After asking questions about their socio-demographic background for quota sampling purposes to ensure representativeness, including age, gender, education and region of residence, respondents were asked in the second part of the survey about their beach and coast visitation behavior, the leisure activities they undertake and their experience with marine litter, the first component of perception that we cover. This was followed in the third part by a section eliciting respondents’ willingness to pay for marine litter clean up measures. The fourth part of the survey contained questions on the other three components of perception: knowledge, concern, and responsibility.

The analysis considers only respondents that had visited a beach or coastal area in 2019. Visual perception is one of the key components of our definition of public perceptions in the marine litter context. We were unable to examine the visual perceptions of respondents who had not visited a coastal area in 2019 for leisure, hence they were excluded from the dataset. 15 Questions were employed to examine the components of public perceptions of MPL identified in the literature review. The 15 survey questions are presented in Appendix Table 1 in the Supplementary Material.

Respondents’ perception related to experience and observation was examined through four statements. Participants were first asked about the water clarity of the visited coastal area and whether they had noticed the presence of MPL. Those indicating that they saw litter were asked to indicate the amount and size of the litter. Concern was operationalized through two questions. The first question asked for the respondent’s level of concern for marine plastics specifically. The second question asked the respondent to rank environmental issues of concern, including water pollution and marine plastic, allowing us to identify the relative importance of MPL in relation to other environmental issues. Similarly, two statements measured the level of responsibility. Respondents were first asked to indicate how responsible they felt personally to reduce plastic pollution, and next to rank listed parties (including plastic consumers) who they felt are responsible for cleaning up MPL. Three statements were included to inquire about respondents’ perceptions of consequences of MPL. Knowledge was measured both subjectively and objectively. Respondents first indicated their self-perceived or subjective knowledge, and this was followed by three right or wrong statements about marine plastic. Their responses to the statements were coded into a single variable for objective knowledge. Including both subjective and objective knowledge allows us to investigate the potential “illusion of knowing”; the result of an overestimation of self-perceived knowledge in relation to actual knowledge (Park et al., 1988).

The software program SPSS (version 26) was employed to quantitatively analyze the survey results and identify salient findings regarding public perceptions of marine litter. An exploratory factor analysis with varimax rotation was performed to examine variable relevance and relationships.

General patterns across respondents’ responses are explored by reporting and comparing mean scores and distributions of responses. Statistical differences between various subsets are examined by means of the non-parametric Mann-Whitney U-test for comparisons of the central tendency of distributions across two subgroups and the Kolmogorov-Smirnov Z-test to test for differences in response distributions (e.g., respondents who noticed the presence of marine litter during their visit versus those who did not) and the Kruskal-Wallis H-test and multiple pairwise comparisons for testing more than two independent samples (e.g., across three marine regions: the Baltic Sea, the North Sea, and the Mediterranean Sea), using a critical p-value to identify statistical significance of 5%.

The analysis applies two geographic scopes. First, comparisons of relevant perception variables are made across the eight surveyed countries. Second, survey outcomes are compared at regional sea level by examining responses from households residing in a specific marine region (Baltic Sea, North Sea, and Mediterranean Sea). Since some surveyed countries border multiple seas (Germany, France), respondents residing in the same country may be allocated to different marine regions based on their region of residence and distance to the marine region. Although France is located along the Atlantic Ocean and Mediterranean Sea, respondents are clustered into the Mediterranean and North Sea region to maintain acceptable sample sizes per marine region. To be more specific, the Baltic Sea region consists of residents from Sweden, Estonia and the eastern part of Germany; the North Sea region of respondents living in Denmark, Netherlands, the western part of Germany, and northern part of France; and the Mediterranean by Italy, Greece and the southern part of France and Germany.

In each of the eight sampled countries we gathered data for approximately 1,000 individuals aged 18 or above. Since the study is focused on those who have had personal experience with marine litter, a subset is used in our study of 4,664 individuals who visited a sea or coastal area in Europe in 2019. The remaining sample of respondents excluded from further analysis here had not visited the beach or coastal zone in their country in the previous year. Socio-demographic characteristics of the survey participants are summarized in Appendix Table 2 in the Supplementary Material. Examination of the data reveals variability in the demographic and socio–economic characteristics across the eight samples. Although respondent numbers vary per country, the variation was limited, ranging from 497 useable respondents in Italy to 657 respondents in Denmark. Germany had almost the same number of respondents as Italy, and Estonia a similar number of respondents as Denmark. With just over 600 respondents, Greece, Sweden, and Netherlands were found somewhere in between these minimum and maximum numbers of respondents.

