Studying geographical distributions of household income data is extremely valuable when researching poverty and its associated harms. However, such distributions are rarely available to researchers due to access and privacy challenges. The generation of synthetic income data is a promising approach to overcome real-world income data collection issues. We draw on several existing income data resources to examine likely income parameters for smaller geographical areas (namely, the UK’s ‘Lower Super Output Areas’: LSOAs), and develop a direct sampling approach to generate synthetic household income distributions in the case city of Nottingham, UK. The resulting LSOA-level household income distributions (Synthetic Nottingham Homes), can aid various poverty studies. The synthetic data are then pseudo-validated to evaluate their proximity to real-world conditions. Subsequently, to demonstrate the utility of such synthetic data, we present an applied example of the Synthetic Nottingham Homes in the context of fuel (energy) poverty estimations.

Loneliness and social isolation are a cause and symptom of poor psychological health, while they also have secondary impacts on physical health, employability and educational outcomes. For a multitude of reasons, these phenomena are challenging to evaluate and understand in tandem; this can be attributed to the latent nature of loneliness and the wider sensitivity and stigma surrounding mental health. Comparatively less is known about the relationship between stigmatised deprivation (i.e., various forms of deprivation that can lead to social exclusion) and loneliness. To address this gap, the current study examines the impact of stigmatised deprivation on the prevalence of loneliness across London, UK. Specifically, we consider two increasingly prevalent forms of stigmatised deprivation: food insecurity and energy insecurity. We develop a hierarchical machine learning model to evaluate the relationship between a wide range of health and socioeconomic features and loneliness among a large sample of London residents (n=2886). Model results indicate that including stigmatised forms of deprivation increases loneliness prediction accuracy. In practice, this corresponds to potential improvements in understanding the drivers of loneliness, as well as how the condition might be remedied. Finally, we suggest policy recommendations in terms of how reducing stigmatised deprivation could also lead to broader mental health benefits.

Introduction & BackgroundThe current approach to measuring fuel poverty in England, Low Income Low Energy Efficiency (LILEE), systematically underestimates the condition in low- and middle-income homes. This is due to its inability to accurately respond to macroeconomic shocks (e.g., energy price inflation) and its overstatement of energy inefficiency as the principal driver of fuel poverty. This inaccuracy poses a critical policy challenge: inaccurate measurement precludes meaningful alleviation. In response, the current study develops an area-level fuel poverty indicator that better captures the realities of fuel poverty by harnessing digital footprints: smart meter data and financial records, acquired from the Smart Energy Data Service (SENSE) and the Financial Data Service (FINDS), respectively. Objectives & ApproachThe aim is to develop a granular, area-level fuel poverty indicator (based on Lower Super Output Areas—LSOAs) through the linkage of smart meter and financial data. This approach facilitates the examination of the spatial and temporal overlap between low absolute energy consumption (SENSE) and low energy expenditure (FINDS). The proposed approach aims to complement or supplant the government’s official fuel poverty statistics, which are typically modelled using relatively small, static survey samples, by providing dynamic and responsive insights into energy affordability at regular intervals (e.g., monthly or seasonal). Relevance to Digital FootprintsEnergy and financial data represent rich digital footprints of sensitive social factors, which are key to a better understanding of fuel poverty in the UK. The SENSE and FINDS data are disparate digital footprints encoding complex consumer behaviours; yet their linkage could possess significant utility as an alternative resource for generating area-level fuel poverty insights. Conclusions & ImplicationsThe SENSE and FINDS collaboration is ongoing; we present a methodological framework for linking two independent datasets and new analytical approaches enabled by the linkage of energy and financial data. The expected outcomes are more accurate and responsive area-level fuel poverty indicators. This is particularly pertinent to England, where competing fuel poverty statistics are needed to overcome the weaknesses of the LILEE approach and to enable targeted measures for a fair energy transition. Further, the proposed indicator could harmonise diverging measurement standards across England and the devolved nations (Scotland, Wales and Northern Ireland), facilitating essential comparative statistics.

