The effectiveness and efficiency of using normative messages to reduce waste: A real world experiment.

Although they are only home to 16% of the global human population, high-income countries produce approximately one third of the world's waste, the majority of which goes to landfills. To reduce pressure on landfills and natural systems, environmental messaging should focus on reducing consumption. Messages that signal social norms have the potential to influence people to reduce their consumption of comfort goods, such as straws, which are not a necessity for most people. We conducted a randomized field-experiment at a marine park in Portugal to test whether different normative messages reduced visitors' paper straw use when compared to non-normative messages. We found that a message framed around a positive injunctive norm significantly reduced straw use compared to a non-normative message. We estimated that using the message at 17 park concession stands could keep over 27500 straws out of landfills annually and save the park money after two years.

Introduction

Humans have a global waste problem. High-income countries, which are only home to 16% of the world’s human population, generate about 34% of the world’s waste each year [1]. In 2018, European Union countries generated 2337 million tons of waste across all economic activities and households [2]. That is approximately five tons of waste per resident of the European Union. Concerningly, only 54.6% of waste was treated in recovery operations, including 37.9% that was recycled, while the remaining 45.4% was either sent to landfills, incinerated, or disposed of otherwise [2]. In environmental circles, the three R’s: reduce, reuse, recycle have long been a common mantra, with most emphasis placed on recycling [3]. However, recycling alone will not solve our global waste problem [4]. To reduce pressure on natural systems and on landfills, messaging should focus on reducing consumption and keeping waste out of the global trash cycle. This can be accomplished by encouraging people to reduce, reuse and refuse, rather than consume and recycle [5].

Single-use plastics have recently become a hot button issue because of their contribution to waste and plastic pollution [6]. In 2015, a video of a distressed sea turtle with a straw being removed from its nostril went viral and became an emblem of the anti-straw movement [7]. Plastic straws, which are difficult to recycle, have been vilified because they often end up in landfills or in the ocean and other waterbodies [8]. A recent study estimated that only 9% of global plastics that have been produced have been recycled as of 2015 [9]. This has led some governmental bodies, including the European Union, to ban single-use plastics, including plastic straws [10]. Instead of rethinking the need for straws, many companies are now switching to paper and cardboard alternatives [11].

While the switch from plastic to paper straws is more sustainable, exchanging one disposable good for another will not solve all environmental problems [11]. For the majority of users, straws are a type of comfort good, a good that provides some benefit, but is not a necessity [12]. While some groups of disabled people need straws [13], the majority of people could forego the use of straws, reducing their contribution to the world’s waste problem. There is an urgent need to understand how to persuade people to reduce their use of unnecessary comfort goods, such as disposable straws.

Environmental groups are increasingly using insights from behavioral science to nudge people toward more pro-environmental behaviors, including waste reduction [14]. A meta-analysis that reviewed behavior-change interventions related to waste production found evidence supporting the use of defaults and commitments to lower amounts of paper, plastic, and food waste [15]. The use of social norm messaging, however, showed mixed results. While social norms reduced plastic bag use and food waste, only one of the three interventions that used norms to reduce paper waste had a significant effect [15-18]. Further research is needed to understand which types of behavioral interventions can effectively reduce the use of paper products.

Normative messages, or messages that signal social norms, have been widely used to influence human behavior [19, 20]. Social norms are unwritten rules about acceptable behaviors in particular settings [21]. These norms have powerful effects on human behavior because people are driven to conform to local customs [21]. In a field experiment, hotel guests were significantly more likely to reuse towels when they received a message stating that 75% of other hotel guests reuse their towels, than when they received a more generic message that did not signal any norms [22]. Normative messaging has been widely used to promote pro-environmental behaviors, such as reducing residential energy and water use [23, 24] and choosing sustainable transportation [25, 26].

Different types of social norms can be activated through messages. Descriptive social norms describe our perceptions of how other people typically behave [27, 28]. While descriptive norms can promote pro-environmental behaviors, they can also reinforce unsustainable behaviors. Cialdini et al. (2006) tested different normative messages as part of campaign to stop visitors from stealing petrified wood from a protected area. They found that messages that described how much wood other visitors were taking tended to increase theft.

Injunctive social norms refer to our perceptions of how others think we should behave [27, 28]. Messages framed around injunctive norms typically express approval or disapproval of certain behaviors. Cialdini et al. (2006) [29] found that a strong injunctive message (“Please don’t remove petrified wood from the park”), resulted in less theft of petrified wood from a protected area than other message conditions. Similarly, de Groot et al. (2013) [30] found that an injunctive message in a supermarket resulted in significantly lower plastic bag use among shoppers than a non-normative message about the environment.

Moral norms are the rules of morality that people in a certain society or group are expected to follow [31]. Some groups view the protection of the planet as a personal responsibility and a moral obligation [32]. For example, some private landowners in the U.S. feel a moral obligation to prevent the extinction of endangered animals [33] and many households in Sweden feel a moral obligation to recycle [34]. Moral norms have received less attention than descriptive and injunctive norms in tests of environmental messaging and may offer a powerful force for promoting environmental actions [35].

Studies examining the relative influence of different types of normative messages have shown that their influence varies across contexts [36, 37]. Whether a message is negatively worded or positively worded is also likely to influence perceptions and behaviors [29, 38]. For example, a positively worded injunctive message that expresses approval of a behavior might have a different effect than a negatively worded injunctive message that expresses disapproval of the opposite behavior.

While many field-experiments have tested the influence of normative messages on water conservation, energy conservation, and littering, few have examined the extent to which normative messages can influence people to reduce consumption of comfort goods, such as a straws [15, 37, 39]. Additionally, a recent meta-analysis of 91 field-experiments that used social norms to promote pro-environmental behaviors found that only 24 experiments took place in a European country other than the United Kingdom [37]. It is important to study messaging in different countries and contexts because culture may influence how people respond to message frames [40]. We conducted two randomized field-experiments at a marine park in Portugal to test whether different normative messages reduced visitors’ paper straw use when compared to non-normative messages. We hypothesize that normative messages will result in significantly less straws taken than non-normative messages.

Study 1: Pilot experiment

Materials and methods

The two experiments took place at Zoomarine, a marine park located in Algarve, the southernmost region of continental Portugal (https://www.zoomarine.pt/en/) (Fig 1). Zoomarine provides entertainment and environmental education to visitors through a combination of marine-inspired educational exhibits and amusement rides. The pilot study (Study 1) sought to identify which type of social norm messaging (injunctive, descriptive, or moral) most effectively nudged park visitors to reduce their use of paper straws when purchasing beverages from a park concession stand. The full experiment (Study 2) built on the results of Study 1 by testing the most effective social norm messages from Study 1 against a control message. This study was approved by the University of Florida’s Internal Review Board (IRB202002244) and the need for consent was waived. IBM SPSS Statistics 26 was used to calculate descriptive statistics for both experiments to compare the mean ratios of different conditions.

