Food interests, preferences and behaviours: a profile of the sustainable food consumer

Purpose – Sustainability challenges are omnipresent. This study aims to identify consumer segments based on food consumption practices from purchasing to disposal. A priori work identified quality attributes, food responsibility,dietarychoicesandfoodorganisationandmanagementwithinthehomeaskeyinfluences.Eachrepresentsanopportunityforconsumerstoemploymoresustainablefoodbehaviours.Aprioriworkidentified severalindicatorsforeachoftheseinfluences.Thisstudyexploredthesuitabilityoftheseindicatorsformeasuringtheidentifiedkeyinfluences. Design/methodology/approach – The authors used an online survey of 324 Irish food consumers in January 2021 to address online food interests, purchasing preferences and purchase behaviour linked to food sustainability. Findings – The authors identify four consumer segments – food dabblers, food appreciators, pro-sus and pressured – which present a holistic view of the sustainable behaviours practiced by food consumers. The findingsprovideinsightintotherangeofsustainability-relatedfoodbehavioursactionedby consumers – from interestinmeat-freeproducts,organicsandlocalproducetohavinganorganisedstockathomeandminimising waste.Thefindingsshedlightonhowconsumersintegrateelementsofsustainabilityintotheirfoodlifestyles. Originality/value – This study captures sustainability-related food behaviours from the point of purchase through to consumption and disposal and identifies four new consumer segments based on interests, preferences and behaviours.


Introduction
Consumers play a significant role in addressing the sustainability challenges of food systems (Hedin et al., 2019;Vermeir et al., 2020;Verain et al., 2015).Globally, 36% of household material footprint (a consumption-based indicator of resource use) (Wiedmann et al., 2015) is attributed to food consumption (Ivanova et al., 2015).The food choices and food consumptionrelated behaviours of consumers is increasingly important (Vermeir et al., 2020).Reported consumer demand for healthy, sustainable food continues to rise (Pricewaterhouse Coopers (PwC) 2021).However, consumers tend to hold producers and manufacturers accountable for making food systems more sustainable (Eurobarometer, 2020;Mintel, 2021) indicating a lack of consumers' sense of responsibility (Luchs et al., 2015).As much as 30% of the environmental impact of European Union (EU) households can be attributed to food consumption practices (Notarnicola et al., 2017).Over half of all food wasted in Europe is concerned with food waste at the household level, mainly due to poor food management habits and behaviours (Principato et al., 2020).Thus, the role of consumers in terms of their product choices and food behaviours (e.g., buying local produce or curtailing meat consumption) is crucial (Vermeir et al., 2020;Verain et al., 2015).Quality attributes, approaches and responsibility around cooking and storing food, dietary choices, and food waste generation are all implicated (Tukker and Jansen, 2006;Reisch et al., 2013).Consequently, private households, as key actors in this context, are essential in emerging solutions (Schanes et al., 2018).According to McKinsey and Company, a new shopping reality exists for consumers with an increase in basket size, a decrease in shopping frequency and an increase in online grocery shopping, alongside a rise of 54% points for cooking within households (McKinsey and Company, 2020).However, it is not clear whether these changes will persist over time (Giudice et al., 2020).Transitioning to circular economy principles has been noted as a means of bolstering the sustainability of food systems.This requires circular food behaviours, such as conscious consumption practices and consumer demand for green products by consumers (do Canto et al., 2021).Fittingly, according to EU communications, 'green recovery' is central to the post-pandemic world (see Giudice et al., 2020).
This study aims to identify consumer segments based on key food consumption practices.Unlike previous studies that focus solely on point-of purchase behaviours (Carrero et al., 2016;Polzin et al., 2023;Su et al., 2019) or routines in the household (Chen, 2023;Jastran et al., 2009;Marshall, 2005), this study considers consumption behaviours ranging from point of purchase to point of disposal.A priori work identified quality attributes, food responsibility, dietary choices, and food organisation and management within the home as key influences.A priori work has also identified several indicators for each of these influences, with this study exploring the suitability of these indicators to measure the identified key influences.Broader willingness to engage in environmentally friendly practices beyond food-related behaviours is also considered.

Sustainability-related food behaviours
Europeans prioritise taste, safety and cost over sustainability concerns (such as climate change, biodiversity loss and carbon emissions) (Eurobarometer, 2020).Mintel's (2021) report supports this by highlighting that "consumers won't compromise on quality or brand familiarity in the name of sustainability" (Mintel, 2021, p. 16).With regard to the range of sustainable consumer behaviours, the most popular are those that are simple and frugal (Mintel, 2021).Notably, Generation Z consumers are mostly aspirational about sustainabilityrelated behaviours, with their widely reported eco-sensitivity appearing to be overrated and ideals are not translated into purchasing behaviour (PwC, 2021).Food considerations, particularly in wealthier societies, are complex (Verain et al., 2012).People may lack a clear understanding of the environmental threats to humanity or who should be responsible Food interests, preferences and behaviours (Thøgersen, 2014).Emphasis has been placed on individuals' dual roles as citizens and consumers to address environmental sustainability issues via green consumerism (e.g., purchasing fair trade and organic, buying less packaging, and supporting local producers) (Barr et al., 2011;De Tavernier, 2012;Tittarelli et al., 2022).However, the checkout is not the only place where consumers can make a difference.The reality of everyday practices (e.g.shopping, cooking and eating) is that habits or routines are embedded in consumers' social, structural and cultural lives (Southerton, 2012).
