Building Student Sustainability Competencies through a Trash-Practice Nudge Project: Service Learning Case Study in Kuwait

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thanks for this. Please see attached for a series of comments and ways to improve this paper. I would very much like to see it in print, as I found it very interesting, but it needs some key things doing to it. Most are minor, but one or two will need a bit more work. I have said "Major Revisions" as there are a lot, but it should not take long to make this publishable.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

Some areas to work on, mostly pointed out in my comments.

Author Response

reviewer1

Comments and Suggestions for Authors

reviewer1:  Thank you very much for your research. The topic of the study is very important.  It contributes to sustainable development and green thinking and makes our world more beautiful and cleaner.

 

Suggestions to the authors for improving the text.

 

response: thanks for the encouragement, and to both reviewers for many comments that helped strengthen the discussion.

 

 

reviewer1:  1)    The aim of the study.

The text does not clearly define the aim of the study or the hypothesis, which are essential elements of a well-structured scientific study. The text discusses the role of service learning in sustainability education and the effectiveness of nudges in changing behaviour. However, it does not clearly articulate the objective of the study, e.g. what specific aspect of service learning or nudging is being examined or what the aim of the study is.

 

From the text it can be established how effective service-learning is in improving students' sustainability competences and show that nudges can improve consumer behaviour in waste management. But this is rather a consequence of the discussion than clearly articulated.

 

response: legitimate concern, the comment indicates we were not clear enough.  We have tried to be a little more explicit.

 

However, we are not clear about the reference to hypotheses for the study.  We are reporting on how we organized the class to do the interventions and build sustainability mindset and competencies.  This is not an experiment in whether the class works better than some other way of organizing a class.  We have no statistical tests, for example, to judge whether students learned or did something better than in some control class.  We do have testimony from the practitioner side: Kuwait Policy Appraisal Lab (KPAL, Kuwait Government); the Qatar Behavioural Insights Unit (QBIU, Qatar Government); and the United Nations Development Programme (UNDP) in Kuwait.  They all participated in the class, liked what the students did and learned from the class, and would more classes organized this way.

 

We note that this special issue is about pedagogies; and the discussion aims to show how teaching / learning can work to build student competencies in sustainability.  It is necessary to report the two interventions carefully, but they are not the main point of the article.  The bullet points after the first paragraph lay out what we discuss.  Possibly they did not see very prominent because the format template seems to have taken out the bullet points and presented these as mini-paragraphs, harder to notice.  We hope they can stay as bullet points.  The first bullet point is about the interventions – they are the tool for training students in the next two bullet points (sustainability mindset and competence), and in orienting the university toward the fourth bullet point. 

 

After the slight revision / addition in the first paragraph and fixing the bullet points, we also have similar slight revisions in several places throughout to bring this out a little more clearly. 

 

 

reviewer1:  2)     Research methods and results:

The text mentions that ANOVA (analysis of variance) is used to determine the significance of the results. ANOVA is an appropriate method for comparing differences between groups and determining whether the observed effect is statistically significant. However, the specific values, statistics or confidence intervals obtained using ANOVA are not given. These would normally be necessary to fully understand the strength of the evidence and the statistical significance of the results.

 

The text provides descriptive statistics, such as the percentage increase in the accuracy of waste removal and sorting but does not provide data variability or distribution (standard deviations, ranges). Without this, it is difficult to evaluate the consistency of observed behavioural changes across different experiments or locations.

 

response: fair enough.  The intervention results themselves are not the main point here, but it is reasonable to want to see a few basic stats.  We have added basic stats next to the two figures, although many nudge researchers worry more about the pattern than statistical significance.  Basically, many nudge researchers say that nudge effects can be small, and are not always statistically significant.  Aljamal et al. (2020) deal with this by using non-parametric stats to assess the pattern of multiple small effects, most of which individually are not significant.  Their brief discussion of the issue (Aljamal et al., 2020, p. 2), among others, briefly discuss this. 