Fig 1

Map of Portugal, which shows where Zoomarine is located.

Reprinted from Zoomarine under a CC BY license, with permission from Zoomarine.

One concession stand in the park was selected for this pilot study and paper straw dispensers and message signs were installed close to the concession stand cashier. We piloted six different message conditions and a control condition over 72 days between July 1st and September 10th, 2018 (Table 1). Each message condition was in place for three days at a time and a control condition (no message) was displayed for three days in between each message condition. All social norm message conditions were on display for two three-day periods over the course of the pilot. Messages were displayed in both English and Portuguese.

Table 1

Messages tested during the pilot experiment.

Condition	Message displayed	Number of days displayed
C—Control	No information	36
(N1)—Negative descriptive social norm	80% of our visitors choose not to use disposable straws with their drinks. The planet thanks you!	6
(N2)—Positive descriptive social norm	80% of our visitors choose to drink directly from the cup or can. The planet thanks you!	6
(N3)—Negative injunctive social norm	Choose not to use disposable straws with your drink. The planet thanks you!	6
(N4)—Positive injunctive social norm	Choose to drink directly from the cup or can. The planet thanks you!	6
(N5)—Negative moral social norm	Protecting our planet is our duty. Choose not to use disposable straws with your drink.	6
(N6)—Positive moral social norm	Protecting our planet is our duty. Choose to drink directly from the cup or can.	6

Data on the number of drinks sold to visitors and the number of paper straws taken by visitors were recorded each day and were used to calculate the ratio of paper straws taken to drinks sold per day.

Results

The experiment ran for 72 days; operator error occurred on 7 of these days, resulting in 65 days of accurate data collection. Operator error means that there were mistakes in the data collection on a particular day that rendered the data unusable. Errors included the implementation of incorrect message signs on a particular day, the temporary absence of straws in a dispenser due to shortages, and errors due to cashier shift turnover. In total, 15,279 drinks were sold over the 65 days and 4,684 paper straws were taken. The ratio of straws taken to drinks sold was calculated for each day and the mean ratio for each condition was calculated (Table 2).

Table 2

Mean ratio of straws taken to drinks sold for six message conditions and a control.

Condition	N° days accurate data collected	Ratio of straws taken to drinks sold x¯ (SD)
C—Control	30	0.347 (0.155)
(N1)—Negative descriptive social norm	6	0.318 (0.129)
(N2)—Positive descriptive social norm	6	0.250 (0.147)
(N3)—Negative injunctive social norm	6	0.215 (0.041)
(N4)—Positive injunctive social norm	6	0.262 (0.097)
(N5)—Negative moral social norm	6	0.272 (0.143)
(N6)—Positive moral social norm	5	0.249 (0.131)

The data did not meet the assumptions for a one-way ANOVA. Two conditions had outliers, as assessed by inspection of a boxplot: the negative injunctive norm condition had a single outlier, while the control condition had three outliers, two of which were extreme outliers (more than 3 box-lengths away from the edge of the box). The results of a Shapiro-Wilk’s test demonstrated that the data were not normally distributed for two conditions: control (p = .001) and negative injunctive norm (p = .034). There was homogeneity of variances, as assessed by Levene’s test for equality of variances (p = .493). Because the data violated these assumptions, a Kruskal-Wallis test was run to determine if there were differences in ratios between the seven conditions. The mean rank of ratios was not statistically significantly different between groups, χ2(6) = 7.971, p = 0.240, ε2 = .125 [41]. This is a moderate effect size. Recognizing that a one-way ANOVA is somewhat robust to violations of these assumptions [42], we also ran a one-way ANOVA and found that there was still no difference in ratios between the seven conditions, F(6,58) = 1.315, p = .265, est. ω2 = .029. This is a relatively small effect size.

Study 2: Full experiment

Building on the results of the pilot, a full experiment was conducted in the summer of 2019. The same concession stand was used for the full experiment, but straws were no longer available in the straw dispenser for free withdrawal due to a larger strategy across the park to nudge visitors to reduce consumption. Visitors now had to ask the cashier for a straw. The message displays were located in a clearly visible area on the counter near the cashier.

The experiment ran for a total of 87 days between June 24th, 2019 and September 20th, 2019 to determine whether different messages had an influence on the ratio of paper straws taken to drinks sold at a concession stand in the park. A message pair using positive and negative injunctive norms was selected for broader testing based on the results of the pilot, which showed that the mean ratio of straws taken to drinks sold was lowest for the negative injunctive social norm condition (; SD = 0.041). Given the lack of a significant result in the pilot study, the choice to move forward with the injunctive message pair was mainly informed by studies that suggest that injunctive norms are more effective than descriptive norms at promoting desired environmental behaviors [28, 29].

Three messaging conditions were tested during the experiment, each in place for 3 days at a time:

N3 Condition: Negative injunctive social norm

Choose not to use straws with your drink.

N4 Condition: Positive injunctive social norm

Choose to drink directly from the glass or can.

Condition: Control

We are changing! 85% of all our disposable materials are already environmentally sustainable.

Messaging conditions were randomly assigned to different three-day periods over the course of the experiment. Data on the number of drinks sold to visitors and the number of paper straws taken by visitors were recorded each day and were used to calculate the ratio of straws taken to drinks sold per day.

Calculation of waste avoided and intervention cost. We estimated the amount of waste that could be avoided through the intervention over the course of a year by comparing the average number of straws taken during the control condition to the average number of straws taken during the positive injunctive norm condition and extrapolating these figures over the 250 days that the park is open on average each year and over the 17 concession stands in the park. We also calculated the cost of running the intervention and experiment, including the fixed cost of staff time to design and promote the intervention, the fixed cost of the displays, and the variable cost of researcher time counting straws, entering data and changing displays each day. We then calculated the cost-benefit ratio for the intervention over time to determine when the park would see a return on investment. We also calculated the cost to avoid one kilogram of waste over time by dividing the net cost of the intervention each year by the number of kilograms of straws avoided each year.