The heterogeneity and richness of pro-environmental behaviour have been well documented in the segmentation literature (Haider et al., 2022;Lavelle et al., 2015;McDonald et al., 2012).Many studies have focused on food consumption (Carrero et al., 2016;Macharia et al., 2013;Su et al., 2019;Verain et al., 2012Verain et al., , 2015)).Environmental values and concerns, socio-demographic variables, and psychological factors have dominated much of this research (Testa et al., 2021).Empirical research tends to focus on point-of-purchase settings (e.g., Carrero et al., 2016;Su et al., 2019) or habitual behaviours (e.g., Chen, 2023;Gilg et al., 2005).Less attention has been paid to understanding sustainable behaviours at various stages of the consumption process.From purchase to disposal, each step represents a touchpoint for enhancing or compromising sustainable food practice.Sustainability-related food behaviours in this study refer to quality attributes (e.g., purchasing local food), food responsibility (e.g., using a shopping list), dietary choices (e.g., adopting vegetarianism) and food organisation and management within the home (e.g., avoiding food waste).
Regarding dietary choices, experts have advised shifting to diets that include more wholefood plant-based and less meat-based foods (Willett et al., 2019;Lea et al., 2006).As a result some citizens are reducing their meat intake and consuming vegetarian meals and associate such actions with being more sustainable (Sanchez-Sabate et al., 2019).In addition to recommendations on what to eat, certain food quality attributes are associated with being more sustainable.Local/domestically produced, seasonality, organic, along with fairtrade have been associated with a more sustainable system (Macdiarmid, 2014;Zepeda and Deal, 2009).Generally, foods promoting these characteristics are perceived as more expensive and as a result may not be chosen by those who are price sensitive (Verain et al., 2012;Aschemann-Witzel and Zielke, 2017).Indeed, consumers may choose engage in one or the othersustainable product choice strategy based on product attributes (e.g.organic, local, free range, etc.), or engage in consumption curtailment strategies (e.g.reduce quantities of meat consumption).Consumers are likely to differ in the strategies which appeal to them most, and thus their sustainable behaviours may centre around one strategy over the other (Verain et al., 2015).Given the consensus around the resource intensity of animal-based products, many advocate curtailment or abstinence (e.g.decreasing portion size, decreasing frequency or adopting vegetarian/veganism) within this food category to address and lower the sustainability impact of one's diet (see Garnett, 2011).In Verain et al.'s (2015) distinction between product-oriented consumers and curtailers, where sustainability motives and subjective knowledge of sustainable food predict both types of behaviour, they found that product-oriented consumers attach more importance to social norms and are subjective in their judgements of sustainable foods, whereas curtailers are usually older, female and have lower incomes.However, eating meat is deeply embedded in the routines of everyday life, and thus reducing or omitting meat poses significant challenges in contexts where meat consumption is expected and desirable (Mylan et al., 2016).Indeed, as attested by Carrigan (2017, p. 17), consumer responsibility 'manifests itself flexibly' and is impacted by complex contextual influences.Empirical evidence suggests that, despite positive attitudes towards 'green' products and intentions to purchase, this does not consistently follow through to the checkout or result in long-term sustainable behaviours, resulting in the well acknowledged attitude-behaviour gap (Moraes et al., 2012;Szmigin et al., 2009;Tittarelli et al., 2022;Yamoah and Acquaye, 2019).
Nonetheless, reducing food consumption, particularly in the Global North, is one of the most critical strategies for sustainable consumption; however, very few studies have investigated this (do Canto et al., 2021).Where the issue of overconsumption is researched it is based primarily on reducing food waste (Gollnhofer, 2017;Makhal et al., 2020) or curtailing the consumption of animal-based products (Verain et al., 2015).Renewed interest in local food supply chains as a means of transitioning to sustainable food consumption offers economic, environmental and social benefits.Purchasing from local, short supply chains 'reduces storage and transportation, provides a better supply-demand balance, creates more transparency and tracking and contributes to waste reduction' (Giudice et al., 2020, p. 11).Indeed, food purchased from local channels is ascribed a higher value by consumers and, consequently, is much less likely to enter the waste stream (O'Neill et al., 2022).
