$val;

}

case/Section/{

$searchResults{$stanza}{SECTION}=$val;

}

case/Publicationtitle/{

$searchResults{$stanza}{PUBLICATION}=$val;

}

case/Sourcetype/{

$searchResults{$stanza}{SOURCETYPE}=$val;

85

}

case/ISSN/{

$searchResults{$stanza}{ISSN}=$val;

}

case/ProQuestdocumentID/{

$searchResults{$stanza}{PROQUESTDOCID}=$val;

}

case/DocumentURL/{

$searchResults{$stanza}{URL}=$val;

}

}

}

if($_=~/(TextWordCount)(\d+)/){

$searchResults{$stanza}{WORDCOUNT}=$2;

}

}

#Nowthatwe'refinishedprocessingtheinputfile...

#Printthecolumnheadingsfortheoutputfileintab‐delimitedformat

printOUTFILE"Inputfile:$inpFile\t\n";

printOUTFILE"NewspaperName\t";

printOUTFILE"Date\t";

printOUTFILE"Section\t";

printOUTFILE"DocumentType\t";

printOUTFILE"ProQuestDocumentID\t";

printOUTFILE"ISSN\t";

printOUTFILE"Headline\t";

printOUTFILE"Author\t";

printOUTFILE"WordCount\t";

printOUTFILE"PageNumber\t";

printOUTFILE"\t";

printOUTFILE"URL\t";

86

#Printthedetailsofeachstanza/articleintab‐delimitedformat

for($i=0;$i<=$stanza;$i++){

my@parts=split/\./,$searchResults{$i}{PUBLICATION};

my$size=scalar(@parts);

$parts[2]=~s/://;

printOUTFILE"$parts[0].$parts[1].\t";

printOUTFILE"$parts[2]\t";

printOUTFILE"$searchResults{$i}{SECTION}\t";

printOUTFILE"$searchResults{$i}{DOCTYPE}\t";

printOUTFILE"$searchResults{$i}{PROQUESTDOCID}\t";

printOUTFILE"$searchResults{$i}{ISSN}\t";

printOUTFILE"$searchResults{$i}{HEADLINE}\t";

printOUTFILE"$searchResults{$i}{AUTHORS}\t";

printOUTFILE"$searchResults{$i}{WORDCOUNT}\t";

printOUTFILE"$parts[3].";

for($j=4;$j<$size;$j++){

printOUTFILE"$parts[$j]";

printOUTFILE"."unless($j==$size‐1);

}

printOUTFILE"\t\t";

printOUTFILE"$searchResults{$i}{URL}\n";

}

closeINFILE;

closeOUTFILE;

87

AppendixC

88

89

90

AppendixD

KappastatisticsSummaryItem Kappa

Score

DepthofDiscussion 0.731

Stigma 1.00

Recovery 0.747

PublicSafety 0.859

Inequity 0.790

AffordabilityofHousing 0.797

IntegrationofServices 0.863

Episodic/Thematicapproach 0.806

Solution(yes/no) 0.853

91

AppendixE

SummaryofP‐Valuesforeachtheme(comparingPreandPost‐SenateReport)Theme P­Value

Recovery 0.0631

Stigma 0.0688

Inequity 0.6087

AffordabilityofHousing 0.7618

IntegrationofServices 0.1156

PublicSafety 0.9776

92

REFERENCES

Adams,C.E.,Pantelis,C.,Duke,P.J.,&Barnes,T.R.E.(1996).Psycopathology,socialandcognitivefunctioninginahostelforhomelesswomen.BritishJournalofPsychiatry,168,82–86

Allen,R.,&Nairn,R.G.(1997).Mediadepictionofmentalillness‐ananalysisoftheuseofdangerousness.Australian&NewZealandJournalofPsychiatry,31(3),375‐81.

Bachrach,L.L.(1990).TheMediaandHomelessMentallyIllPersons.HospitalandCommunityPsychiatry,41(9),963‐964.

Bachrach,L.L.(1992).Whatweknowabouthomelessnessamongmentallyillpersons:Ananalyticalreviewandcommentary.InH.R.Lamb,L.L.Bachrach,&F.I.Kass(Eds.),Treatingthehomelessmentallyill:AtaskforcereportoftheAmericanPsychiatricAssociation(pp.13–40).Washington,D.C.:AmericanPsychiatricAssociation.