Introduction & BackgroundPerfume purchasing is a highly multisensory process, relying on olfactory, tactile and visual cues. Although online retail already restricts direct olfactory engagement, the COVID-19 pandemic heightened these constraints by temporarily removing in-store sampling and therefore limiting physical interaction with products. Despite the growing interest in sensory marketing, little is known about how consumers' real-world purchasing behaviour changes when key sensory inputs are restricted. Objectives & ApproachThis research examines how consumers’ perfume purchasing behaviour shifts in limited multisensory retail environments by adopting a sensory deprivation perspective. Specifically, it investigates whether there are changes in purchasing behaviour across pre-COVID, during-COVID and post-COVID periods, including potential shifts in reliance on brand, price, product descriptions and visual cues. Relevance to Digital FootprintsThis research contributes to digital footprint literature by showing how purchasing behaviour may change under sensory restrictions. By analysing behavioural patterns captured in large-scale transactional data alongside product-level sensory cues, the study explores whether consumers adapt their purchase decisions when direct sensory evaluation is restricted. Overall, the study highlights the value of digital footprint data in examining purchasing behaviour in sensory-restricted retail environments. Conclusions & ImplicationsConceptually, this study advances sensory marketing research by shifting the focus from sensory optimisation to sensory constraint, positioning sensory deprivation as a meaningful condition that shapes consumer behavioural patterns. Empirically, it offers preliminary insights into how consumers’ perfume purchasing behaviour may shift when multisensory access is disrupted. The research demonstrates how natural disruptions can be used to deepen understanding of consumer psychology in digital and physical retail marketplaces.

Hyperlocal sharing platforms are gaining traction as a sustainable alternative for distributing resources, particularly in the case of food sharing, where success relies heavily on community engagement and local interactions. Despite their growing popularity, there is little understanding of what contributes to the growth of sharing platforms and their evolution over time. In collaboration with the UK's largest food sharing platform, this study explores over five million anonymised sharing instances across 312 districts in England over a 51-month period. Building on prior literature, we empirically examine platform growth using temporal network structures, user behavioural variation, and local demographic and environmental characteristics, to model user acquisition and retention. Using a machine learning approach, we utilise SHAP variable importance to determine the most impactful attributes for platform growth and identified key predictors of the proliferation of food sharing platforms. This includes the distribution of super-users, the presence of active volunteers, and the formation of structured communities. Our findings demonstrate how simultaneously combining data from users, networks and geographical dimensions provides a more useful explanation of growth than any isolated disciplinary theory. The results contribute to the theorisation of growth in sharing platforms, offering managerial insights that support the development and sustainability of food sharing networks within the sharing economy.

The definition and quantification of fuel poverty remain contentious, particularly within the English policy landscape. The current official definition—Low Income Low Energy Efficiency (LILEE)—systematically underestimates the condition in low- and middle-income households, effectively obfuscating changing energy affordability norms. This paper presents a sensitivity analysis of fuel poverty trajectories in Nottingham, UK, across competing definitions and future scenarios. We analyse the temporal composition of fuel poverty in diverging 'optimistic' (domestic efficiency upgrades) and 'pessimistic' (energy price inflation) scenarios. Results demonstrate that LILEE overstates the efficacy of efficiency upgrades while failing to adequately reflect the impact of price shocks. To address these deficiencies, we propose the Temporal Equity Framework (TEF), a budget standard-based approach paired with a proportional energy expenditure indicator that captures shifting affordability norms more equitably. The TEF also introduces continuous monetary depth indicators to evaluate the entrenchment of fuel-poor households (the 'fuel poverty gap') and the resilience of non-fuel-poor households (the 'fuel poverty buffer'). We argue that responsive definitions which also recognise fuel poverty as a continuum of vulnerability, rather than a static, dichotomous state, are paramount to ensuring the equitability of energy transitions. Further, we suggest that the early identification of fuel poverty and spatial targeting of remedial resources could be markedly improved by adopting the TEF. We conclude that the adoption of a multidimensional budget standard-based definition, as proposed by the TEF, is essential to facilitate a more accurate and just evaluation of energy transitions in England.

Alcohol-specific mortality has risen in England, yet behavioural drivers remain hard to measure due to stigma and survey bias. We investigate spatial inequalities in alcohol mortality across English Lower-tier Local Authorities by integrating novel supermarket transaction data with established socioeconomic predictors. Using machine learning with spatially stratified cross-validation, we test whether shopping behaviours improve prediction beyond traditional demographics. Starting from a combined 28-feature set (13 demographic, 15 shopping), a refined 15-feature model achieved R^2=0.374 , outperforming both a demographic-only model and the full combined set. Key shopping features–spirits purchasing intensity (quadratic), store-level alcohol concentration, and spatial consumption spillovers–ranked among the strongest predictors. A Bayesian bootstrap with a ± 0.01 R^2 ROPE indicated practical superiority over the full model. Retail behavioural signals capture consumption patterns that demographics miss, offering policymakers timelier tools to identify at-risk communities and target interventions.

