The attitudes and practices of United Arab Emirates consumers towards food waste: A nationwide cross-sectional study

1.1 Literature review

The terms food loss waste (FLW), food loss (FL) and food waste (FW) are all used interchangeably in the literature, however Food Loss is commonly used to refer to upstream losses, the reduced amount of edible food, quantity or quality, occurring during initial production, postharvest and processing stages of the food chain; whilst the term Food Waste, refers to any raw or cooked quality foods, including inedible items (i.e., skin, seeds, etc.) that is produced for human consumption, and then discarded at the procurement and consumption stages.^11,14 For the present paper we focus on FW, however it is important to consider how FW relates to the wider elements of this concept. Lipinski et al. (2013), characterised the term FLW as occurring across the entire food chain, with amounts lost or wasted varying both by country and region, or by commodity (key food groups, bread, starchy cereals, fruits, vegetables, etc.). More importantly, whether FLW is classified further i.e., as FW or FL appears to be influenced by several factors, namely the characteristics of a population, a country’s stage of economic development (developed, developing) or measure of GDP, climate, and political stability. Low-income or developing countries for example are more likely to incur the greatest losses upstream, i.e. Food Loss (FW), at pre or post-harvest/pre-processing stages i.e. Food Loss, primarily due⁹ to issues relating to inhospitable climates, drought, water scarcity, inadequate storage and distribution infrastructures, war and civil unrest. In contrast, industrialised or developed countries, including countries in the MENA region, such as the UAE, typically experience the greatest food wastage at post-production – i.e. downstream loss, in the food retail and hospitality sector and at the household consumption stage. Almost a third (28%) of the FLW produced by developed countries, including the UAE, occurs at the consumption stage, and consists mostly as household waste, compared to just 7% consumer stage waste for developing countries.⁹ A level associated with food insecurity, a lack of available foods, and in response more frugal food management strategies. Many developed countries, particularly those located in the Northern hemisphere, have successfully implemented policies to reduce waste in the retail and hospitality sectors, and with some success in changing consumer behaviour, resulting in FWL at the consumption stage, similar to developing countries.

This is not the case however for some countries in the southern hemisphere, which have experienced increased development and affluence. Considerable progress has been made by the UAE government in addressing FW within the food retail and hospitality sectors. Upstream action directed at addressing FW, i.e. at the consumption stage of the food chain, including household food waste (HFW), is however an under-developed area.¹⁵ The UAE government has invested substantial funds to support multi-sectoral partnerships to create innovative and sustainable food waste solutions, with additional investment promised.¹⁶ One example, Winnow,¹⁷ has helped save an estimated three million meals associated with FW. Significant success with food repurposing efforts is also evident.¹⁸ Whilst this action is commendable, the savings are insignificant compared with the overall costs associated with downstream losses, i.e., FW, in the UAE, since the value-added lost to waste is the highest at the consumption stage; moreover, when approximately 90% of all food consumed in the UAE is imported the value-added losses are even more considerable and therefore critical to address. Thus, calls have been made for government and industry to partially shift some of the current emphasis on targeting hospitality and retail, to focus on the neglected component of FW arising at the household or consumer end of the food chain. Such calls have been echoed by other countries with similar FW profiles, i.e. countries that have experienced rapid economic development, such as Saudi Arabia and Qatar, who rely on importing the majority of its food. In order to effectively address FW in the UAE, a better understanding of consumers FW behaviour and the various factors influencing this is a fundamental requirement. As part of the present study, we have reviewed the FW literature, with specific reference to studies undertaken in countries similar to the UAE, in the MENA or Arab speaking region.