Benkert,O.,Graf‐Morgenstern,M.,Hillert,A.,Sandmann,J.,Ehmig,S.C.,Weissbecker,H.,Kleppingeer,H.M.,&Sobota,K.(1997).Publicopiniononpsychotropicdrugs:ananalysisofthefactorsinfluencingacceptanceorrejection.JournalofNervous&MentalDisease,185(3),151‐158.

Brosius,H.B.&Kepplinger,H.M.(1990).Theagenda‐settingfunctionoftelevisionnews:Staticanddynamicviews.CommunicationResearch,17(2)183‐211.

CanadianInstituteforHealthInformation(CIHI)(2007).ImprovingtheHealthofCanadians:MentalHealthandHomelessness(Ottawa:CIHI).

CentreforMentalHealthServices(CMHS)(2003).Homelessness­ProvisionofMentalHealthandSubstanceAbuseServices.Maryland:UnitedStatesDepartmentofHealthandHumanServices.RetrievedJanuary1,2009fromhttp://mentalhealth.samhsa.gov/publications/allpubs/homelessness/

Cohen,B.C.(1963).Thepressandforeignpolicy.Princeton:PrincetonUniv.Press

Collins,P.A.,Abelson,J.,Pyman,H.&Lavis,J.N.(2006).Areweexpectingtoomuchfromprintmedia?AnAnalysisofnewspapercoverageofthe2002Canadianhealthcarereformdebate.SocialScienceandMedicine,63,89‐102.

93

Cook,F.L.,Tyler,T.R,Goetz,E.G.,Gordon,M.T.,Protess,D.,Leff,D.R.&Molotch,H.L.(1983).MediaandAgendaSetting:EffectsonthePublic,InterestGroupLeaders,PolicyMakers,andPolicy.PublicOpinionQuarterly,47,16‐35.

Culhane,D.P.,Metraux,S.,&Hadley,T.(2002).PublicServiceReductionsAssociatedwithPlacementofHomelessPersonswithSevereMentalIllnessinSupportiveHousing.HousingPolicyDebate13(1)107‐163.

Davis,K.E.,&O’Neill,S.J.(2005).Specialsectiononrelapseprevention:Afocusgroupanalysisofrelapsepreventionstrategiesforpersonswithsubstanceuseandmentaldisorders.PsychiatricServices(56),1288‐1291.

Day,D.M.,&Page,S.(1986).PortrayalofmentalillnessinCanadiannewspapers.CanadianJournalofPsychiatry,31(9),813‐17.

Dear,M.&Gleeson,M.(1991).CommunityAttitudesTowardtheHomeless.UrbanGeography,12(2),155‐76.

Dietrich,S.Heider,D.,Matschinger,H.&Angermeyer,M.C.(2006).Influenceofnewspaperreportingonadolescents’attitudestowardpeoplewithmentalillness.SocialPsychiatryandPsychiatricEpidemiology,(41),318‐322.

Drake,R,McHugo,G.,Bebout,R.,&Becker,D.(1999).Arandomizedclinicaltrialofsupportedemploymentforinner‐citypatientswithseverementaldisorders.ArchivesofGeneralPsychiatry,56,627‐633.

Dorfman,L.(2003).StudyingtheNewsonPublicHealth:HowContentAnalysisSupportsMediaAdvocacy.AmericanJournalofHealthBehavior,27(S3),S217‐S226.

Dorfman,L.&Wallack,L.(2008).Provokingthough,changingtalk:puttingitintoPractice.Youcangettherefromhere.PortlandStateUniversity.RetrievedJanuary1,2010fromwww.bmsg.org/documents/TalkingInequality‐1.pdf.

Dorfman,L.,Woodruff,K.,Chavez,V.&Wallack,L.(1997).YouthandViolenceonLocalTelevisionNewsinCalifornia.AmericanJournalofPublicHealth,87(8),1311‐1316.

Echenberg,H.&Jensen,H.(2008).DefiningandEnumeratingHomelessnessinCanada.Ottawa:LibraryofParliament.

Entman,R.M.(1993).Framing:TowardClarificationofaFracturedParadigm.JournalofCommunication.(43)4,51‐58.

FrancisC,PirkisJ,DuntD,BloodRW(2001)Mentalhealthandillnessinthemedia:Areviewoftheliterature.CommonwealthDepartmentofHealthandAgeingCanberra.

Folsom,D.&Jeste,D.(2002).Schizophreniainhomelesspersons:Asystematicreviewoftheliterature.ActaPsychiatricaScandinavia,105,404‐413.