Introduction & Background Traditional poverty classifications rely on relatively small samples; for example, the main resource used to model fuel poverty (i.e., an inability to afford sufficient energy services) in England is the English Housing Survey (~30,000 homes). The superior coverage of financial datasets, e.g., aggregate banking data, may allow more precise estimations. Further, the current approach to measuring fuel poverty in England, Low Income Low Energy Efficiency (LILEE), has been shown to significantly underestimate the true rate of fuel poverty. To reconcile these matters, the current study explores the merits of financial data for fuel poverty classifications and also provides evidence for a set of fuel poverty measurement criteria that remedy the shortcomings of LILEE while also being compatible with adjacent modern agendas. Objectives & Approach This study analyses financial data, specifically an income volatility dataset provided by Smart Data Foundry that was derived from NatWest accounts (drawn from an original sample of ~5 million customers), to explore alternative fuel poverty measurement approaches in England. The analysis involves the estimation of fuel poverty according to a range of existing definitions, e.g., LILEE, 10% and Low Income High Costs, as well as emerging methods, e.g., the Minimum Income Standard (MIS)-based approach, which is able to account for living standards and was recently adopted as the official fuel poverty definition in Scotland. Following this, the merits of using financial data for poverty classifications are discussed, and the modern appropriateness of competing fuel poverty definitions is evaluated. Relevance to Digital Footprints The relevance to digital footprints is inherent to the study’s purpose: that is, to improve the identification of fuel poor homes through the utilisation of financial sector digital footprints. Results Initial results provide further evidence that the English (LILEE) approach to measuring fuel poverty underestimates fuel poverty in many low-income homes. We suggest that using financial data for fuel poverty classifications can provide more granular and accurate estimates. In addition, we suggest that MIS-based approaches to measuring fuel poverty are more accurate and better suited to modern needs. Conclusions & Implications The results can be used to inform a more effective approach to measuring fuel poverty in England: first, in terms of the data used for fuel poverty modelling; and second, concerning the most appropriate fuel poverty definition.

Introduction & Background As digital footprints data continue to grow in complexity and volume, understanding and summarising large, high-dimensional time series is becoming increasingly important for analysing behavioural patterns. In various domains, mass data collection is routine and crucial to detecting breakpoints (significant statistical changes) and underlying patterns. Examples of such data include: transactional records; internet history records; social media data. And yet, despite this explosion of data, it remains challenging to uncover commonalities across real-world time series that are both high-dimensional and noisy. Objectives & Approach We introduce a new method for deciphering complex digital footprints data named ALI (Automatic Lifestate Identification), a parameter-free algorithm which aims to create interpretable summaries of the data. ALI efficiently identifies breakpoints, and clusters resulting segments, referred to as lifestates, across multiple time series. ALI uses an Expectation Maximisation (EM) algorithm that iteratively searches for breakpoints and lifestates, meaning that each is informed by the other and improved on each iteration. The result is an efficient and parsimonious characterisation of the time series; that is, a collection of lifestates describing the dataset and its patterns. Relevance to Digital Footprints ALI has the potential to improve how digital footprints are analysed across many different domains. Results This work demonstrates the practical utility of the ALI algorithm by applying it to two real-world datasets: a transactional dataset from a UK-based pharmaceutical company, and a dataset from a Sri Lankan telecommunications company. The two datasets represent the purchase history for over 18,000 new mothers, and the mobile application usage of a ‘super-app’ which has over 1 million downloads. This work demonstrates how digital footprints data can be used to understand shared experiences across multiple domains. Understanding cross-individual patterns can be useful for providing tailored services to people, particularly for understanding when individuals transition from one stage of their life to another. Conclusions & Implications The experiments show that ALI is successful in identifying transitions into parenthood more accurately and reliably when compared to previous methods. Furthermore, ALI finds shared states across the whole set of mothers in the dataset, allowing for comparisons between the retail behaviours of different mothers before, during, and after pregnancy. The results of this work show ALI's capability to extract meaningful lifestates in challenging, real-world scenarios.

