Behavior Matters: An Alternative Perspective on Promoting Responsible Data Science

Although an increasing number of ethical data science and AI courses is available, with many focusing specifically on technology and computer ethics, pedagogical approaches employed in these courses rely exclusively on texts rather than on algorithmic development or data analysis. In this paper we recount a recent experience in developing and teaching a technical course focused on responsible data science, which tackles the issues of ethics in AI, legal compliance, data quality, algorithmic fairness and diversity, transparency of data and algorithms, privacy, and data protection. Interpretability of machine-assisted decision-making is an important component of responsible data science that gives a good lens through which to see other responsible data science topics, including privacy and fairness. We provide emerging pedagogical best practices for teaching technical data science and AI courses that focus on interpretability, and tie responsible data science to current learning science and learning analytics research. We focus on a novel methodological notion of the object-to-interpret-with, a representation that helps students target metacognition involving interpretation and representation. In the context of interpreting machine learning models, we highlight the suitability of “nutritional labels”—a family of interpretability tools that are gaining popularity in responsible data science research and practice.

Data science is an emerging discipline that offers both promise and peril. Responsible data science refers to efforts that address both the technical and societal issues in emerging data-driven technologies. How can machine learning and AI systems reason effectively about complex dependencies and uncertainty? Furthermore, how do we understand the ethical and societal issues involved in data-driven decision-making? There is a pressing need to integrate algorithmic and statistical principles, social science theories, and basic humanist concepts so that we can think critically and constructively about the socio-technical systems we are building. In this talk, I will overview this emerging area.

Data integration has been extensively studied by the data management community and is a core task in the data pre-processing step of ML pipelines. When the integrated data is used for analysis and model training, responsible data science requires addressing concerns about data quality and bias. We present a tutorial on data integration and responsibility, highlighting the existing efforts in responsible data integration along with research opportunities and challenges. In this tutorial, we encourage the community to audit data integration tasks with responsibility measures and develop integration techniques that optimize the requirements of responsible data science. We focus on three critical aspects: (1) the requirements to be considered for evaluating and auditing data integration tasks for quality and bias; (2) the data integration tasks that elicit attention to data responsibility measures and methods to satisfy these requirements; and, (3) techniques, tasks, and open problems in data integration that help achieve data responsibility.

BackgroundRecent EULAR guidelines for the management of fibromyalgia syndrome (FMS) strongly recommend aerobic and strengthening exercise programmes, which demonstrate positive effects on symptoms and physical function.1 Despite these benefits, physical...

ABSTRACTObjective: This paper describes the theoretical basis of the Link program, a world-first dedicated online navigation tool to facilitate mental health help-seeking for young adults.Method: Several theories of help-seeking and behaviour change were explored to determine a useful theoretical framework for the Link program, a world-first dedicated online navigation tool to facilitate mental health help-seeking for young adults.Results: Four commonly used theories of behavioural change were examined with the Theory of Planned Behaviour found to best match the objectives of the Link program. Previous interventions aimed at increasing help-seeking behaviour and intentions using the Theory of Planned Behaviour found that attitudes, subjective norms, and perceived control beliefs have a role in help-seeking. Combined with elements of the Help Seeking Model, due to its specificity in mental health help-seeking, the Theory of Planned Behaviour was selected as the theoretical framework of the Link program. The resultant prototype allows young adults to enter information about an issue they are facing, the severity of the issue, and the type of services they prefer. Recommendations for information or services are then presented based upon these individual needs.Discussion: Little is known about the effects online services have on emotional affect and whether these services facilitate help-seeking in young people. Detailed description of this developmental work will assist in understanding the mechanisms underpinning the Link program. Further research will evaluate the effectiveness of the Link program in improving help-seeking, emotions, and mental health outcomes.

Primary care health professionals frequently encounter patients suffering from illnesses associated with lifestyle choices. Behaviours such as smoking, drinking alcohol, engaging in poor diets and physical inactivity are well known to have contributed to the increase in lifestyle-related illnesses like coronary heart disease, hypertension and diabetes. Behaviour change theories and interventions have attempted to reverse thistrend yet lifestyle-related diseases remain the biggest killers in society. The current study aimed to explore health professionals’ views of their role in behaviour change in order to reduce the existing gap between behaviour change theory and implementation of behaviour change interventions within routine clinical practice. Theoretical sampling in accordance with grounded theory methodology guided a purposive sample (N=14) which included medical trainees and qualified general practitioners. Individual semi-structured interviews explored participants’ views of behaviour change management in primary care. Results showed that participants were largely sceptical over the types of techniques they believed to be effective to change patients’ unhealthy behaviour. Their skills in facilitating healthier behaviours had primarily occurred through personal experience rather than formal training. The results of this study indicate the importance of delivering consistent training to health care professionals about effective, evidence-based behaviour change strategies. Further work is required to investigate the barriers to behaviour change interventions and the most effective methods for training medical practitioners in effective theoretically informed behaviour change.