The experiment ran for 87 days; operator error occurred on 10 of these days, resulting in 77 days of accurate data collection. Operator error means that there were mistakes in the data collection on a particular day that rendered the data unusable. In total, 11,346 drinks were sold over the 77 days and 1,597 paper straws were taken. Over the course of the experiment, data were collected for positive injunctive norm message for 25 days, for the negative injunctive norm message for 26 days, and for the control message for 26 days.

A Welch’s ANOVA was conducted to determine if the mean ratio of paper straws taken to drinks sold was different for the different messaging conditions because the data did not meet all the assumptions for a one-way ANOVA. There were no outliers in the data, as assessed by inspection of a boxplot for values greater than 1.5 box-lengths from the edge of the box. The results of a Shapiro-Wilk’s test also demonstrated that the ratio values were normally distributed for the negative injunctive norm condition (p = 0.101) and the control condition, (p = 0.077), while the distribution of ratio values for the positive injunctive norm condition was on the edge of normal distribution (p = 0.05). The assumption of homogeneity of variances was violated, as assessed by Levene’s test for equality of variances (p = 0.004), which led to the use of a Welch’s ANOVA. The mean ratios of paper straws taken to drinks sold were statistically significantly different between the different conditions, Welch’s F(2, 45.39) = 5.85, p = 0.006, est. ω2 = 0.112. This is a relatively large effect size.

The mean ratio of straws taken to drinks sold per day decreased from the control condition (, SD = 0.094), to the negative injunctive norm condition (, SD = 0.076), to the positive injunctive norm condition (, SD = 0.045), in that order (Table 3). A Games-Howell post hoc analysis revealed that the mean decrease from the positive injunctive norm condition to the control condition (-0.062, 95% CI [-0.1116, -0.0114]) was statistically significant (p = 0.013), Cohen’s d = 0.835 (Fig 2). In this case, we reject the null hypothesis and accept the alternative hypothesis that a positive normative message resulted in significantly less straw use than a non-normative message. The mean decrease from negative injunctive norm condition to the control condition (-0.0208, 95% CI [-0.0782, 0.0366]) was not significant (p = 0.659), Cohen’s d = 0.243. The mean difference between the negative injunctive norm and the positive injunctive norm was not significant (0.0407, 95% CI [-0.0017, 0.0832], p = 0.063), Cohen’s d = 0.649.

Table 3

Mean ratio of straws taken to drinks sold for two message conditions and a control.

Condition	N° days data collected	N° straws taken	N° drinks sold	Ratio of straws taken to drinks sold x¯ (SD)
Positive injunctive social norm	25	3888	431	0.111 (0.045)
Negative injunctive social norm	26	3611	548	0.152 (0.076)
Control	26	3847	618	0.172 (0.094)

Fig 2

Results of a Games-Howell post-hoc analysis shows the mean differences between conditions with 95% confidence intervals.

Note. * Denotes a statically significant difference at the 0.05 level between groups.

Calculation of waste avoided through intervention. Over the 26 days that the control message was in place, an average of 23.77 straws were taken per day. Over the 25 days of the positive injunctive norm message was in place, an average of 17.24 straws were taken per day. We estimated the number of straws saved per year at one concession stand by multiplying the average number of straws taken per day for each condition by the 250 days that the park is open on average annually: 5942.5 straws would be taken if the control message was used and 4310 straws would be taken if the positive injunctive norm message was used. If the positive injunctive norm message was used across the 17 concession stands in the park, the park could potentially keep 27752.5 straws out of its waste stream each year. Paper straws weigh approximately 1.1 grams each [43], which means that 30.53 kilograms of trash removed from the waste stream each year.

The experiment had two fixed costs amounting to 685€. This figure includes researcher’s time to design and promote the project (640€) and the cost to produce and design signage (45€). The variable cost of counting straws, entering data, and changing displays amounted to 1.5€ per day at the concession stand. In total, it cost approximately 815.5€ to run the full experiment for 87 days at a single concession stand. To expand the intervention to the other 16 stands in the park, there would be a fixed cost of 480€ (including 30€ per stand for display signage). The total cost in Year 1 would be 1295.50€. Straws cost 0.019€ each, which means that 527.30€ could also be saved on straws annually.

In Year 3, the park would save 286.40€ on the purchase of straws, with a return on investment of 22.11%. The cost of avoiding one kilogram of waste would be 25.16€ in Year 1 and 7.89€ in Year 2. In Year 3, the financial benefits of the intervention would outweigh the costs and the park would save 9.38€ for every kilogram of waste avoided.

Discussion

This field experiment measured actual behavior as well as randomly assigned experimental and control conditions in a public setting. We tested whether normative messages could reduce visitors’ use of paper straws at a marine park and found that the positively worded injunctive message resulted in significantly less paper straw use than the non-normative message and that there was a relatively large effect size. We also found that the intervention could be a cost-effective way to reduce waste and could potentially save the marine park money after only two years.

It is surprising that a positive injunctive norm message was most effective. Past research on environmental messaging suggests that negatively worded injunctive messages may have a greater impact on behavior than positively worded injunctive messages [29]. Similarly, a recent study on the promotion of healthy food consumption found that a negative injunctive message was more effective than a positive injunctive message [38]. Negative information is often afforded greater attention and weight in a person’s consciousness than positive information [44, 45]. Our study, which found the positively framed message to be more effective, suggests that further research is needed on how framing influences the effects of injunctive norms [38]. It is important to acknowledge that we only tried one version of a message for each norm condition and that our sample sizes were limited, which means the randomization process may not have fully balanced all variables (e.g., respondent profile) across all conditions. It is possible that different messages within the different conditions could have influenced behavior in other ways. Framing effects are also likely to vary across cultures and countries. When examining framing effects across seven Central and Western European countries, one study found differences in the way that consumers from different countries, even within the same region, responded to positive and negative message frames [40].

While we did not collect information on participant characteristics, which could help better explain our results, the psychological literature provides some potential explanations as to why the positive injunctive message may have been more effective. One possible explanation is that the negatively worded message may have triggered psychological reactance in some visitors [46]. Reactance is a form of motivational arousal that occurs when someone feels that their personal freedoms are threatened by rules and restrictions [46]. Resistance can occur when injunctive social norms (or proscriptive messages) appear to impinge on personal freedoms and may even lead people to increase the undesirable behavior [47]. While this is speculative, reactance may have decreased the effect of the negative injunctive message, effectively erasing any signal when compared to the control group.