94

Funkhouser,G.R.(1973).Theissuesofthesixties:Anexploratorystudyinthedynamicsofpublicopinion.PublicOpinionQuarterly,37(1),62–75.

GelbergL,LinnLS(1988).SocialandPhysicalHealthofHomelessAdultsPreviouslyTreatedforMentalHealthProblems.HospitalandCommunityPsychiatry(39),510‐516.

Glynn,C.J.,Herbst,S.,O’Keefe,G.J.,&Shapiro,R.Y.(1999).Communicatingwiththepublic.InPublicopinion(pp.381–414).Boulder,CO:WestviewPress.

Golden,A.,Currie,W.,Greaves,E.,&Latimer,J.(1999).Takingresponsibilityforhomelessness:AnactionplanforToronto.ReportoftheMayor’sHomelessnessActionTaskForce:Toronto,ON.

Granello,D.,Pauley,P.,&Carmichael,A.(1999).Relationshipofthemediatoattitudestowardpeoplewithmentalillness.JournalofHumanisticCounselingEducation&Development,38(2),98‐103.

Gross,K(2008).FramingPersuasiveAppeals:EpisodicandThematicFraming,EmotionalResponse,andPolicyOpinion.PoliticalPsychology,29(2),169‐192.

Hazelton,M.(1997).Reportingmentalhealth:adiscourseanalysisofmentalhealthrelatednewsintwoAustraliannewspapers.Australian&NewZealandJournalofMentalHealthNursing,6(2),73‐89.

Homan,S.M.,Flick,L.H.,Heaton,T.M.,&Mayer,J.P.(1993).Reachingbeyondcrisismanagement:Designandimplementationofextendedshelter‐basedservicesforchemicallydependenthomelesswomenandtheirchildren:St.Louis.AlcoholismTreatmentQuarterly,10,101–112.

Howlett,M.(2008).OneYearLater–theMentalHealthCommissionofCanada.ProceedingsoftheAnnualProvincialMentalHealthandAddictionsConferenceforNewBrunswick,Miramichi,N.B.

Hwang,S.W.(2001).HomelessnessandHealth.CanadianMedicalAssociationJournal(164)2,229–233.

Iyengar,S.(1990a).FramingResponsibilityforPoliticalIssues:TheCaseofPoverty.PoliticalBehavoir,12(1),19‐40.

Iyengar,S.(1990b).Theaccessibilitybiasinpolitics:televisionnewsandpublicopinion.InternationalJournalofPublicOpinionResearch,2(1),1‐15.

Iyengar,S.(1991).Isanyoneresponsible?Howtelevisionframespoliticalissues.Chicago,Ill:UniversityofChicagoPress.

Iyengar,S.(1996).FramingResponsibilityforPoliticalIssues.AnnalsoftheAmericanAcademyofPoliticalandSocialScience,546,59‐70.

95

Iyengar,S.(1997).Overview.InIyengar,S.&Reeves,R.(Eds.),DotheMediaGovern?Politicians,VotersandReportersinAmerica(pp.211–216).ThousandOaks,CA:SagePublications.

Iyengar,S.,&Simon,A.(1993).NewscoverageoftheGulfcrisisandpublicopinion:Asurveyofeffects.CommunicationResearch,20(3),365–383.

Jacobs,F.,Little,P.M.D.,&Almeida,C.(1993).Supportingfamilylife:Asurveyofhomelessshelters.JournalofSocialDistressandtheHomeless,2,269–288.

Kirby,M.(2008).MentalhealthinCanada:outoftheshadowsforever.CanadianMedicalAssociateJournal,178(10),1320‐1322.

Krieger,J.&Higgins,D.L.(2002).Housingandhealth:Timeagainforpublichealthaction.AmericanJournalofPublicHealth,92(5),758‐768.

Laird,G.(2007).Shelter–HomelessnessinaGrowthEconomy:Canada's21stCenturyParadox.Alberta:SheldonChumirFoundationforEthicsinLeadership.RetrievedJanuary1,2009,fromww.chumirethicsfoundation.ca/files/pdf/SHELTER.pdf

Lee,B.A.,Link,B.G&Toro,P.A.(1991).Imagesofthehomeless:Publicviewsandmediamessages.HousingandPolicyDebate,2(3),649‐682.

Levine,I.S.,Rog,D.J(1990).MentalHealthServicesforHomelessMentallyIllPersons:FederalInitiativesandCurrentServiceTrends.AmericanPsychologist,(45)8,963‐968.

