Intention to Purchase Active and Intelligent Packaging to Reduce Household Food Waste: Evidence from Italian Consumers

Reviewer #1

Recommendations

Response

1. There is a lack of legal requirement presentation concerning active and intelligent packaging. 

We thank the referee for his/her suggestion. According to the referee suggestion, we integrated the manuscript with the legal requirement concerning active and intelligent packaging. Referee can see Pag. 2, lines 55-62. The additional text is highlighted in red in the revised version of paper and reported below for your convenience:

“Active and intelligent packaging are defined in Europe by the Regulation 450/2009/EC which integrated the Regulation 1935/2004/EC [26]. The European regulation states that the individual substances, or the combination of substances, used to make active and intelligent components should be subjected to the European Food Safety Authority (EFSA) authorization which performs the risk assessment of such substances. Also, the regulation requires that active and intelligent packaging should be labelled with the words “do not eat”, allowing consumers the identification of non-edible parts [12,26,27]”

2. The problem is researched group that is quite small, homogeneous. There is no proof that researched group is representative (in researched groups persons with higher education was 66,6% of researched population while according to Eurostat data from 2016 in Italy there was 26,2% of population with higher level of education).

Referee is correct by pointing out that the sample used is not representative of the Italian population as well as that it encompasses a large share of educated respondents. We acknowledge

this limitation in the revised version of the manuscript. Referee can see Pag. 10, lines 366-369. 

In those lines we point our as “…given the small sample size, caution must be used in assessing our results as might not be transferable to the Italian population as well as cannot be generalized in other geographical contexts. Also, in our sample, two out of three respondents have a high level of education which may significantly affect the consumers’ awareness for FW as well as their attitudes and subjective norms related to individual intention to lower the household FW.” Hope we have the clarified referee’s concern.

3. The results would be much interesting concerning international aspect.

We thank the referee for his/her suggestion. The revised version of the manuscript discusses our results by comparing the latter with Tiekstra et al.’s (2021) findings which investigated millennials' preferences for intelligent packaging and active packaging by sampling 1249 individuals across Europe.  The inclusion of Tiekstra et al. (2021) paper allowed a discussion of our results in an international perspective. Referee can see Pag. 8, lines 269-272, of the revised version of the manuscript. The text added is in red and reported below for your convenience: “This finding was also supported by a recent study by Tiekstra et al. (2021), who interviewed 1249 individuals from 16 European countries, showing that active packaging proved more difficult, rather than the intelligent one, to be successful into the market due to a lower consumers acceptance about this technology [61].”

The reference added is:

Tiekstra, S., Dopico-Parada, A., Koivula, H., Lahti, J., & Buntinx, M. (2021). Holistic Approach to a Successful Market Implementation of Active and Intelligent Food Packaging. Foods, 10(2), 465.

4. In my opinion the level of education significantly influence research results and conclusions were awareness and attitudes are taken into consideration. Conclusions, although properly formulated, have a limited cognitive value.

We thank the referee for his/her comment. We addressed the referee’s comment above as well as referee can see Pag. 10, lines 366-369 of the revised version of the manuscript in the conclusion section.

Reviewer #2

Recommendations

Response

1. This study is very well written and explained, the manuscript is nice to read. In my opinion, the subtitle '1.1 literature review' is not necessary as it could be included in the Introduction.

We thank the referee for his/her suggestion. However, considering the adjustment that we did in the Introduction section to address the request of one of the Reviewer with the inclusion of the legal requirement concerning active and intelligent packaging, we decided to maintain the subparagraph “1.1 literature review” avoiding having a long Introduction potentially boresome for the reader.

2. Is it possible to mention the range of the age of the respondents? Now I only see: above 18 y (median 35.8).

We thank the referee for the suggestion. The revised version of the manuscript includes, in Table 2, the age range along with its mean and the standard deviation. Also, the range of other continuous variables is reported in the revised version of Table 2. Please, the referee can see the updated version of Table 2, Pag. 4-5.

