Message Framing for Alcohol Awareness

Sip N’ Paint into Alcohol Awareness

Reconfiguring the thought behind a behavior is imperative when trying to change it. Message framing is an impactful way of influencing behavior

(Gallagher & Updegraff, 2012)

. Within message framing there are two different subframes, loss-frame and gain-frame. When a message is proposed in a loss-frame, it will focus on the harmful or negative side of doing or not doing a specific behavior (Gallagher & Updegraff, 2012). The message could be framed to show the bodily harm a substance can cause or the harmful effects it could have on individuals and students. This works like a scare tactic, in the way that it shows people the negative side of the issue. The other form of message framing, gain-frame, deals with the positive effects of either doing a specific behavior or not doing a specific behavior (Gallagher & Updegraff, 2012). This framing technique could be used to show the benefits of either performing a beneficial behavior or stopping a nonbeneficial behavior.

The researcher will apply these two framing techniques, gain- and loss-frame, to alcohol consumption and attitudes about future alcohol consumption in students at Youngstown State University (YSU). Colleges are heavily associated with drinking, especially binge drinking (NIAAA, 2018). Many students may not know the health risks behind drinking alcohol. These messages may teach students the harmful effects of binge drinking and the positive effects of not binge drinking. In the experiment, the researcher will explore the impact of differing frames on student alcohol consumption.

Previous Research

Message framing was developed from Prospect Theory (Gallagher & Updegraff, 2012). The theory argues that when faced with a decision between an outcome of little risk or an outcome of greater risk, a person will make their decision based on the way the two messages are framed (Detweiler, Bedell, Salovey, Pronin, & Rothman, 1999). Detweiler and colleges (1999) studied how individuals responded to health messages. They found that a gain-framed message was more influential to promote health behaviors perceived to be slightly risky. On the other hand, loss-framed messages were better to persuade an individual to perform a riskier health behavior (Detweiler et al., 1999).

In a recent study, students read either a gain- or a loss-framed message on the social or health consequences of drinking

(Kingsbury, Gibbins, & Gerrard, 2015)

. There were four groups: gain/loss-frame social and gain/loss-frame health consequences of drinking. After reading the excerpts that described the health/social consequences, the participants took a drinking interest survey. From these surveys the researchers found that, for the social consequences, a loss-frame was more effective in reducing heavy drinking intentions than the gain-frame. For the health consequences, the researchers found that heavy drinking intentions were lowered more for a gain-frame than a loss-frame. Lastly, the researchers found that the participants who had a previous history of heavy drinking were more susceptible to the test, regardless of the framed message. This meant that the participants with a history of heavy drinking had lowered drinking intentions compared to the other participants without a history of heavy drinking (Kingsbury et al., 2015). The participants with a history of heavy drinking reported lower intentions of drinking regardless of the frame of message they received.

Another study that supports the hypothesis being researched focused on willingness to exercise. In this study, the participants were split up into four groups: overweight gain/loss and normal-weight gain/loss

(Kozak, Nguyen, Yanos, & Fought, 2013)

. The groups then received messages and attended an exercise instruction session depending on what groups they were assigned. The researchers found that for the gain-frame message, the overweight/obese group was the only group to show any difference in behavior. The gain-frame message being the most influential on exercise behavior shows, once again, that gain-frame messages lead to a higher level of influence (Kozak et al., 2013).

Van ’t Riet and colleagues (2010) conducted a study where he and his research team studied the persuasive effects of combining different affects, either positive or negative, to gain- and loss-framed messages. The reason the researchers were interested in negative or positive affect is because they thought it would put the participants in a state of “action readiness.” Previously, they found that positive affect was related to gain-framed messages due to the beneficial outcomes the messages show, while the loss-framed messages were linked to negative affect due to their cost-related framing. In the first experiment, the researchers gave questionnaires to college students during their lunch time. On the top of every page there was a statement that the participants had to agree or disagree with and write their level of positive or negative affect. There were loss- and gain-framed questionnaires. They found that the gain-framed messages elicited a higher level of information acceptance and that the gain-framed messages were reported to having a higher positive affect than the loss-framed messages. The second experiment focused on reducing salt intake and they, once again, found that gain-framed massages influenced a greater amount of change in intent (Van ‘t Riet, Ruiter, Werrij, Candel, & de Vries, 2010).