Introduction & Background The study of loneliness and social networks has largely involved social media analysis for both social circle insights and loneliness predictions from language. While geospatial data have also been used to investigate loneliness, the approach has predominantly involved a traditional investigation of geolocations per se. Few studies have focused on the factors contributing to the success of human relationships and the associated loneliness reduction – which was the objective of this project. Objectives & Approach The data were obtained from B:friend – a wellbeing charity dedicated to tackling loneliness through befriending schemes for Older Adults – and joined with the Index of Multiple Deprivation (2019) for feature engineering. The objective was to identify features which contributed to a successful friendship. The dataset included neodemographic information: geospatial and demographic data for both sides of each pair (the Older Adult (Older Neighbour) and the paired Volunteer); as well as preference features for a potential pairing. Machine learning classification models (logistic regression, decision trees, random forest, support vector machines and xgboost) were used to determine which features contributed to successful short- and long-term befriending schemes. A total of 805 pairings were analysed. Relevance to Digital Footprints The project used geospatial information paired with demographic records collected by B:friend digitally to identify the features which contributed to a successful friendship. Results The preliminary results offer insight into the features which contribute to short-and long-term friendships, with variables such as house suitability, third COVID-19 lockdown and income decile difference driving the predictions. On average, a third of the friendships appear to succeed. Additionally, we use a series of variable importance measures to rank order variables which contribute to successful befriending of the pairings. Conclusions & Implications Combining demographic and geospatial information for machine learning helps identify important factors which contribute to short- and long-term relationships, and, by extension, to tackling loneliness. This novel approach can benefit the public, as well as professionals, by offering tailored suggestions for loneliness interventions, utilising factors contributing to the specific requirements of various individuals, as well as their preferences, thus further expanding on the pre-established findings such as homophily.

Increasing food insecurity (FIS) in the UK presents a major challenge to public health. Universal Credit (UC) claimants are disproportionately impacted by FIS but research on socio-demographic factors and consequent nutritional security is limited. A cross-sectional online survey (September 2021 - April 2022) assessed FIS in UC claimants (males and females, n = 328) (USDA 10 question module), dietary intake (females, n = 43; 3–4 × 24-hour dietary recalls) and coping strategies. Binary logistic regression tested sociodemographic variables influencing the odds of food insecurity. Diets ofUC were compared with national diet and nutrition survey (NDNS) participants and thematic analysis conducted for drivers and impacts of FIS. FIS was experienced by 84.8

The way that fuel poverty (or energy poverty) is defined and quantified continues to divide opinion. In England, modern fuel poverty definitions, such as Low Income High Costs (LIHC) and Low Income Low Energy Efficiency (LILEE), have increasingly emphasised energy-efficiency-induced fuel poverty, to the detriment of income-poverty- and energy-costs-induced fuel poverty. Moreover, LIHC and LILEE assume that fuel poverty is a relative condition, such that fuel poverty rates are necessarily capped below the median household; in effect, this also implies that fuel poverty is ineradicable. The LIHC and LILEE methodology's assumption of relativity is potentially invalid and highly likely to induce the underestimation of fuel poverty in low- and middle-income homes, particularly amid periods of economic recession and high consumer energy costs. In this article, we scrutinise the evolution of fuel poverty politics in England to better understand its modern problematisation as an issue that is primarily caused by energy inefficiency. Subsequently, existing fuel poverty definitions are critically examined via the application of a novel Fuel Poverty Definition Evaluation Framework (FPDEF). We conclude that the majority of existing fuel poverty definitions have limited modern applicability, especially LIHC and LILEE, which explicitly obfuscate fuel poverty in some low-income, high-efficiency homes. Policy recommendations are provided for a fairer fuel poverty definition that satisfies modern needs. Specifically, we argue that budget standard approaches to measuring fuel poverty can better account for the potentially adverse impacts of energy transitions on fuel poverty and living standards.