With regards to food organisation and management, consumers do not hold a careless disregard for food they waste at home (Evans, 2012a).Rather, food-related routines such as meal planning, shopping, cooking, eating, storing and cleaning routines play a major role in food waste (Stancu et al., 2016) and "[. ..] food waste is a largely unintended outcome of entangled daily routines revolving around food" (Dobernig and Schanes, 2019, p. 480).Impulse buying is a significant antecedent of food waste (Lahath et al., 2021;Porpino et al., 2015;Schmidt, 2016).For food responsibility, consumers who do not plan their grocery needs and who shop without shopping lists are more exposed to marketing cues and internal stimuli such as emotions which may result in impulsive buying (Massara et al., 2014).Specifically, making shopping lists, meal planning and checking food stocks before shopping have all been shown to reduce food waste (Quested et al., 2013).Additionally, activities around waste minimisation also represent an action that households can take to be more sustainable (Thøgersen and Grunert-Beckmann, 1997;Aschemann-Witzel et al., 2015).In the case of food, the use of leftovers, how people deal with foods past its best-before-date and attitude to generating food waste could influence this.Where prior planning doesn't occur and impulse buying leads to surplus, consumers try to manage the "anxiety laden" process of wasting food via complex procedures that allow them to dispose of surplus or leftovers with as little guilt as possible (Evans, 2012a, b).While the majority of the food surplus ends up in the waste stream, it is usually not immediately put there.Rather, procrastination around wasting food means that it is stored, usually in the fridge, and then "quietly but actively forgotten' (Evans, 2012b(Evans, , p. 1130) ) until rendered inedible and placed in the bin (Evans, 2012a, b).
Food marketing and retailers contribute to consumer-related food waste via expiration date confusion, pricing strategies and encouragement to impulse purchase large pack sizes (Aschemann-Witzel et al., 2016;Davenport et al., 2019).Purchasing food online may lead to an increased likelihood of wasting food due to psychological distance (Ilyuk, 2018) or may lead to less waste due to less impulse purchasing and more structured eating habits (Berg and Henriksson, 2020).Here, recognising the material context of food practices is important, as it highlights how the infrastructure of provision, and indeed the material properties of the food itself, affects food waste (Evans, 2012a;Mylan et al., 2016).

Procedure and respondents
We collected data through an online survey of Irish food consumers conducted in January, 2021.The survey formed part of the Horizon 2020 ERA-NET SUSFOOD (SUStainable FOOD) (Horizon, 2020) project "PLATEFORMS" and included questions addressing online food purchase behaviour, purchasing preferences and food choice motivations, attitudes and behaviours that are linked to food sustainability.The draft online survey was piloted with a sample of 40 respondents.Ethical approval for this study was obtained from the Social Research Ethics Committee of the University.

Food interests, preferences and behaviours
According to Berry (2018) online grocery shopping in Ireland accounts for a small proportion of purchase at 2.2% (of sales) with an average grocery spend per online purchase in 2017 of V66.43.At the point of undertaking the survey in January 2021 this had changed due to the lockdown, with for example, McKinsey and Company (2022) suggesting that sales had increased by 50% in North America during the pandemic.However, even with such an increase, the overall percentage of the population engaging in online grocery shopping remains low.This presented a survey recruitment challenge.As a result the survey was delivered online via the Qualtrics platform (QualtricsXM, Dublin, Ireland) to their consumer panel.Standard quality control protocols were in place to ensure the integrity of the data (e.g., avoidance of over-surveying participants and professional survey takers).As an addition precaution, to capture a diverse set of users, quotas were set to ensure a balanced sample on gender (minimum to 40% for male or female) and age (50% under 45 and 50% 45 and older).We also sought to recruit a mix of urban and rural dwellers.Descriptive statistics for the sample are presented in Table 1.A total of 400 respondents were recruited for the study.Data cleaning excluded 66 participants from the analyses because of a lack of dispersion in their responses to a range of variables.With regard to adequacy of sample size for factor analysis, consideration was given to accepted guideline which indicates that a sample size should ensure 10 to 20 observations per variable (Nunnally, 1978;Arrindel and van der Ende, 1985;Pett et al., 2003).For the purposes of exploring key influences on sustainable food consumption practices, 17 variables were identified and included in this study (see Table 2), consequently sample size was set at 400 to ensure adequacy of sample size.Data analysis was conducted using Statistical Package for the Social Sciences (SPSS) version 28.

Measures
Purchasing behaviour questions.We measured the frequency of online purchases for 14 food categories with a five-point frequency ordinal scale ranging from once a week to never.These frequency data were used to profile the identified segments.The 14 categories were dairy products and eggs, meat and meat products, fish and seafood, bread and bakery products, fresh fruit and vegetables, oils, cans/bottled food products, seeds, dried fruits and pulses, nonalcoholic beverages, alcoholic beverages, coffee and tea, snacks, chocolate and candies, pasta, rice and grains, and products for special dietary needs.