The positive injunctive message may have been more effective because it reminded visitors of their pro-environmental attitudes and values. While some visitors to zoos and aquariums come solely for entertainment, many visitors are interested in learning about animals [48] and about ways to make a difference for conservation [49]. The positive injunctive message may have reminded visitors of their pre-existing environmental attitudes, nudging them to act in a way that is consistent with these attitudes. Self-perception theory suggests that people are generally motivated to maintain consistency in their behaviors and beliefs [50]. Cornelissen et al. (2008) [51] found that cueing behaviors as pro-environmental increased the likelihood that people would engage in those environmental behaviors, particularly if people perceived of themselves as being environmentally responsible. It would be interesting to test similar messages at a concession stand in a setting that has no environmental associations, such as a sporting event or a musical concert.

When comparing the positive and injunctive norm messages, salience bias might also help explain why the negative injunctive message resulted in higher paper straw use than the positive injunctive message [52]. The negative injunctive message included the word ‘straw’ while the positive injunctive message did not. Simply seeing the word could have triggered some people to think of straws and to ask for one [52]. Framing theory suggests that even minor changes in the presentation of an issue can result in significant changes in how that issue is perceived [53]. In future experiments, matching the positive injunctive and negative injunctive messages as closely as possible could help eliminate any influence of word choice and sentence structure on outcomes.

The experimental setting may also have influenced the results. The Focus Theory of Normative Conduct suggests that normative elements are likely to more be effective if they are focal, or salient, when the behavioral decision is being made [27]. In this experiment, explicit normative messages were prominently placed in the behavioral setting, right next to the cashier. The broader saliency of the issue in people’s minds could have also influenced behavioral decisions. Straws have been in the Portuguese news and the international news over the past few years as pressure has mounted for people around the world to reduce their use of single-use plastics [54]. Given the widespread discussion around straws, it is possible that visitors had already formed associations with straws (e.g., using straws is not environmentally friendly) that influenced their behavior.

In the full experiment, the fact that people were required to ask the cashier for a straw, rather than passively taking straws from a dispenser, may have also influenced results. This change also means it is not possible to compare effect sizes between the pilot experiment and the full experiment, which is a limitation of the study. Changes in default options can have a strong influence on behavior [55, 56]. Having to request a straw could also amplify social norm effects because perceived social disapproval can generate strong effects on behavior [57]. A field experiment found that people were more likely to choose reusable takeaway boxes when they witnessed others using a reusable takeaway box [58]. In this experiment, asking for a straw after seeing the signage required visitors to directly flout explicit social norms, both in front of the cashier and in front of other customers in line. Having to ask for a straw may also disrupt automated choices, increasing the likelihood that consumers take the time to think about whether they actually need a straw [59].

Future research at Zoomarine could also explore spatial and temporal variations by carrying out this experiment at more stands for a longer period of time. Studies could also explore how social norms can influence the consumption of other comfort goods, including other single-use products. It would also be interesting to examine the interactions between positive and negative framing and social norms in more detail. Beyond Zoomarine, more real-world experiments should test how environmental messaging can influence people’s environmental behaviors [60] and how normative messages are affected by framing [38].

Costs and benefits of interventions. Moving forward, it is critical that more field behavioral experiments examine the financial efficiency of interventions, as well as their effectiveness [61]. Interventions are more likely to be adopted by companies and governments if they make financial sense. In their review of interventions designed to increase pro-environmental behaviors, Byerly et al. (2018) found that only 15 of the 72 studies they reviewed examined cost-effectiveness. Similarly, few studies have examined the potential cost savings of social norm interventions [61, 62].

We estimated that displaying the positive injunctive message at 17 park concessions stands could keep approximately 27500 straws out of the park’s waste stream each year. After an initial investment of 815.5€ to run the experiment, the only cost to implement this intervention would be the cost of the displays at each concession stand. Our cost-benefit calculation is based on some key assumptions, including the assumption that all concession stands will sell similar numbers of drinks and the assumption that visitation rates and visitor behavior will remain similar. While some events, such as the COVID-19 pandemic, show that this is not always the case, we feel that these estimates are still informative. Accepting these assumptions, our calculations project that the return on investment could be approximately 22% if the intervention ran for three years, and that rate would continue to increase over time. The cost to avoid one kilogram of waste would also drop over time, from approximately 25€ in Year 1 to 8€ in Year 2. The few waste-related interventions that include social norms and cost-benefit calculations indicate that these campaigns are a promising direction for waste reduction. In Port Colborne, Ontario, Canada, a campaign and program to divert organic waste from a landfill had a fixed cost of $269,500 and an ongoing annual cost of $23,000. Even with these costs, the campaign had a four-year pay-back period and a return on investment of approximately 12% over the first ten years [63]. Similarly, a program to reduce household energy consumption found that norms-based messaging could reduce electricity consumption in the average household by over 2% in a randomized control trial [61]. The study calculated the cost effectiveness of the program and showed that it compared favorably to the estimated cost effectiveness of similar energy-efficiency programs. More businesses might be convinced to implement interventions to reduce waste if they understood the potential costs and benefits over time.

Conclusions

Across the globe, there is an urgent need to find strategies to reduce waste production. Our research demonstrates that minor changes in the wording of a normative message can significantly influence behavioral outcomes, moving individuals and companies toward more sustainable practices. While straws may seem like a minor contribution to the waste stream, a report estimated that the countries of the European Union consumed 36.4 billion drinking straws annually [64]. Furthermore, interventions like ours could potentially reduce waste from other comfort goods, such as takeout containers and single-use bags, in ways that not only support environmental sustainability but also make financial sense.

15 Oct 2021

PONE-D-21-25759The effectiveness and efficiency of using normative messages to reduce waste: A real world experimentPLOS ONE

Dear Dr. Neves,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Nov 29 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Valerio Capraro

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

I have now collected two reviews from two experts in the field, whom I thank for their detailed and thoughtful reviews. Both reviewers found the paper interesting and easy to read. However they both expressed a number of concerns that should be addressed in a major revision. Therefore, I would like to invite you to revise your paper following their comments. Needless to say that all comments should be addressed or rebutted. When uploading your revised manuscript, please upload also a response letter containing a point-by-point response to all the reviewers' comments.

I am looking forward for the revision.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please ensure that you include a title page within your main document. We do appreciate that you have a title page document uploaded as a separate file, however, as per our author guidelines (http://journals.plos.org/plosone/s/submission-guidelines#loc-title-page) we do require this to be part of the manuscript file itself and not uploaded separately.

Could you therefore please include the title page into the beginning of your manuscript file itself, listing all authors and affiliations

3. Please clarify in your Methods section that IRB waived the need for consent

4. Thank you for stating the following financial disclosure:

“This work was funded by national funds through Fundação para a Ciência e a Tecnologia (FCT) as part the project CIP - Refª UID/PSI/04345/2020 (Jean-Christophe Giger)”

Please state what role the funders took in the study.  If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

If this statement is not correct you must amend it as needed.

Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.

5. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

6. We note that Figure 1 in your submission contain map images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

1. You may seek permission from the original copyright holder of Figure(s) [#] to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

2. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors conducted two social norm intervention studies aimed at reducing consumption of paper straws at one concession stand in a marine park. The first study was a shorter pilot, with the goal of narrowing down the type of social norm messages that would be included in the larger second study. The second study compared a control message, a positive injunctive norm message, and a negative injunctive norm message. Results indicated that the positive injunctive norm message significantly reduced paper straw consumption compared to the control condition. For each day of data collection, the authors collected a count of the number of drinks that were sold at the concession stand as well as a count of the number of paper straws that were requested. These values were transformed into a ratio and each condition was associated with a single value: the mean of these ratios across all days of data collection in that condition. Finally, the authors provided a cost-benefit analysis that extrapolated the results of this study to determine how much money this social norm intervention would yield the marine park over time.

I thought this manuscript was clearly presented and easy to read. The introduction has a nice overview of relevant literature and the discussion section thoughtfully brings up many other psychological factors that could be contributing to the results.

Major Comments:

The use of ratios obscures differences in base rates that may have varied by condition. For example, were there any differences in the number of drinks that were sold across conditions? Relatedly, was any other information about the day recorded? Number of visitors? Temperature? I’m wondering if it’s possible to rule out environmental factors that may have covaried with condition.

I would be interested in seeing what the data that will be made available upon publication looks like. Will it simply be the averages of the ratios for each condition? Or will the data file include the number of drinks ordered and the number of straws requested for each day in each condition? I think the latter would be more useful for people interesting in further exploring this interesting dataset.

Minor Comments:

Page 2, line 30: “Concerningly, only 37.8% was recycled, while 45.7% was sent to landfills.” What accounts for the remaining 16.5%?

Typo page 2 line 39: “Single-use plastics have recently become a become a hot button issue…” ‘become’ is repeated twice.

Page 8: “A message pair using positive and negative injunctive norms was selected for broader testing based on the results of the pilot, which showed that the mean ratio of straws taken to drinks sold was lowest for the negative injunctive social norm condition (x̄ = 0.214; SD = 0.041).” However, the results of the pilot study indicate that the negative injunctive social norm condition was not statistically different from any other condition (including the control condition). Might be more clear that the choice of the injunctive norms was based on the literature because the pilot study didn’t provide any clear winners.

Describe the nature of the operator error for both Study 1 and Study 2.

Why do you use different statistical tests to answer the same question for Study 1 and Study 2? Study 1 uses a Kruskal-Wallis test to determine if there was a difference in ratios by condition. Study 2 uses a Welch’s F test to determine if there was a difference in ratios by condition. Either use the same test for each or provide a justification for why one is more appropriate for each case.

Page 11: When you inspect the normality of the distributions for each condition, are you looking at the distribution of ratio values over the 25-26 days for each condition? Regardless, make it clear what the values are that you are inspecting.

Page 12 “In this case, we reject the null hypothesis and accept the alternative hypothesis that normative messages resulted in significantly less straw use than non-normative messages.” Should be more specific and say ‘a positive normative message’.

Looks like the biggest intervention happened when the park decided that patrons had to ask for straws (a change that happened between Study 1 and Study 2). Would be good to discuss the effect sizes of the norm intervention in terms of the change in control condition values from Study 1 to Study 2.

In Discussion section, be clear that you only tried ONE version of a message for each norm condition. It’s possible that different messages that fit within the different norm definitions may have impacted behavior differently.

The reactance suggestion in the discussion section would predict that negatively worded normative messages would encourage MORE straw consumption than the control condition. This did not happen.

Page 16: “salience bias might also help explain why the negative injunctive message resulted in higher paper straw use” Be clear that you mean higher than the positive injunctive message, not higher than the control condition.

Reviewer #2: This paper described a simple experiment testing the effectiveness of normative messaging on the use of straws. Overall, it was clear and easy to read, however I felt the paper was missing many details to help us understand what was done, why, and the relevance of the findings. I outline these below.

Main concerns

- Why is the data not provided to the reviewers? Data availability is required to publish in PLOS one and is highly relevant for the review process. Please make this available.

- I would like to see the results of the pilot study below the study description. Understanding what was found in that experiment is important to understand the methods of the main experiment of the paper.

- The authors did not find any difference between conditions in the pilot study

- Why use a Kruskal-Wallace test for the pilot? The non-parametric test was justified for the main experiment, but not the pilot. If it passes parametric assumptions, a two-way ANOVA would be appropriate (except for control condition), where you compare Positive/Negative manipulation, and manipulation type (descriptive, injunctive, social norm). Additionally, why not compare each individual group to the control, as you did in the main experiment? This result section was lacking, in both justifications and statistics.

- All test decisions should be justified through the manuscript

- What is operator error? Describe it in detail (can be in a supplement)

- Effect sizes should be interpreted throughout the manuscript

- Data was collected over time, why was this not analysed? Was the straw to drink ratio consistent over time, or did it increase/decrease? This analysis is necessary to make any argument about effectiveness of the intervention. Without it, the discussion about savings over time is premature.

- Why are there no comparisons to other stands in the park over the same time period?

- The authors state this experiment is important because it is conducted in a European country and has been understudied, yet does not discuss cultural effects on messaging. This should be discussed in the introduction and in the discussion – it is very possible that the effectiveness of the messages (positive/negative) may be related to cultural factors

Minor comments

- Describe the Welch’s F test as non-parametric (line 243)

- Interpret effect size of Welch’s F test (line 245)

- Give actual numbers in Table 3, in addition to what is presented (number of drinks sold, number of straws taken)

- Acknowledge the discussion section on reactance as speculative

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

18 Nov 2021

Reviewer #1:

The authors conducted two social norm intervention studies aimed at reducing consumption of paper straws at one concession stand in a marine park. The first study was a shorter pilot, with the goal of narrowing down the type of social norm messages that would be included in the larger second study. The second study compared a control message, a positive injunctive norm message, and a negative injunctive norm message. Results indicated that the positive injunctive norm message significantly reduced paper straw consumption compared to the control condition. For each day of data collection, the authors collected a count of the number of drinks that were sold at the concession stand as well as a count of the number of paper straws that were requested. These values were transformed into a ratio and each condition was associated with a single value: the mean of these ratios across all days of data collection in that condition. Finally, the authors provided a cost-benefit analysis that extrapolated the results of this study to determine how much money this social norm intervention would yield the marine park over time.

