5.6 Measures to assess the impact of anti-tobacco public ......Figure 5.32 WHO FCTC Article 12: Education, communication, training and public awareness section5.6janvier12102:Layout

5.6 Measures to assess the impact ofanti-tobacco public communication campaigns

319

IInnttrroodduuccttiioonn

Public communication campaignsare used to improve awareness,knowledge, and understanding ofan issue, in an attempt to influenceindividual behaviour, build supportfor, and contribute to policy andsocial change. Carefully monitoringthe implementation and outcomesof campaigns is essential toensuring their effectiveness anddemonstrating their contribution to aspecific public health outcome. Thissection summarizes the main com-ponents of individual behaviourchange and public will campaigns,briefly describes the theory andpractice of public communicationcampaigns and their evaluation, andprovides approaches for evaluatingeach component to determineimpact, from planning and develop-ment through implementation anddemonstrating results. Specificmeasures are identified for use asindicators of the achievement ofproximal and intermediate out-comes of public communicationcampaigns. However, the key tomeasuring the impact of publiccommunication campaigns is arti-culating clearly at the outset whatthe campaign is intended toaccomplish, who the campaign isintended to reach, what thecampaign is intended to cause, and

what communication and evaluationstrategies will be used.

A comprehensive public com-munication campaign will includemultiple components and demandextensive resources, particularly formedia production and placement(Atkin, 2001; Coffman, 2002;Dorfman et al., 2002). Thesecomponents may include resourcesfor advertisement production andplacement across a range of media;development and use of pressmaterials and press events;advocacy activity to influence howmessages are framed andinterpreted; and community actionto make messages locally relevant,compelling, and supportive ofcampaign goals. However, specificcampaign components may beimplemented independently and,depending on the desired out-comes, may be nearly as effectiveas a comprehensive campaign.Depending on the aims of thecampaign, and the resources andopportunities of the local jurisdiction(nation, province, state or com-munity), specific components orcombinations of components will bemore relevant. This section providesguidance on evaluation methods foruse in planning and implementing apublic communication campaign inorder to increase the likelihood ofsuccess. It will also serve as

guidance on measures to be usedto demonstrate the effectiveness ofthe campaigns in achieving moreproximal outcomes associated withthe WHO FCTC Article 12 directives(WHO, 2003; Figure 5.32). Selecting measures of effectivenessand demonstrating them are easiestwhen a campaign is grounded in achange theory that describes alogical progression from activities tooutcomes. Measures of effective-ness then can be selected tocoincide with specific expectedoutcomes, as described in Figures5.33 and 5.34. For example, apublic communication campaigndesigned to increase support forand promote the enactment andeffective implementation of asmoke-free air law might include:1. Television, radio, and print

advertising about the healthhazards associated with ex-posure to tobacco smoke, withmeasures of effectivenessdemonstrating that the targetaudience saw or heard andunderstood the message andassimilated the information (i.e.,awareness, attitudes, beliefs, orknowledge increased or werereinforced).

2. Contacts with news, health,community reporters, and edi-torial staff to encourage news,editorial, and community interest

section5.6janvier12102:Layout 1 12/01/2009 14:54 Page 319

IARC Handbooks of Cancer Prevention

320

stories about the dangers oftobacco smoke and conveyingsupport for smoke-free poli-cies. Proximal outcomes mightbe the news and specialinterest stories and editorialsthat are printed or airedaddressing the policy goals.

3. Media advocates might usesimilar public relationsstrategies focused on mediaoutlets in particular legislativedistricts that are known to beaccessed by influential lea-ders. Outcomes might bedocuments from records ofpublic comments by thetargeted decision makers.

4. Community groups and mem-bers may be organised to hostcommunity education events,meet with political represen-tatives, offer personaltestimonials of the value of

smoke-free air policies oradverse impacts of tobaccosmoke exposure. Proximal out-comes of these strategiescould include media coverageof community events, opinionpolling, intercept interviews, orother indicators of communityattitudes, and meetings with orother engagement of localdecision makers.

Together, these coordinated ac-tions, promulgating a clear andconsistent message and demandfor policy action, constitute acomprehensive public commu-nication campaign to advance thepublic health as outlined in theWHO FCTC; specifically, asdirected in Article 12 (Figure 5.32).This section describes the use ofpublic communication campaignsto advance these Article 12directives and measures of

whether the campaign hascontributed to specific goals.

CCoommppoonneennttss ooff aa ppuubblliiccccoommmmuunniiccaattiioonn ccaammppaaiiggnn

Public communication campaignstend to be divided into two types,each emphasizing somewhatdifferent strategies and outcomes:individual behaviour change cam-paigns, and public will or publicengagement campaigns (Coff-man, 2002). Individual behaviourchange campaigns seek tochange the types of behavioursthat lead to personal or socialproblems or instill behaviours thatwill improve individual or socialwell-being (Coffman, 2002). Publicwill campaigns, on the other hand,focus on motivating public officialsto take policy action, which in turnwill motivate, support, or enhance

Each Party shall promote and strengthen public awareness of tobacco control issues, using all available communicationtools, as appropriate. Towards this end, each Party shall adopt and implement effective legislative, executive,administrative or other measures to promote:

(a) broad access to effective and comprehensive educational and public awareness programmes on the health risksincluding the addictive characteristics of tobacco consumption and exposure to tobacco smoke;

(b) public awareness about the health risks of tobacco consumption and exposure to tobacco smoke, and about thebenefits of the cessation of tobacco use and tobacco-free lifestyles as specified in Article 14.2;

(c) public access, in accordance with national law, to a wide range of information on the tobacco industry as relevantto the objective of this Convention;

(d) effective and appropriate training or sensitization and awareness programmes on tobacco control addressed topersons such as health workers, community workers, social workers, media professionals, educators, decision-makers, administrators and other concerned persons;

(e) awareness and participation of public and private agencies and nongovernmental organizations not affiliated withthe tobacco industry in developing and implementing intersectoral programmes and strategies for tobacco control;and

(f) public awareness of and access to information regarding the adverse health, economic, and environmentalconsequences of tobacco production and consumption.

WHO (2003)

FFiigguurree 55..3322 WWHHOO FFCCTTCC AArrttiiccllee 1122:: EEdduuccaattiioonn,, ccoommmmuunniiccaattiioonn,, ttrraaiinniinngg aanndd ppuubblliicc aawwaarreenneessss


Measures to assess the impact of anti-tobacco public communication campaigns

321

health and healthy behaviours.Public will campaigns are used to“…legitimize or raise the impor-tance of a social problem in thepublic eye as the motivation forpolicy action or change.” (Coff-man, 2002). Evaluation chal-lenges, strategies, and measuresare somewhat different for eachtype of campaign. Ideally, gover-nmentled individual behaviourchange campaigns will raiseawareness, produce behaviouralchange, revise the social contextwithin which behaviour occurs,and produce new demands on thegovernment to further advanceenvironmental shifts to reinforceand produce new behaviourchange. For example, a govern-ment-sponsored campaign on thehealth risks of tobacco use couldlead to public demands forgovernment services to treattobacco dependence, and a newtax on tobacco products to pay forthe services. The public willcampaign for a higher tobacco tax,and dedicated use of the newresources may be coordinated bynongovernmental organisations,but may eventually lead to agovernment-sponsored campaignto increase access to tobaccodependence treatment.

Individual behaviour changecampaigns (“public educationcampaigns”) emphasize adver-tising and marketing as a mainstrategy. Campaign planners andevaluators must have a clearsense of what the campaign willcause to happen, why it willhappen, and who it will happen to,based on some theory ofbehaviour change (described

below in the section). Measures ofcampaign effectiveness will centeron what members of the targetgroup will be aware of, know, anddo as a result of the com-munication campaign that isdifferent from what they wereaware of, knew, and did before thecampaign (National CancerInstitute, 2002), or that is differentfrom what a comparable group isaware of, knows, and does relatedto topics addressed in thecampaign.

Public will or engagementcampaigns are used to buildpublic demand (“will”) to addressa particular problem through policyand social action. Public willcampaigns focus on the public’sresponsibility to create thesupportive environment that willallow or promote a desired be-haviour change (Coffman 2002;National Cancer Institute, 2005).The key strategies of public willcampaigns are media advocacyand public relation, with re-inforcing and supporting com-munity action, including com-munity organising and policyadvocacy. Public will campaignsseek to set the public agenda byinfluencing the media agenda(and the way people and decisionmakers are exposed to andprocess issue information)through media advocacy. But theultimate objective of policy orsocial change is achievedbecause the public will campaignprompts people to act, not byadopting a particular healthbehaviour, but by supporting(demanding) a particular policychange.

Public communication cam-paigns include a variety ofcommunication, and other stra-tegies, to educate the targetpopulation and disseminate infor-mation in compelling andengaging ways to raise the level ofdiscomfort individuals have with aparticular behaviour (e.g. tobaccouse). They also pressure decisionmakers on specific issues for thepurpose of changing (or ad-vancing) policies. Types of publiccommunication include paid (or“mass”) media, public relations,media advocacy, and communityaction implemented discreetly orin combination (Coffman, 2002;Dorfman et al., 2002). Thus, thepublic communication campaigncomponents shown in Figure 5.33can be implemented andevaluated as a multi-componentintervention, with the interventionsand outcomes in each “row”influencing outcomes in otherrows, or as discreet campaigns,with outcomes following linearlyfrom the specific intervention.