In Van ‘t Riet’s further research, he studied the difference between message framing and perceived risk

(Van ’t Riet et. al., 2014)

. The risk-framing hypothesis argues that the perceived risk is important when trying to persuade with different famed messages. With a perceived higher risk, a loss-framed message would be more persuasive, while a perceived lower risk would pair better with a gain-famed message (Van ‘t Riet et al., 2014). The participants were given the descriptions of different lotions. There was a gain and loss-frame massage for each lotion. The first two lotions were grouped since one had temporary side effects and the other did not. The other two lotions were grouped since one was supposed to prevent skin cancer while the other was supposed to detect it. The researchers were measuring the participants intent to use the lotion. To find this, they asked the participants how likely they would be to do a number of things relating to the product, like purchasing it or asking their doctor about it (Van ’t Riet et al., 2014). While none of their results were statistically significant, there was a trend in the data that showed the opposite of what the risk framing hypothesis would suggest. The gain-frame was slightly more influential for the high-risk messages, while the loss-frame was slightly more influential for the low-risk messages.

In this next study, the researchers wanted to decrease the amount of binge drinking on their college campus

(de Graaf, van den Putte, & de Bruijn, 2015)

. The researchers conducted a three-wave study where the participants were questioned about their attitudes, intentions, and behaviors towards responsible drinking. The first wave measured baseline attitudes, intentions, and behaviors towards responsible drinking. The participants were given the gain or loss-framed message during the second wave of the study, two weeks after wave one. Lastly, for wave three, the participants had to once again report their attitudes, intentions, and behaviors towards responsible drinking. The results showed that a gain-frame message led to more positive attitudes and intentions toward responsible drinking. For participants with high issue involvement loss-framed messages led to more positive attitudes and intentions toward responsible alcohol use (de Graaf, van den Putte, & de Bruijn, 2015).

In 2013, researchers studied counterfactual thinking and message framing and how the messages can affect binge drinking behavior in college students. Counterfactual thinking is “the process of mentally undoing the outcome of an event by imagining alternate antecedent states”

(Baek, Shen, & Reid, 2013)

. The researchers found in previous studies that this counterfactual thinking can help message framing influence behaviors. The authors describe two main types of counterfactual thinking: additive and subtractive. Additive counterfactual thinking is focused around promotion while subtractive counterfactual reasoning is focused around prevention. The participants first completed a cognitive task designed to induce the different types of counterfactual thinking. The participants then viewed either a gain or a loss-framed anti-binge drinking Public Service Announcement (PSA). Finally, the participants ended by completing a questionnaire. The researchers found that the two message frames influenced the participants similarly on attitude toward binge drinking by type of counterfactual thinking. Additionally, the gain-framed message resulted in lower binge drinking intention for subjects who engaged in additive counterfactual thinking (Baek, Shen, & Reid, 2013).

Message framing and message polarity were incorporated in a study attempting to get children to eat healthier

(Wyllie, Baxter, & Kulczynski, 2015)

. Message polarity is linked to psycholinguistics because it deals with words that may subtly give the impression of affirmation or negation, meaning that the words would either give the impression of continuing/starting a behavior or stopping/not doing a behavior. The researchers wanted to determine if phrasing messages in a way to convey affirmation or negation would be better in influencing the participant to eat healthier. The participants were asked to indicate the degree to which they enjoyed eating healthy foods. The children were then shown an image of a child eating a slice of watermelon with a different caption. The captions were framed and polarized with either negation or affirmation. After being shown the image the participants then reported their attitude toward fruit and whether they would try to eat more fruit in the future. Results showed that children reported a more positive attitude toward a PSA that was gain-framed. They also showed that the PSAs with the affirmation linguistic devices rated higher than the negation linguistic devices (Wyllie, Baxter, & Kulczynski, 2015).

In the Pavey and Churchill (2014) study, they focused on the possible effect of different priming tasks on participants’ intentions to avoid high-calorie snacks and their resulting behavior involving snacks. Self-determination Theory suggests that autonomy, the experience of voluntarily engaging in a behavior based on personal interests, is a basic psychological need, which when frustrated can lead to maladaptive psychological functioning and lower well-being (Pavey & Churchill, 2014). Those with high levels of autonomy are likely to exhibit high levels of autonomous motivation and intentions to reduce a certain health behavior. The researchers found that the persuasiveness of gain-framed, or loss-framed messages depends on the autonomy the individual possesses. For participants primed with autonomy, the gain-framed message was more effective in promoting the avoidance of high-calorie snacks, compared to the loss-framed message

(Pavey & Churchill, 2014)

. This study supports the researcher’s hypothesis that gain-framed messages will have a greater influence on the behavior of the individuals who attend the Sip N’ Paint event.