Sustainability related practices indicators: food interests, preferences and behaviours indicator statements.In addition to literature informing the broad food sustainability activities of households, as previously discussed, an informant perspective was used to identify the behaviour and practice characteristics that were relevant to this study.An earlier stage in the Food interests, preferences and behaviours work involved in depth interviews to identify the most salient behaviours and preferences that consumers associate with sustainable practices.These behaviours and preferences relate to pre-purchase planning, acquisition, dietary preferences, product attribute preferences, food management in the household, and dealing with waste and surplus.This work was undertaken by research partners in the Plateforms SUSFOOD2 funded research project.Following an initial analysis of country specific data (Norway, Sweden, German, Italy and Ireland) the project team jointly discussed participant behaviours and characteristics that were connected to sustainability and how this translated into measures.Dietary choices (e.g.avoiding meat, eating vegetarian meals), product attribute choices (e.g seasonality, origin) and organisation and management of food related activities (e.g.responsibilities, shopping planning, and managing leftovers) were common themes across all partner countries.The statements allowed for conflicting perspectives on the same characteristic created due to issues such as cognitive dissonance.Table 2 presented these statements.
Indicators of sustainability were developed through the measurement of 17 food behaviour and practice characteristics.Two statements per characteristic (one positive and one negative), which have been associated with food sustainability in consumers everyday food lives, were employed.Using 7-point Likert type scale response statements, respondents were asked to indicate how well each statement described them from not at all (1) to very well (7).Each indicator score was calculated by subtracting the negative from the positive item, with scores ranging from 6 to À6.A higher score indicate that the consumer is displaying more of the measured characteristic.Table 2 presents the statements for each of these indicators.
"Being more sustainable" measures.Two measures of 'being more sustainable' were included.The first was an intention to change food behaviours for sustainability reasons and reflects a widely used stages-of-change measure.Respondents were asked to indicate which of six behaviour change options best described them (the options are presented in Table 5).The second represented willingness to be environmentally friendly and included six statements from the scale designed by Kautish et al. (2019), based on the work of Abdul-Muhmin (2007) and Zabkar and Hosta (2013).These measures were used to profile the segments.
As outlined in the results section, exploratory factor analysis using principal component analysis with varimax rotation was applied to identify the underlying structure within a set of observed variables (Stewart, 1981).Interrelationships among statements are identified and represented through underlying dimensions (factors) (Hair et al., 1998).This method is commonly used in similar segmentation studies (Grunert et al., 1998;Ryan et al., 2002;Buckley et al., 2007).K-means clustering was conducted to identify segments.Both parametric and nonparametric tests were applied with similar results observed.In such situations, as suggested by Turner et al. (2020), it is reasonable to report non-parametric equivalent.Thus ANOVA (analysis of variance) with post-hoc Bonferroni tests were employed to identify significant differences based on attitudinal and behavioural preferences between these segments.

Results
Exploratory factor analysis, using principal component analysis with varimax rotation, was applied to identify the underlying structure within a set of observed variables (Stewart, 1981).Interrelationships among statements are identified and represented through underlying dimensions (factors) (Hair et al., 1998).This method is also used in many segmentation studies (Grunert et al., 1998;Ryan et al., 2002;Buckley et al., 2007, Wongprawmas et al., 2021).
The seventeen sustainability related practices indicators (presented in Table 2 and discussed in the methods section) based on various stages of food consumption (from purchase planning to waste management) were included in the factor analysis.Statements with a <0.40 factor loading on more than one factor were excluded as scores of over 0.4 are considered stable (Guadagnoli and Velicer, 1988).Following this approach, four factors (measured by 12 indicators) that explained 60.86% of the data variation were identified.The Kaiser-Meyer-Olkin measure of sampling adequacy (0.78) and Bartlett's test of sphericity (χ 2 5 1351.28.45; p-value <0.001) confirmed the suitability of these data for factor analysis.The final factor solution made sense in the context of food practices.Reliability analysis using Cronbach's α was performed to test the reliability and internal consistency of each of the four factors (see Table 3).Cronbach's coefficients range from 0.54 0.85.Based on Streiner's (2003) observations, these coefficients were deemed satisfactory.Three of the four factors identified have coefficient alpha of 0.8, however one factor, "organisation", has 0.54.We retained this due to its relevance.While less than ideal, "organisation" captures a number of relevant variables (minimising waste, managing leftovers, organising purchasing).Further, presentation of the items used may serve to prompt future research that explores this "organisation' construct.The factors, statements, factor loadings and reliability coefficients are presented in Table 3.
The first factor, Quality Attributes, represents the potentially important product attributes associated with sustainability.These include local trade, seasonal trade, fair trade and country of origin.The second factor, Food Responsibility, includes statements addressing the level of responsibility the individual has in the provisioning and preparation of food in the household.The third factor, Dietary Preferences, relates to meal choices associated with more environmentally sustainable practices.This factor focused on avoiding meat and consuming vegetarian meals.The fourth factor, Organisation, represents the degree to which food practices are planned and organised.This includes the tendency to be organised when purchasing food, managing leftovers and avoiding food waste.
To identify the existence of segments, K-means clustering was conducted using the food provisioning practices factor scores.A range of cluster solutions (2-5) was evaluated by comparing the mean variable scores across the identified clusters.Based on this assessment, a four-cluster solution was identified as the best explanation for these data and reflected the differences in the sample population.The clusters were named based on their distinguishing features.The segments identified are labelled, "Food Dabblers", "Food Appreciators", "Pro-Sus" and "Pressured".