Paid media:

Paid or mass media is often themost expensive component of apublic communication campaign,and yet may be the one thatreaches the greatest number ofpeople. It can be effective incommunicating a tightly controlledmessage, creating an image,brand, theme, or call to action forthe overall campaign, and canchange attitudes, beliefs, andknowledge in the target popu-lation. Paid media, also known asadvertising, introduces an issue or



322

PPuubblliiccCCoommmmuunniiccaattiioonn

CCaammppaaiiggnn

PPrrooxxiimmaall OOuuttccoommeess

AAwwaarreenneessss AAttttiittuuddeess aannddkknnoowwlleeddggee

PPuubblliicc HHeeaalltthhIImmppaaccttss

AAccttiioonn BBeehhaavviioouurr

IInntteerrmmeeddiiaattee OOuuttccoommeess DDiissttaall OOuuttccoommeess IImmppaaccttss

Paid or massmedia

Increasedawareness amongtarget population:generalpopulation,smokers, youth,decision makers:recall, receptivity,responsiveness

Increaseawareness amongtarget population:news media,advocacy groups,key opinionleaders (businessleaders), decisionmakers

Changed/reinforcedattitudes and beliefsamong targetpopulation;Increasedknowledge;Policy discussions;increased decisionmaker support forpolicies and otheractions.

Policy enactment: - increased availability of

tobacco dependencetreatment;

- higher prices for tobaccoproducts;

- smoke-free laws; - tobacco-free outdoor areas- marketing restrictions- funding increases

Changed/reinforcedattitudes andbeliefs amongtarget population;increaseknowledge;increased supportfor policies

Calls toquitline, quitattempts;home smokingpolicies;response to“call to action”

- Successfulquits;

- Initiationaverted;

- Increasedcompliancewith policies

Decreasedmorbidityandmortalitydue tosmoking andtobacco use

Publicrelations

Mediaadvocacy

CommunityAction• Community,

health, socialworkers

• Media,decision-makers,administrators

• Educators,concernedpersons

Increaseawarenessamong (local)target population:Communitygroups,communityopinion leadersand decisionmakers,community newsmedia

Engagedcommunitymembers andleaders; - increased

knowledge; - increased

support forpolicies

Voluntarypolicies adopted;Local policiesenacted;Increasedinterventionsprovided (e.g.,smokingcessation)

Effective policyimplementation;High com-pliance withpolicies;New demandsfor policy/action

FFiigguurree 55..3333 FFllooww ddiiaaggrraamm ooff ppuubblliicc ccoommmmuunniiccaattiioonn ccoommppoonneennttss aanndd pprrooxxiimmaall aanndd ddiissttaall oouuttccoommeess



323

IInnccrreeaassee aawwaarreenneessss aammoonngg ttaarrggeettppoolluullaattiioonn::

news media, advocacy groups, keyopinion leaders (business leaders),decision makers

Changed/reinforced attitudes andbeliefs among target population;increased knowledge; policydiscussions; increased decisionmaker support for policies and otheractions

Changed/reinforced attitudesand beliefs among targetpopulation; increaseknowledge increased supportfor policies

Calls to quitline, quit attempts;home smoking policies;response to “call to action”

OOuuttccoommeess- Successful quits;- Initiation averted;- Increased compliance with policies

CCoommmmuunniiccttyy aaccttiioonnMore engaged and activelocal communicties with aninterest in the issue

SSoocciieettaall ppoolliiccyy eennaaccttmmeenntt::- increased availability of tobacco

dependence treatment; - higher prices for tobacco

products;- smoke-free laws;- tobacco-free outdoor areas- marketing restrictions- funding increases

IInnccrreeaassee aawwaarreenneessss aammoonnggttaarrggeett ppooppuullaattiioonn::

general population, smokers,youth, decision makers:recall receptivity,responsiveness

MMaassss aaccttiioonn• Paid media• Public relation• Media advocacy

FFiigguurree 55..3344 CCoonncceeppttuuaall ffrraammeewwoorrkk ffoorr eevvaalluuaattiioonn ooff aannttii--ttoobbaaccccoo ppuubblliicc ccoommmmuunniiccaattiioonn ccaammppaaiiggnnss



324

concept, delivers it to a largeaudience, and, if done effectively,raises awareness, increasesknowledge, creates interest,engagement, concern, and stimu-lates conversation and action(Centers for Disease Control andPrevention, 2003). Paid media isnot a necessary part of a publiccommunication campaign. How-ever, if resources are available, itcan dramatically expand the reachof a campaign and reinforce andsupport the public relations andcommunity action components.Paid media also may be used asa media advocacy strategy, withkey messages strategically placedin print, electronic, and othermedia, to reach influential decisionmakers and opinion leaders,including policy makers. Massmedia that is not paid for (e.g.media campaigns that rely ondonated time and the use of publicservice announcements), canserve the same function as paidmedia in an overall publiccommunication campaign, but isunlikely to have the reach of paidmedia or the target specificity; inaddition, placement most likely willbe outside the control of thecampaign. Depending on theresources available, and thespecific targets of the campaignmessage, paid media campaignscan feature a variety of mediachannels including television,radio, print, transit, billboards,Internet, brochures, and others.

Public relations:

The goal of public relations is todisseminate public communication

campaign messages throughothers, specifically the news media,opinion leaders, and those whomay be perceived as having morecredibility or objectivity thancampaign sponsors or paid mediamessages. Exposure obtained frompublic relations is “earned”coverage; “earned” because it isnot paid for but obtained throughstrategic advocacy efforts, in-cluding working with news mediaoutlets, community leaders, policymakers, and others with influenceto disseminate key messages.Public relations provides oppor-tunities to reach the target audiencethrough sources that appear morelegitimate, and allows the provisionof more detailed information thanpaid media, all while positioning thecampaign positively and potentiallyinfluencing the policy debate(Centers for Disease Control andPrevention, 2003). Public relationsalso provides the opportunity to“localize” national and internationalnews, events, and research(Chapman & Dominello, 2001;Niederdeppe et al., 2007), andbring to life local stories of personaltragedy (e.g. related to tobaccouse) that can stand on their own orbe coordinated with and reinforcepaid media messages.

Public relations involves estab-lishing relationships with membersof the press, and other influentialmembers of the community,developing supporting materialsincluding press releases and presskits, and staging community eventsand press conferences, amongother strategies. News and othermedia play a large role indetermining what the public thinks

about (agenda setting), howinformation is organised andpackaged for public consumption(framing), and focuses the public onparticular information at particulartimes for use in decision making(priming). Thus, public relationsstrategies are key elements ofpublic communication campaignsand should be vigorously imple-mented as part of publicengagement campaigns, inparticular, that seek policy or socialchange (Wallack et al., 1993, 1999;Coffman, 2002). While publicrelations strategies are employed toset the public agenda and keepissues in the public eye, they areoften directed at specific policymakers and become part of amedia advocacy strategy.

Media advocacy:

Media advocacy is an effort to usethe tools of mass media and publicrelations to reframe the publicdebate, encourage a communityto rethink its norms, and reachdecision makers who have thepower to transform the communityenvironment through the adoptionof policies that enhance publichealth (Wallack et al., 1993, 1999;WHO, 2004). Media advocacydiffers from paid media in that itsmain target is comparatively small(and could be only one individual),and the goal is policy change thatwill promote, support, or reinforceindividual behaviour change andthe public health agenda.However, media advocacy canuse paid media as one strategy toaccomplish advocacy objectives.In order to reach those individuals



325

with the power to make the policychange, media advocacy effortscan target highly organised andmotivated individuals (or organi-sations) who can pressure policymakers to make the desiredchange. Media advocacy mayeven target the general public in aneffort to set the public agenda andreframe an issue. In this case, paidmedia is a tool of media advocacy;communicating a message topolicy makers through engagedcitizens as the target audience.Mass media campaigns show-casing responsible tobaccocompany behaviour are likely mediaadvocacy campaigns targeted atpoliticians and voters in an effort torecast the company’s public image,earn the respect of the public, andrelieve public pressure on policymakers to take action that wouldconstrain the industry. Just as paidmedia, targeted at specific groupsof individuals, may use a variety ofmessages that cajole, engage,cause fear, or provoke anger in aneffort to stimulate behaviourchange, media advocacy uses bothpositive and negative tactics to exertpressure on decision makers andprovoke political action.

Community action:

In the context of public com-munication campaigns, commu-nity action engages the com-munity in defining a problemlocally and taking community-specific steps to advance abehavioural, normative, or policyshift at the local level or in supportof state or national goals.Community action is linked to, and

increases the resources of, thelarger public communicationcampaign, raising awareness,engaging local news media,organising community events,disseminating information throughlocal channels, and meeting with(and advocating with) local of-ficials (Pierce et al., 1990; Bracht,2001). These community effortsare often legitimized, reinforced,and supported by paid media.Where paid media may not bepossible, community action be-comes a crucial component ofpublic communication campaigns,often incorporating communityorganising tactics to advancemedia and policy advocacyobjectives. Community action isboth an extension of the publiccommunication campaign to thelocal level, and a strategy insupport of key public com-munication campaign compo-nents. It can take the form ofcommunity advocacy, publicrelations, participation in govern-ment processes, decision makereducation, leadership training,staged events (e.g. press events,media advocacy, and grassrootsmobilization), and communityorganisation to demand change(Niederdeppe et al., 2007).Community action also increasesthe likelihood that the publiccommunication campaign mes-sages and results will endure longpast the formal end of thecampaign (Bracht, 2001).

TThheeoorreettiiccaall uunnddeerrppiinnnniinnggss

Grounding a campaign in one ormore theories of behaviour change

enables campaign planners toexplain why and how a campaignshould work, thus assessing thecampaign’s progress throughoutthe health communication process,not just at the end of the campaign(Atkin, 2001; Coffman, 2002;National Cancer Institute, 2002;Randolph & Viswanath, 2004).Assessing progress enablesplanners to improve the campaignas it is developed and imple-mented, before more resourceshave been invested in a campaignthat may not succeed. Publiccommunication campaigns that aregrounded in theory are easier toevaluate over the lifetime of thecampaign (and easier to causallylink to outcomes), as planners areable to identify at the outset themore immediate or proximalindicators of whether a campaignis on track, as well as the longer-term indicators of campaigneffectiveness. Change theoriesrelevant to public communicationcampaigns include: the theory ofreasoned action, social cognitivetheory, the health belief model, thetrans-theoretical model (“stages ofchange”), consumer informationprocessing model, organisationalchange theory, community orga-nisation theory, and diffusion ofinnovation theory (among others),each described briefly below. Rea-ders are referred to Connell &Kubisch (1998), Atkin (2001),Bracht (2001), Coffman (2002),and the National Cancer Institute(2002) for additional information,bibliographies, and primarysources.