The overall gender ratio (female = 49.5%; male = 50.4%; other = 0.1%) is considered representative for the whole population of the participating countries, although in some countries female respondents were slightly overrepresented (e.g., Estonia) and in other countries male respondents (e.g., Greece and Netherlands). All adult age groups were represented; the respondents’ age ranged between 18 and 92 years, with an average age of 46 years. Greek respondents are on average significantly younger (40.59 years) than the sample mean, while the Swedish respondents are significantly older (48.41).

First, we calculated the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and conducted Bartlett’s sphericity test to determine whether the data is suitable for a factor analysis. We excluded the two ranking questions and recoded the three questions to examine one’s objective knowledge into a single variable, leaving 11 perception components. The KMO measure has a value of 0.809, which is considered “meritorious” according to Kaiser (1974). Bartlett’s test of sphericity indicated a significant correlation between the variables (p < 0.001).

Second, we conducted Spearman’s multivariate factor analysis to define the underlying structure in the data and to model interrelationships among the perception variables. Analysis of the communalities of the perception variables indicated that one variable of the component group “experience and observation,” namely responses to the question about perceptions of water clarity, did not contribute as much as the other components and was therefore eliminated. The scree plot test identified three eigenvalues greater than one, suggesting that the remaining 10 components could be organized into three component groups: experiences and observation, concern and consequences, and knowledge and responsibility (see Table 1). Experience and observation are related to one’s visual perceptions and can be explained as the process of absorbing what one sees, including memories and experiences, and organizing it in the brain. Respondents’ observations of the frequency, amount and size of marine litter are part of this first factor component group. The second component group encompasses psychological factors that influence one’s level of concern and perceived consequences associated with MPL. The third component group consists of subjective and objective knowledge, as well as perceived personal responsibility. Responsibility was spread across two component groups, which was not unexpected since the mean score analysis showed high deviations across responses for self-responsibility (Appendix Table 3 in the Supplementary Material). The 10 perception variables in the three component groups explain 63.6% of the variation within the data. The participants’ responses per component group will be discussed next.

Considering public concerns across the eight countries, the results in Figure 4 show that the vast majority of respondents selected “totally agree” and “somewhat agree” to the statement “I am very worried about plastic pollution in seas and oceans.” In total, 83% of the respondents confirmed being concerned about MPL, whilst only 4% expressed not to feel concerned. It was noted that the Greek sample was most concerned about the negative impacts of MPL with 92% confirming concern and only 1% disagreeing, whereas the Scandinavian countries were least concerned. Nevertheless, still 77% of the respondents in Sweden and 80% of the respondents in Denmark confirmed being concerned about MPL. These results are confirmed when comparing the mean scores across the five-point scale statements (1 = strongly disagree, 5 = strongly agree). These reveal that the respondents are generally highly concerned about marine litter (Mean score across all respondents = 4.26) and are aware of the associated negative impacts of MPL on marine organisms (Mean score = 4.58), followed by coastal appearance (Mean score = 4.53), and human health (Mean score = 4.39).

The study also sought to explore how concern about MPL relates to other environmental challenges. Survey respondents were asked to select the top three environmental issues that worry them most from a list of 14 (and an open-ended category where they could specify another environmental concern) and rank them in order of importance. Figure 5 shows the weighted average of the highest ranked issues.¹ Respondents’ concern about water pollution and plastics in oceans is high compared to other environmental issues. They are ranked as the second and fourth most worrisome issues, respectively, with about 10% of the total number of points being allocated to these issues. However, the ranking of the environmental issues differ across countries. For example, whilst a significant share (14%) of the points is allocated to plastics in oceans by the French, only 5% of the total points are assigned to this issue by respondents living in Estonia, who find air pollution more worrisome (17%) (a country by-country presentation of the distribution is presented in Appendix Table 4 in the Supplementary Material).

Figure 6 presents a comparison of mean scores associated with perceived consequences across countries and reveals a similar pattern per country: consequences of marine litter for marine organisms are considered most severe (except in France), followed by consequences for natural beauty and scenery and, to a lesser extent, consequences for human health. The highest mean scores for the three consequences are found in Greece, and the lowest mean scores are reported by Dutch respondents.

The final cluster of variables addresses subjective and objective knowledge, as well as the sense of personal responsibility. Objective knowledge, also referred to as factual knowledge, is measured by means of three statements about the sources of plastic pollution and recycling practices, with a high average mean score indicating a high level of knowledge. Figure 7 indicates that knowledge about the recyclability of plastics is low in all countries, whereas knowledge about the contribution of washing synthetic clothing and the role of consumers in the production of plastic pollution is considerably higher.