Lopez,L.R.(1991).Adolescents'attitudestowardmentalillnessandperceivedsourcesoftheirattitudes:anexaminationofpilotdata.ArchivesofPsychiatricNursing,5(5),271‐80.

Martison,B.E.,&Hindman,D.B.(2005).Buildingahealthpromotionagendainlocalnewspapers.HealthEducationResearch,20,51‐60.

Matas,M.,el‐Guebaly,N.,Harper,D.,Green,M.,&Peterkin,A.(1986).Mentalillnessandthemedia:partII.Contentanalysisofpresscoverageofmentalhealthtopics.CanadianJournalofPsychiatry,31(5),431‐33.

McCombs,M.E.&Shaw,D.L.(1972).TheAgenda‐SettingFunctionofMassMedia.PoliticalOpinionQuarterly.36:2,176‐187.

McCombs,M.,Shaw,D.L.&Weaver,D(1997).CommunicationandDemocracy:ExploringtheIntellectualFrontiersinAgenda‐SettingTheory.Mahway,NJ:Erlbaum.

McManus,J.&Dorfman,L.(2003).YouthViolenceStoriesFocusonEvents,NotCauses.NewspaperResearchJournal,(23)4,6‐12.

96

MentalHealthCommissionofCanada(MHCC)(n.d.).AbouttheCommission.RetrievedFebruary1,2009fromhttp://www.mentalhealthcommission.ca/English/Pages/Background.aspx.

Metraux,S.,Marcus,S.C.,Culhane,D.P.(2003).TheNewYork‐NewYorkHousingInitiativeandUseofPublicSheltersbyPersonswithSevereMentalIllness.PsychiatricServices,(54)1,67‐71.

Mitchell,L.N.(2007).Theirimages,ourselves:Canadianprintmedia’sconstructionoffeminismsurroundingthecutstotheStatusofWomenCanada.(MastersThesis,QueensUniversity,2007)RetrievedJanuary1,2009fromQspace.http://hdl.handle.net/1974/881

Min,E.(Ed)(1999).Readingthehomeless:themedia'simageofhomelessculture.Westport:Praeger.

Minkoff,K.,&Drake,R.E.(1992).Homelessnessanddualdiagnosis.InH.R.Lamb,L.L.Bachrach,&F.I.Kass(Eds.),Treatingthehomelessmentallyill:AtaskforcereportoftheAmericanPsychiatricAssociation(pp.221–248).Washington,D.C.:AmericanPsychiatricAssociation.

Mojtabai,R.(2005).PerceivedReasonsforLossofHousingandContinuedHomelessnessamongHomelessPersonswithMentalIllness.PsychiatricServices,(56)2,172‐178.

Mosher‐Ashley,P.,Henrikson,N.,&French,E.(2000).MeetingtheneedsofthementallyillhomelessinMassachusetts‐basedemergencyshelters.JournalofSocialDistressandtheHomeless,9,1‐17.

NADBANK.(2007).2007ReadershipHighlights.RetrievedFebruary10,2009,fromhttp://www.nadbank.com/en/study/readership.

Nairn,R.(1999).Doestheuseofpsychiatristsassourcesofinformationimprovemediadepictionsofmentalillness?Apilotstudy.Australian&NewZealandJournalofPsychiatry,33(4),583‐89.

North,C.S.,Smith,E.M.,Pollio,D.E.,&Spitznagel,E.L.(1996).Arethementallyillhomelessadistincthomelesssubgroup?AnnalsofClinicalPsychiatry,8,117–128.

OfficeoftheAuditorGeneralofBritishColumbia(OAGBC).(2009).Report16:Homelessness:ClearFocusNeeded.BritishColumbia:AuditorGeneralofBritishColumbia.

Paterson,B.(2006).Newspaperrepresentationsofmentalillnessandtheimpactofthereportingof‘events’onsocialpolicy:the‘framing’ofIsabelSchwarzandJonathanZito.JournalofPsychiatricandMentalHealthNursing,(13),294–300.

97

Patterson,M.,Somers,J.M.,McIntosh,K.,Sheill,A.&Frankish,C.J.(2008).HousingandSupportforAdultswithSevereAddictionsand/orMentalIllnessinBritishColumbia.CentreforAppliedResearchinMentalHealthandAddiction.RetrievedJanuary1,2009,fromhttp://www.carmha.ca/publications/index.cfm?contentID=29.

Philo,G.(1996).Themediaandpublicbelief.InG.Philo(Ed.),MediaandMentalDistress.Essex:AddisonWesleyLongman.