The theory of reasoned actionpostulates that attitudes and



326

norms create behavioural inten-tions, which in turn causebehavioural outcomes. A publiccommunication campaign may bedesigned to change or reinforcespecific attitudes and norms forthe purpose of causing behaviourchange. An evaluation of such acampaign would assess rein-forcement of or shifts in attitudesand norms, and would only expectbehavioural change where atti-tudes were or became consistentwith the desired behaviourchange. If such attitudinal shiftsfailed to occur or were notreinforced, the campaign wouldlikely be revised.

Social cognitive theory postu-lates that behaviour changeresults from motivation to changeand the acquisition of skills andabilities (self-efficacy) to change,within a given environmentalcontext. A public communicationcampaign grounded in this theorywould try to attract the targetaudience’s attention, convey acompelling message, impartspecific skills, and providemotivation to undertake behaviourchange (preferably in conjunctionwith a reinforcing environmentalchange, such as a price increaseon cigarettes, or the adoption of asmoke-free policy). An evaluationof such a campaign would assessattitudes and knowledge (skills) inthe target population and desire tochange the behaviour. In addition,a firm understanding of theenvironmental context would helpshape the development of thecampaign.

The health belief modelsuggests that people change

behaviour when they feelsusceptible or vulnerable as aresult of a given behaviour, andbelieve that the costs of continuingthe behaviour outweigh the costsof changing the behaviour.

The trans-theoretical model(“Stages of Change”) posits thatpeople proceed through (linearlyor cyclically) a readiness conti-nuum of behaviour change stagesfrom pre-contemplation to main-tenance of the behaviour change.Public communication campaignsbased on this theory will identifythe specific stages of the targetpopulation and attempt to movethem to the next stage, will havedifferent messages for audiencesin the different stages, or,perhaps, will target people at onestage only. Evaluation outcomemeasures will be determined bythe purpose and target audienceof the campaign, and may belimited to shifts along the readi-ness to change continuum (e.g.from “happy to smoke” to “thinkingabout quitting”).

The consumer information pro-cessing model suggests that howmuch and what kind of informationpeople have and how theyprocess it, are determinants ofwhether people will use infor-mation to inform and motivatebehaviours or behaviour change.To increase the chances thatinformation will be used indecision-making, public commu-nication campaigns must makeinformation available, package itas innovative and useful, andensure that it is accessible to (ableto be processed by) the targetpopulation. Tenets of this theory

are particularly helpful for evalu-ating campaign messages,materials, and delivery mediaduring the planning phases andearly implementation to ensurethat messages are understood byand resonate with the targetaudience.

The principles of communityorganisation theory are based oncommunity empowerment andcapacity building. In order to besuccessful and have a sustainedimpact, public communicationcampaigns must include partner-ships with community members,organisations, and governments,and mobilize communities todevelop and implement strategiesin support of campaign goals.Evaluation of a campaign basedon this theory would includestakeholder interviews, measuresof community competence, moni-toring of community activities, andother community evaluation tech-niques.

Diffusion of innovation theorydescribes how new norms, ideas,products, and practices diffusethrough communities and becomeaccepted or established in society.The theory focuses on charac-teristics of the innovation, as wellas characteristics of the com-munity, social networks, andcommunication systems throughwhich the innovation is spread.Cigarette use is a primaryexample of how a new product“catches on” and diffuses throughcommunities. Currently, smoke-free norms are being re-established, with the “diffusion”explained by this and othertheoretical models.



327

Often, public communicationcampaigns are grounded inseveral theories in order toaccount for the complexitiesinvolved in behavioural and socialchange enterprises. They mayeven adopt new theoreticalapproaches as the campaignproceeds, based on evaluationfindings, which might revise theirunderstanding of the local (oraudience-specific) change pro-cess, or provide new informationabout population attitudes andbeliefs.

TThheeoorryy aanndd pprraaccttiiccee ooff ppuubblliicc ccoommmmuunniiccaattiioonn ccaammppaaiiggnn eevvaalluuaattiioonn aannddaapppprrooaacchheess ttoo eevvaalluuaattiinnggeeaacchh ccoommppoonneenntt

In the context of the WHO FCTC,evaluation of public commu-nication campaigns should assesswhether the campaign is meetingits objectives as it is beingplanned, developed, and imple-mented in order to best ensuresuccess, and demonstrate that thecampaign has indeed achievedthe expected outcomes. Thus,evaluation resources should beinvested at the planning anddevelopmental stages to ensurethat specific interventions arecustomized to the target popu-lation and are culturally specificand appropriate. They should alsobe invested over the life of thecampaign, and beyond, to assesswhether proximal and moreintermediate outcomes are beingmet. Particularly where innovativeor unproven strategies are beingimplemented, or new theoretical

models are being tested, moreformal outcome studies may beappropriate. Public communica-tion campaigns may be discreetinterventions with a beginning, anend, and a predictable sequenceof events in the middle. Often,however, they are more accu-rately described as a “messysocial process” (Hornik, 2002),diffused by multiple strategies,through multiple channels, acrossindividuals, communities, andinstitutions, with direct and indirecteffects and diffuse outcomes thatmay reverberate long past theofficial “end” of the campaign(especially if policy change ob-jectives were achieved) (Freimuthet al., 2001). As a result,evaluation resources are appro-priately invested in ongoingsurveillance, point in timemonitoring, special studies toidentify opportunities for im-provement, confirm that progressis being made, identify mediationaland moderator effects (seeSection 3.2), and link interventionsto specific milestones andoutcomes. Experimental designsand controlled trials often are notpossible or appropriate (Balch &Sutton, 1997; WHO, 1998b), butinstead a collection of information,existing data, and specific studiesare needed to fully understandwhether and how a campaignworked. Tightly linking campaignobjectives to proximal outcomescan help demonstrate impact and,in particular, can help rule outcompeting explanations for ob-served change. Table 5.40 listsmethods to assess the effec-tiveness of each public com-

munication campaign componentat various levels of evaluation,including establishing proximaloutcomes.

At the outset of a publiccommunication campaign, theproblems and issues to be tackledand the baseline situation willhave been established throughongoing surveillance or, at thecommunity level, a needs as-sessment. The programmaticevaluation typically is concep-tualized and implemented in fourstages (described below)throughout the life span of theintervention and beyond. For-mative evaluation begins ascampaign concepts are beingdeveloped and summative evalu-ation focuses on the overall valueof the campaign in terms ofaccomplishing its stated ob-jectives. At the front end,evaluation includes testing andverification of campaign concepts(“formative”) and careful moni-toring of campaign activities andresources (“process”) to ensurethe campaign is being developedand implemented appropriately,efficiently, and with somelikelihood of success (Atkin &Freimuth, 2001). At the back end,“outcome” evaluation answers thequestions of whether the cam-paign has achieved its short- andlong-term objectives, and hasvalue to the community in terms ofadvancing public health goals (amajor focus of this volume). Thepoint is that evaluation should bewell integrated into all phases ofthe public communication cam-paign, and this information shouldbe well-utilized throughout the life



328

CCoommmmuunniiccaattiioonnSSttrraatteeggyy

LLeevveell ooff EEvvaalluuaattiioonn

Ongoing Surveillance

FFoorrmmaattiivveeDo the messages,materials, strategies“work?” Are theytailored to theintended audience?

PPrroocceessss

Implementationprocess: what andhow much wasdone? Distribution,effort expended,resources committed

PPrrooxxiimmaall aannddIInntteerrmmeeddiiaattee OOuuttccoommeess Knowledge, attitude,policy, normativeshifts

DDiissttaall OOuuttccoommeess Behaviour change,disease rate change

PPaaiidd mmeeddiiaaPPuubblliicc sseerrvviiccee

aannnnoouunncceemmeennttss

Focus groupdiscussions;Internet panelstudies;Marketing surveys;Document analysis

Gross rating points(GRPs)/Target ratingpoints (TRPs), whichare available frommedia buying firmsand media channels,provide indicators ofreach, frequency,exposure, andimpressions.

Populationbased/random digitdial (RDD) surveys(in-person, mail,telephone, Internet)of knowledge,attitudes, beliefs,behaviours; calls totelephone help line;web site visits,measures of res-ponses to specificcalls to action.

PPuubblliicc rreellaattiioonnss aannddeeaarrnneedd mmeeddiiaa

Key informant in-depth interviews;document analysis

News mediatracking: count ofstories run; tobaccocontrol advocacygroups cited; contentanalysis, slant.Case study

Special populationsurveys/keyinformant interviews;Official records ofgovernment policyand NGO policy.

MMeeddiiaa aaddvvooccaaccyy aannddggoovveerrnnmmeenntt

rreellaattiioonnss

“Who do you know?”inventory; keyinformant in-depthinterviews

News media framinganalysis

Document analysisof legislative records.Case study.

CCoommmmuunniittyy aaccttiioonn

Community needsassessment; community capacityanalysis; strengths,weaknesses,opportunities, threats(SWOT) analysis;health risk profile

Activity logs;meeting minutes

Community policydatabase: voluntary, statutory/regulatory;Case study.

Cigarette tax and sales records; behavioural risk factor and disease surveillance; diseaseregistries; vital records

Structured analysis of data from existing surveillance systems

Table 5.40 Methods to Assess the Effectiveness of Public Communication Campaigns by CampaignComponent and Level of Evaluation



329

cycle of the campaign, andbeyond, to make judgments aboutcampaign progress, improve itseffectiveness, and inform deci-sions about its future (Patton,1997). Tables 5.41 and 5.42 listindicators of the effectiveness ofeach public communication com-ponent and corresponding out-comes by evaluation level. Eachlevel is described more fullybelow.