 

“None of these effects was very large, but “nudge” perspectives from behavioral economics (e.g., [10,11]) are used to this. Small effect sizes are common in nudge research. For example, a recent review of nudges to reduce excessive food consumption found that many nudge effect sizes are not even statistically significant [12]. In their meta-analysis, Arno and Thomas [13], report an average effect size of about 15 percent in nudge interventions aimed at fostering healthy dietary behaviors. Because “nudges do not always create large absolute shifts in behavior, scholars and policymakers may underappreciate their value” [14] (p. 1042). Nudge tools may have smaller impact than big programs, but they do not usually cost much, so they can be very cost effective.”

 

Aljamal, A., Speece, M., & Bagnied, M. (2020). Sustainable policy for water pricing in Kuwait. Sustainability, 12(8), 3257.

 

10. Thaler, R.H. Behavioral economics: Past, present, and future. Am. Econ. Rev. 2016, 106, 1577–1600.
11. Thaler, R.H.; Sunstein, C.R. Nudge: Improving Decisions about Health, Wealth, and Happiness; Yale University Press: New Haven, NJ, USA, 2008.
12. Wright, B.; Bragge, P. Interventions to promote healthy eating choices when dining out: A systematic review of reviews. Br. J. Health Psychol. 2018, 23, 278–295.
13. Arno, A.; Thomas, S. The efficacy of nudge theory strategies in influencing adult dietary behaviour: A systematic review and meta-analysis. BMC Public Health 2016, 16, 676.
14. Benartzi, S.; Beshears, J.; Milkman, K.L.; Sunstein, C.R.; Thaler, R.H.; Shankar, M.; Tucker-Ray, W.; Congdon,W.J.; Galing, S. Should governments invest more in nudging? Psychol. Sci. 2017, 28, 1041–1055.

 

Note that this is simply to let the reviewer know that we do consider these issues.  In the case discussed in this article, the nudge effects are relatively large, and statistically significant.  Thus, there is no need to put this discussion of small effects lacking statistical significance into this article.

 

 

reviewer1:  It is not mentioned how large a sample was used in each experiment and whether a power analysis was carried out. A power analysis is important to ensure that the sample size is sufficient to detect significant effects.

 

response: the cinema sample was 73 shows with the intervention and 49 shows in the control, already stated in the original on p. 8 line 384.  We did not specifically calculate power, but followed practitioner advice about balance between textbook ideal (equal proportions) to maximize power, and real-world preference to understand the intervention better with heavier weight on the intervention.  White (2018) demonstrates that not much power is lost with only 30% in the control, but power drops off quickly after that.  Our proportion is 32% control.  Brief note of this has been added, but this is not a methodology article, we do not want to get into detailed discussion of methodology. 

 

https://www.markhw.com/blog/control-size

White, M.H. (2018)  How Big Should the Control Group Be in a Randomized Field Experiment?

note: the colors in the graph are for different effect sizes in White’s extensive simulations.

 

 

reviewer1:       3)     About the Conclusions

The conclusions should more clearly summarise the effectiveness of the incentives, perhaps by repeating the main statistical results or outcomes. At present, the effectiveness of incentives is only mentioned, but this could be highlighted more strongly by providing specific data from experiments to strengthen the conclusions.

 

response: useful suggestion, which goes back to the first bullet point in the introduction.  A brief summary of the results is added to demonstrate that the article did address that first bullet point (influencing consumer trash behaviors in public places). 

 

 

reviewer1: 4) Further attention should be paid to data visualisation. All data must be presented in English in the figures

 

response: not quite sure what this is about, but perhaps it indicates that the content of the two graph figures needs a little stronger support.  The tables with each figure should help.

 

Submission Date           29 July 2024

Date of this review         15 Aug 2024 14:14:59

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you very much for your research. The topic of the study is very important.  It contributes to sustainable development and green thinking and makes our world more beautiful and cleaner.

Suggestions to the authors for improving the text.

1   1)    The aim of the study.

The text does not clearly define the aim of the study or the hypothesis, which are essential elements of a well-structured scientific study. The text discusses the role of service learning in sustainability education and the effectiveness of nudges in changing behaviour. However, it does not clearly articulate the objective of the study, e.g. what specific aspect of service learning or nudging is being examined or what the aim of the study is.