3. The sample size is small and it's a fact that >25% of the respondents are very high educated, but that's mentioned in the discussion. That's a critical remark for the study.

The referee is correct by pointing out that the sample used is small, not representative of the Italian population and it encompasses a large share of educated respondents. We acknowledge

this limitation in the revised version of the manuscript. Referee can see Pag. 10, lines 366-369.  In those lines we point our as “…given the small sample size, caution must be used in assessing our results as might not be transferable to the Italian population as well as cannot be generalized in other geographical contexts. Also, in our sample, two out of three respondents have a high level of education which may significantly affect the consumers’ awareness for FW as well as their attitudes and subjective norms related to individual intention to lower the household FW.” Hope we have the clarified referee’s concern.

4. Would it be possible to compare the results with the recent study of Tiekstra et al., 2021, in Foods, who surveyed 1249 European millennials and also concluded a preference for intelligent packaging over active packaging.

We thank the referee for his/her valuable suggestion to compare our results with those of

Tiekstra et al. (2021) which explored millennials preferences for active and intelligent packages.

The revised version of our manuscript encompasses a results comparison of our study to those of Tiekstra et al., (2021), an additional text is included in the revised version of the paper and reported below for your convenience:

“This finding was also supported by a recent study by Tiekstra et al. (2021), who interviewed 1249 individuals from 16 European countries, showing that active packaging proved more difficult, rather than the intelligent one, to be successful into the market due to a lower consumers acceptance about this technology [61].”

Please referee can see the revised version of paper on Pag. 8, lines 269-272.

5. Paragraph 3. Results should be not after the text, but as a subtitle.

We thank the reviewer for pointing out it. We fixed the mistake in the revised version of the manuscript.

Reviewer #3

Recommendations

Response

1. Line 35: “…99.8% of all foods and beverages are sold packaged…” does this refer to the percent by weight, by calories or by product codes?

We thank the referee for the suggestion. The 99.8% of all foods and beverages refers to the percent by items.  We clarified referee’s concern in the revised version of the manuscript. Referee can kindly see can see Pag. 1, line 35.

2. Line 117: ‘over 18 years…’ or ’18 and older’? Were 18 years olds excluded?

We thank the reviewer for pointing out it. We fixed the mistake in the revised version of the manuscript. Referee can see Pag. 4, line 140. The text added is in red and reported below for your convenience: “The survey was targeted to Italians aged 18 years old and over, who are responsible for the food shopping in their household”.

3. Line 147 “too much” should be “more”

We thank the reviewer for pointing out it. We reformulate the sentence following the referee’s comment. Please referee can see Pag. 6, line 173 in the revised version of the manuscript.

4. 166-167: How were active and intelligent packaging explained to respondents? Given the novelty of these terms, how did you validate if the respondents understood these terms?

We thank the reviewer of his/her suggestion. Active and intelligent packaging were explained to respondents before to start the survey. However, once provided to respondents the information about these two completely different technological solutions, we didn’t validate the respondents understood. This is because our goal was not to analyze the consumers knowledge about active and intelligent packaging as well as to what extent the knowledge could affect the consumers’ willingness to purchase these packaging solutions. We address the referee’ s comment including this clarification in the revised version of the manuscript. Please referee can see Pag. 4, lines 141-143. The text added is in red and reported below for your convenience: “Before starting the survey, a brief description of active and intelligent packaging was provided to respondents, as showed in Appendix A - Table 1.”

5. 188-89: Separate ‘3. Results onto its own line

We thank the reviewer for pointing out it. We fixed the mistake in the revised version of the manuscript.

6. Table 3. It looks like respondents didn’t understand the difference between active and intelligent packaging, as the two models seem to be feature nearly indistinguishable coefficients. From the point of view of the technology, each should drive different relationships as active technology simply extends shelf-life, but intelligent technology merely better informs participants about the shelf-life that exists.  Please conduct a Likelihood Ratio test to see if pooling the two models can be rejected.