Formative evaluation:

Formative research and evalu-ation identify the causal pathwaythrough which an intervention islikely to work, and facilitatescampaign improvement as it isbeing developed and imple-mented. It does not speak to thecampaign’s value or impact, butidentifies its strengths andweaknesses, and aspects of thecampaign that are not working asplanned or are not likely tosucceed (Mark et al., 2000). It canprovide information about keymessages that are or are notresonating, and the types ofindividuals who are or are notresponding to the campaign,among other variables importantto its success. Information fromformative evaluation is used bycampaign planners and staff tosolve problems, address weak-nesses, revise expectations,revamp the campaign conceptsand executions, or otherwiseimprove conceptualization andimplementation (Patton, 1997).Evaluation and research, such asmarketing surveys, that inform thecreative process, also serve as

baseline measures of attitudes,beliefs, and norms the publiccommunication campaign isattempting to change.

Process evaluation:

Process evaluation is applied toprogramme implementation andanswers the question how well thecampaign is being delivered to theintended audience. Process toolsmeasure effort and activity andhelp inform whether a campaign isbeing delivered as intended, and,if not, where the shortfalls areoccurring. Retrospectively, pro-cess evaluation can shed light onwhat went wrong, if a particularcampaign fails to meet itsobjectives, and identify lessons onhow to make future campaignsmore effective. Process evaluationinvolves monitoring resources,activities, and inputs includingmaterials produced and dis-tributed, news contacts made,meetings held, and a variety ofinformation related to theplacement of paid media. Processevaluation does not address theachievement of campaign out-comes or impacts, but can beused to link campaign activities tothose outcomes by quantifying the“dose” of the campaign over timeand in different communities.

Outcome evaluation:

Evaluation strategies for proximaloutcomes are used by publiccommunication campaign plan-ners and evaluators to determinewhether the shorter-term out-comes the campaign was

designed to achieve have actuallybeen met. As outlined in Figure5.33, outcome evaluationgenerally requires more resourcesthan formative or process evalu-ation, and, depending on theavailability of financial andscientific resources, may beaccomplished by special studiesor by accessing information fromroutinely collected data sources.Outcomes of public commu-nication campaigns vary fromcognitive shifts (proximal) throughsocial normative and behaviouralshifts (distal), including individualknowledge, beliefs, awareness,attitudes, self-efficacy, beha-vioural intentions, behaviourthrough environmental changes,media frames, policy enactment,and normative change (measuredpolicy enactment and compliancewith policies).

Evaluation of more distal out-comes assesses achievement ofpublic health goals, which almostcertainly do not result from publiccommunication campaigns alone.Impacts would include changes inhealth behaviours (e.g. tobaccouse), tobacco-related diseaserates (e.g. lung cancer incidence),and, ultimately, rates of death dueto tobacco use. Outcome evalu-ation can be the most rigorous,complex, and resource intensivelevel of evaluation, and should beconsidered carefully at theprogramme (not the campaign)level. Public communicationcampaigns, after all, constituteonly one component of the WHOFCTC effort to transform societyand “reaffirm the right of all peopleto the highest standard of health.”



330



FFoorrmmaattiivvee

Do the messages,materials, strategies“work?” Are they tailored to the intendedaudience?

PPrroocceessss

Implementationprocess: what andhow much wasdone? Distribution, effortexpended,resources committed

PPaaiidd oorr mmaassss mmeeddiiaaPPuubblliicc sseerrvviiccee


How is the messagelikely to make theaudience feel?What message isunderstood and willthe audience takeaway?What part of themessage pleases,annoys, angers,scares the audience?Is the ad believable?Does the messagespeak to “people likeme?”Is the message cul-turally appropriate?Is the message com-pelling? What is the appro-priate channel for themessage?What are the com-peting messages?

Number of adsrunning, placement,impressions, Grossrating points(GRPs)/ Targetrating points (TRPs),money spent,location of out-of-home media, timelines met


What is the current“informationenvironment?”Will the messagechange the“informationenvironment?”What kind of news tomake (how to framethe message)?

News media tracking:count of stories run;tobacco controladvocacy groupscited; case study;content analysis;framing analysis(point of view,accuracy, slant,agenda setting)


rreellaattiioonnss

Is the policy/legislative environ-ment hostile orhospitable to themessage?What are the com-peting priorities?Who are allies?What are obstacles?

Indicators of decisionmaker interest andaction from publichearings and officialmeetings


Strengths,weaknesses,opportunities,threats (SWOT)analysis results;meetings withcommunity membersand leaders;formation ofcommunity advisorygroup

Community meetingsheld; coalitionsformed; organi-sations involved;number of activities,trainings and eventsplanned/imple-mented;number ofpeople who participate;resources invested inoutreach (money,time, personnel);number of materialsproduced anddistributed

Table 5.41 Formative and Process Indicators of the Effectiveness of Public Communication Campaignsby Campaign Component and Level of Evaluation



331



SShhoorrtt--tteerrmmOOuuttccoommeess

Awareness,knowledge, Attitude shifts

IInntteerrmmeeddiiaattee OOuuttccoommeess

Knowledge, attitude/policy shifts

DDiissttaall OOuuttccoommeess

Behaviour normativechange, disease rate change

PPaaiidd mmeeddiiaaPPuubblliicc sseerrvviiccee


Confirmed aware-ness (discrimination)of media message;level of receptivity tomedia message (e.g.talked to othersabout it); support forspecific policy:increased availabilityof tobacco depen-dence treatment;higher prices fortobacco products;smoke-free air laws;marketing restric-tions; fundingincreases

Number of respon-ses to call to action: calls to quitline,visits to web site,other); knowledgeand attitude shifts:reduced accep-tability of smoking/exposure to tobaccosmoke; increasedawareness of harmfrom smoking/tobacco smoke;increased intentionsto quit; increasedknowledge of how toquit


Did the issue get onthe media agenda?Was the issueframed according tothe campaignobjectives?Did the mediacoverage advancethe message?Public/decisionmaker support forspecific policy

Policy enactment


rreellaattiioonnss

Did the issue get onthe public agenda?Support for policies;legislative proposalssubmitted; legislativebills introduced

Policy enactment;amount of cigaretteor other taxes.


Voluntary policiesadopted; health carepolicies to providetobacco dependencetreatment; betterinformed profes-sionals; improvedhealth care services;availability of cessa-tion support

Community laws andregulations enacted;community servicesand programmesestablished

Table 5.42 Outcome Indicators of the Effectiveness of Public Communication Campaigns by CampaignComponent and Level of Evaluation

Per capital consumption of cigarettes; smoking prevalence, use of other tobacco products;exposure to tobacco smoke; incidence of tobacco caused disease



332

If done well, they will most likelycontribute to population or targetgroup changes in, or reinfo-rcement of, attitudes, knowledge,and beliefs that contribute topolicy, environmental, and nor-mative improvements thatadvance the public health.

MMeeaassuurreess ttoo aasssseessss pprrooxxii--mmaall aanndd ddiissttaall oouuttccoommeess

Of all the public communicationcampaign components, evaluationindicators are probably the mosthighly developed, or as least themost familiar, for paid media.Indicators include results of focusgroup testing of media messagesto ensure they “speak” to or“resonate” with the targetaudience, convey the intendedmessage, and are likely toprovoke the desired attitude andbehaviour changes. Indicators ofthe campaign’s reach into thetarget population and the fre-quency with which campaignmessages were aired (usuallyquantified as gross rating points(GRP) or target rating points(TRP) for television ads andviewer “impressions” for printmedia) are common processmeasures. Reach also can bequantified by means of consumersurveys designed to elicitawareness of the campaign (i.e.aided or unaided recall of specificcampaign ads, messages, andthemes). Proximal measures ofcampaign effects, for examplechanges in awareness, attitudes,and knowledge about the issuebeing promoted, can be obtainedby in-person, mail, telephone, or

Internet surveys of the targetpopulation, where outcome mea-sures can be linked to awareness.More distal outcomes, like actualbehaviour change, also aretypically measured by some kindof survey of the target population.Similar survey tools and methodsare used to assess populationsupport for specific policyinitiatives. Depending on thepurpose of the campaign, othermeasures of campaign effective-ness may be appropriate. Forexample, indicators of effective-ness of a paid media campaigndesigned to promote telephone-based cessation services couldinclude the number of calls to ahelpline or number of calls amongthose aware of the campaign.Paid media campaigns promotingother calls to action (e.g. to visit aweb site, sign a petition, or send aletter), would be similarlyevaluated by the number ofindividuals who respond by takingthe requested action. Systemswould have to be established tocompile and count these actions,and may be as simple asmonitoring “hits” to a web sitebefore and after a call to action,tallying the number of signatureson a petition, or including postagepaid (addressed) envelopes in aletter writing campaign, with thepostage charge providing infor-mation on the number of letterssent.

For process and outcomemeasures, in order to support aclaim that changes in awareness,attitudes, and behaviours resultfrom the campaign itself, evalu-ators need to demonstrate that the

effects occurred uniquely, tem-porally, or to a greater extent inthe target population. This may beachieved by identifying acomparison community not re-ceiving the public communicationcampaign. This could either be thetarget population, prior toimplementation of the paid mediacampaign (good), a similarcommunity not receiving thecampaign at the same time as thetarget population (better), or byvarying the dose of the campaignacross jurisdictions (Farrelly et al.,2005b). In some cases, interceptsurveys, or surveys of availablemembers of the target population(“convenience” samples), will bethe only practical means ofgleaning the potential impact of acampaign. Such surveys may beuseful for obtaining anecdotalinformation, identifying problems,or fleshing out the details offindings from larger populationsurveys, but generally are notconsidered to be robust indicatorsof population level outcomes.