From the text it can be established how effective service-learning is in improving students' sustainability competences and show that nudges can improve consumer behaviour in waste management. But this is rather a consequence of the discussion than clearly articulated.

2)     Research methods and results:

·       The text mentions that ANOVA (analysis of variance) is used to determine the significance of the results. ANOVA is an appropriate method for comparing differences between groups and determining whether the observed effect is statistically significant. However, the specific values, statistics or confidence intervals obtained using ANOVA are not given. These would normally be necessary to fully understand the strength of the evidence and the statistical significance of the results.

·       The text provides descriptive statistics, such as the percentage increase in the accuracy of waste removal and sorting but does not provide data variability or distribution (standard deviations, ranges). Without this, it is difficult to evaluate the consistency of observed behavioural changes across different experiments or locations.

·       It is not mentioned how large a sample was used in each experiment and whether a power analysis was carried out. A power analysis is important to ensure that the sample size is sufficient to detect significant effects.

3)     About the Conclusions

The conclusions should more clearly summarise the effectiveness of the incentives, perhaps by repeating the main statistical results or outcomes. At present, the effectiveness of incentives is only mentioned, but this could be highlighted more strongly by providing specific data from experiments to strengthen the conclusions.

4) Further attention should be paid to data visualisation. All data must be presented in English in the figures

Comments on the Quality of English Language

 About Language

Content is more important here. The English is mostly without major errors, but the text is sometimes difficult to understand. The text uses a lot of elements of direct translation, which hinders the perception of the content.

Author Response

reviewer2

Comments and Suggestions for Authors

reviewer2: Thanks for this. Please see attached for a series of comments and ways to improve this paper. I would very much like to see it in print, as I found it very interesting, but it needs some key things doing to it. Most are minor, but one or two will need a bit more work. I have said "Major Revisions" as there are a lot, but it should not take long to make this publishable.

 

response: thanks, we will be happy if the article is useful.  also, thanks for so much careful detail in the review.

 

 

reviewer2: peer-review-39204137.v1.pdf   extensive comment in the pdf

 

response: (the whole list)

 

1. 1 line 20 clarified, the is the UNDP’s Kuwait office, not just the UN in general. The UNDP Kuwait office is mentioned several times, but we agree it is more clear to note it directly here, also.

 

1. 1 line 26 the reviewer commented on lack of an article in several places. One of the authors here is a native speaker, we judged those places to represent generic references to something, in which case no article is required https://academicguides.waldenu.edu/writingcenter/grammar/articles . 

We do acknowledge that in some of those places it is open to interpretation whether the noun is generic or specific, and we added an article where there might be dispute. 

 

Sustainability’s English final edit is usually pretty good, we defer to their English editor in the final version. 

 

This comment is sufficient for several notes about articles, so we don’t specifically comment on this issue again

 

p1 line 28          [1] Thaler & Sunstein, 2008 are cited a couple sentences later, but we agree, we have briefly defined what nudges do here (same ref).  We don’t want to get into too much detail in the introduction, it is discussed more thoroughly in the literature, and analysis.

 

1. 1 line 31 yes, the numbering was an oversight, we don’t finalize the numbering and put them in order in the refs until we get ready to submit. The embedded cite system and alphabetical refs is more common in most social science work, and thus easier (for us) to keep track of in our pre-submission multiple revisions.  Fixed now. 

 

Several additional comments about numbering later, but we do not specifically respond after this one.

 

1. 1 line 40 this was originally bullet points, but got formatted into paragraph form in the template. the reviewer is correct, it looks odd, and we hope it can remain bullet points

 

1. 1 line 44 the next paragraph (p. 1 line 46) points out that waste management is a problem in many developing countries, and the following paragraph (p.2 line 59) ties the service learning elements into the waste management problem. (possibly the comments is meant to point out that the problems are not exclusive to developing countries – true, but the focus here is a (high income) developing economy example.