Philo,G.,McLaughlin,G.,&Henderson,L.(1996).Mediacontent.InG.Philo(Ed.),MediaandMentalDistress.Essex:AddisonWesleyLongman.

Riordan,T.(2004).ExploringtheCircle:MentalIllness,HomelessnessandtheCriminalJusticeSysteminCanada.Ottawa:LibraryofParliament.RetrievedJanuary1,2009,fromhttp://www2.parl.gc.ca/Content/LOP/ResearchPublications/prb0402‐e.htm

Ropers,R.H.(1988).Theinvisiblehomeless:Anewurbanecology.NewYork:HumanSciencesPress.

Rosenheck,R.,Kasprow,W.,Frisman,L.,&Liu‐Mares,W.(2003).Cost‐effectivenessofsupportedhousingforhomelesspersonswithmentalillness.ArchivesofGeneralPsychiatry,60,940‐951.

Scheufele,B.(2006).Frames,schemata,andnewsreporting.Communications,31,65‐83.

Scheufele,D.A.(1999).FramingasaTheoryofMediaEffects.JournalofCommunication,49(1),103‐122.

Scott,J.(1993).Homelessnessandmentalillness.TheBritishJournalofPsychiatry,162,314‐324.

Smith,E.M.,North,C.S.,&Spitznagel,E.L.(1992).Asystematicstudyofmentalillness,substanceabuse,andtreatmentin600homelessmen.AnnalsofClinicalPsychiatry,4,111–120.

Sullivan,G.,Burnam,A.,Koegel,P.&Hollenberg,J.(2000).QualityofLifeofHomelessPersonsWithMentalIllness:ResultsFromtheCourse‐of‐HomelessnessStudy.PsychiatricServices,15(9),1135‐1141.

SusserE,ConoverS,StrueningEL:Problemsofepidemiologicalmethodinassessing

thetypeandextentofmentalillnessamonghomelessadults.HospitalandCommunityPsychiatry40:261–265,1989

Susser,E.Moore,R.&Link,B.(1993).RiskFactorsforHomelessness.AmericanJournalofEpidemiology,(15)2.

98

Takahashi,L.M.,&Dear,M.J.(1997).Thechangingdynamicsofcommunityoppositiontohumanservicefacilities.JournaloftheAmericanPlanningAssociation,(63)1,79

TheStandingSenateCommitteeonSocialAffairs,ScienceandTechnology(TheSenate).(2006).Outoftheshadowsatlast:transformingmentalhealth,mentalillnessandaddictionservicesinCanada.Ottawa:TheSenate.RetrievedJanuary1,2009,fromwww.parl.gc.ca/39/1/parlbus/commbus/senate/com‐e/soci‐e/rep‐e/rep02may06‐e.htm

Tsemberis,S&Eisenberg,R.F(2000).PathwaystoHousing:SupportedHousingforStreet‐DwellingHomelessIndividualsWithPsychiatricDisabilities.PsychiatricServices(51)4,487‐493

Viera,A.J&Garrett,J.M.(2005).UnderstandingInterobserverAgreement:TheKappaStatistic.FamilyMedicine,37(5),360‐363.

Wahl,O.(1992).Massmediaimagesofmentalillness:areviewoftheliterature.JournalofCommunityPsychology(20)4,343‐352.

Wahl,O.(1996).Schizophreniainthenews.PsychiatricRehabilitationJournal,20(1),51‐54.

Wahl,O.,Borostovik,L.,&Rieppi,R.(1995).Schizophreniainpopularperiodicals.CommunityMentalHealthJournal,31(3),239‐48.

Wahl,O.E.,Wood,A.&Richards,R.(2002).NewspaperCoverageofMentalIllness:IsItChanging?AmericanJournalofPsychiatricRehabilitation(6)1,9‐31

Ward,G.(1997).Makingheadlines.Mentalhealthandthenationalpress.London:HealthEducationAuthority.

Weaver,D.H.(1996).WhatVotersLearnfromMedia.AnnalsofAmericanAcademyofPoliticalandSocialSciences(546),34‐47.

Weaver,D.H.(2007).ThoughtsonAgendaSetting,FramingandPriming.JournalofCommunication,57,142‐147

Wolitski,R.J.,Kidder,D.P.&Fenton,K.A.(2007).HIV,Homelessness,andPublicHealth:CriticalIssuesandaCallforIncreasedAction.AidsBehavior11,S167‐S171