The scores of objective knowledge across the three themes presented in Figure 7 are computed into one variable indicating the average level of objective knowledge per country. Figure 8 presents the mean scores for objective and subjective knowledge across the eight European countries. On the one hand, the mean scores of objective knowledge do not differ much and vary from a high average score of 3.2 for the French and Greek samples as the most knowledgeable respondents to 2.9 for the German sample as the least knowledgeable respondents about MPL. On the other hand, the average scores for respondents’ subjective self-reported knowledge show much more pronounced differences, with the Greek (Mean score = 4.0) and Italian (Mean score = 3.9) respondents considering themselves most knowledgeable, while the respondents from Denmark and Sweden ascribe significantly lower levels of knowledge about MPL to themselves with average scores of 2.96 and 3.12, respectively. Paired-samples t-tests indicated that scores of subjective and objective knowledge were statistically significantly different from each other at the 5% significance level in each surveyed country, except for Sweden; t-test statistic = −0.065 (p = 0.948). Appendix Table 6 in the Supplementary Material provides all test results of the paired-sample t-test to compare subjective and objective knowledge scores within countries and Appendix Tables 7, 8 present the test statistics of the Kruskal-Wallis H-test and pairwise comparisons to test whether and how levels of objective or subjective knowledge significantly differ across countries.

Respondents were asked who they think should be made responsible for the clean-up of MPL. As shown in Figure 5, companies and consumers are considered most responsible, whilst the fishing and tourism industry are considered least responsible. A country-by-country comparison (Appendix Table 5 in the Supplementary Material) reveals similar patterns within individual countries with two notable deviations. First, Estonians hold their national government mainly responsible for cleaning MPL (21% compared to the average of 12% for all eight samples together in Figure 5) and to a much lesser extent plastic-producing companies (12% compared to the average of 20% in Figure 5) or plastic-using companies (11% compared to the average of 16% in Figure 5). Second, German respondents do not ascribe much responsibility to consumers and companies, but instead defer mainly to the EU and their national government to manage MPL.

Figure 9 shows the sense of responsibility that respondents feel regarding efforts to reduce MPL in their own country. Wide-ranging mean scores (3.17–4.43 on a five-point Likert-scale) and relatively high standard deviations indicate that personal responsibility to reduce MPL differs significantly across countries and within countries (see in the Appendix Table 3 for detailed information on mean scores and standard deviations per country and Appendix Table 9 for the outcome of the Kruskal-Wallis H-test and pairwise comparisons across countries). The share of respondents feeling personally responsible differ significantly across countries: 89% of the Greek respondents reported to feel personally responsible, whilst only 43% of the Dutch respondents agreed on being personally responsible. Greek, Swedish, and Danish survey participants barely reported disagreement to the statement of being personally responsible (respectively 1, 7, and 3% of all the Greek, Swedish, and Danish respondents). These percentages are significantly higher (p < 0.001 for all comparisons with Greece, Sweden, and Denmark) in Netherlands, Italy and Germany, where more than 20% reported not to feel personally responsible.

The next series of analyses focus on public perceptions of marine litter at the level of the regional seas: the Baltic Sea (N = 1,832), the North Sea (N = 1,534), and the Mediterranean Sea (N = 1,298). The non-parametric Kruskal-Wallis H-test is applied to examine whether statistically significant differences can be detected among the three selected regions, accompanied by pairwise comparisons using the Mann-Whitney U-test.

The Kruskal-Wallis H-test results revealed that there are statistically significant differences (at least at the 5% significance level) for all selected perception variables across the three marine regions (see Appendix Table 10 in the Supplementary Material for the test statistics). Table 2 shows the results of the pairwise comparisons across the three regional seas. First, testing for differences in respondents’ perceptions of frequency and magnitude reveals that respondents from the Mediterranean indicated that they saw MPL more often (U = 806316.000, p < 0.001) and observe larger quantities (U = 449297.000, p < 0.001), significantly more so than respondents from the Baltic Sea region. The size of this observed litter, however, is significantly smaller than the size of litter observed by residents from the North Sea region (U = 581673.000, p < 0.05). Second, comparisons indicate that the respondents from the Mediterranean Sea region are significantly more concerned than residents from the Baltic (U = 840597.000, p < 0.001), and North Sea region (U = 972319.000, p < 0.001), about MPL. Third, Table 2 shows that perceived negative consequences of marine litter are rated significantly lower by respondents from the North Sea, whilst mean scores of participants from the Baltic and Mediterranean Sea regions are only significantly different for perceived consequences on human health (U = 938988.500, p < 0.05). Fourth, the Kruskal-Wallis H-test showed statistically significant differences for the respondents’ self-responsibility (H(2) = 71.737, p < 0.001), self-perceived knowledge (H(2) = 342.156, p < 0.001) and actual knowledge (H(2) = 49.155, p < 0.001) across the three selected regional seas. Table 2 presents that the highest mean scores were found in the Mediterranean Sea region, whilst significantly lower scores were found for the respondents residing in the North Sea and Baltic Sea region. In general, Mann-Whitney U-tests results presented in Table 2 show that respondents from the North Sea region rated in general lower on the three perception component groups than respondents from the Baltic Sea region (with the exception of frequency of seeing MPL and subjective knowledge) and the Mediterranean Sea region. Respondents from the Mediterranean Sea, in turn, score comparably high to those from the Baltic Sea region.