 

1. 2 line 49 or works best, since our sorting experiment uses plastic, paper, and metal; so etc. would not work well.

 

1. 2 line 50 yes, is, thanks

 

1. 2 lines 58-71 this is new here, but content is not changed much from what it was in the previous draft at p. 7 lines 308-317, where the reviewer suggested moving it to define nudging earlier. good idea. this passage defines nudges, and having it earlier will help in following some of the later discussion a little better.  it is slightly revised to fit well here, with an additional sentence in the next paragraph to also connect it.

 

there was also a comment suggesting that the quote should be stated in our own words.  (The original format was indented as a quote, the template seems to have turned it into a paragraph.  We hope it is clearly identified as a quote.)  In this case, Thaler (Nobel prize for this behavioral econ stuff) and Sunstein are among the most prominent nudge experts, and are about the most credible sources possible for defining nudging.  They are frequently quoted in defining nudges.

 

1. 2 line 73 slightly reworded, added which were a cinema and the university.

 

p 2 line 80         yes, thanks.  to benefits for students, we added, and useful small projects that help community organizations improve operations.  Our two cites here [10, 12] already cover this, no need for an additional cite.

 

1. 2 line 87 we left this as is. It is probably not a majority view that WEF + waste management is a four-element nexus, that’s why we said “some observers talk about it” this way.   However, it is very common to include waste management issues in the discussion of the WEF nexus. 

 

a simple Google Scholar search [ water energy food waste management nexus ] turns up 186,000 results.  of course, not all say much about this, but if even 10 % or so do, there is substantial discussion that waste management is intimately connected to WEF, whether or not one wants to explicitly call it one of the nexus elements.

 

1. 2 line 89 parentheses taken out

 

1. 3 line 103 already cited [23] just a couple sentences before, but fair enough to make sure readers who might want to check for the details find them quickly. We also put [5] here, which is for the broader similarity across the GCC, but includes some Kuwait discussion.

 

1. 3 line 103 rather, added

 

1. 3 line 133 good idea to bring in wicked problems. we don’t want a detailed discussion of definitions, but we put in a short paragraph, and slightly edited around it for smooth flow.  brief mention of the concept, with a few relevant cites, makes it clear that there is not likely to be a simple easy solution, and wicked problems rarely get completely solved.  This helps the argument for nudging.  Nudges rarely completely solve problems, but if a problem is basically impossible to completely solve, then measures that reduce the problem are quite useful.

 

1. 3 line 148 agree, readers can follow the water example better with slightly more discussion. it is added. 

 

1. 3 line 150 the definition has been moved up (see comment above)

 

1. 4 line 170 TOMA (top of mind awareness) is more often used in connection to a brand. it is the same concept, but we don’t want to imply that we are talking about brands.   We prefer to tie it to more general “conscious level thinking / System 2” rather than to a specific branding context.

 

1. 4 Fig. 1 agree that it isn’t the best visually. We think the current format is a little better at visually conveying the message.  We wish we had some real artistic talent on this paper.

 

1. 4 line 185 saving the planet is not small! Probably the reviewer is being sarcastic here, but basically that’s the point of why nudge is useful.  consciously people might think “we need to save the planet”, but that thought is not connected to every detail of their daily routine.   For most consumers, it is “glad I had a few minutes for a cold drink before I walk over to the library on this 50 degree C day.  I need to toss the can now, the library won’t let me bring a drink in.”  very, very strong environmental attitudes might be top-of-mind here, but that’s not most people.  nudging can get more of the tosses in the right sorting bin.

 

1. 5 line 189 nudging is now better defined earlier, but it is still a good idea to briefly mention gamification. It is not the topic here, so we don’t want extensive discussion, but it is useful to note that some nudges can be gamified, and gamification is sometimes used foster environmentalism.  However, it is not a conceptual issue in this discussion, so we mention it a little later, when we discuss the sorting nudge.

 

1. 5 line 196 rephrased the short passage on education. we added in “or for things that temporarily catch their interest” on top of consider important, to tie in with the reviewer’s (good) suggestion that we mention gamification.