We thank the referee for his/her useful comment. However, as we wrote above, active and intelligent packages are based on two completely different technologies. Thus, according to the theory, should be empirically treated separately regardless of whether consumers understand the difference between the two. Also, we want to point out that our research goal is to explore the consumer willingness to purchase those packages to lower food waste and not to assess the level of what consumers understand about these two packages. To accommodate the referee’s suggestion, we performed the Likelihood Ratio test LR which gave us a p-value equal to 0.10 rejecting the null hypothesis that one model performs better than two separate models with the 10% of significance level. The statistics, besides the theory, also support our decision to use two separate models instead of one unique model.

7. 202-203 “…and larger than that ….to purchase active packaging…” Unless it Is significantly larger as determined by an appropriate test, please do not discuss the difference in magnitudes. Again, you must convince the reader that your participants know the difference between the two terms.

We thank the referee for his/her suggestion. We thank the referee for his/her suggestion. We performed a coefficients equality test across the two models for the parameters suggested by testing whether the 0.812 was statistically different from 0.679. The statistic for the equality test was F(2, 256)=3.555 with Prob>F 0.03. Thus, we reject the null hypothesis that the two coefficients are statistically equal at 5% of the significance level.

8. 207-08 “…higher than that…” Please report the test statistic and p-value that informs this comparison.

Thanks for the comment. We performed the paired means t-test getting a statistic of the p-values for the two-tailed t-test.  Because our p-value is 0.00004014 which is less than the standard significance level of 0.01, we can reject the null hypothesis of means equality. Thus, we stand with our claim that “…the mean value for the intention to buy intelligent packaging (6.29) higher than that for the intention to buy active packaging (5.81)”. Hope we have clarified the referee's concern.

9. 228. ‘More willing to purchase…’ strike unless statistical test supports this at traditional levels of significance.

Please see our comment above. Hope we have clarified the referee’s concern.

10. “…more likely to accept intelligent packaging…” strike this unless you can establish convincingly that your respondents understood the difference between the two types of packaging.

Thanks for the comment. However, once again, the aim of this paper is not to explore whether respondents understood the differences between the two types of packaging but whether such packages are likely to be used by consumers interested to lower the food waste at home level.

11. 306-328. Implications. I’m not sure how your study can be used to speak to the likely efficacy of a symbol or logo for advanced packaging or for informational campaigns. No logos or differential informational messages are considered by participants in this paper, nor do you even establish how well participants understand the technology behind the packaging.  While promoting a symbol and informational campaigns may seem logical, it is not supported by anything in this study and should be struck as an implication.

We thank the reviewer of the suggestion.

We suggest the use of a logo or symbol as a tool to facilitate consumer recognition of active and intelligent packages, thus promoting their adoption. Such statement is in the policy suggestions as we do not directly test their efficacy in promoting purchases of active and intelligent packaging.

It is worth saying that the focus of the study is not to analyze the consumer’s level of information related to these two packages either nor whether a logo/symbol may enhance the willingness to purchase. Once again, as we wrote in the original version of the manuscript, the aim of the study is “to explore to what extent consumers willing to reduce FW are also willing to purchase products packaged with active and intelligent packaging to achieve their scope”.

Hope we have the clarified referee’s concern.

12. Limitations. The Theory of Planned Behavior should be mentioned separately as a limitation of the study. The intention-behavior gap continues to be a great issue in the realm of behavioral science, including for the study of food waste behavior, and TPB is particularly problematic as it ignores the role of stimuli that are not front of mind (habits and routines that don’t avail themselves to conscious articulation, nudges).

The referee is correct by pointing out this further limitation that we acknowledged in the revised version of the manuscript. Referee can see Pag. 10, lines 377 – 384.