In another study, researchers attempted to evaluate the effects of congruence of measurement and message on binge drinking behavior. Participants received these messages with a set of items that were used to measure changes in intervening variables such as attitudes and intentions. Depending on the wording of the post-message items, there could either be congruence or incongruence of message and measurement. Thus, a person might be given a loss-framed message and either approach (engaging in) or avoid (refraining from) the measurement items (Hutter, Lawton, Pals, O’Connor, & McEachan, 2015). When an engage measure, such as asking a question, is followed by a loss-framed message, it can lead to a congruous pair. A congruous pair was more likely to lead the participant to intend to engage in binge drinking behavior after reading the loss-framed message. However, when an engage measure followed a gain-framed message, an incongruent pair was created. This means that those who read the gained framed messages, about how refraining from binge drinking is better for your health, did not have as high of an intent to binge drink (Hutter et al., 2015).

The last article of focus is a large meta-analysis that reviewed 94 studies to compare the influential power of gain and loss-framed messages

(Gallagher, K., & Updegraff, J., 2012)

. It was found in the analysis that gain-frame messages were much more influential in encouraging health protective behaviors. The analysis also found no significant persuasive effects when looking at attitudes or intentions.


There have been many studies involving message framing and health behaviors, but there are few on the immediate influence the messages may have on the participants. In the proposed experiment, the researcher will examine the number of drinks consumed between individuals who are exposed to a gain/loss-frame message or no message at all. This will show if there is an immediate difference in intent to drink. In the United States, 80% of college students participate in drinking alcohol and 40% engage in binge drinking (Pavey & Churchill, 2014). According to the National Institute on Alcohol Abuse and Alcoholism (NIAAA), binge drinking is consuming enough alcohol, in a two-hour period, that raises one’s blood-alcohol concentration (BAC) over 0.07. For men it may take five drinks, while women may only need four to raise their BAC high enough

(NIAAA, 2018)


These messages could inform college students about the effects of excessive drinking in the most effective way. As a result, this could influence the student body to become healthier.

The researcher hypothesizes that the messages, either gain or loss, will lower future intent to drinking compared to the control group. This will be measured by giving the participants a post-test survey. The researcher further hypothesizes that the gain-frame will result in greater decreases in intent of future drinking than the loss-frame. This is hypothesized because information from previous studies have shown that gain-frame messages result in a higher rate of change in behavior compared to the loss-frame and the control group.

Method and Design


Traditional college students, 21 years or older, who attend a Sip N’ Paint event held at Youngstown State University (YSU). This event will not be open to the public, so all participants will be students from the university. The participants will have signed up for the event ahead of time via an online entry system. This online sign up list can then be cross-referenced through YSU’s database to ensure the students are 21 or older. To lessen liability concerns, the event will be hosted by a club/organization that has a faculty advisor who can be present during the event. The organization will also recruit the participants to allow for the researcher to stay as anonymous as possible. The organization will release announcements to the YSU campus advertising the Sip N’ Paint event. The organization will also sit in the Kilcawley Center spreading the world about the event and getting participants to sign up.

On the online form, there would be a brief message that the participant will agree to before they will submit the form. If they sign up in the Kilcawley center, the club/organization members will ask them to sign a short, informed consent document. The document, on both the online and in-person forms, will say that the participant is comfortable with the researcher collecting data on alcohol consumption throughout the event and that there will be a post-test questionnaire following the Sip N’ Paint event. It will also explain that all data collected will be completely confidential and no names will be reported in the final research study results. If they do not wish to sign the informed consent sheet, they can still participate in the event, but their data will not be collected. The informed consent willl also have a statement on it say that if the researcher or any others think the participant is too intoxicated to make it home safely, they reserve the right to call a taxi in order to transport the participant home safely.