 

1. 5 line 199 we took out the short repetition

 

p5 line 204        briefly added.  service learning is already noted earlier (now slightly better than in the earlier version), and in more detail a little later, but it is a good idea to reiterate the basics here where we move toward a little more detailed discussion. 

 

p.6 line 250       only 40%

 

1. 6 line 254 we use quotes in places to unambiguously establish that it is not just us saying this, but it appears in well-established journals. i.e., some of these points are not really our original points which simply get some support from the lit.  We often check when reading articles, and often see that the author has put their own nuance on the point.  that’s legitimate if they are saying the same thing the cite said.  not as good if they are extending beyond what the cite said.  Checking the cite, it somewhat agrees, but not exactly.  The language in such citing needs to be careful, “tends to support”, “hints at”, “suggests”, or something, and not imply that the cite exactly says this.  A direct quote is useful for distinguishing “they said this” from “their discussion tends to support this”

 

In this particular place, however, we follow the reviewer recommendation because have several quotes in this section, better not to have too many all at once.

 

1. 6 line 272 close ) added

 

1. 6 line 289 thanks for catching this – we missed it in several final checks. entirely “consistent with sustainability concepts” got edited out somewhere in the process.

 

p.7 lines 308-317           this passage defines nudges.  it was in the discussion about the class content here, but the reviewer suggested moving it to define nudging earlier.  good idea.  it is now at p. 2 line 59-71, and slightly revised to fit there well.

 

p 7 line 332       changed to follow suggestion

 

1. 8 line 350 changed. we were not very clear in stating this.  It is not a law about whether a cinema can insert something into a movie.  It is illegal because the Kuwait government can be tough about enforcing contracts.  The contract with the movie supplier does not allow the cinema to insert things between the movie and the credits (or anywhere in the film). 

 

1. 8 line 356 good wording – we used it

 

1. 8 line 384 why not 50/50 for control vs sample? in textbook theory, this gives the best power, because confidence intervals around the control vs sample means are very similar (assuming variance does not change much from the treatment).  however, in real-world experiments, there are trade-offs. 

 

We are more interested in what the treatment does than in the case where we don’t do anything (control).  We have 32% of the trials are control.  White (2018) discusses this in some detail.  With control group at 30%, we don’t sacrifice much power, but we have more feel for how the treatment group behaves. If our control was about 10-20% of trials, that might be somewhat problematic, although how much of a problem would depend on the degree of overlap in the confidence intervals between control & treatment.  “A 25% to 30% range is a good compromise, as this exposes 70% of the sample to the treatment, yet still does not harm power terribly” (White, 2018).

 

but it is a legitimate issue for reviewers to check on, and we cite White (2018) with a sentence about this, but we don’t want to get into this discussion very deeply in this paper.  The paper is not about statistical methodology, and arguing for even or unbalanced control / treatment ratios belongs in a methodology article.  This statistical detail has little to do with the topics in the paper, as long as the basic stats are sound.   Our 30% control gives us the real-world benefit of more data on the treatment, without much sacrifice of the textbook preference for better power from equal sizes. 

 

https://www.markhw.com/blog/control-size

White, M.H. (2018)  How Big Should the Control Group Be in a Randomized Field Experiment?

note: the colors in the graph are for different effect sizes in White’s extensive simulations.

 

 

1. 9 line 389 Basically, any field research in social sciences needs to be flexible because things rarely work exactly as the textbook might like it to. (One of the authors of this paper has written three applied marketing research books (long ago), based on extensive MR consulting.)  The need to be flexible and adapt to specific conditions that might not have been foreseen is common in the applied MR discussion, but as far as I (the MR author) know, there is no textbook or journal article about it.  It is often noted in passing.  E.g., after some detailed discussion about how to set up a field experiment, Malodia et al (2023, p. 1956), have one sentence: “Finally, the study design must be sufficiently flexible to respond dynamically to situational changes, and researchers must be prepared to adapt accordingly.”

 

Malodia, S., Dhir, A., Hasni, M. J. S., & Srivastava, S. (2023). Field experiments in marketing research: a systematic methodological review. European Journal of Marketing, 57(7), 1939-1965.