I thought this manuscript was clearly presented and easy to read. The introduction has a nice overview of relevant literature and the discussion section thoughtfully brings up many other psychological factors that could be contributing to the results.

Response 1: Thank you for your comments. We have replied in detail below.

Major Comments:

Comment 1: The use of ratios obscures differences in base rates that may have varied by condition. For example, were there any differences in the number of drinks that were sold across conditions? Relatedly, was any other information about the day recorded? Number of visitors? Temperature? I’m wondering if it’s possible to rule out environmental factors that may have covaried with condition.

Response 1: We allocated the conditions randomly, so this should theoretically not impact our results. However, we recognize that our sample size for each condition is limited and so have added a sentence in the discussion to acknowledge this (Line 406): “It is important to acknowledge that we only tried one version of a message for each norm condition and that our sample sizes were limited, which means the randomization process may not have fully balanced all variables (e.g., respondent profile) across all conditions.”

Comment 2: I would be interested in seeing what the data that will be made available upon publication looks like. Will it simply be the averages of the ratios for each condition? Or will the data file include the number of drinks ordered and the number of straws requested for each day in each condition? I think the latter would be more useful for people interesting in further exploring this interesting dataset.

Response 2: The data we have available for each day includes the number of drinks ordered and the number of straws requested. You can now find the dataset here: https://figshare.com/s/648611d23cb654d0dfcf. The public DOI is: https://doi.org/10.6084/m9.figshare.15134754.v1

Minor Comments:

Comment 3: Page 2, line 30: “Concerningly, only 37.8% was recycled, while 45.7% was sent to landfills.” What accounts for the remaining 16.5%?

Response 3: We have updated these statistics with the latest published numbers from 2018 and have included percentages adding up to 100%. See Page 2, Line 49: “In 2018, European Union countries generated 2337 million tons of waste across all economic activities and households [2]. That is approximately five tons of waste per resident of the European Union. Concerningly, only 54.6% of waste was treated in recovery operations, including 37.9% that was recycled, while the remaining 45.4% was either sent to landfills, incinerated, or disposed of otherwise [2].”

Comment 4: Typo page 2 line 39: “Single-use plastics have recently become a become a hot button issue…” ‘become’ is repeated twice.

Response 4: We have corrected this typo.

Comment 5: Page 8: “A message pair using positive and negative injunctive norms was selected for broader testing based on the results of the pilot, which showed that the mean ratio of straws taken to drinks sold was lowest for the negative injunctive social norm condition (x̄ = 0.214; SD = 0.041).” However, the results of the pilot study indicate that the negative injunctive social norm condition was not statistically different from any other condition (including the control condition). Might be more clear that the choice of the injunctive norms was based on the literature because the pilot study didn’t provide any clear winners.

Response 5: We agree. We have clarified the text starting on Line 246 to say: “Given the lack of a significant result in the pilot study, the choice to move forward with the injunctive message pair was mainly informed by studies that suggest that injunctive norms are more effective than descriptive norms at promoting desired environmental behaviors [28,29].”

Comment 6: Describe the nature of the operator error for both Study 1 and Study 2.

Response 6: We have added the following explanation to the text starting on Lines 202, “Operator error means that there were mistakes in the data collection on a particular day that rendered the data unusable. Errors included the implementation of incorrect message signs on a particular day, the temporary absence of straws in a dispenser due to shortages, and errors due to cashier shift turnover.” A brief explanation is also on Line 298.

Comment 7: Why do you use different statistical tests to answer the same question for Study 1 and Study 2? Study 1 uses a Kruskal-Wallis test to determine if there was a difference in ratios by condition. Study 2 uses a Welch’s F test to determine if there was a difference in ratios by condition. Either use the same test for each or provide a justification for why one is more appropriate for each case.

Response 7: We used a Kruskal-Wallis Test for the pilot because the data did not meet all of the assumptions for a parametric test, such as a one-way ANOVA. Two conditions had outliers and the results of a Shapiro-Wilk’s test demonstrated that the data were not normally distributed for two conditions. We have more clearly justified this decision from Lines 213 - 222: “The data did not meet the assumptions for a one-way ANOVA. Two conditions had outliers, as assessed by inspection of a boxplot: the negative injunctive norm condition had a single outlier,

while the control condition had three outliers, two of which were extreme outliers (more than 3 box-lengths away from the edge of the box). The results of a Shapiro-Wilk’s test demonstrated that the data were not normally distributed for two conditions: control (p = .001) and negative injunctive norm (p = .034). There was homogeneity of variances, as assessed by Levene's test for equality of variances (p = .493). Because the data violated these assumptions, a Kruskal-Wallis test was run to determine if there were differences in ratios between the seven conditions. The mean rank of ratios was not statistically significantly different between groups, χ2(6) = 7.971, p = 0.240, ε2= .125 [41] This is a moderate effect size.”

Recognizing that a one-way ANOVA is somewhat robust to violations of the assumptions, we also ran a one-way ANOVA and found that there was still no difference between conditions. We have added these results on Line 222: “Recognizing that a one-way ANOVA is somewhat robust to violations of the assumptions [42] we also ran a one-way ANOVA and found that there was still no difference in ratios between the seven conditions, F(6,58) = 1.315, p = .265, est. ω2 = .029. This is a relatively small effect size.”

Comment 8: Page 11: When you inspect the normality of the distributions for each condition, are you looking at the distribution of ratio values over the 25-26 days for each condition? Regardless, make it clear what the values are that you are inspecting.

Response 8: Yes, we were inspecting the distribution of ratio values and have added this detail to the text to clarify. The sentence starting on Line 307 now reads: “The results of a Shapiro-Wilk’s test also demonstrated that the ratio values were normally distributed for the negative injunctive norm condition (p = 0.101) and the control condition, (p = 0.077), while the distribution of ratio values for the positive injunctive norm condition was on the edge of normal distribution (p = 0.05).”

Comment 9: Page 12 “In this case, we reject the null hypothesis and accept the alternative hypothesis that normative messages resulted in significantly less straw use than non-normative messages.” Should be more specific and say ‘a positive normative message’.

Response 9: We have made this edit in the text. The sentence beginning on Line 326 now reads, “In this case, we reject the null hypothesis and accept the alternative hypothesis that a positive normative message resulted in significantly less straw use than a non-normative message.”