Regardless of social context, FW at the household level (HFW) occurs at three stages, (a) after purchase but before preparation, (b) between preparation and serving, or (c) after serving (as leftovers).^19,20 In each of these stages a variety of factors have been identified as influencing food waste and consumer’s behaviour, which in turn is directly influenced by a range of socio-demographic and environmental factors.²¹ All human behaviour including HFW is socially and culturally nuanced, so research aimed at understanding HFW must be socially situated, in order to provide insight into the sociocultural practices in that specific context. Moreover, the UAE is socially and culturally unique with an extremely diverse population; depending on the specific Emirate, 80-90% of the population are expatriates, originating primarily from India, Pakistan, Philippines, USA, Canada, and Western Europe, respectively. Some of these cultures, including Emirate households, typically cohabit with non-family members, as intergenerational families, whilst others live as smaller and homogenous units. Household composition affects how food is accessed, procured, stored, cooked, consumed, and wasted. In order to understand how the sociocultural context and demographic factors influence HFW in the Arab region, and the UAE specifically, we created a conceptual framework from HFW studies conducted in The Near East and North Africa (NENA) and MENA regions. There was a distinct paucity of studies pre-2010, before the establishment of the UN SDGs and calls²² for more research were made. In the past decade the number of published studies has increased, including several systematic reviews.^7,23 The majority of these studies are undertaken in Egypt or Saudi Arabia with only limited research undertaken by the smaller Arab countries. This particular body of literature also examines a wide range of factors relating to HFW: levels of HFW; consumer or HFW practice (i.e. behaviours relating to or resulting in HFW) and the knowledge, attitudes, intentions, motivations, relating to HFW practice; household food management practices; and strategies associated with reducing HFW. More research is needed to increase understanding about consumer behaviour as it relates to HFW, including individual’s attitudes, beliefs, and motivations towards HFW behaviour, but especially studies specific to the UAE’s heterogenous population and socio-cultural context.

It is widely accepted that socio-cultural, economic, and socio-demographic factors are directly linked with HFW. Household income, size and composition are all acknowledged as key determinants of HFW²⁴ as age, level of education, and income significantly influence intention to reduce food waste. Households in rural areas are likely to consume food prepared at home, compared with those living in urban areas who favour convenience, processed i.e. value-added foods and imported foods; being employed outside the home, or self-employed, is positively correlated with food waste; eating out and purchasing food on special promotion have been linked with regular and higher levels of FW.²⁵

Understanding consumer behaviour is also key to developing a better understanding of HFW and how to prevent or reduce this. Many attempts to understand consumer behaviour, in any area, have utilised behavioural theories as a conceptual framework, with varying success. Several Arab-based studies of HFW have used Ajzen’s (1991) Theory of Planned Behaviour (TPB). The basic theory of TPB is an explanatory or predictive model that hypothesises consumer behaviour is directly influenced by an individual’s ‘attitudes towards a particular behaviour’ (positive or negative evaluations of self-performance relating to a behaviour; the influence of ‘subjective norm’ or an individual’s response towards social pressure to conform, or significant others, to conform to particular behaviour; ‘perceived behavioural control’ or a belief in one’s ability to perform the (desired) behaviour (i.e. self-efficacy) and perceived behavioural control or the perceived ease of difficulty in behaving a specific way.²⁶ For example, an individual’s motivation or intention to reduce FW might be determined by their predisposition towards waste control (subjective norm) i.e. whether society or significant others value FW avoidance behaviour. As one study in Tehran²⁷ reported, the TPB was helpful in directing attention on factors influencing practice and they found that good food management skills (food procurement, food storage and handling, etc.,), may be a strong predictor of HFW behaviour; this however might be moderated by the trade-off between knowing what is considered socially acceptable (social norms) and confidence in their ability to adopt and carry out good management practices (self-efficacy). Whilst the TPB has been successfully used to help explain consumer behaviour in the field of FW, some researchers have found that the model only partially explains variations in participants behaviours. For example, Karim-Ghani et al., 2013, explained 13.7% of variation in consumers’ intention to adopt waste separation behaviour to avoid FW. Whilst Chalek et al. 2016, cited in Aktas et al. 2018, used TPB to help determine regional differences in FWL, estimating almost 65% of the variation between country-level food waste could be determined by gross national income.^28–30 Researchers have argued that consumer management skills (CMS) and specifically food management strategies, directly influence FW behaviour, for example, consumers who make shopping lists are less likely to over-purchase³¹ whilst consumers who regularly over-purchase and buy more food than needed, are enticed by special offers and promotions and potential economies of scale from buying in bulk, are more likely to waste food, due to poor ‘food control management’ practices (FCM); this however may be moderated by certain attitudes for example anxiety relating to the cost of FW and feeling guilty as a result of over-buying, is negatively associated with household FW loss.³² With the latter found to be important in studies conducted in the Arab region, as mentioned below. Consumers who tend to adopt less healthy dietary patterns have been shown to positively influence FW, yet higher economic status may increase the relative frequency or occurrence of household food waste linked with over-purchasing behavior and food abundance due to increased affordability of food for high-income households.³³ Since our initial review of the literature suggested a contradiction around affluence and financial attitudes, including food management behaviour, associated with cultural traditions around hospitality, we considered utilising TPB as a general framework, whilst also including focus on the wider socio-cultural literature on FW behaviour, particularly relevant in studies undertaken in the Arab region, and not restricting our study to this framework. The development of the present study was therefore influenced by the wider literature on FW practices and behaviour, values, beliefs, and attitudes across MENA.

