Message specificity

Thomson et al. (2012) found that messages that were matched to gender and drink were more acceptable. Jarvis and Pettigrew (2013) explored health outcomes and drink driving warnings using a choice task and found that the messages with the greatest utility differed across gender. Jarvis et al. (2015) found gender to be a significant covariate in a choice task. These authors also found a complex pattern of results such that for some drinks (pre-mixed beverages and wine) a negative effect of choice was found with larger font sizes. Further for rum drinkers preceding the statement with “WARNING” had a negative impact on choice.  However for beer drinkers no effects were found.


Conclusion: Gender needs to be considered carefully in the design of warning messages and more research is needed to assess the interaction between type of drink and warning statement. Studies that focus on placing messages on a particular drink need to be cautious in generalizing their results to other types of beverage.


Pettigrew et al. (2014) found in a study comparing 12 warnings that general warnings about cancer versus specific warnings about types of cancer were more believable, convincing and seen as personally relevant. Creyer et al. (2002) tested the USA warning against “Alcohol is a drug” in the USA and Australia and found that the “Alcohol is a drug” warning resulted in greater risk perceptions than the USA warning. Creyer et al.’s results are explained by the associative memory perspective which suggests that because the word drug is associated in memory with negative concepts such as addiction, presenting alcohol as a drug creates a link to these more negative concepts.


Conclusion: More evidence is needed in comparing across types of message taking into account other design features such as the use of signal word. Research so far indicates that general messages might work best in enabling processing of the message but it isn’t clear which messages might work best in encouraging responsible consumption.