Comment 10: Looks like the biggest intervention happened when the park decided that patrons had to ask for straws (a change that happened between Study 1 and Study 2). Would be good to discuss the effect sizes of the norm intervention in terms of the change in control condition values from Study 1 to Study 2.

Response 10: Thank you for this helpful comment. It is true that the effect sizes of the pilot and the main study are not directly comparable because of the change that happened between Study 1 and Study 2 where patrons had to start asking for straws. This change was beyond our control. We have more clearly acknowledged this limitation in the discussion section with a sentence starting on Line 469: “This change also means it is not possible to compare effect sizes between the pilot experiment and the full experiment, which is a limitation of the study.”

Comment 11: In Discussion section, be clear that you only tried ONE version of a message for each norm condition. It’s possible that different messages that fit within the different norm definitions may have impacted behavior differently.

Response 11: We have added the following sentences to the discussion, starting on Line 406 to address this point: “It is important to acknowledge that we only tried one version of a message for each norm condition and that our sample sizes were limited, which means the randomization process may not have fully balanced all variables (e.g., respondent profile) across all conditions. It is possible that different messages within the different conditions could have influenced behavior in other ways.”

Comment 12: The reactance suggestion in the discussion section would predict that negatively worded normative messages would encourage MORE straw consumption than the control condition. This did not happen.

Response 12: This is an interesting point and we agree that would expect reactance to run contrary to the effect of the norm. However, this could simply reduce the impact of the norm if the effect of reactance is smaller than the effect of the norm. The negatively worded normative message would only lead to more consumption if the effect of reactance was larger than the effect of the norm. We have explored this nuance further in the discussion and have also more clearly acknowledged that this explanation is speculative. See the paragraph starting on Line 419: “While we did not collect information on participant characteristics, which could help better explain our results, the psychological literature provides some potential explanations as to why the positive injunctive message may have been more effective. One possible explanation is that the negatively worded message may have triggered psychological reactance in some visitors [46]. Reactance is a form of motivational arousal that occurs when someone feels that their personal freedoms are threatened by rules and restrictions [46]. Resistance can occur when injunctive social norms (or proscriptive messages) appear to impinge on personal freedoms and may even lead people to increase the undesirable behavior [47]. While this is speculative, reactance may have decreased the effect of the negative injunctive message, effectively erasing any signal when compared to the control group.”

Comment 13: Page 16: “salience bias might also help explain why the negative injunctive message resulted in higher paper straw use” Be clear that you mean higher than the positive injunctive message, not higher than the control condition.

Response 13: Good point. We have made this clearer. This is the updated sentence starting on Line 447: “When comparing the positive and injunctive norm messages, salience bias might also help explain why the negative injunctive message resulted in higher paper straw use than the positive injunctive message [52].”

Reviewer #2:

This paper described a simple experiment testing the effectiveness of normative messaging on the use of straws. Overall, it was clear and easy to read, however I felt the paper was missing many details to help us understand what was done, why, and the relevance of the findings. I outline these below.

Main concerns

Comment 1: Why is the data not provided to the reviewers? Data availability is required to publish in PLOS one and is highly relevant for the review process. Please make this available.

Response 1: You can now find the dataset here: https://figshare.com/s/648611d23cb654d0dfcf. The public DOI is: https://doi.org/10.6084/m9.figshare.15134754.v1

Comment 2: I would like to see the results of the pilot study below the study description. Understanding what was found in that experiment is important to understand the methods of the main experiment of the paper.

Response 2: We made this change as requested.

Comment 3: The authors did not find any difference between conditions in the pilot study

Response 3: That is correct. We’ve highlighted this lack of difference starting on Line 220 of the results and again on Line 246: “Given the lack of a significant result in the pilot study, the choice to move forward with the injunctive message pair was mainly informed by studies that suggest that injunctive norms are more effective than descriptive norms at promoting desired environmental behaviors [28,29].”

Comment 4: Why use a Kruskal-Wallace test for the pilot? The non-parametric test was justified for the main experiment, but not the pilot. If it passes parametric assumptions, a two-way ANOVA would be appropriate (except for control condition), where you compare Positive/Negative manipulation, and manipulation type (descriptive, injunctive, social norm). Additionally, why not compare each individual group to the control, as you did in the main experiment? This result section was lacking, in both justifications and statistics.

Response 4: The goal of the pilot was to determine which treatment performed best overall, so we made a comparison between all conditions and the control, rather than comparing each individual treatment to the control. We have more clearly justified our use of the Kruskal-Wallis test by adding the following details, starting on Line 213: “The data did not meet the assumptions for a one-way ANOVA. Two conditions had outliers, as assessed by inspection of a boxplot: the negative injunctive norm condition had a single outlier, while the control condition had three outliers, two of which were extreme outliers (more than 3 box-lengths away from the edge of the box). The results of a Shapiro-Wilk’s test demonstrated that the data were not normally distributed for two conditions: control (p = .001) and negative injunctive norm (p = .034). There was homogeneity of variances, as assessed by Levene's test for equality of variances (p = .493). Because the data violated these assumptions, a Kruskal-Wallis test was run to determine if there were differences in ratios between the seven conditions. The mean rank of ratios was not statistically significantly different between groups, χ2(6) = 7.971, p = 0.240, ε2= .125 [41] This is a moderate effect size. Recognizing that a one-way ANOVA is somewhat robust to violations of the assumptions [42] we also ran a one-way ANOVA and found that there was still no difference in ratios between the seven conditions, F(6,58) = 1.315, p = .265, est. ω2 = .029. This is a relatively small effect size.”

Comment 5: All test decisions should be justified through the manuscript

Response 5: We have more clearly justified our test decisions throughout the manuscript. See Lines 213 - 222 for a justification of the Kruskall-Wallis test in the pilot study and Lines 304 - 319 for a justification of the Welch’s ANOVA in the main experiment.

Comment 6: What is operator error? Describe it in detail (can be in a supplement)

Response 6: We have added the following explanation to the text starting on Lines 202, “Operator error means that there were mistakes in the data collection on a particular day that rendered the data unusable. Errors included the implementation of incorrect message signs on a particular day, the temporary absence of straws in a dispenser due to shortages, and errors due to cashier shift turnover.” A brief explanation is also on Line 298.

Comment 7: Effect sizes should be interpreted throughout the manuscript

Response 7: Thanks for the reminder. We have added the following sentence to the results, starting on Line 317: “The mean ratios of paper straws taken to drinks sold were statistically significantly different between the different conditions, Welch’s F(2, 45.39) = 5.85, p = 0.006, est. ω2 = 0.112. This is a relatively large effect size.”