Figure 1 illustrates the relative positioning of each identified segment on the segmenting variables, while Figure 2

Food interests, preferences and behaviours
Dabblers are distinguished from other segments by their lower levels of responsibility for food and meal provisioning.Food Appreciators display strong interest in quality-related food choice attributes.The Pro-Sus segment tends to engage in a range of practices linked to positive sustainability outcomes.Their dietary choices and organisation of food activities, to a lesser extent, suggest that they are taking personal responsibility for food sustainability.Finally, the Pressured segment, while responsible for household food provisioning, is not interested in food-related practices, and food is a chore for them.
The detailed profiling of each segment follows and is supported by the information presented in Tables 4-7.ANOVA-analysis with post-hoc Bonferroni tests was employed to identify significant differences based on attitudinal and behavioural preferences (Table 4).This profile examines differences at the factor level and further explores this by considering the core measures that primarily represent each factor.Segments were also profiled based on their intention to change food behaviours for sustainability reasons and online food purchase behaviours in food categories (Tables 5 and 6).Finally, segments were profiled based on demographic characteristics: age, gender, occupation, number of adults and dependents in the household, education levels, social classification and dwelling location (urban/rural).The characteristics of the significant differences are presented in Table 7.No significant differences were observed in education level, number of dependent children and dwelling location (p > 0.05).
Segment 1: Food Dabblers (n 5 104, 32%) The "Food Dabblers" segment is less likely than the other segments to take responsibility for acquiring and cooking food in their homes (p-value < 0.001).For example, their mean scores of 0.54 and 0.77 for food preparation and food purchase respectively compared to 5.24 and 4.97 in the case of the Pro-Sus segment (Table 4).They are most likely not the primary food À1.017 c Note(s): * Mean Score range from 6 to 6.A higher score means that the segments displays more of the characteristic **The superscripts "a-d" indicates significant mean differences between segments at the 95% confidence interval (p < 0.05) based on ANOVA test with Post hoc Bonferroni.Superscript letters should be read by row.If letters differ between two segments the mean scores are significantly different.For example, in the case of quality attributes, 'food appreciators" mean score is significantly higher than the other three segments, as indicated with the "a" superscript, the remaining three segments means scores are not significantly different from each other as indicated with common "b" superscript annotation.In each row the green shaded cells highlight the segment with the highest mean score on the item and the yellow indicates the lowest mean score Source(s): Authors work Food interests, preferences and behaviours decision makers in their households.In comparison, Food Appreciators and the Pressured are more likely to engage in sustainable dietary choices (p-value <0.001).They were more likely to reduce their meat consumption [1] than both of these segments (p-value <0.001) and more likely to eat vegetarian meals frequently compared to Food Appreciators (p-value <0.001).Additionally, they were more organised in their food behaviours than the Pressured and less organised than the other two segments (p-value <0.001).While their ability to manage leftovers is similar to the Pressured (p-value >0.05), they are less likely to create waste and be disorganised in their shopping than this segment (p-value 0.001).Their outlook on quality attributes is generally similar to that of the Pressured and Pro-Sus.On closer inspection, they place less importance on all quality attributes (local, domestic country of origin (COO), seasonality and fair price for producers) except one (price-quality) than on the Pro-Sus (pvalue <0.001).Their attitudes towards the range of quality attributes examined were not significantly different from those of the Pressured.

Food interests, preferences and behaviours
Their online purchase frequency for specific food categories also differentiates them from the other segments (Table 6).They were more likely (75%) than the Pro-Sus (61%) to purchase fish and seafood online; additionally, their purchase of oils was noteworthy, with 18% indicating that they buy this product weekly, compared to 2% for the Pressured.Interestingly, they are also more likely (p-value <0.05) to be online purchasers of products that meet special dietary needs (70%) than Food Appreciators (40%) and the Pressured (43%).
A higher proportion of this segment (36%) identified themselves as upper or upper middle class compared to the other segments, Food Appreciators (5%), Pro-Sus (12%) and Pressured (15%) (Table 7).Food Dabblers are more likely to be composed of a younger cohort than other segments.Indeed, almost 30% of this segment is aged 18-29 which compares to 12% for Food Appreciators.Compared to the other segments, they were more likely to live in households with three adults (p 5 0.048).Indeed, 38% of this segment lived in 3þ adult households in comparison to 22%, 27%, and 32% for Food Appreciators, Pro-Sus and Pressured respectively.They are more likely to be in some form of employment (full-time or parttime), with 80% falling into one of these categories compared to 64% of the Pressured.