Diaz-Ruiz et al. (2018), identify six predictors of HFW: environmental awareness; materialism (consumerism); purchasing behaviour; dietary choice; attitudes towards waste recycling; and waste prevention.³⁴ Abdelradi (2018) adopts this framework, extending this to include knowledge of food waste as a problem, personality traits and religion; this aligns with Elshaer et al. (2021) who posited that religion is a key predictor of adverse HFW in Saudi and therefore the wider Muslim region.³⁵ One explanation for religiosity as a predictor of adverse FW behaviour is the importance of frugality, morality and altruism in religious codes,³⁶ which reportedly contradicts with respondents aspiration to be perceived as a ‘good provider’, as a precursor to increased food waste. The cultural importance of Arab traditions rooted in food, hospitality and expressions of generosity may also accentuate the influence of religiosity as a determinant of HFW in the UAE and especially in countries where the population is predominantly Muslim; whereby people who have a positive and moral attitude towards the environment and the importance of food waste have a greater intention to reduce food waste, yet this contradicts with pressures associated with cultural traditions (hospitality). This aligns with Hedari et al. (2020), who suggested that this influence is developed through the effect on subjective norm and attitude, and waste-avoidance or prevention behavior may be a stronger predictor of intention to reduce household food waste than previously thought; therefore, interventions to reduce consumer food waste should concentrate on establishing waste avoidance attitudes.³⁷ This concurs with a study conducted in the U.S.,³⁸ whereby consumers who are aware of the consequences of food waste are also more receptive towards food waste reduction. Suggesting that considering the consequences of food waste, i.e. economic cost, needless hunger, and climate change, may encourage consumers to act sustainably and avoid waste. A sense of community has also been found to encourage food waste reduction, repurposing and recycling behavior. However, a study of youth in Pakistan suggests people who perceive they are excessively busy are significantly less likely to engage in the same food waste avoidance behaviors.³⁸ Although knowledge, attitudes and practice influence the participants’ decisions in determining what types of food are appropriate for them and their families to consume and which foods should be discarded. Despite this understanding of food labels with respect to food shelf-life is limited amongst consumers, thus, more food than necessary is discarded.³⁹ Having a strong sense of social responsibility, including particular religious values or environmental concerns, is also reported with conserving food and reducing food waste behaviours.⁴⁰ Food-related practices (procurement, cooking, storage and consumption) of a household vary according to cultural values, beliefs and socio-economic status (SES) i.e., educational attainment, occupational and income status.⁴¹ Research also suggests that food waste may be higher in households where the woman is employed outside of the home⁴² and whilst women may traditionally stay at home, in the UAE, where the majority are expatriates, a high proportion of women are in employment.

The rapid societal and economic transition experienced by oil-producing Arab states, from traditional nomadic to wealthy industrialized societies, may compound consumer food waste behavior in countries such as the UAE. Baig et al. (2019), reported how HFW in Saudi Arabia, the leading producer of HFW regionally, is positively associated with economic affluence. In addition to cultural traditions (i.e. the need to express hospitality), widespread lack of awareness about issues relating to FW and the absence of policies targeting consumers to reduce FW, is believed to compound food waste behavior.⁴³ As a study undertaken in KSA⁴⁴ explains, how surplus food is classified as ‘unclean’ or ‘bayt’ by its people is an important determinant and predictor of HFW. This is especially relevant for families who eat from individual rather than traditional practice of communal plates. The sensory properties of food are key when deciding whether to use ‘bayt’ food (cooked food that is kept for at least one night). Leftovers are sometimes deemed unwanted because of the desire to avoid eating the same food on consecutive days. Although ‘bayt’ food is perceived as less desirable, it is considered edible and so it is haram (forbidden) to waste it. This raises a dilemma, in choosing enjoyment by serving fresh food vs. performing duties as Muslims by reusing edible leftover, or ‘bayt’, food. To escape the sin of waste, some interviewees reported passing unwanted food to their house staff or forcing themselves to eat it. During Ramadan meat-less surplus food becomes bayt and is unwanted because it is a ‘light’ dish. The authors also explain that rapid economic development, resulting in an affluent population has directly encouraged surplus generation and the attractiveness of leftovers. Where a preference for abundance and the multiplicity of choices, but also having the option not to consume leftovers, are related to this rapid rise in affluence. Social norms, as our data suggests, are more important than economic status in the rejection of surplus food. In Saudi, eating flavorless food was once common, today, however, it is no longer the norm to eat and serve such food – even amongst less affluent interviewees. Reliance on restaurants while eating outside the home during Ramadan also results in greater amounts of home-cooked foods being rejected. Whereas the fear and embarrassment of being criticized or bullied can deter some people from repurposing household food and taking leftovers to consume outside of the family home.⁴⁴