Perhaps the most commonindicators of effective publicrelations result from news mediatracking. This is both a simplecount of news stories related tothe tobacco topic promoted by apublic relations effort (or a ratio ofsuch stories to other health-related stories), and a contentanalysis of those articles todetermine characteristics such asthe message, accuracy, slant,point of view, and prominence ofmessage, among others (Henry &Gordon, 2001; Durrant et al.,2003; Clegg-Smith et al., 2005;Neiderdeppe et al., 2007). It may



333

even include responses to thestory, such as letters to the editor,news media-sponsored Internetpolls, and whether the story was“picked up” by other mediachannels and outlets. Togetherthese indicate intermediateoutcomes; demonstrating first thatthe public relations efforts resultedin increased news mediacoverage, and second that thecoverage conveyed the infor-mation and point of view promotedby the public relations campaign.News media tracking effortstypically are limited to print newsmedia, which has been shown tobe a marker of media coverageoverall (e.g. electronic media)(National Cancer Institute, 2006).However, coverage of televisionmedia can be tracked as well, withvolunteers or paid viewers sys-tematically viewing and catalogingtelevision news coverage ofspecific issues for content andcharacteristics. While countingnews stories and describingcharacteristics relevant to tobaccocontrol is straight-forward, con-necting them to specificprogrammatic outcomes may bemore difficult, as noted byevaluators of the American StopSmoking Intervention Study:

The challenge in evaluation,however, is demonstratingthat news media coveragedoes in fact influence thethinking, decisions, andbehaviour of the public andof policy makers. Althoughdetermining such a cause-and-effect relationship forsome very focused and

geographically limited topicsmight be possible, re-searchers in the field ofevaluation are still grapplingwith how to do so for wide-scale public health inter-ventions (National CancerInstitute, 2006).

Process indicators becomeparticularly important in making thelink, as well as understanding howsome strategies might be improvedin the event that anticipated resultsare not achieved. Anecdotes andpersonal statements may beparticularly relevant to under-standing the influence of newsstories on decision maker action.

The longer-term goals of theWHO FCTC are to changeindividual behaviour as a result ofmodifications to the social, eco-nomic, and health environment,which in turn result fromgovernment intervention (WHO,2003). These modificationsprovide the conditions withinwhich people can be healthy(Institute of Medicine, 1988). Theimplementation and success ofthese interventions are based inpart on popular expectations anddemands. Media advocacystrategies put these policy changedebates on the public and policymaker agendas. By focusingmedia attention on specific publichealth issues (agenda setting),and focusing the debate to reflectthe public health perspective(framing), media advocates seekto influence the information thepublic uses to make decisions(priming), and reach opinionleaders and policy makers to

change public policy (Wallack etal., 1993). Indicators of mediaadvocacy success include: mea-sures of whether the campaignissue has become part of themedia agenda and is framedaccording to the public healthperspective (established throughcontent analysis of news mediaprogramming and print newsarticles), whether media supportthe particular policy agenda (e.g.in newspaper editorials) andwhether their support is asso-ciated with policy maker support(e.g. through key informantinterviews), and whether thatparticular agenda is advanced(e.g. in legislative debate, theintroduction of legislative bills orthe enactment of legislation).Areas of exploration for formativeand process evaluation of mediaadvocacy include: an assessmentof the message’s connection withpeople at the community level, themedia’s understanding of theissue, how the issue can beframed to capture media andpublic attention and focusattention on larger public healthvalues (e.g. how the tobaccocontrol issue can be framed toemphasize social accountabilityrather than personal responsibility(Wallack & Dorfman, 2001)), howrelationships have been de-veloped with community advocacygroups and the media, and aquantification of the actions ofthese groups. Measures ofproximal outcomes associatedwith media advocacy include:public support for specific policygoals as measured by in-person,mail, telephone or Internet



334

surveys, and political polling, andframing analysis to monitor andassess news media reports ofspecific policy initiatives and of theissue or problem the policyinitiative seeks to address.Standard methods include: keyinformant and opinion leaderinterviews, political polling, newsmedia tracking, and contentanalysis. Since the key targets ofmedia advocacy are the policymakers (i.e. organisations orlegislative bodies that have thepower to make the policy change),indicators of success will be drawnfrom official meeting minutes andtranscriptions, key opinion leaderinterviews focusing on specificinitiatives, and official records ofpolicy enactment, as well as newsmedia reports.

Public communication cam-paigns are most effective inchanging community and socialnorms, and building support forand actually prompting theenactment of public health policy,when they incorporate communityaction (Hopkins et al., 2001).However, what constitutes com-munity action, how communityinitiatives are described, quanti-fied, and measured, and whatchange theories underlie theirdevelopment and success, havebeen topics of ongoing debate(Connell & Kubisch, 1998; Bracht,2001; Connell et al., 2007).Successful community actioninvolves change at many levels(individual, family, personal net-work, institutional, and com-munity), including many domains(economic, social, physical, andcommunity), and evolves over

different time periods (near-term,interim, long-term, or ultimate)(Connell & Kubisch, 1998). Thus,the measurement challenge issubstantial. As with publiccommunication campaigns over-all, evaluation of communityinitiatives is easier when anexplicit change theory is specifiedat the outset from which to identifyspecific indicators of the develop-ment and implementation of theinitiative and progress towardanticipated short-, intermediate,and longer-term outcomes. Indi-cators should reflect theprocesses through which activitiesare developed and planned (com-munity needs assessment,meetings held, individuals pre-sent, organisations involved), theimplementation of those activities(e.g. meetings with decisionmakers, community forums, pressevents), and proximal results ofthese activities (the adoption ofresolutions, community partici-pation in events, news coverage,improvements in awareness ofproblems and solutions, increasein community member knowledgeabout the specific issue andsupport for specific action,evidence of decision-maker sup-port), as well as longer-termchange (policy enactment or theachievement of a specificobjective, such as provision ofcommunity cessation services orremoval of pro-tobacco adver-tising at a specific location)(Gambone, 1998). Methods fordescribing, monitoring, and cap-turing the effects of communityaction, community interventions,and measures of short- and

longer-term outcomes are onlypoorly developed at this time.Nonetheless, cataloging the inputthat supports community action(financial, in-kind, and personnelresources), quantifying activitylevels (number of meetingsconvened, contacts made, eventsheld), and documenting processthrough case studies, can behelpful in discerning whether andhow community actions contributeto public communication goals.

Table 5.40 summarizes me-thods to assess the effectivenessof public communication cam-paigns by campaign component(paid media, public relations,media advocacy, and linkedcommunity action), and level ofevaluation (formative, process,outcome), as described in thissection. Not all jurisdictions willhave the resources to implementa multi-component public com-munication campaign or fieldpopulation-based surveys toassess campaign outcomes. Forsuch jurisdictions, the methodswithin each column may be useddiscreetly for each level ofevaluation within a particularcomponent. It is not necessary tomount a population level survey inorder to demonstrate a populationlevel impact, but some populationlevel data base (like emergencyroom admissions for acutemyocardial infarction, calls to aquitline, or sales of cessationmedication), with informationproximally related to the campaignresult, is needed.

Tables 5.41 and 5.42 listformative and process indicatorsand proximal and distal outcome



335

indicators, respectively, of theeffectiveness of public commu-nication campaigns, by campaigncomponent and level ofevaluation. Evaluation indicatorsof proximal campaign effectsshould be specific to the individualcampaign message andcommunication component. Forexample, survey questions de-signed to understand the targetaudience’s awareness of, orreactions to, a set of commercialsor advertisements describing thehealth risks of exposure tosecondhand smoke, will be spe-cific to the content of the messageand the goals of the campaign(e.g. to increase support for theenactment of or compliance with asmoke-free workplace law), andwill be different from surveyquestions associated with acampaign to promote cessationamong current smokers. In bothcases, campaign planners andevaluators should test themessages with members of thetarget audience, monitorimplementation of the campaign,assess exposure to and under-standing of the messages,determine attitudes, beliefs, andknowledge related to the topic,before and after the campaign (oramong those exposed and notexposed to the campaign), andassess changes (increases) in thelikelihood of engaging in theparticular campaign “call to action”(e.g. refrain from smoking in publicplaces, demand no smoking inpublic places, think about quittingsmoking, make a quit attempt).The specific questions used willbe determined by the content and

goals of the public communicationcampaign. Examples of specificquestions are provided below.

A helpful source of indicatorsto measure outcomes associatedwith public communication cam-paigns is the Centers for DiseaseControl and Prevention (CDC)manual, Key Outcome Indicatorsfor Evaluating ComprehensiveTobacco Control Programmes(Starr et al., 2005). This user-friendly, fairly comprehensiveguide compiles and providesinformation on 120 outcomeindicators for use in evaluating theshort-, intermediate, and longer-term impacts of comprehensivetobacco use prevention andcontrol programmes. Indicatorsare organised according to threeprogrammatic areas (preventinginitiation, promoting cessation,and eliminating exposure totobacco smoke), and grounded inevidence-based logic models.Detailed information is providedfor each indicator including indi-cator definition, example datasources, specific measures (e.g.question wording), and overallquality of the indicator. Thoseindicators useful for monitoring theoutcomes of public commu-nication campaigns, specifically,are listed in Table 5.43. The guideis available online (in English) athttp://www.cdc.gov/tobacco/Indicators/KeyIndicators.htm.

The Question Inventory onTobacco, formerly known as TheSurvey Questionnaire DesignResource (available online (inEnglish) at http://apps.nccd.cdc.gov/QIT/), is another importantresource for identifying survey

questions that contribute to eff-ective measures of interventionoutcomes. Also developed by theCDC's Office on Smoking andHealth, the online resourcecategorizes more than 1700tobacco-related questions from 13United States national and statesurveys that have been usedstarting in 1990. The QuestionInventory on Tobacco resourceprovides easy-to-use searchcapabilities to locate surveyquestions, including possibleanswer formats, and identifies thespecific surveys in which thequestions have been used.