Reviewer #3

Recommendations

Response

First, a general issue that concerns the value of the research.  While your intent was not to test whether respondents understood the two technologies, the lack of documentation of whether respondents understand these technologies holds major implications for the value of the research. Suppose respondents don't really understand these technologies and their responses were just 'yea-saying' behavior that will diminish once they better understand the technologies. Then the implication is that firms may use your research to change their packaging thinking that consumers are willing to purchase these reformulated packages. However, if consumers don't really understand the product, they may be no longer interested in the products once they learn more, and the firms will have undertaken expensive changes based upon fleeting interest ('yea-saying' behavior) that is sparked by a brief technology description. Your LR test for differences between the packaging types has a p-value of 0.10, which is the border of statistical significance and not a compelling basis upon which to say that respondents were processing the information with any depth. Hence, documenting whether respondents understand the item being assessed should not be viewed as an 'option' for such research that would be nice to add, but rather an essential element that will determine whether readers and users should engage with the research. 

As described in the manuscript our goal was to assess whether consumers are willing to purchase active and intelligent packaging to reduce household food waste. Authors firmly believe that the objectives, the entire methods section, results, and discussion follow a clear, sound reasoning complying with scientific practices of consumer studies. It is customary of studies that investigate consumer intention to use/buy/purchase/pay do not assess whether consumers understand the objective of the study (e.g., as in our case active and intelligent packages). For instance, the following papers published in high-raked journal assess consumer’s intention to choose selected products for which authors do not assess to what extent consumer’ understand the object in analysis:

1) Gómez-Luciano, C. A., de Aguiar, L. K., Vriesekoop, F., & Urbano, B. (2019). Consumers’ willingness to purchase three alternatives to meat proteins in the United Kingdom, Spain, Brazil and the Dominican Republic. Food quality and preference, 78, 103732. https://doi.org/10.1016/j.foodqual.2019.103732

2) Melbye, E. L., Hansen, H., & Onozaka, Y. (2015). Advertising functional foods: The effects of physical body size and appeal type on ad credibility and purchase intentions. Journal of International Food & Agribusiness Marketing, 27(2), 142–154. https://doi.org/10.1080/08974438.2014.918916.

3) Roosen, J., Bieberstein, A., Blanchemanche, S., Goddard, E., Marette, S., & Vandermoere,F. (2015). Trust and willingness to pay for nanotechnology food. Food Policy, 52, 75–83. https://doi.org/10.1016/j.foodpol.2014.12.004.

4) Kraus, A. (2015). Factors influencing the decisions to buy and consume functional food. British Food Journal, 117(6), 1622–1636. https://doi.org/10.1108/BFJ-08-2014-0301

5) Krutulyte, R., Grunert, K. G., Scholderer, J., Lähteenmäki, L., Hagemann, K. S., Elgaard, P., et al. (2011). Perceived fit of different combinations of carriers and functional ingredients and its effect on purchase intention. Food Quality and Preference, 22(1),11–16. https://doi.org/10.1016/j.foodqual.2010.06.001.

None of the studies above measuring consumer intention to purchase assessed a priori whether respondents understood the product or the product technologies tested. Thus, we stand that our research approach complies with scientific practices in the field of consumer studies. Assessing the level of consumer understanding of active and intelligent packages will be certainly explored in the future. Thanks for the referee suggesting this good point for future study.

Also, what the referee calls “'yea-saying' behavior” leading to a mismatch between consumers’ positive attitudes, or purchase intentions, towards a product and the relatively low level of action when it comes to purchasing decisions is referred to in the literature as the “attitude-behaviour-gap” or “attitude-intention-behaviour gap”. Such drawbacks affect most of the studies based on the survey approach and it is not possible to properly account for unless of using reveled data or secondary data. The following papers describe the issue:

1) Balcombe, K., I. Fraser, and E. McSorley (2015). Visual Attention and Attribute Attendance in Multi-Attribute Choice Experiments. Journal of Applied Econometrics 30 (3): 447–67.

2) Osman, M., & Nelson, W. (2019). How can food futures insight promote change in consumers’ choices, are behavioural interventions (e.g., nudges) the answer?. Futures, 111, 116-122.