As the current literature suggests many factors are involved in determining consumer food waste and encouraging food waste avoidance behaviour. These factors are also socially and culturally situated and therefore require context specific research. Efforts to reduce household food waste therefore require research that helps illuminate the determinants of FW behaviours, including knowledge and attitudes towards household FW, amongst consumers living in the UAE. The present study will address this by contributing to the understanding of consumer knowledge, attitudes and behaviours relating to FW in the region. Based on our review of the current FW literature in this region, we can assume that FW is positively associated with certain socio-demographic characteristics, such as age, social status and household income, including anxiety relating to food insecurity by low-income households; household composition, ethnicity; religious beliefs; cultural values; knowledge and attitudes towards food waste and food waste avoidance. FW behaviour is also likely to be directly influenced by consideration of social norms related to FW; consumers who adopt highly processed, less traditional middle eastern or Mediterranean dietary patterns are likely to be positively influenced to FW compared with those who adopt traditional dietary patterns and traditions, including acceptance for repurposing foods i.e. ‘bayt’; as is with lack of knowledge and awareness of food and nutrition labelling or the nutrition content of foods; and poor household food-management skills. As stated, the purpose of the present study is to provide much needed baseline data collected in the UAE on consumers attitudes, opinions, knowledge and practice (behaviours), relating to household food waste and is guided by the above conceptual framework.

2.1 Study design and sample

A cross-sectional study was undertaken to collect data on household food waste knowledge, attitudes and practices (KAP) from people within the UAE. For inclusion, participants needed to be UAE residents aged 18 or above. The minimum sample size was calculated to be 637 with a population of 9.89 million, with a margin of error of 5%, 95% confidence interval (95% CI) and response rate of 40%.⁴⁵

2.2 Sampling and recruitment strategy

To obtain a representative sample of the population in the United Arab Emirates, a stratified sampling technique was employed. The population of the emirates was divided into three geographical areas: Abu Dhabi, Dubai, and the Northern Emirates. Each geographical area was considered a stratum. Participants were randomly selected from each geographical area in proportion to their population. To increase the representativeness of the sample with respect to households in the UAE, survey responses were collected from the UAE resident population in the three main zones: Abu Dhabi, Dubai and Northern Emirates. The survey was created using google forms. The survey was distributed via email and social media platforms to all participants. The survey was accessible via an anonymous link between 14^th December 2021 and 8^th February 2022. At the end of the study period, the survey was automatically disabled. Collaborators from each emirate collected the data to ensure that it was representative of the population. A question in the survey response inquired about the participants’ current residence.

2.3 Ethical approval

The Social Sciences Research Ethics Committee (REC) from the United Arab Emirates University (UAE-U) approved the study (reference number is ERS_2021_8380). The study was conducted in accordance with Helsinki Declaration guidelines.⁴⁶ Informed consent was obtained from all participants involved in the study prior to commencing the online survey. They were provided with clear and detailed information about the purpose of the study, the procedures involved, and how their data would be used and published. Participants were explicitly informed that their identities would remain confidential and would not be disclosed in any publications or presentations resulting from the study. The focus would solely be on the aggregated data and group-level findings, ensuring anonymity and privacy for each individual participant.