Positioning our results in the literature is hampered by the absence of comparative country studies and diverging research methods and approaches. Nevertheless, some interesting comparisons can be made. General patterns in our results, regardless of the country of investigation, show that public knowledge about recyclability is low, but general concern about MPL is high, and the impacts of MPL are perceived to be most threatening for marine life. These findings about public concern and impacts of MPL are in line with previous findings. Forleo and Romagnoli (2021) found that Italians consider the threats of MPL for marine life most harmful, whilst Hartley et al. (2018) also confirm that the negative effects of marine litter on the marine environment are perceived more severe than for coastal appearance, human health and the fishing industry.

We also identified several differences across countries. In the present study, 70% of the respondents noticed the presence of MPL, with shares of respondents observing MPL ranging across countries from 54% in Estonia to 79% in Netherlands. These numbers are lower than those reported by Hartley et al. (2018), where 95% of the respondents in 16 European countries reported seeing litter when visiting the coast. However, our study asked about plastic litter specifically, whereas Hartley et al. (2018) considered all types of litter. Both studies found that a high prevalence of plastic was observed in European marine and coastal ecosystems, reiterating the scale of the MPL problem and need for action to reduce the amount of MPL. On the one hand, seeing marine litter did not influence respondents’ concern or awareness of consequences. On the other hand, knowledge and personal responsibility to reduce MPL were affected by the presence of litter. These results could indicate that those who have more knowledge or feel more responsible are more aware of the presence of MPL, or vice versa. As argued by Rayon-Viña et al. (2019), seeing MPL could motivate the sense of responsibility and search for information about MPL.

While levels of objective knowledge hardly differ across countries, ratings of subjective knowledge vary considerably. An illusion of knowledge was most evident among Italian and Greek respondents, as were high levels of concern about MPL and its associated negative effects, with almost 90% of the Greek and Italian respondents indicating concerns about MPL. These findings suggest that considering yourself knowledgeable about MPL contributes to feelings of concern. However, more research is needed to further explore the causality of these relationships.

With regard to responsibility, significant differences were found, not only in terms of responsibility to reduce MPL, but also to clean-up MPL. Although less than half of the respondents in Netherlands and Italy confirmed feeling responsible to reduce MPL, up to 90% of the respondents acknowledged this in the other investigated countries. In general, companies, consumers and the EU were identified as being responsible to clean up MPL, while the tourism and fishing industry are considered least responsible for the perceived littering. These findings correspond to some extent to the results by Hartley et al. (2018), who found that retailers, industry and government were perceived as main responsible for littering. The general public believed that consumers carry responsibility for cleaning up MPL, but it is not exactly clear whether they consider themselves part of this group. It is remarkable that respondents from all countries ascribe low responsibility to fisheries, although the fishing industry is widely recognized to be a significant contributor to MPL entering the oceans. For example, Consoli et al. (2018, 2020) found that derelict fishing gear represented 32% of the overall litter in a coastal area of the central Mediterranean Sea, and 97% on the seabed of the deep water of Malta, which is exploited by local fisheries using fish aggregation devices. Furthermore, we detected some differences in perceptions of assigning responsibility across countries: Germany and Estonia have a somewhat different view as they ascribe more responsibility to governmental institutions and agencies, such as the EU, national governments and waste management companies, and significantly less to private actors, including companies and consumers.