Introduction & Background The perfume industry faces significant ethical scrutiny ranging from environmental degradation and labour concerns to the carbon costs of global logistics. Despite these concerns, consumer purchase decisions are more often guided by sensory cues such as scent, branding and packaging design. Such attributes are routinely prioritised in marketing strategies, yet empirical research quantifying their effect on consumer buying behaviour remains scarce. Furthermore, little is known about how these sensory elements interact with buyer characteristics such as location, age or personality types. Objectives & Approach This study addresses this knowledge gap by combining two rich sources of digital footprint data: (1) transactional sales records from the UK’s largest fragrance retailer, including loyalty card linked purchases across physical and online stores and (2) user generated reviews from Fragrantica, a leading global perfume database. The two datasets were linked by brand and product name. This enabled the merge of behavioural purchase data with perceptual attributes such as longevity, sillage, gender perceptions and perceived price value. This, therefore, allowed the possibility to analyse the relationship between sensory attributes and consumer purchase behaviour at national scales. Relevance to Digital Footprints This research contributes to the growing digital footprint literature in two ways. First, it leverages large-scale real-world behavioural data to provide a national lens on perfume consumption. Second, it demonstrates a novel integration of perceptual review data with sales data offering insights into how sensory perceptions drive consumer behaviour and enabling interdisciplinary approaches to behavioural research. Results Data drawn from the Fragrantica database includes 104,894 perfumes and 2,112,790 perfume reviews generated by 1,528,705 users. This was linked to sales dates for >1.7k over 4 years. Results emphasise how important sensory elements such as scent, sillage, and packaging are in influencing consumers' decisions to buy perfume and point to socio-demographic trends that may reflect purchase motivations. The study also highlights the value of combining transactional and sensory data to investigate consumption patterns in the field, opening the door for further research that may help guide consumer-focused and sustainable perfume creation approaches. Conclusions & Implications This study highlights the value of combining transactional and perceptual data to gain a deeper understanding of consumer psychology in perfume consumption. While not traditionally framed within the ‘public good’ field the research contributes in several meaningful ways. First, by identifying how individuals respond to sensory marketing, it provides opportunities for guiding consumers towards more sustainable or ethically produced perfumes without compromising on sensory appeal. Second, given the emotional and identity-based nature of scent, the findings may inform future research into how perfume consumption supports individual wellbeing and self-expression. Overall, the study encourages more ethical, data-informed innovation in an industry increasingly under pressure to balance beauty with responsibility.

Reading and evaluating product reviews is central to how most people decide what to buy and consume online. However, the recent emergence of Large Language Models and Generative Artificial Intelligence now means writing fraudulent or fake reviews is potentially easier than ever. Through three studies we demonstrate that (1) humans are no longer able to distinguish between real and fake product reviews generated by machines, averaging only 50.8 overall - essentially the same that would be expected by chance alone; (2) that LLMs are likewise unable to distinguish between fake and real reviews and perform equivalently bad or even worse than humans; and (3) that humans and LLMs pursue different strategies for evaluating authenticity which lead to equivalently bad accuracy, but different precision, recall and F1 scores - indicating they perform worse at different aspects of judgment. The results reveal that review systems everywhere are now susceptible to mechanised fraud if they do not depend on trustworthy purchase verification to guarantee the authenticity of reviewers. Furthermore, the results provide insight into the consumer psychology of how humans judge authenticity, demonstrating there is an inherent 'scepticism bias' towards positive reviews and a special vulnerability to misjudge the authenticity of fake negative reviews. Additionally, results provide a first insight into the 'machine psychology' of judging fake reviews, revealing that the strategies LLMs take to evaluate authenticity radically differ from humans, in ways that are equally wrong in terms of accuracy, but different in their misjudgments.

Food systems are a major contributor to environmental impacts such as greenhouse gas emissions and biodiversity loss, with widespread dietary changes required to avoid surpassing safe planetary boundaries by 2050. To promote dietary shifts among the public it is crucial to understand how people perceive the environmental impact of food products. However, prior investigations into this topic have covered only a narrow range of product types and elicited perceptions using researcher-imposed, single-item environmental friendliness scales-which might not align with consumers' underlying mental representations. We conducted a card sorting study in which UK participants (n = 168) organised a diverse range of supermarket food products into environmental impact categories that they created and labelled themselves. Participants subsequently viewed product-level scientific impact estimates and reported whether they were surprised by how high or low each impact was. Multidimensional scaling of card sorting data indicated a two-dimensional solution, with participants primarily distinguishing products by animal vs. plant-origin and level of processing. Category labels assigned during the sorting task and participants' self-reported surprise at scientific impact estimates suggested they tended to overestimate the impact of highly processed foods and underestimate the impact of waterintensive products (e.g., nuts). Furthermore, mixed-effects regression analyses indicated that surprise at how high the scientifically estimated impact of a given product was predicted intentions to consume less of that product in the future. Results provide novel insight into consumers' mental representations of food sustainability, with implications for the development of information-provision strategies such as eco-labelling and public awareness campaigns.