The WHO FCTC Article 12describes five topics about whichthe public should be made awarethrough public communicationtools: 1) health risks of tobaccoconsumption (including addiction),2) health risks of tobacco smoke,3) benefits of quitting, 4) aspectsof the tobacco industry, and 5)adverse health, economic, andenvironmental consequences oftobacco production (this topic isnot addressed in this section). Inaddition, Article 12 specificallyaddresses awareness of tobaccoissues among media pro-fessionals, decision makers,community health and socialworkers, educators, and con-cerned individuals. Variousstrategies can be used to achievepublic awareness, from traditionalpaid media campaigns utilizingtelevision, radio and/or printtargeting the general population orpopulation subgroups (likesmokers or youth), to strategic,targeted public relations andcommunity action (or education)



336

campaigns targeting communityleaders, health care providers,business leaders, nongovern-mental organisation directors,tribal leaders, and others. Specificmeasures to ascertain populationlevel awareness and knowledge of

topics included in Article 12 aredescribed below.

Awareness of paid mediacampaigns by the generalpopulation or specific targetedsubgroups (e.g. smokers, youth) isgenerally ascertained by popu-

lation level surveys, includingtelephone surveys (care should betaken with written surveys toensure that responses are notcued by response categories orother prompts). Awareness can beascertained by general questions

OOuuttccoommee LLeevveell IInnddiiccaattoorr IInnddiiccaattoorr ddeessccrriippttiioonnNNuummbbeerr

SShhoorrtt--tteerrmm 1.6.1 Level of confirmed awareness of anti-tobacco media messages 1.6.2 Level of receptivity to anti-tobacco media messages1.6.4 Level of support for policies 1.6.5 Level of support for increasing excise tax on tobacco products

IInntteerrmmeeddiiaattee 1.10.1 Proportion of young people who think that smoking is cool and helps them fit in

1.10.2 Proportion of young people who think that young people who smoke have more friends

1.10.5 Proportion of young people who are susceptible never-smokers1.12.1 Amount of tobacco product excise tax 3.11.1 Proportion of adult smokers who have made a quit attempt 3.11.2 Proportion of young smokers who have made a quit attempt3.11.3 Proportion of adult and young smokers who have made a quit attempt

using proven cessation methods

LLoonnggeerr--tteerrmm 2.7.1 Proportion of the population reporting exposure to secondhand smoke in the workplace

2.7.2 Proportion of the population reporting exposure to secondhand smoke in public places

2.7.3 Proportion of the population reporting exposure to secondhand smokeat home or in vehicles

3.13.1 Proportion of smokers who have sustained abstinence from tobacco use3.14.1 Smoking prevalence3.14.2 Prevalence of tobacco use during pregnancy3.14.3 Prevalence of postpartum tobacco use 3.14.4 Per capita consumption of tobacco products

From Starr et al. (2005; http://www.cdc.gov/tobacco/Indicators/KeyIndicators.htm)

Table 5.43 Example Indictors from Key Outcome Indicators for Evaluating Comprehensive TobaccoControl Programmes Relevant to Monitoring the Effects of Public Communication Campaigns



337

that require the participant to fill indetails of the campaign, or ofspecific ads, or by providing theparticipant with some generalinformation about a campaign orad in an effort to prompt orfacilitate recall (Table 5.44).Unaided recall is generally con-sidered to be a superior methodfor accurately estimating exposureto and awareness of a campaignor message (Sly et al., 2001b), butaided recall has been shown to bean effective measure as well(Niederdeppe, 2005). Measures ofawareness are designed todetermine exposure to a specificmessage or advertisement, sothat respondents can becategorized according to exposurelevel, and differences in attitudesand behaviours can be correlatedto awareness. Some more generalmeasures of awareness areintended to ascertain the amountof anti-tobacco advertising towhich subjects are exposed, inorder to make more generalinferences about the relationshipbetween anti-tobacco messagingin general, and attitudes andbehaviours in general. Finally,measures of the relevance orsalience of the media message tothe individual provides keyinformation on whether thecampaign is effectively commu-nicating the message, the types ofindividuals who are more likely torespond to the message, and theutility of the campaign incontributing to programme goals.This information can be used tostrengthen a poorly performingcampaign in progress or, at least,provide useful information for

developing the next campaign. Inorder to assess exposure to andawareness and salience ofmessages to survey participants,measures need to be customizedto the specific media campaign.

Once exposure, awareness,and salience are assessed andquantified, an analysis can beundertaken as to whether thosewho were exposed to thecampaign message were aware ofit or receptive to it in some way,are more likely to be aware of andunderstand the key messages ofthe campaign, and whether thisnew awareness or knowledge isassociated with specific proximaloutcomes (e.g. attitudes andbeliefs). Measures of awarenessand knowledge of specificcampaign messages should beconstructed to closely matchspecific messages being deliveredand the overall intent of thecommunication campaign. Themeasures described in Table 5.45relate specifically to awarenessand knowledge of the topicsdescribed in Article 12. Forcampaigns that address othertopics (e.g. motivating tobaccousers to quit, issues related to“light” and “low tar” cigarettes,increasing support for specificpolicies), readers are referred tosurveys and measures describedin this Handbook and elsewhere(National Cancer Institute, 2002;Centers for Disease Control andPrevention, 2003; Starr et al.,2005). Measures of behaviours(e.g. quitting, uptake, abstinence)that a campaign may seek toinfluence, as well as mediatorsand moderators of these

behaviours (e.g. perceptions ofrisk, cooccurring disorders), arediscussed in Section 3.1.

Measures of effectiveness forpublic relations, media advocacy,and community action efforts areless well developed and generallyhave not been collected andcatalogued in the form of surveys,interviews, and question lists.Readers are referred to Tables5.41 and 5.42 for listings of thetypes of information that should becollected in order to assess theeffectiveness of these efforts.Additional information, resources,and “how to” instructions forassessing the effects of publicrelations and media advocacyefforts, in particular, are availablefrom Radke (1998), NationalCancer Institute (2002), Centersfor Disease Control and Pre-vention (2003) and the WHO(2004), among others.

MMoonniittoorriinngg ootthheerr ccoommmmuunnii--ccaattiioonn ccaammppaaiiggnnss

Public communication campaignsconstitute an effort to control theinformation environment (Ran-dolph & Viswanath, 2004), tomake specific informationavailable to the target population,influence the public agenda, andframe the policy debate from apublic health perspective, with theobjective of changing behaviours,norms, and policies to advancepublic health. However, publiccommunication campaigns oftentake place in a cluttered andcompetitive environment. Theyare competing for attention withother communication campaigns,



338

CCoonnssttrruucctt ((aa)) AAwwaarreenneessss ooff SSppeecciiffiicc AAnnttii--TToobbaaccccoo MMeeddiiaa MMeessssaaggeess

MMeeaassuurree “Have you recently seen an anti-smoking or anti-tobacco ad on TV that shows-- [brief description of ad]?” (Yes, Maybe, No)

“What happens in this advertisement?”

“What do you think the main message of this ad was?”

SSoouurrccee LMTS, 2003 (http://americanlegacy.org)

VVaalliiddiittyy Established validity (Sly et al., 2001b)

VVaarriiaattiioonnss Variations are possible in the amount of prompting provided to the respondent (e.g. “Are you aware of any advertising or campaign against smoking or about or against cigarette companies that is now taking place?” (Yes, No) from [source?])“What is the theme/slogan of this advertising or campaign?”

CCoommmmeennttss It may be necessary to assess overall TV viewing/radio listening patterns to understandwhether participants had the opportunity to be exposed to the media message.

Expected response categories should be pre-determined, but should not be read to therespondent. Responses are categorized as accurately describing the ad (indicatingawareness) or not.

CCoonnssttrruucctt ((bb)) AAwwaarreenneessss ooff GGeenneerraall AAnnttii--TToobbaaccccoo MMeeddiiaa MMeessssaaggeess

MMeeaassuurree 11:: AAdduulltt “Now I would like you to think about advertising or information that talks about the dangersof smoking, or encourages quitting. In the last 6 months - [since…] - how often, if at all, haveyou noticed such advertising or information?” (Never, Rarely, Sometimes, Often, Very often)

SSoouurrccee The ITC Project, 2007

VVaalliiddiittyy Face validity.

VVaarriiaattiioonnss The time period of interest, the medium specified, and the types of ads described can all vary(e.g. “During the past 7 days, how many commercials have you seen on TV about NOTsmoking cigarettes?” (None, One, Two or three, Four to six, Seven or more), from Global Adult Tobacco Survey (GATS, 2007).

In this question, one can substitute (or add) “heard on the radio” or “seen on a billboard,” asappropriate, for “seen on TV.”

CCoommmmeennttss The use of this general item is helpful to characterize level of exposure to the broad rangeof state/provincial and national or other media-based anti-tobacco education campaigns.Such questions may be particularly helpful for pre-campaign surveys to quantify the amountof “background” anti-tobacco advertising to which the population is exposed.

Table 5.44 Measures to Assess Population Level Awareness and Knowledge of Public CommunicationCampaign Paid or Mass Media Components



339

MMeeaassuurree 22:: YYoouutthh “During the past 30 days (one month), how many anti-smoking media messages (e.g.,television, radio, billboards, posters, newspapers, magazines, movies) have you seen orheard?” (A lot, A few, None)

SSoouurrccee GYTS, 2007


VVaarriiaattiioonnss “When you go to sports events, fairs, concerts, community events, or social gatherings, howoften do you see anti-smoking messages?”(I never go to sports events, fairs, concerts, community events, or social gatherings, A lot,Sometimes, Never)

CCoommmmeennttss The use of this general item is helpful to characterize level of exposure to the broad rangeof state/provincial and national, or other media-based, anti-tobacco education campaigns.Such questions may be particularly helpful for pre-campaign surveys to quantify the amountof “background” anti-tobacco advertising to which the population is exposed. The variationmay be useful for assessing awareness of general anti-smoking messages in non-electronicvenues.