 

I (the MR author here) learned this very early interviewing farmers in Sudan in 1985 – no matter how thoroughly you plan from the textbook, something isn’t going to work that way.  With more experience, you can factor in all the things that might not work.  You still won’t get everything.  (I tell my MR students – if you ever do a perfect project, please get in touch and tell me how you did it.  I’ve done hundreds of MR projects over decades – I never had a perfect project.)  That’s not something a manual can fix, researchers need to be flexible and think quickly.

 

Well, probably more than the reviewer want to know.  Bottom line, we added this cite / ref (Malodia) to the article, but this is another methodological issue we don’t want to spend much time on, since the article is not about methodology. 

 

1. 9 line 399 reviewer1 also wanted a little more detail on basic stats (see above). A table has been added after each of Fig. 3 & Fig. 4, and we incorporated the ANOVA significance figures mentioned in this note into the tables, rather than listing them in the text.

 

1. 9 line 411 we changed “will get” to “can shift behavior toward”

 

1. 10 line 420 sorry, we do not have pictures of either original bins or the treatment labeling. Those records are with the (now graduated) students, it was an oversight not to keep a copy during the term.

 

1. 12 top, extra figure yes, accidental duplicate

 

1. 14 reviewer2 comment: You need two discrete and important sections: one is the limitations of the study (any comparable article will have this that you can use as a model)

 

1. 14 reviewer2 comment: The other is a recommendations section, where you give your key recommendations for 1) sustainability educators, 2) students, 3) corporate partners, 4) policymakers and 5) return to MENA, as in your abstract, to comment on how this will help at a much wider level.

 

response: we changed “3.0 Conclusions” to 3.0 Discussion, and added a section 3.1 Conclusion to briefly address these final two comments.

 

Submission Date           29 July 2024

Date of this review         06 Aug 2024 15:54:05

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thanks for this - you have acted on almost all the recommendations, and I hope you agree that the paper is better for it. I much prefer your redesigned table. Hope to see this in publication soon.

Author Response

editor comments for minor revision

 

note: we took out all of the yellow highlight from the previous revision1 – it has already been looked at.  Only changes in this revision2 are highlighted, mostly in the conclusion, but also a very few small details we noticed reading through again.

 

 

editor: 1) The Conclusions part is, for sure, one of the most important sections of any research manuscript. Thus, in my view, don’t make sense to include it as a sub-section (3.1). Please consider the Conclusions as a main section, in this case Section 4.

 

response: The former 3.1 conclusions section is now section 4.  The section is reorganized, with a little bit of new material, following the suggestion in point 2 below. 

 

 

editor: 2) Limitations of any research work should also be provided, but I believe they are not the main conclusions. Thus, why starting the conclusions by mentioning these limitations?! I suggest organizing the Conclusions section by starting with some generic concluding remarks, followed by some specific concluding remarks (e.g., Recommendations) and to conclude with the limitations (also future research works, if there is any prediction about it).

 

response: Yellow highlight with red font is the new material, following this suggestion.  This new material is mostly the short “generic concluding remarks” and a little bit about how limitations might suggest a few future research issues.  There is very slight revision to adapt the flow to the shift in organization. 

 

Yellow highlight with standard black font is just moved, not changed.  This is the limitations, which now come after the other parts.

 

 

editor: 3) The reference list should be carefully revised, since there are several errors, e.g., repeated references ([19], [20], [24], [8] in lines 806-812, etc.), references without numbering (lines 795-798, 813-814, etc).

 

response: not sure what happened here.  The pdf file we got from the article site (which corresponds to the line numbers noted here) has lots of repetition and unnumbered references.  These are from the revision process before we finalized the reference list.  The Word doc we got from the article site does not have these problems, it is the finalized revision as submitted.  Possibly the pdf included deleted things from the tracking, or possibly a Sustainability editor already fixed it.  In any case, for this slight revision, we checked the doc version again, and don’t see any of these problems.

Author Response File: Author Response.docx