Segment 2 food appreciators (n 5 94; 29%) The "Food Appreciators" are different from the other segments given their focus on quality attributes that have been associated with sustainability (p < 0.001).In comparison to the other  12) 4( 7) 18(6) Total 104( 100) 94( 100) 66( 100) 60( 100) 324(100) Note(s): *Each subscript letter denotes a sub-set of cluster number of case categories whose column proportions do not differ significantly from each other at the 0.05 level Source(s): Authors work Table 7. Segment demographic profile BFJ 125,13 three segments they appear to have a stronger interest in food placing greater emphasis on quality over price (x -5 2.30; p < 0.001) and prefer/are more interested in local (x -5 3.58; p < 0.001) and domestic foods (x -5 3.15; p < 0.001).This suggests that provenance appears to play a notable role in their food choice processes, and this differentiates them from other segments.With regard to organisation, they are second only to the Pro-Sus in the management of leftovers (x -5 3.68), and trying not to be overly disorganised in their purchases (x -5 1.58) (Table 4).The importance of a fair price for producers and in-season purchases is similar for these and the Pro-Sus (p-value <0.05) and means more to them compared to the other two segments.They are the second most organised segment, with the Pro-Sus paying more attention to waste minimisation and managing leftovers.Their commitment to a meat diet was evident in that they were the least likely to eat vegetarian meals (x -5 À2.58) or meals without meat (x -5 À2.47) (p < 0.001).
It is interesting to note that 62% of this segment has changed their food behaviours to become more sustainable.They and the Pro-Sus displayed a similar behaviour change profile which was significantly different from the Food Dabblers and Pressured segments (p < 0.001) in that they embedded some food-related sustainable actions in their everyday lives (Table 5).
There are a few notable differences between online food purchase behaviour and the sample average (Table 6).They have the highest proportion of individuals who never buy dietary and special requirement foods at 60%, which is comparable to the sample average of 46% (p-value<0.05).Additionally, they were significantly more likely to never purchase fish online (44%) than Food Dabblers (25%) and Pressured (27%) (p < 0.05).
Food Appreciators were more likely to include those aged >60 years (p < 0.001) (Table 7).Indeed, 29% of this segment were 60 plus compared to 14% for the sample average and 4% for Food Dabblers.They were the least likely of all segments to live in a household with three or more adults (p 5 0.048) and were significantly more likely to be male compared to the other segments at 59% compared to a sample average of 42%.
Segment 3: pro-sus (n 5 66, 24%) The "Pro-Sus" appear to be the sustainability early adopters amongst our food consumers.They are very much engaged with sustainability initiatives with their high mean scores for dietary change (reducing meat x -5 3.33; increasing vegetarian x -5 3.27) and organisation (minimise creating waste x -5 3.00; increasing vegetarian x -5 4.7) (Table 4).They were more likely to try to avoid meat-based meals and eat vegetarian meals than the other three segments (p < 0.001).Their organisation patterns also differ in that they are more likely to minimise waste (p < 0.001) and manage leftovers (p ≤ 0.047) than the other segments and are more likely to engage in organised in their purchases than Food Dabblers and Pressured segments (p < 0.001).Additionally, while placing less emphasis on quality attributes than Food Appreciators they are more interested in these than the other two segments.This segment displays a strong disposition towards a range of activities and preferences associated with a more sustainable food system.
This segment displays a similar pattern of responses to food-related sustainability behavioural intentions (Table 5) as the Food Appreciators with 38% of this segment indicating that they have already embedded some food-related sustainable actions in their everyday lives for more than six months and a further 27% having changed their behaviours within the last six months.Unsurprisingly, only 11% of this segment suggested that they had no intention of taking sustainable action within the next six months.This was compared with 18% for the total sample and 27% for the Pressured (p < 0.001).This segment's online food purchase behaviour reflects their tendency to avoid meatbased meals, as 39% never purchase meat or fish and seafood fish on such platforms (Table 6).Meat purchase behaviour is particularly noteworthy when compared to the other segments, where 7-19% of these segments never buy meat online (p-value <0.001).

Food interests, preferences and behaviours
While Pro-Sus spanned across the age categories, they were more likely to be female (75%) than male, compared to 53 and 41% for the Food Dabblers and Food Appreciators, respectively (Table 7).Additionally, compared to the other segments, there is a disproportionate number of Pro-Sus in the part-time employment category at 21%.This is compared to 10 and 12% for Food Appreciators and Pressured.
Segment 4: pressured (n 5 60, 19%) The "Pressured" maintain the least sustainable household food practices amongst our food consumers.They are not focused on sustainability, as evident in their lack of interest in avoiding waste (x -5 À1.85), and attention to the management of leftovers (x -5 0.87) (Table 4).While responsible for household food provisioning, of the 4 segments they are the least organised in their purchases (x -5 À1.02, p < 0.001).They are more likely to create waste and make unplanned food purchases (p < 0.001) than the other segments.Their management of leftovers was similar to Food Dabblers (p > 0.05) and significantly poorer than that of the other two segments (p < 0.001).They were among the least concerned with quality attributes, displaying a similar pattern to the Food Dabblers' segment (p > 0.05).Their dietary choices differ significantly from those of the Pro-Sus and fall between the Food Dabblers and Food Appreciators.Their position on avoiding red-meat-based meals was similar to that of the Food Appreciators (p > 0.05).This position on meat is further evidenced when one examines online meat purchases by just 7%, indicating that they never buy from this category (Table 6).This is comparable to the 39% for Pro-Sus.In contrast, in keeping with their profile, 57% never bought dietary and special requirements.This is compared to 30% for the Food Dabblers' segment.
