Previous publications on the project Elisabet E

Previous publications on the project • Elisabet E. Storvoll, Ingeborg Rossow & Jostein Rise (2010). Alkoholpolitikken og opinionen. Endringer i befolkningens holdninger til alkoholpolitikken og oppfatninger om effekten av ulike virkemidler i perioden 2005 -2009. SIRUS-rapport 1/2010. • E. E. Storvoll, I. Rossow & J. Rise (2010) Changes in public opinion on alcohol policy 2005 -2009. Presented at KBS, Lausanne, 2010.

Public opinion and alcohol policy in Norway: the relation between perceived effectiveness and attitude towards alcohol policy measures from 2005 -09 Jostein Rise Norwegian Institute for Alcohol and Drug Research

Purpose • Address conceptual and theoretical issues using individual level analyses at each survey points and explore – stability of the attitude structure over time, – dimensionality of attitudes

Data and method • 6 nationwide surveys commisioned by The Norwegian Directorate of Health • In each survey new respondents were drawn from Synovate’s web panel • Internet • Response rate= 41 -55% • The analyses comprised 20 -69 -year-olds (N=8256) – Weighted to ensure similar age and gender distributions

Mean scores of attitudes towards various alcohol policy measures according to year: 1 (fully agree) to 4 (fully disagree) 2005 (Aug) 2005 (Dec) 2006 2007 2008 ”Alcohol too expensive” 1. 9 2. 1 2. 2 2. 4 ”Too difficult to buy” 3. 2 3. 3 3. 4 ”Age limits too high” 3. 7 3. 8 ”Wine in grocery stores” 2. 1 2. 2 2. 1 2. 3 2. 4 ”Liquor in grocery stores” 3. 2 3. 3 3. 4 ”Legal advertising” 3. 2 3. 3 3. 4 ”Bloodalcohol limit too strict” 3. 3 3. 4 3. 5 3. 6

Mean scores of beliefs about effectiveness of various measures to reduce harm: 1 (to a very high extent) to 4 (to a very low extent) 2005 (Aug) 2005 (Dec) 2006 2007 2008 2009 ”Age limits” 2. 5 2. 4 2. 6 2. 7 2. 6 ”Legal BAC levels” 1. 9 2. 0 2. 1 ”Rules for serving” 2. 8 2. 6 2. 7 2. 5 2. 6 2. 5 ”Wine, liquor at Vinmonp” 2. 9 2. 8 2. 7 ”High prices/taxes” 3. 0 2. 9 2. 8 ”Information /education” 2. 0 2. 1 2. 0 ”Parents” 1. 6 1. 6

Concept of attitudes • General, relatively enduring evaluation of objects – General: overall summary evaluation – Enduring: represented in long-term memory; some stability over time – Evaluative: reflects good-bad, favourableunfavourable, positive-negative responses

Attitudes • Bipolar dimension: -3 (bad), 0 (neutral), +3 (good) Vary in: Valence (+, -) Extremity or intensity Structural properties (stored in memory)

• Attitudes should be distinguished from beliefs – Beliefs • not evaluative • Estimates of subjective probabilites that an object (a policy measure) will lead to a certain state of affairs (reduce harm) • Building blocks of attitudes (attitude structure)

Attitude strength • Strong association between attitudes and beliefs – More stable – More predictive of behaviour – More resistant to change – Selective information processing

Correlations between beliefs about effectiveness (vertical row) and corresponding attitudes according to year 2005 (Aug) 2005 (Dec) 2006 2007 2008 2009 ”Age limits” -. 17 -. 31 -. 24 -. 27 ”Legal BAC levels” -. 12 -. 20 -. 15 -. 11 ”Rules of serving” -. 33 -. 42 -. 34 -. 37 -. 40 -. 36 ”High prices(tax es” -. 45 -. 50 -. 45 -. 51 -. 45 ”Wine/liqu or at Vinmonp” -. 54 -. 55 -. 56 -. 60 -. 55 -. 58

Conclusion: attitude structure • Attitude towards universal measures are embedded in a firm and stable structure – Sudden and abrupt changes are unlikely

Factor loadings of attitudinal items according to year 2005( Aug) 2005 (Dec) 2006 2007 2008 2009 ”Alcohol too expensive” . 74 . 75 . 74 74 ”Too difficult to buy” . 64 . 66 . 67 . 66 . 73 ”Wine in grocery stores” . 78 . 76 . 75 . 82 . 75 . 76 ”Liquor in grocery stores” . 77 . 75 . 77 . 76 . 75 . 79 ”Legal advertising” . 70 . 72 . 66 . 42 . 67 . 68

Effect indicator model of attitudes I 1 I 2 Latent variable: Attitudes towards alcohol policy measures I 3 I 4 I 5

Effect indicator model of attitudes (factor loadings 2009). 74. 73 Latent variable: Attitudes towards alcohol policy measures . 76. 79. 68

The content of the attitudinal dimension • Universal policy measures – ”Alcohol is too expensive to buy” – ”It is too difficult to buy alcohol” – ”It should be possible to buy wine in grocery stores” – ”It should be possible to buy liquor in grocery stores” – ”It should be allowed to advertise alcohol”

What underlying process may have contributed to the responses of the five indicators? • A speculative idea – People have a representation of the policy implications of the total consumption model (TCM) with its universalistic principles emphasising regulation and control of price and availability measures directed towards the general population

Social representation theory (Moscovici) • Distinction between the reified (world of sceince) and the consensual (world of common sense) universes – The consensual universe is comprised of social representations which are created, used and reconstituted to make sense of everyday life – Shared by a collective group or larger population (consensual) – Generated in communication (interpersonal and media) and social interaction – Abstract notions, ideas and images are transformed into concrete and objective common sense realities

• The proliferation of expert knowledge throughout all sectors of society has made the lay public amateur scientists and more competent consumers • Expert knowledge has become an integral part of mass culture and ultimately what can be regarded as common sense • This process of objectification may also have occurred with the TCM and thus become a collective representation, i. e. a common sense reality

A possible conclusion • If the present analysis makes sense, then the stability of the attitude structure observed in the present study may partly be explained by the social representation of the TCM (or its policy implications) • The observed stability will continue until the TCM is challenged by another model • The challenge has to start in the reified universe and then be transformed into the consensual universe, which obviously will take some time