MMeeaassuurree 33:: LLooccaattiioonnss “In the last 6 months, have you noticed advertising or information that talks about thedangers of smoking, or encourages quitting in any of the following places?” (Yes, No) –READ OUT EACH STATEMENT• on television• on radio• at the cinema [US/Canada/AUS: at the movies]• on posters or billboards • in newspapers or magazines• on shop/store windows or inside shops/stores where you buy tobacco• on cigarette packs• leaflets• on the Internet• anywhere else? (specify)



VVaarriiaattiioonnss Locations listed should be relevant to the campaign and the jurisdiction and should vary, asappropriate.

CCoommmmeennttss Understanding where consumers are exposed to anti-tobacco media may help in planninga public communication campaign, or may help identify specific ads or campaigns to whichconsumers have been exposed.




340

CCoonnssttrruucctt ((cc)) AAwwaarreenneessss ooff SSmmookkiinngg RReellaatteedd NNeewwss SSttoorriieess

MMeeaassuurree “Now I want to ask you about the media more generally. First, thinking about news storiesrelating to smoking or tobacco companies that might have been on TV, radio, or in thenewspapers. In the last 6 months, that is, since [6 month anchor], about how often, if at all,have you seen or heard a news story about smoking?” (Never, Rarely, Sometimes, Often,Very often)

“On balance, how did the news stories portray smoking? Were they All pro-smoking, Mostlypro-smoking, Equally pro- and anti-smoking, Mostly anti-smoking, All anti-smoking?”



VVaarriiaattiioonnss

CCoommmmeennttss These questions can help assess the effects of public relations and media advocacy efforts,when compared over time and referencing periods of campaign activity. Results should becross referenced with news media tracking to better understand how people’s perceptionscorrespond to actual reporting.

CCoonnssttrruucctt ((dd)) SSaalliieennccee ooff tthhee AAnnttii--TToobbaaccccoo MMeeddiiaa MMeessssaaggee

MMeeaassuurree 11:: AAdduulltt “This ad said something important to me. Would you say you… (Strongly agree, Agree,Disagree, Strongly disagree, Have no opinion, Don’t know?”“After seeing this ad, did you talk to anyone about not smoking?” (Yes, Maybe, No)

SSoouurrccee Wakefield et al., 2003b

VVaalliiddiittyy Established validity.

VVaarriiaattiioonnss This question set focuses on not smoking. Depending on the content and purpose, othertopics could be inserted in place of “not smoking.”

CCoommmmeennttss This type of question is used following the respondent’s description of a specific ad to gaugewhether the respondent found the ad to be salient to his or her situation, and whether thead prompted the respondent to think more about the topic.

MMeeaassuurree 22:: YYoouutthh aanndd “On a scale from 1 to 5, where 1 means you don’t like this ad at all and 5 means you like the yyoouunngg aadduulltt ad very much, how much do you like this ad?” (One, Two, Three, Four, Five)

“Would you say the ad grabbed your attention?” (Yes, No)

“Did you talk to your friends about this ad?” (Yes, No)

SSoouurrccee Legacy Media Tracking survey (LMTS; http://americanlegacy.org)




341

typically commercial advertisingand marketing campaigns, andthey are competing for salience,relevance, and resonance withother efforts to promote beha-viours and norms and control theterms of the policy debate. Thesuccess or failure, and the relativeimpact, of a public communicationcampaign will be dependent tosome degree on what is going onin the larger information environ-ment. This is particularly importantfor public communication cam-paigns focused on tobacco controlissues. Tobacco and phar-maceutical companies use thesame communication strategies,

including paid media, publicrelations, media advocacy, and (tosome extent) community action topromote their products andperspectives. Monitoring andunderstanding the larger infor-mation environment allows publiccommunication campaigns toadapt strategies to respond to orreflect the realities of thisenvironment and better under-stand and document the chal-lenges and constraints thatthreaten the success of a publiccommunication effort.

Tobacco and pharmaceuticalcompany communication cam-paigns can be monitored with

many of the same tools andindicators as a public commu-nication campaign. However, keysteps, processes, and informationwill be unavailable to publiccommunication campaign plan-ners and evaluators, such as theexact target and objectives of thecampaign. For example, the targetmay appear to be youth, but isactually voting adults; it mayappear to be smokers, but isactually policy makers; it mayappear to be concerned adults,but is actually potential membersof the jury. The objectives mayappear to be preventing youthinitiation or promoting adult

VVaalliiddiittyy Established validity. Thrasher et al., 2006b.

VVaarriiaattiioonnss “Tell me how much you agree or disagree with the following statement: This ad is convincing.Would you say you… (Strongly agree, Agree, Disagree, Strongly disagree, Have no opinion,Don’t know?”

“Would you say the ad gave you good reasons not to smoke?” (Yes, No)

“Would you say the ad makes you question the motives of cigarette companies?” (Yes, No)

“Did you talk to your friends about this ad?” (Yes, No)

From the Legacy Media Tracking Survey

CCoommmmeennttss The selection of appropriate questions for measuring salience depends on what the studyis most interested in understanding. Question sets, as opposed to individual questions, aretypically necessary to understand this construct. The examples provided here measureslightly different issues: was the message noticed and did it “create a buzz” versus did themessage impart information that was integrated into the respondents thinking on the topic.

LMTS: Legacy Media Tracking Survey ITC: International Tobacco Control Policy Evaluation StudyGATS: Global Adults Tobacco SurveyGYTS: Global Youth Tobacco Survey




342

CCoonnssttrruucctt:: ((aa)) HHeeaalltthh RRiisskkss ooff TToobbaaccccoo CCoonnssuummppttiioonn

MMeeaassuurree 11:: AAwwaarreenneessss ”To what extent, if at all, has smoking damaged your health?” (Not at all, Just a little, A fairAAdduullttss amount, A great deal)

””HHooww wwoorrrriieedd aarree yyoouu,, iiff aatt aallll,, tthhaatt ssmmookkiinngg will ddaammaaggee your hheeaalltthh iinn tthhee ffuuttuurree??”

”TToo wwhhaatt eexxtteenntt,, iiff aatt aallll, has ssmmookkiinngg lloowweerreedd your qquuaalliittyy ooff lliiffee? How worried are you, ifat all, that smoking will lower your quality of life in the future?”




CCoommmmeennttss

MMeeaassuurree 22:: AAwwaarreenneessss “Do you think cigarette smoking is harmful to your health?” (Definitely not, Probably not,YYoouutthh Probably yes, Definitely yes)




CCoommmmeennttss

MMeeaassuurree 33:: KKnnoowwlleeddggee “I am going to read you a list of health effects and diseases that may or may not be causedAAdduullttss by smoking cigarettes. Based on what you know or believe, does smoking cause the

following (Yes, No to each question):• heart disease in smokers• stroke in smokers• impotence in male smokers• lung cancer in smokers• lung cancer in nonsmokers from secondhand smoke?”



VVaarriiaattiioonnss As far as you know, are each of the following chemicals included in cigarette smoke? (Yes,No)• cyanide• mercury• arsenic• carbon monoxide?”

From the International Tobacco Control Policy Evaluation Survey

Table 5.45 Measures to Assess Population Level Awareness and Knowledge of Public CommunicationCampaign Messages



343

CCoommmmeennttss Depending on the purpose of the campaign and the selected key messages, questions willneed to be modified to be relevant. As a baseline indicator, prior to campaign implementation(or as general surveillance), these measures estimate population level knowledge of healthrisks (but not perceptions of personal risk).

MMeeaassuurree 44:: KKnnoowwlleeddggee “It is safe to smoke for only a year or two, as long as you quit after that? Would you say you…YYoouutthh Strongly agree, Agree, Disagree, Strongly Disagree, No opinion?”

SSoouurrccee Legacy Media Tracking Survey (LMTS, (http://americanlegacy.org)


VVaarriiaattiioonnss “Do you think it is safe to smoke for only a year or two as long as you quit after that?”(Definitely not, Probably not, Probably yes, Definitely yes)

From the Global Youth Tobacco Survey CCoommmmeennttss

MMeeaassuurree 55:: AAddddiiccttiioonn “Do you consider yourself addicted to cigarettes?” (Not at all, Yes–somewhat addicted, Yes–AAwwaarreenneessss AAdduullttss very addicted)

SSoouurrcceess The ITC Project, 2007



CCoommmmeennttss

MMeeaassuurree 66:: AAddddiiccttiioonn If you started smoking regularly, do you think you could stop smoking anytime you wanted?” AAwwaarreenneessss YYoouutthh (Definitely yes, Probably yes, Probably not, Definitely not, No opinion)



VVaarriiaattiioonnss “Once someone has started smoking, do you think it would be difficult to quit?” (Definitelynot, Probably not, Probably yes, Definitely yes)

Source: Global Youth Tobacco Survey

CCoommmmeennttss




344

CCoonnssttrruucctt ((bb)) HHeeaalltthh RRiisskkss ooff TToobbaaccccoo SSmmookkee EExxppoossuurree

MMeeaassuurree 11:: AAwwaarreenneessss “Do you think that breathing smoke from other people's cigarettes is… Very harmful to one'sAAdduullttss health, Somewhat harmful to one's health, Not very harmful to one's health, Not harmful at

all to one's health, Don’t know/Not sure?”

SSoouurrccee GATS, 2007

VVaalliiddiittyy


CCoommmmeennttss The perception that environmental tobacco smoke (ETS) is harmful can be an importantfactor for gauging public support for tobacco control efforts. This question also can be anindicator of the effects of ETS education efforts.

MMeeaassuurree 22:: AAwwaarreenneessss “Do you think the smoke from other people’s cigarettes is harmful to you?” (Definitely not, YYoouutthh Probably not, Probably yes, Definitely yes)




CCoommmmeennttss The perception that environmental tobacco smoke (ETS) is harmful can be an importantfactor for gauging public support for tobacco control efforts. This question also can be anindicator of the effects of ETS education efforts.

MMeeaassuurree 33:: KKnnoowwlleeddggee “Would you say that breathing smoke from other people's cigarettes causes… (Yes, No to AAdduullttss each question)

[RANDOMIZE ORDER]…Lung cancer in adults…Heart disease in adults…Colon cancer in adults…Respiratory problems in children…Sudden infant death syndrome?”