2.4 Data collection instrument

The design of the semi structured survey was informed by relevant existing validated questionnaires from the literature.^7,35,47–51 Four experts from the field were requested to rate the questions on a scale of 1 to 10 for clarity, validity (ability to meet study objectives) and reliability. Any question which scored below seven on any aspect was removed. Feedback from the experts was addressed to strengthen the survey design. Reliability was tested using Cronbach Alpha. A reliable score of 0.80 was observed. The questionnaire was bilingual, English and Arabic; with forward and backward translation from English to Arabic undertaken by bi-linguists as per the methodology.⁵² The final survey, with Arabic translations, was piloted on 30 candidates for comprehensiveness and clarity. No major issues were received. The structure of the survey included a clear statement of the details of the study, necessary for informed consent, which was obtained from participants prior to starting the questionnaire. Participants were advised that they could exit the study by quitting the survey at any time, without consequence.

The questionnaire collected data on sociodemographic characteristics, and food waste knowledge, attitudes, and practices (KAP). For KAP-related questions, participants were asked to score their responses based on a five-point ordinal Likert scale which was later computed (mean) to estimate knowledge and awareness, attitude, and practice scores (Supplementary Table 1).

2.4.1 Knowledge and awareness

The knowledge subdomain had two items while the awareness subdomain had a single item (Supplementary Table 1). Participants knowledge, and participants knowledge and awareness, were assessed by calculating the mean score of the two knowledge items and the three items of the knowledge and awareness subdomains, respectively. Higher scores reflected better knowledge and awareness regarding food waste (i.e. desirable behavior).

2.4.2 Attitude

Validated questions related to personal, emotional, behavioral, and financial subdomains of the attitude towards food waste were used and contained five, three, two, and four items, respectively (Supplementary Table 1).^35,47,48,50 Participants ‘personal, emotional, behavioral, and financial attitudes were assessed by calculating the mean score of the items included in their respective subdomains. The overall attitude was assessed by calculating the mean score of all fourteen items included in all four attitude subdomains. Higher scores reflected better attitude regarding food waste (i.e., desirable).

2.4.3 Practice

The practice domain had fifteen items (Supplementary Table 1), and their mean score was used to assess participants’ practice of food waste. Higher scores reflected worse practice of food waste (i.e., undesirable).

2.5 Data analysis

Data was analyzed using the statistical software Statistical Package for the Social Sciences (SPSS) (version 28).⁵³ The mean and standard deviation (SD) were used to summarize continuous data. The mean percentage was calculated as a percentage of the maximum possible score. Categorical data were presented using frequencies and percentages. Spearman rank correlation coefficients were utilised to examine correlations between practice and each of knowledge and awareness, total attitude, and their subdomains. Furthermore, multiple regression model was used to examine the association of all independent variables (knowledge, awareness, personal attitude, behavioral attitude, emotional attitude, and financial attitude) with practice. In a separate analysis, multiple regression model was used to examine the association of sociodemographic characteristics (gender, age, number of adults in household, marital status, area of residence, number of children, educational level, and household income) with practice. The 95% confidence intervals (95% CIs) were used to determine the strength and direction of the associations. Type I error was fixed at 5% for all tests.

3.1 Sociodemographic characteristics and household composition

A total of 1052 participants were included in the study. A total of 70.9% of participants were from Abu Dhabi, 17% from Dubai, and 12% from Northern Emirates (Sharjah, Ajman, Umm Al Quwain, Ras Al Khaimah and Fujairah). The distribution of the survey responses matches the distribution of the population and the area of the UAE (Supplementary Figure 1). The majority of participants were female (76%) aged between 18 and 30 years (58.2%) (Supplementary Table 2). The lowest response rates 1.3%, 6.6%, and 6.6% were received from ages 61-70, 51-60 years, and 41-50 years respectively. The most common household compositions reported by participants were homes of seven or more adults (25%) and two adult households (19.8%). Additionally, six (12.1%), five (13.6%), four (15.5%), three (10.1%) and one adult (4%) household occupancy were reported. Over half (54.8%) of the participants were single compared with 42.6% married. Whilst only 0.3% and 2.4%, were reported to be widowed or divorced, respectively. The majority of participants lived in the city (87.4%), while 7.3%, and 5.3% of participants living in towns and villages, respectively (Supplementary Table 2).