Interestingly, over a quarter of this segment (27%) suggested that they had no intention to change their food behaviours for sustainability reasons within the next six months (Table 5).This was 11 and 12% for Pro-Sus and Food Appreciators, respectively (p < 0.001).An additional 28% of this segment indicated that they intended to take action within the next six months.This suggests that the level of procrastination was not as evident in the other segments (p 5 0.001).This female (75%) dominated segment displayed no other notable demographic characteristic differences from the sample population (Table 7).

Discussion
The aim of this study was to identify online consumer segments based on key food consumption practices, from purchasing to disposal, with a view to providing a holistic view of the range of sustainable behaviours practiced by food consumers.The consumer segments emerging from the findings are Food Dabblers, Food Appreciators, Pro-Sus and Pressured.Quality attributes, food responsibility, dietary choices and levels of food organisation and management are embedded in a complex mix of sustainability-related food behaviours practiced by the different consumer segments.Therefore, these segments present a picture of food consumers relating to not only their food purchase behaviours but also their interests, preferences and habitual behaviours in the home.
The mix of sustainable behaviours actioned in each segment may shed light on a broader willingness to engage in environmentally friendly practices beyond food-related behaviour.Considering the stated food behaviour change intentions due to sustainability concerns for the four segments along with their current behaviours in terms of quality attributes, level of food responsibility in the home, dietary choices and preferences, and levels of food organisation and waste management offers insight into overall dispositions towards engaging with sustainability.The quadrants in Figure 3 indicate willingness to engage (high or low) with sustainability-related behaviours (high or low).
From Figure 3, we can see that where Food Dabblers do not necessarily display behaviours consistent with pro-sustainability, their willingness to engage is high and therefore presents an opportunity to encourage more active purchasing and behaviour change towards sustainability.Pressured individuals display low sustainability-related behaviours along with an overall low willingness to engage, although there is some potential to draw on their good intentions so long as they fit within their pressured lifestyle.Pro-Sus consumers score high in both willingness and behaviour, making them the most engaged with sustainability, while the Food Appreciators display a good degree of sustainable behaviours; they may be unwilling to change certain aspects of their consumption, for example, their meat consumption.
The findings presented in this paper offer four distinct segments with different approaches to sustainable food consumption, integrating elements of quality attributes sought (e.g.local, seasonal, origin), level of food responsibility in the home (e.g.food preparation), dietary choices and preferences (e.g.meat avoidance, vegetarian), and levels of food organisation and waste management (e.g.best-before dates, using leftovers, food waste).
Food Dabblers are most distinguished from the other segments by their lower levels of responsibility for food and meal provisioning and are more likely to comprise of younger consumers than other segments.While not taking primary charge for food provisioning, relative to some of the other segments, they have a greater disposition towards some sustainable food behaviours.In comparison to Food Appreciators and the Pressured they are more likely to engage in sustainable dietary choices and are more likely to reduce their meat than both of these segments.Interestingly, the Food Dabblers had the highest proportion (20%) of consumers who intend to take food-related sustainability actions within the next 30 days, but they are the least likely to have changed their behaviours for at least six months, suggesting that this young cohort is indeed aspirational about sustainability-related behaviourstheir widely reported eco-sensitivity (PwC, 2021) appears to be either overrated or, more promisingly, not yet reaching the point where they have the capacity and/or need to action out their preferences, that is, this segment may be pro-active once they leave the family home and become responsible for their own food choices.However, based on their attention to eating less meat and more vegetarian meals, thus engaging in consumption curltailment strategies (Verain et al., 2015), the low percentage indicating that they have taken food related actions to become more sustainable may infer that, for now, they tend to follow certain trends and are less concerned about the environment.
Food Appreciators displayed a strong interest in quality-related food choice attributes, which distinguishes them from other segments.This segment has a strong interest in food, valuing quality over price via their preferences for origin, seasonality and fair trade.Given their appreciation for their food, particularly local food, this segment is more likely to be older males, manage their waste very well and not shop on impulse.Being second only to the most sustainable group in terms of waste minimisation and managing leftovers, Food Appreciators however, do not compromise meat consumption.They were the least likely group to go Food interests, preferences and behaviours without meat or eat vegetarians.These meat eaters, however, do not disregard their responsibilities when it comes to food sustainability and are, in fact, performing favourably both in terms of purchase-related behaviours (e.g.local) and habitual behaviours (e.g. using leftovers).