This study presents a unique database allowing for geographical comparisons of MPL perceptions, but it also has some limitations. First, despite the fact that the survey was distributed online by a professional survey company and completed anonymously, we must recognize that self-reported answers may lead to social desirability biases, potentially resulting in higher and more favorable results related to public perceptions. Moreover, variabilities in sample size and gender distribution of the national respondent groups, as well as cross-cultural survey response patterns, may influence the survey outcomes. For example, van Herk et al. (2004) found that respondents in the Mediterranean countries generally scored higher on acquiescence and had more extreme response styles than those in the Northwestern European countries. Other external factors possibly influencing responses could be the fact that the survey was carried out in times of a global pandemic. Concerns about the impact of the COVID-19 pandemic could either crowd out an individual’s worry about other issues, also referred to as the Finite Pool of Worry (Weber, 2010; Botzen et al., 2021), or lead to a spill-over effect by generalizing one’s concern to other worries, referred to as the Theory of Affect Generalization (Johnson and Tversky, 1983). The exact impact of the pandemic on the general public’s concern toward environmental problems is as of yet unknown and needs to be explored more empirically.

Additionally, it must be noted that visual perceptions of MPL cannot be compared with data about actual presence of MPL. Comparisons with observed MPL could improve the understanding of an individual’s processing of visual perceptions and allow for further examination of the influence of psychological and external components on one’s own observations and experiences. Since our study did not ask for specific locations of litter observations, such as coordinates, we cannot compare observations with monitored amounts of MPL. Furthermore, while this study covers a range of EU countries, when it comes to MPL, a global strategy to prevent and reduce MPL is needed. Research into public perceptions in countries outside Europe, such as the study of Arulnayagam (2020) focusing on public perceptions of MPL in Sri Lanka and the research by Van Rensburg et al. (2020) examining beachgoers perception of single-used plastic (SUP) use in South Africa, contribute to a more thorough understanding of regional differences and similarities in perceptions of MPL.

Finally, the results of this study already point to differences in perceptions not only between but also within countries. Therefore, future research should explore these intra-country differences in more detail to contribute to a better understanding of the differences between the interests, attitudes, behaviors and information needs of different social groups, which could lead to more specific communication and engagement initiatives to further incentivize participation in the fight against MPL.

The findings of this study have several policy implications. First, since knowledge is an important predictor of perceptions and actions (Vicente-Molina et al., 2013; Soares et al., 2021), enhanced educational initiatives aimed at increasing the general public’s understanding of MPL could translate into notable changes in desirable individual behavior. Results of this research have shown that the general public’s understanding of recyclability of plastics is particularly low in all surveyed countries, which indicates a clear international need for more information dissemination on this topic. Increased knowledge on recyclability could stimulate pro-environmental behavior and increase support for implementation of existing and new European policies, such as the EU Directive 2015/720, which sets targets to reduce the consumption of lightweight plastic carrier bags, and the upcoming EU Directive 2019/904, also known as the SUP ban.

Second, variations in levels of self-responsibility to reduce MPL across countries indicate that some nations should increase efforts to enhance citizen’s sense of personal responsibility and to facilitate their engagement in tackling marine litter. The Dutch and Italian government authorities, in particular, should design engagement programs that address the key role of citizens’ contributions to combat marine litter, since the sense of self-responsibility is particularly low in these countries. Moreover, Estonians and Germans appear to rely on and potentially wait for actions taken by the EU and their national government. Although governmental actors must indeed take their responsibility in developing strategies, setting targets and making policies, they should also take steps to make citizens and companies aware of their contribution and to incentivize them to prevent and reduce MPL.

Third, since the results showed that high levels of responsibility are ascribed to companies producing and using plastics, extended producer responsibility measures should be promoted. The upcoming single-use plastic ban (European Commission, 2019/904) includes extended producer responsibility schemes covering the costs to clean up litter, as well as the costs of awareness raising measures to prevent and reduce marine litter. These measures apply to various products, such as tobacco filters, balloons, and fishing gear and are not only in line with the polluter pays principle, but, as our research shows, also in line with the general public’s opinions. The fact that the public holds companies responsible supports the EU’s efforts to develop extended producer responsibility schemes to encourage innovation, product development and the use of sustainable alternatives.

Finally, since certain perceptions are similar across nations, such as lack of knowledge of recyclability, decision-makers could collaborate to develop joint strategies to target regions as a whole. For example, the content of particular educational initiatives could be shared throughout EU Member States. However, decision-makers should also be cautious to use a “one-size-fits-all” approach, since similarities in perceptions do not necessarily mean that strategies and measures can successfully be duplicated or have similar impacts in different countries. Differences in perceptions across surveyed countries indicate a need for the development of differentiated national strategies, tailored to match the needs of the general public in specific countries and regions. However, previous studies (e.g., Soares et al., 2021) have shown that, in additional to one’s nationality, other socio-demographic and -economic factors also influence individuals’ perceptions. Therefore, insights in perceptions across countries are an important initial step, but not a final one, for developing international and country-specific policy and communication efforts to encourage community participation in addressing MPL pollution.