3) Bimbo, F., Bonanno, A., & Viscecchia, R. (2016). Do health claims add value? The role of functionality, effectiveness and brand. European Review of Agricultural Economics, 43(5), 761-780.

Thus, future studies will focus on the subject by collecting behavioral data, rather than approaching the subject by using survey techniques.

Do you have the amount of time participants spent on the page with the technology descriptions?  Perhaps you could use this as an instrument for documenting respondent understanding of the technology and see if results are robust among only the most informed respondents.

We thank the referee for the suggestion. Unfortunately, we did not collect information on the time each participant spent on the technology description page.

7 & 8. Details of test statistics supporting key discussion points like those referenced in 7 and 8 should be included in the text or footnote of the revised manuscript. You report the test statistics in the response to the reviewer, but good statistical practice dictates these should be shared with all readers (i.e., in the revised manuscript).

We accommodate referee’s suggestion. Then, we included the test statistics in the revised version of the manuscript. Please referee can see Pag. 7, lines 219-223 and lines 226-229. The text added is in red and reported below for your convenience: “Also, we performed a coefficients equality test across the two models to test whether the parameter 0.812 was statistically different from 0.679. The statistic for the equality test was F(2, 256)=3.555 with Prob>F 0.03. Thus, we reject the null hypothesis that the two coefficients are statistically equal at 5% of the significance level.” and “Also, in this case, we performed the paired means t-test to analyze whether the means were statistically different. We obtained the p-value equal to 0.00004014 which is less than the standard significance level of 0.01. Then, we can reject the null hypothesis of means equality.”

11. Again, this paragraph states that you are sharing implications of your research. You predict that an information intervention (logo/symbol) will improve adoption despite readily admitting that your research design does not address the role of information in participant's response.  Therefore, it cannot be an implication of your research. Rather this suggestion is speculation, and should be labeled as a speculation rather than an implication, or better yet, omitted from the manuscript because casual readers may still implicitly assume it is an implication of the research.  Perhaps instead you could speak of future research designs that could be used to explore the efficacy of logos/symbols in promoting adoption.

We accommodate referee’s suggestion. Then, we removed the discussion of the logo/symbol as an implication of our research from the revised version of the manuscript. Please referee can see the Conclusions section from Pag. 9 to 10.

12. You respond "...we acknowledged in the revised version...lines 377-384..."  I cannot find any revisions in my copy that acknowledge the highlighted weaknesses of the Theory of Planned Behavior/   

We accommodate referee’s suggestion. Then, we clarified referee’s concern in the revised version of the manuscript. Please referee can see Pag. 10, lines 357-361. The text added is in red and reported below for your convenience: “Therefore, the use of the TPB could be considered as another limitation of the study considering that it does not take into account environmental or economic factors that may influence the personal intention to perform a behavior as well as it ignores the role of individual's emotions and unconscious stimuli during the decision-making process.”

Point 6. Please 

We accommodate referee’s suggestion. Then, we expand our discussion related to point 6 of the previous referee’s report including the following statements: “Also, we performed the Likelihood Ratio test LR which gave us a p-value equal to 0.10 rejecting the null hypothesis that one model performs better than two separate models with the 10% of significance level. The p-value of 0.10, which is border of being statistically significant, may indicates that respondents in our sample not clearly distinguish the difference between the two packages.” Please referee can see Pag. 7, lines 209-213. Hope we clarified the referee’s concerns.

Reviewer #3

Recommendations

Response

1.  Several of the newly added sentences will need moderate editing for English usage.

We thank the referee for his/her suggestion. We made some corrections to English usage for the last added sentences, as reviewer suggested. Revisions are reported in red colour in the revised version of the manuscript. Please Referee can see Pag. 7, lines 209-213, lines 219-223; lines 226-229 and Pag. 10, lines 357-361.  Hope we clarified the referee’s concerns.