There will be 15 circular tables spread out around the room with six chairs placed at each table. There will be different messages on some of the tables that will have either a gain- or loss-framed message on them. There will also be tables without message on them to act as the control group. There will be five tables in every group, five tables with gain-framed messages, five with loss-framed messages, and five tables without any messages. The messages will be obtained from a previous experiment on the effects of alcohol on one’s health (Kingsbury, Gibbins, & Gerrard, 2015). In the previous experiment, the researchers used both social and health focused gain/loss-framed messages and split their participants into four separate groups. In the current study, the researcher will adapt the test to only include the health gain/loss-framed messages. There may not be enough participants to adequately split up between four group, so two experimental groups will work better for the current study.


This post-test questionnaire would measure the immediate influence on future drinking and be designed to only take five minutes to complete. It would also include easy to comprehend questions, due to the fact that some of the participants will be under the influence of alcohol and may be inebriated. There would be questions pertaining to the demographics of the participants, including age and sex, as well as other questions about the participant’s attitudes and intentions towards alcohol.

The questions will be answered by either circling a number on a Likert scale of whether the participant strongly agrees, agrees, disagrees, or strongly disagrees, or filling in the blank response. The only fill in the blank questions will be the number of drinks the participant consumed throughout the course of the event and the age of the participant. Some of the questions that could will be asked are, “I will drink alcohol this upcoming weekend,” “Growing up, there were members of my family that drank alcohol more than twice a week,” “When I am offered several free alcoholic drinks, I drink more than usual,” and “I drink alcohol as a way to reduce stress.” These questions will show how the participant feels about their future drinking habits, the level of involvement they had around alcohol as a child, the availability of alcohol possibly being a reason for binge drinking behavior, and if alcohol is seen as a stress relieving activity among college students. The last two questions will be more beneficial for the college in order to gain insight on the growing binge drinking culture.


The Sip N’ Paint event will be presented in an identical fashion to each participant, except for the fact that there would be gain-framed message, loss-framed messages, or no messages on the tables in the room, depending on what group the participants were in. The 15 tables will be set up with the painting materials and the messages already on them before the students arrive. The researcher will already be in the room at a table filled with other confederates, acting as a participant, to not stand out. The researcher and confederates will not interact with the participants. The participants will randomly enter the room, which will act as the random assignment because there will be no way for the participants to know which table will have which messages, or no messages, until they sit down. The tables will also be spread out so to discourage the participants from talking with other tables and possibly talking about the framed messages.

The participants will enter the room randomly (in order of when they arrived to the event) and will be told that they can sit anywhere they like. The event will begin as soon as everyone has found their seat. A speaker, from the organization that would be running the event, will introduce the event and go over how the event will run. They will talk about how the instructor will move from one step to the next slowly and be able to come around the room and help anyone who is struggling. They also will explain where the alcohol will be located and that they can get up and refill their glass when they want to. There will be a limit of six drinks per participant.

When the event comes to an end the researcher and confederates will pass out the post-test questionnaire and be available for any questions. This process will take no more than five to eight minutes in order to get the essential information and not fatigue the participants since they may be inebriated. The researcher will then speak to the participants, thanking them and once again reminding them that all the information they put down will be confidential and no names will be reported in the final research document. The researcher will also tell the participants that there is a short debriefing form on the last page of the questionnaire that the participants can rip off and take home with them. The participants will then leave their post-test questionnaires on their tables after they are finished and exit the room, thus ending the experiment.

Data Analysis

For the descriptive statistics, the averages (means) and standard deviations of the three groups will need to be calculated in order to compare which group drank the most on average. Since there will be a cap on the amount of drinks the participants can have, there will most likely be no outliers. For the inferential statistics, an ANOVA test will be conducted since the researcher will be analyzing more than two groups, so an f ratio will need to be computed. This test will show the variance between the groups and show if there were statistically significant differences between the amount of alcohol consumed based upon messages placed at the table (gain-frame, loss-frame, or control).


While I feel like this experiment could shed some light on the drinking culture at YSU and possibly other campuses across the county, there could be several limitations or concerns that would impact our results. Getting approval from the IRB would be the first challenge with an experiment like this because the participants, depending on their alcohol tolerance, could be leaving the experiment site intoxicated and unable to drive or walk home safely. To try to alleviate some of the responsibility from the researcher, the club hosting the event would have their faculty sponsor present during the event. The researcher would have a document signed by the faculty sponsor and the president of the club indicating that the researcher would not be responsible for the participant’s actions after they left the event. The participants also would sign an informed consent sheet that would state that the club reserves the right to call the participant a taxi if they are deemed too intoxicated to make it home safely.