Introduction & Background Parliaments play a significant role in democratic decision-making, drawing on evidence from a wide range of stakeholders. This evidence data, published on official government websites, is a valuable archive of digital traces reflecting the policymaking process. Among the key data sources is Hansard, the UK Parliament’s official verbatim record which captures the full transcripts of parliamentary debates and discussions, revealing diverse perspectives, stances and interactions shaped by participants’ political and social contexts. While datasets like Hansard offer rich insights into the policymaking process, their scale and varied dialogic structure present challenges for traditional analysis methods. Recent studies indicate large language models (LLMs) can exhibit intelligent and collaborative decision-making behaviours in social simulations. This makes them a useful tool for simulating complex discussions and analysing group dynamics. Objectives & Approach This study explores the potential of LLMs to simulate UK parliamentary debates, with a focus on speaker roles and stance-taking. The immediate objective is to assess whether LLMs can mimic the structure and dynamics of real debates. Longer-term, this work aims to lay the foundation for a more comprehensive deliberation sandbox which provides an experimental environment that supports multi-stakeholder negotiation and collaborative problem-solving in a low-risk setting. Such a tool could enhance transparency and support more inclusive, evidence-informed policymaking. We evaluate model performance by comparing LLM-generated debates with real parliamentary discussions. The study tests general-purpose LLMs (including Gemini 2.5 and ChatGPT-o3) and GovernmentGPT (an open-source LLM fine-tuned on Hansard). All models were prompted on the same topic (i.e., ultra-processed foods) to enable comparison. Each generated output includes a speaker’s party affiliation and speech content. Relevance to Digital Footprints The Hansard data in this study qualify as digital footprints, capturing the digital traces of stakeholder engagement in shaping public policy. By analysing and simulating these footprints, this study shows how such data can be used to model complex social interactions and support public understanding of policy discussion. Results Two key patterns emerge from analysis. First, the simulated debates approximate the party composition of real parliamentary debates, i.e., the proportion of utterances by party broadly aligns with actual distribution. Second, many statements exhibit neutral or unclear stances, highlighting challenges in generating clear argumentative positions. Conclusions & Implications These initial results suggest that LLMs can reproduce key elements of real parliamentary debates, offering a promising step toward simulating more complex, multi-actor policy dialogues. Future work will seek to connect these outputs to public understanding more directly by introducing new stakeholder voices, simulating responses to citizen concerns, and identifying opportunities for consensus or clarification in contested policy areas such as sustainable food systems, public health, and wellbeing.

Introduction & Background Ethical consumption has, generally, increased over the past 20 years in line with access to information. However, little research exists that explicitly ties access to information with ethical purchasing. What research does exist suggests a major obstacle to ethical purchasing is financial means, but this explanation leaves a “intention-behaviour gap” that is typically explained as a lack of moral motivation, but this may not be the case. Objectives & Approach The purpose of this study was to explore how key deprivation indicators, such as the Index of Multiple Deprivation (IMD), and IMD subdomains of Income and Education, affect one well-known, highly visible aspect of ethical purchasing: Fairtrade (specifically, chocolate and coffee products across all price points available in most stores). By utilising anonymised customer data from a major UK supermarket and mapping this data onto neighborhood-level deprivation scores, a probabilistic assessment of the effects of these areas of deprivation was carried out. Relevance to Digital Footprints By using transactional data from over 450,000 customer’s supermarket loyalty cards, this study shows how access to education may play a role in promoting ethical consumerism. This work was carried out with the goal of promoting a more environmentally and socially conscious type of consumption. Results Overall IMD and Income deprivation indices were inversely related to fairtrade purchasing up to the 5th decile of deprivation (i.e. most deprived areas purchased fewer Fairtrade products) before plateauing. Education deprivation showed an inverse relationship across all deciles, demonstrating that educational deprivation acts as a major predictor of Fairtrade purchasing even when the effect of income is kept separate. More educational access means more Fairtrade purchasing across coffee and chocolate products. Conclusions & Implications This analysis of real-world shopping data shows that educational deprivation, which can affect access to information, is a major factor that predicts ethical purchasing. Though we can’t infer causation, these findings suggest that organisations wishing to promote ethical consumption may benefit from addressing informational/educational roadblocks rather than assuming ethical consumption is just a problem of wealth and moral motivation.