We have also added a sentence to the first paragraph in the discussion that interprets the effect size of est. ω2 = 0.112 for the main experiment. See the sentence starting on Line 392: “We tested whether normative messages could reduce visitors’ use of paper straws at a marine park and found that the positively worded injunctive message resulted in significantly less paper straw use than the non-normative message and that there was a relatively large effect size.”

Additionally, we have interpreted the effect sizes for the pilot study results, starting on Line 219.

Comment 8: Data was collected over time, why was this not analysed? Was the straw to drink ratio consistent over time, or did it increase/decrease? This analysis is necessary to make any argument about effectiveness of the intervention. Without it, the discussion about savings over time is premature.

Response 8: We allocated the conditions randomly, and so would expect base conditions to be similar across treatments. However, we recognize that extrapolating the data from the experiment over a longer time frame is fraught with challenges, as our assumption that conditions will remain similar might not hold in practice. We have more clearly acknowledged these limitations starting on Line 501: “Our cost-benefit calculation is based on some key assumptions, including the assumption that all concession stands will sell similar numbers of drinks and the assumption that visitation rates or visitor behavior will remain similar. As the COVID pandemic has shown this is not always the case, however we feel that these estimates are still informative. Accepting these assumptions, our calculations project that the return on investment could be approximately 22% if the intervention ran for three years, and that rate would continue to increase over time.”

Comment 9: Why are there no comparisons to other stands in the park over the same time period?

Response 9: We were not able to collect the same data at other stands in the park over the same time period due to limitations of staff and funding.

Comment 10: The authors state this experiment is important because it is conducted in a European country and has been understudied, yet does not discuss cultural effects on messaging. This should be discussed in the introduction and in the discussion – it is very possible that the effectiveness of the messages (positive/negative) may be related to cultural factors

Response 10: This is an important point and we have added a sentence and reference to the introduction starting on Line 153: “It is important to study messaging in different countries and contexts because culture may influence how people respond to message frames [40].” We have also added a sentence to the discussion to address this point starting on Line 411: “Framing effects are also likely to vary across cultures and countries. When examining framing effects across seven Central and Western European countries, one study found differences in the way that consumers from different countries, even within the same region, responded to positive and negative message frames [40].”

Minor comments

Comment 11: Describe the Welch’s F test as non-parametric (line 243)

Response 11: The Welch’s F test is a parametric test. It is also known as Welch's ANOVA. It is an alternative to the one-way ANOVA and is used when the assumption of homogeneity of variances is violated. We have changed the name in the text to Welch’s ANOVA.

Comment 12: Interpret effect size of Welch’s F test (line 245)

Response 12: We have added an interpretation of the effect size after reporting the effect size starting on Line 317: “The mean ratios of paper straws taken to drinks sold were statistically significantly different between the different conditions, Welch’s F(2, 45.39) = 5.85, p = 0.006, est. ω2 = 0.112. This is a relatively large effect size.”

Comment 13: Give actual numbers in Table 3, in addition to what is presented (number of drinks sold, number of straws taken)

Response 13: We have updated Table 3 to include this data.

Comment 14: Acknowledge the discussion section on reactance as speculative

Response 14: Starting on Line 419, we changed the wording to more clearly indicate that the discussion section on reactance is speculative. “While we did not collect information on participant characteristics, which could help better explain our results, the psychological literature provides some potential explanations as to why the positive injunctive message may have been more effective. One possible explanation is that the negatively worded message may have triggered psychological reactance in some visitors [46]. Reactance is a form of motivational arousal that occurs when someone feels that their personal freedoms are threatened by rules and restrictions [46]. Resistance can occur when injunctive social norms (or proscriptive messages) appear to impinge on personal freedoms and may even lead people to increase the undesirable behavior [47]. While this is speculative, reactance may have decreased the effect of the negative injunctive message, effectively erasing any signal when compared to the control group.”

Editor Comments

Comment 1: I have now collected two reviews from two experts in the field, whom I thank for their detailed and thoughtful reviews. Both reviewers found the paper interesting and easy to read. However they both expressed a number of concerns that should be addressed in a major revision. Therefore, I would like to invite you to revise your paper following their comments. Needless to say that all comments should be addressed or rebutted. When uploading your revised manuscript, please upload also a response letter containing a point-by-point response to all the reviewers' comments.I am looking forward for the revision.

Comment 2: Thank you for the opportunity to submit a revision.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Response: We have updated the formatting following the guidelines.

2. Please ensure that you include a title page within your main document. We do appreciate that you have a title page document uploaded as a separate file, however, as per our author guidelines (http://journals.plos.org/plosone/s/submission-guidelines#loc-title-page) we do require this to be part of the manuscript file itself and not uploaded separately.

Could you therefore please include the title page into the beginning of your manuscript file itself, listing all authors and affiliations

Response: Yes, we have made this change.

3. Please clarify in your Methods section that IRB waived the need for consent

Response: We have made this change, starting on Line 171: “The full experiment (Study 2) built on the results of Study 1 by testing the most effective social norm messages from Study 1 against a control message. This study was approved by the University of Florida’s Internal Review Board (IRB202002244) and the need for consent was waived.”

4. Thank you for stating the following financial disclosure:

“This work was funded by national funds through Fundação para a Ciência e a Tecnologia (FCT) as part the project CIP - Refª UID/PSI/04345/2020 (Jean-Christophe Giger)”

Please state what role the funders took in the study. If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

If this statement is not correct you must amend it as needed.

Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf.

Response: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript." We have amended this statement in our cover letter and would appreciate it if you could change the online submission form on our behalf.

5. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

Response: We have made the data available. You can now find the dataset here: https://figshare.com/s/648611d23cb654d0dfcf. The public DOI is: https://doi.org/10.6084/m9.figshare.15134754.v1

6. We note that Figure 1 in your submission contain map images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

Response: We have obtained permission from the original copyright holder and have uploaded the form as an “Other” file with our submission. We have also changed the figure caption to read “Reprinted from Zoomarine under a CC BY license, with permission from Zoomarine.”

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

9 Dec 2021

The effectiveness and efficiency of using normative messages to reduce waste: A real world experiment

PONE-D-21-25759R1

Dear Dr. Neves,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Xingwei Li, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

13 Dec 2021

PONE-D-21-25759R1

The effectiveness and efficiency of using normative messages to reduce waste: A real world experiment

Dear Dr. Neves:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Prof. Dr. Xingwei Li

Academic Editor

PLOS ONE