Source(s): Authors work
Notable in the character of Food Appreciators is their careful selection and use of resources and their general avoidance of waste.This segment displays personal norms around frugality, which ties with a quality consciousness that minimises impulsive purchases (through organised purchasing) and consumption.The attention given to enjoyable eating experiences by this segment reflects their interest in good food, where values are attained through careful curated choices which emphasises a preference for quality over quantity.A willingness to spend more per unit combined with a minimisation of waste appears as their modus operandi and ties with concepts associated with frugality (Bouckaert et al., 2008;Farrell, 2010).Frugality is also relevant to Food Dapbblers given their younger profile.Naderi and van Steenburg (2018) highlight frugality, along with future orientation, as a significant determinant of sustainable behaviour among millennials.In this context, connecting positive emotions, possibly through peer influence, to concepts of frugality and reinforcing these through social approval could offer a useful avenue to prompt greater responsibility for all aspects of food in everyday life and more sustainable behaviours.This type of action could be further supported by drawing attention to both immediate and long-term benefits for the self and environment.Frugality in this group may be due to limited resources, time and money.
The Pro-Sus segment displayed a tendency to engage in a range of practices linked to positive sustainability outcomes and thus represents the most sustainability-focused segment of the four.Indeed they appear to be at the forefront of behavioural change as they embrace many of the suggested food actions that support a more sustainable diet.This segment is the most notable for meat and vegetarian meal preferences.Their dietary choices and organisation of food activities suggest that they are taking personal responsibility in response to sustainability-related concerns such as carbon emissions, climate change and biodiversity loss (Luchs et al., 2015).This segment, more likely to be female, avoids meatbased meals in favour of vegetarian meals, more so than the other three segments and equally minimises waste and manages their food leftovers.These early adopters embrace a variety of positive purchase-related and habitual behaviours, with indications that many have already changed many of their behaviours to be more sustainable.
Finally, the Pressured segment, who are responsible for household food provisioning, are not interested in food-related sustainability practices.This segment does not display sustainable behaviours but rather appears to be time-poor and prioritises getting food on the table.They are likely to be female, employed outside the home and have full responsibility for food provisioning but lack any obvious interest in the food they provide beyond meeting functional needs.This group are disorganised in their purchasing behaviour and tend to lack in the area of managing leftovers and food waste.Although not concerned with quality attributes, they are similar to Food Appreciators in terms of their preference for meat.Interestingly, in this segment, almost a third suggested intentions to change their food behaviours towards sustainability in the next six months.This suggests that they are aware of the unsustainable aspects of their consumption, but have not yet managed to change their daily consumption purchases/habits.These are an important group to recognise in addressing the sustainability of the food system, as they appear open to change but procrastinate due to other lifestyle demands.Given the competing demands on time and energy experienced, attention could be given to creating conditions conducive to their engagement in more sustainable food behaviours.Innovations in how food producers and retailers engage with and deliver food to this group could help mitigate barriers to sustainable practices; these could involve innovative packaging and more malleable ordering delivery services.Such developments would evolve with consumer needs, offering a service that responds to ebbs and flows of food consumption patterns within the household.As this segment already engages in online shopping services, the utilisation of a range of digital technologies could support more personalised shopper experiences that aim to fit within the household's food lives while supporting sustainable food practices.
Triggering and reinforcing emotional connections with sustainable food practices and attributes warrants consideration.Ila and Sanjaya (2019) highlight the potential role of gratitude, pride and happiness in embedding sustainable practices.They suggest that it is important to "capitalise" on positive feelings triggered through the individual's engagement with these sustainable behaviours.Through the leveraging of emotion, specific sustainable practices could become more meaningful to groups, such as Food Dabblers and Pressured groups.For the latter, this may relate to next-generation security and quality of life with communications that trigger emotions such as guilt to initiate a behaviour and expression of gratitude that triggers pride and happiness to reinforce these.For the former, the emotions of joy and pride could be triggered by conveying what they can do with this, reinforced by confirming that these actions meet social/peer approval.

Conclusion
Food consumption and related behaviours is complex (Verain et al., 2012).Research has tended to focus on point-of-purchase settings (Carrero et al., 2016;Su et al., 2019) or habitual and routine behaviours (Chen, 2023;Gilg et al., 2005) but less attention has been paid to understanding sustainable food behaviours at various stages of the consumption process.Based on literature and prior research, sustainability-related food behaviours in this study referred to quality attributes (e.g., purchasing local food), food responsibility (e.g., using a shopping list), dietary choices (e.g., adopting vegetarianism), and food organisation and management within the home (e.g., avoiding food waste).This study offers insight into the sustainability-related food behaviours of four consumer segments -Food Dabblers, Food Appreciators, Pro-Sus and Pressured.This holistic view of the range of sustainable behaviours practiced by food consumers offers insight into the transition to "green recovery" in our post-pandemic world (Giudice et al., 2020) and the mix of sustainable behaviours actioned in each segment may shed light on a broader willingness to engage in environmentally friendly practices beyond food-related behaviour.Note 1. Table 2 presents the base statements used in the creation of indicators and forms the basis for reporting on each of the segments.Table 4 present the significant difference associated with these indicators.
Figure 1.Radar plot of mean factor scores by segmenting variable Figure 3. Disposition towards engaging with overall sustainability Question: How well do the following statements describe you as a food consumer?Answer on a 7-point scale, where 1 is DOES NOT DESCRIBE ME AT ALL and 7 is DESCRIBES ME VERY WELL Source(s): Authors work provides this detail at the sustainability indicator level.Food