Nearly a third (27.5%) of participants lived in households without children, while 56.1% lived in households with 1, 2 or 3 children (19.9%, 18.7%, and 17.5%, respectively). A total of 17% of the participants lived in households with more than three children (four, five, or six children, 9.5%, 3.7%, and 3.1%, respectively). The majority of participants were reportedly educated to degree level, with most having undergraduate degrees (65.8%), whilst almost a quarter of the sample (23.3%) reported having postgraduate degrees and 8.7% had a PhD or higher. A minority (0.6%) reported having attained less than secondary school education, 1.7% less than high school education. Over half of participants (55.7%) reported basic middle income, 13.4% marginal middle income, 24.6% upper middle income, 2.4% lower income, and 4.2% upper income (Supplementary Table 2). The full anonymised dataset can be found under Underlying data.⁶⁶

3.2 Knowledge and awareness, attitude, and practice

In Supplementary Table 3, we present the descriptive data and analysis of estimated self-reported food waste-related Knowledge and Awareness, Attitude, their respective subdomains, and Practice.

Among those surveyed, mean (95% CI) knowledge, awareness and knowledge and awareness combined were 83.00 (81.91- 84.10), 63.29 (61.92-64.65), and 76.43 (75.56-77.30), respectively. Mean (95% CI) personal, behavioural and emotional attitude “toward avoiding” food waste were all above 80% [84.32 (83.55-85.09), 87.60 (86.81- 88.40), and 83.95 (83.01-84.90), respectively], with mean (95% CI) financial attitude of 68.89 (68.26-69.53). Mean (95% CI) overall attitude was 80.56 (80.00-81.13). Practices of food waste had a mean (95% CI) of 19.92 (18.77-21.07).

3.3 Spearman’s Rank Correlations

In Table 1, we show the correlations between practice as a response variable and knowledge, awareness, knowledge and awareness, personal attitude, behavioural attitude, emotional attitude, financial attitude and overall attitude, as independent variables. The correlation coefficients indicate the strength and direction of the relationship between two variables.

The undesirable food waste practice was negatively and significantly correlated with (Spearman’s rho correlation coefficient, p-value) knowledge (-0.155, <0.001), knowledge and awareness (-0.123, <0.001), personal attitude (-0.303, <0.001), behavioural attitude (-0.237, <0.001), financial attitude (-0.165, <0.001) and overall attitude (-0.276, <0.001). The significance of all the correlations continued even after controlling for the effect of age, as shown in Table 1. This indicates that age did not have a significant impact on the observed associations between the variables. These results indicate that an increase in each of these variables (knowledge, knowledge and awareness, personal attitude, behavioural attitude, financial attitude and overall attitude) is correlated with a reduction in food waste practice. The correlation between practice and emotional attitude was not significant (-.051, 0.051), however; it reached significance after adjusting for age (-0.075, 0.015). Overall, the results suggests that better attitude, knowledge and awareness are correlated with reduction in food waste practice.

3.4 Correlation between Practice and Knowledge, Awareness, Attitudes

3.4.1 Multiple Regression Models

We ran a multiple regression model to further clarify the associations of all independent variables (knowledge, awareness, personal attitude, behavioural attitude, emotional attitude, and financial attitude) with practice as the dependent variable (Table 2). Analysis of variance (ANOVA) for the multiple regression model indicated that at least one of the independent variables (the six subdomains) is significantly associated with the predictor variable (practice domain) (F=21.563, p-value <0.001; Supplementary Table 4).

Results of the multiple regression models showed that (regression coefficient, p-value) knowledge (-0.077, 0.003), personal attitude (-0.282, <0.001), and financial attitude (-0.150, 0.001) were significantly and negatively associated with food waste practice, indicating that better knowledge, personal attitude, or financial attitude was associated with reduction in the undesirable food waste practice. While awareness (0.046, 0.021) and emotional attitude (0.074, 0.029) had positive significant association with food waste practice.

We ran a seperate multiple regression model to assess the associations of sociodemographic variables (gender, age, number of adults in household, marital atatus, area of residence, number of children, educational level, household income) with practice as the dependent variable (Table 3). Analysis of variance (ANOVA) for the multiple regression model indicated that at least one of the independent variables (sociodemographic variables) is significantly associated with the predictor variable (practice domain) (F=5.983, p-value <0.001; Supplementary Table 5).

Results from the second multiple regression model (regression coefficient, p-value) indicated that being 18-30 years old is significantly and positively associated with higher food waste practice compared to those 51 years old and above (0.211, 0.046). In addition, having 1-3 (-0.247, 0.001) or 4-6 (-0.204, 0.001) adults in the household was associated with less food waste practice compared to households with 7 or more adults. No other associations reached statistical significance.