SSoouurrcceess CDC Adult Tobacco Survey; 1987 National Health Interview Survey(http://www.cdc.gov/nchs/nhis.htm)




345



CCoommmmeennttss These items can gauge the level of public understanding of the health effects of tobaccosmoke on nonsmokers. Colon cancer is included in this series as an indicator for "over-reporting" in order to estimate the possible magnitude of over-reporting.

MMeeaassuurree 44:: SSuuppppoorrtt ffoorr • In the indoor dining area of restaurantsPPoolliiccyy AAdduullttss • In indoor shopping malls

• In public buildings• In bars and cocktail lounges• In day care centers• In indoor sporting events and concerts“… do you think that smoking should be allowed in all areas, some areas, or not allowed atall?”




CCoommmmeennttss Programmatic focus and activities, the goals of the communication campaign, and the localjurisdiction will determine which environments need to be included in the survey and whetheradditional environments are added. Such questions provide information on attitudes towardsrestrictions on exposure to secondhand smoke; a measure of social norms.

MMeeaassuurree 55:: SSuuppppoorrtt ffoorr “Are you in favour of banning smoking in public places (such as restaurants, buses, PPoolliiccyy YYoouutthh “streetcars, trains, in schools, on playgrounds, in gyms and sports arenas, discos)?” (Yes,

No)




CCoommmmeennttss

CCoonnssttrruucctt:: ((cc)) BBeenneeffiittss ooff QQuuiittttiinngg

MMeeaassuurree 11:: AAdduullttss “How much do you think you would benefit from health and other gains if you were to quitsmoking permanently in the next 6 months?” (Not at all, Slightly, Moderately, Very much,Extremely)






346

VVaarriiaattiioonnss “If a person has smoked a pack of cigarettes a day for more than 20 years, there is littlehealth benefit to quitting smoking.”

(Strongly agree, Agree, Disagree, Strongly disagree, Don’t know/Not sure)

Used by CDC Adult Tobacco Survey; COMMIT evaluation

CCoommmmeennttss Recognition of the health benefits of cessation may be an important determinant of quitattempts, and an early indicator of the effects of health education efforts.

MMeeaassuurree 22:: “What was the main reason you decided to stop smoking?” (SELECT ONE RESPONSEYYoouutthh ONLY)

a. I have never smoked cigarettesb. I have not stopped smoking c. To improve my healthd. To save moneye. Because my family does not like it f. Because my friends don’t like itg. Other




CCoommmmeennttss Recognition of the health benefits of cessation may be an important determinant of quitattempts, and an early indicator of the effects of health education efforts.

MMeeaassuurree 33:: “Are you aware of assistance that might be available to help you quit smoking, such asAAwwaarreenneessss ooff telephone quitlines, local health clinic services, and…?” (Yes, No)SSppeecciiffiicc RReessoouurrcceess



VVaarriiaattiioonnss “In the last month, that is, since [date], have you noticed any advertisements for stop-smoking medications?” (Yes, No)

Used by the International Tobacco Control Policy Evaluation Survey

CCoommmmeennttss The “?” refers to (and should be replaced by) locally specific help promoted in the specificcommunication campaign. Awareness of smoking cessation resources increases thelikelihood that smokers will make quit attempts. Information on the reach of interventionsenables states to assess and improve the delivery of available resources.




347

CCoonnssttrruucctt ((dd)) TToobbaaccccoo IInndduussttrryy –– AAwwaarreenneessss aanndd KKnnoowwlleeddggee

MMeeaassuurree 11:: AAdduullttss “I am going to read you some statements about tobacco companies. Please tell me if youstrongly agree, agree, neither agree nor disagree, disagree or strongly disagree with eachof the following statements:

• Tobacco companies should be allowed to advertise and promote cigarettes as they please.• Tobacco products should be more tightly regulated. • Tobacco companies can be trusted to tell the truth about the dangers of their products.• Tobacco companies should take responsibility for the harm caused by smoking.• Tobacco companies have tried to convince the public that there is little or no health riskfrom secondhand smoke.• The government should do more to tackle the harm done by smoking.• The government doesn’t really care about people smoking because it makes so muchmoney from tobacco taxes.”




CCoommmmeennttss Recommended for use as single items when space is limited: • Tobacco companies can be trusted to tell the truth about the dangers of their products.• Tobacco companies have tried to convince the public that there is little or no health risk

from secondhand smoke.

MMeeaassuurree 22:: YYoouutthh “People have different views about the issue of smoking and cigarette companies. Howmuch do you agree or disagree with the each of the following (strongly agree, agree,disagree, strongly disagree):(RANDOMIZE ORDER)• Cigarette companies should have the same right to sell cigarettes as other companies

have to sell their products. Would you say you… strongly agree, agree, disagree orstrongly disagree?

• Cigarette companies lie.• Cigarette companies deny that cigarettes cause cancer and other harmful diseases.• Cigarette companies deny that cigarettes are addictive.• Cigarette companies have done some really bad things.• Cigarette companies try to cover-up all the bad things they have done.• I would not work for a cigarette company• The people who run cigarette companies know what they are doing is wrong• No other companies act as badly as cigarette companies.• I would like to see cigarette companies go out of business.• Cigarette companies target teens to replace smokers who die• Cigarette companies get too much blame for young people smoking.• Anti-smoking advertisements are no more honest than cigarette ads.• Cigarette companies should have the same right to make money as any other type of

company.• The government should let companies sell whatever they want.• Cigarette companies try to get young people to start smoking• Cigarette companies target minority groups.”



348

cessation, but the actual objec-tives are to cast the company in asympathetic light, change adultopinions about the culpability ofthe company in promoting tobaccouse, and reinforce policy makeropinions about the company associally responsible. Pharma-ceutical company campaigns maybe more transparent than tobaccocompany campaigns; that is,pharmaceutical campaigns thatappear to promote the use of aparticular cessation medicationmay be attempting to do exactlythat. Public communication cam-paign planners may want torespond directly to tobaccocompany campaigns by coun-tering or exposing the mainpurpose of tobacco companymessaging in their own publiccommunication campaigns, andmay want to build on, reinforce, oravoid direct competition with

pharmaceutical ads promotingproven cessation strategies. Whiletobacco and pharmaceutical com-panies have vastly greaterresources to invest in marketingand communication, frequentlypublic health programmes andgovernmental and nongovern-mental organisations have accessto channels that are off limits totobacco companies (e.g. tele-vision and radio). Publiccommunication campaign plan-ners should avoid directcompetition with tobacco com-panies, and instead utilize toolsand strategies that give publiccampaigns the advantage (e.g.electronic media not available totobacco companies and com-munity action that exposes thehuman face of the tobaccotragedy).

To monitor the largerinformation environment, public

communication campaign plannersand evaluators may includeindicators of awareness of andreceptivity to tobacco and phar-maceutical company advertisingon the same population surveysused to monitor campaignindicators (Farrelly et al., 2002,2003b), as well as attitudestoward, salience of and per-ceptions, beliefs and behavioursassociated with exposure to thetobacco or pharmaceutical com-pany campaign. Reach andfrequency of these campaignsmay be gleaned, imperfectly, byidentifying print advertising andcalculating impressions, andmonitoring the airwaves for theappearance of ads and calculatingexposure based on observationsof placement. Tobacco andpharmaceutical company publicrelations and media advocacyefforts may be monitored through


VVaalliiddiittyy Established validity. Hersey et al., 2005; Thrasher & Jackson, 2006.


CCoommmmeennttss A five item scale measuring perceptions of the tobacco industry, based on “Cigarette companieslie,” “Cigarette companies try to get young people to start smoking,” “I would like to see cigarettecompanies go out of business,” “I would not work for a cigarette company,” and a fifth item“How much do you like cigarette companies?” (5 point scale: I like them a lot [1] to I don’t likethem at all [5]) showed small but significant improvement following the introduction of anti-tobacco industry media campaigns in selected US states (Hersey et al., 2005).

LMTS: Legacy Media Tracking Survey ; ITC: International Tobacco Control Policy Evaluation Study; GATS: Global Adults TobaccoSurvey; GYTS: Global Youth Tobacco Survey; COMMIT: The Community Intervention Trial for Smoking Cessation



349


the same news media trackingsystems and content analysisundertaken to monitor imple-mentation and outcomes of thepublic communication campaign(National Cancer Institute, 2005,2006). While exact quantificationof tobacco and pharmaceuticalcompany campaigns may beunnecessary (or impossible) toobtain, a realistic understanding ofthe content and purpose of thesecompeting messages is essentialto crafting a meaningful andrelevant public communicationcampaign that will be effective in acluttered and contentious infor-mation environment.

SSuummmmaarryy aanndd rreeccoommmmeennddaa--ttiioonnss

This section provides a frameworkfor developing, implementing, and

evaluating public communicationcampaigns. These multicom-ponent interventions seek toimprove awareness and know-ledge of tobacco-related issueswith the intention of promotingindividual behaviour change andsupport for and progress towardpolicy and social change. Thepurpose of evaluating thesecampaigns is to inform thedevelopment of effective cam-paigns, to identify and correctproblems while the campaign is inprocess, and document the publichealth impact of the campaign.Core methods include testingcampaign messages during thedesign phase, monitoring thereach of the campaign duringimplementation, and assessingcore constructs, including aware-ness, knowledge, attitudes,beliefs, and support for policies

and tobacco-related behaviourchange. The measures describedhere, like the campaigns them-selves, need to be customized tothe specific content, purpose, andmessage of the communicationeffort being implemented.

Regardless of the results of thepublic communication campaign(and particularly if it failed to showresults), evaluations should bemade publicly available. A systemto collect and document campaignresults would enhance ourunderstanding both of how publiccommunication campaigns workand how to make them better.


Documents

5.6 Measures to assess the impact of anti-tobacco public ......Figure 5.32 WHO FCTC Article 12: Education, communication, training and public awareness section5.6janvier12102:Layout