Sample size would be another possible concern for this experiment because 90 people in a room together might be logistically impossible for a Sip N’ Paint. There is usually one instructor at an event like this and with 15 tables spread out, the participants might not be able to see or hear the instructor as they try to lead the group. Not being able to see or hear may cause the participants to lose interest or get upset, which may cause them to drink more regardless of what framed messages are on the table. If the number of participants dropped under 60 people, the sample size might not reach solubility and the experiment would not be generalizable.

The last major concern would be the messages themselves. The participants could talk to the different tables and read other table’s messages. If this were to happen, the entire study would collapse because the researcher would have no way of telling which participants saw which messages and if they had an impact on their immediate drinking response or their future intentions towards drinking. Regardless of which message the participants saw, the fact that there is a message telling them to not drink might make them drink more out of spite. This may only happen to a few participants, which is why it would be important to have a large sample size.

Finding the underlying reasons why traditional college students drink is important when researching the current drinking culture. In the setting of a Sip N’ Paint, the students will be more relaxed and more willing to open up about their drinking habits. Drinking can lead to different illnesses later in life and if too much alcohol is ingested at one time, the bodies organs can fail (NIAAA, 2018). In order to understand why college students drink, we must conduct further research.


  • Baek, T. H., Shen, L., & Reid, L. N. (2013). Effects of message framing in anti–binge drinking PSAs: The moderating role of counterfactual thinking.

    Journal of Health Communication



    (4), 442-458.
  • De Graaf, A., van den Putte, B., & de Bruijn, G. (2015). Effects of issue involvement and framing of a responsible drinking message on attitudes, intentions, and behavior.

    Journal of Health Communication



    (8), 989-994.

  • Detweiler, J. B., Bedell, B. T., Salovey, P., Pronin, E., & Rothman, A. J. (1999)

    . Message framing and sunscreen use: Gain-framed messages motivate beach-goers. Health Psychology, 18(2), 189–196.
  • Gallagher, K. M., & Updegraff, J. A. (2012). Health message framing effects on attitudes, intentions, and behavior: A meta-analytic review. Annals of Behavioral Medicine, 43, 101-116.

  • Hutter, R. C., Lawton, R., Pals, E., O’Connor, D. B., & McEachan, R. C. (2015)

    . Tackling student binge drinking: Pairing incongruent messages and measures reduces alcohol consumption.

    British Journal of Health Psychology



    (3), 498-513.
  • Kingsbury, J. H., Gibbons, F. X., & Gerrard, M. (2015). The effects of social and health consequence framing on heavy drinking intentions among college students.

    British Journal of Health Psychology



    (1), 212-220.
  • Kozak, A. k., Nguyen, C., Yanos, B. R., & Fought, A. (2013). Persuading students to exercise: What is the best way to frame messages for normal-weight versus overweight/obese university students?

    Journal of American College Health



    (5), 264-273.
  • National Institute on Alcohol Abuse and Alcoholism. (2018, August). Alcohol facts and statistics. Retrieved November 22, 2018, from

  • Pavey, L., & Churchill, S. (2014). Promoting the avoidance of high-calorie snacks: Priming Autonomy moderates message framing effects.

    Plos ONE



    (7), 1-8.
  • Van ’t Riet, J., Cox, A. D., Cox, D., Zimet, G. D., De Bruijn, G., Van den Putte, B., & … Ruiter, R. A. (2014). Does perceived risk influence the effects of message framing? A new investigation of a widely held notion.

    Psychology & Health



    (8), 933-949.

  • Van ‘t Riet, J., Ruiter, R. C., Werrij, M. Q., Candel, M. M., & de Vries, H. (2010)

    . Distinct pathways to persuasion: The role of affect in message-framing effects.

    European Journal of Social Psychology



    (7), 1261-1276.
  • Wyllie, J., Baxter, S., & Kulczynski, A. (2015). Healthy kids: examining the effect of message framing and polarity on children’s attitudes and behavioral intentions.

    Journal of Advertising



    (2), 140-150.