Few previous reviews have identified the characteristics of effective digital health behaviour change interventions for children and/or adolescents. This systematic review aimed to identify and report the characteristics, including the theoretical bases and the incorporated behaviour change techniques (BCTs). CINAHL Ultimate (EBSCO), MEDLINE (Ovid), PubMed, ProQuest, and Web of Science were searched to identify intervention studies published in international academic peer-reviewed journals between January 2002 and May 2024 inclusive. Only studies that used a randomised controlled trial (RCT) study design to evaluate interventions were eligible for inclusion in this review. Studies were reviewed independently by two researchers to assess the risk of bias and extract data. There were 17 study articles that met the eligibility criteria for inclusion. Twenty interventions were evaluated in the 17 articles. Ten interventions were considered to be effective and incorporated one or more of the following BCTs: (1) shaping knowledge, (2) goals and planning, (3) feedback and monitoring, (4) social support, and (5) reward. Eight effective interventions were based on one or more of the following behaviour change theories: (1) Social Cognitive Theory, (2) Theory of Planned Behaviour, (3) Transtheoretical Model-Stages of Change, (4) I-Change Model, and (5) Health Action Process Approach. Effective interventions incorporated one or more BCTs and most were based on behaviour change theories. SO WHAT?: Findings can inform the selection of appropriate behaviour change theories and techniques in the development of future interventions for children and/or adolescents.

Behavior change support systems (BCSSs) have the potential to help people maintain healthy lifestyles and aid in the self-management of coronary heart disease (CHD). The Persuasive Systems Design (PSD) model is a framework for designing and evaluating systems designed to support lifestyle modifications and health behavior change using information and communication technology. However, evidence for the underlying design principles behind BCSSs for CHD has not been extensively reported in the literature. This scoping review aims to identify existing health BCSSs for CHD, report the characteristics of these systems, and describe the persuasion context and persuasive design principles of these systems based on the PSD framework. Using the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines, 3 digital databases (Scopus, Web of Science, and MEDLINE) were searched between 2010 to 2022. The major inclusion criteria for studies were in accordance with the PICO (Population, Intervention, Comparison, and Outcome) approach. Searches conducted in the databases identified 1195 papers, among which 30 were identified as eligible for the review. The most interesting characteristics of the BCSSs were the predominant use of primary task support principles, followed by dialogue support and credibility support and the sparing use of social support principles. Theories of behavior change such as the Social Cognitive Theory and Self-Efficacy Theory were used often to underpin these systems. However, significant trends in the use of persuasive system features on par with behavior change theories could not be established from the reviewed studies. This points to the fact that there is still no theoretical consensus on how best to design interventions to promote behavior change in patients with CHD. Our results highlight key software features for designing BCSSs for the prevention and management of CHD. We encourage designers of behavior change interventions to evaluate the techniques that contributed to the success of the intervention. Future research should focus on evaluating the effectiveness of the interventions, persuasive design principles, and behavior change theories using research methodologies such as meta-analysis.

BackgroundConversational agents (CAs) are increasingly used in health care to deliver behavior change interventions. Their evaluation often includes categorizing the behavior change techniques (BCTs) using a classification system of which the BCT Taxonomy v1 (BCTTv1) is one of the most common. Previous studies have presented descriptive summaries of behavior change interventions delivered by CAs, but no in-depth study reporting the use of BCTs in these interventions has been published to date.ObjectiveThis review aims to describe behavior change interventions delivered by CAs and to identify the BCTs and theories guiding their design.MethodsWe searched PubMed, Embase, Cochrane’s Central Register of Controlled Trials, and the first 10 pages of Google and Google Scholar in April 2021. We included primary, experimental studies evaluating a behavior change intervention delivered by a CA. BCTs coding followed the BCTTv1. Two independent reviewers selected the studies and extracted the data. Descriptive analysis and frequent itemset mining to identify BCT clusters were performed.ResultsWe included 47 studies reporting on mental health (n=19, 40%), chronic disorders (n=14, 30%), and lifestyle change (n=14, 30%) interventions. There were 20/47 embodied CAs (43%) and 27/47 CAs (57%) represented a female character. Most CAs were rule based (34/47, 72%). Experimental interventions included 63 BCTs, (mean 9 BCTs; range 2-21 BCTs), while comparisons included 32 BCTs (mean 2 BCTs; range 2-17 BCTs). Most interventions included BCTs 4.1 “Instruction on how to perform a behavior” (34/47, 72%), 3.3 “Social support” (emotional; 27/47, 57%), and 1.2 “Problem solving” (24/47, 51%). A total of 12/47 studies (26%) were informed by a behavior change theory, mainly the Transtheoretical Model and the Social Cognitive Theory. Studies using the same behavior change theory included different BCTs.ConclusionsThere is a need for the more explicit use of behavior change theories and improved reporting of BCTs in CA interventions to enhance the analysis of intervention effectiveness and improve the reproducibility of research.

Sedentary lifestyles and bad eating habits influence the onset of many serious health problems. Healthy behavior change is an arduous task, and requires a careful planning. In this thesis, we propose that behavior recommenders can help their users achieve healthy behavior change. Such a system should inspire its users with small, incremental and achievable goals. For this, it must resolve a trade-off between two opposing objectives: help the user achieve a steady improvement in target behavior, and avoid extreme goals that may injure or discourage the user. This is an unprecedented challenge in the recommender systems research. If the system understands the impacts of past interventions for behavior change, it can determine its usersâ behavioral responses to its own recommendations. This implies a specific data curation, in which we not only measure people's behavior but also deliberately introduce an intervention to monitor its effect on people's patterns. In turn, the system can use these existing users' information to derive the right procedure for effective recommendations. In this study we capitalize on this insight and develop InspiRE - our behavior recommender framework. Through InspiRE we propose the following contributions: 1) We design the data curation. 2) We develop the novel approaches for behavior profiling 3) We develop an evaluation process for this novel type of recommender system, and also compare it with traditional, similarity-based recommendation approach. We curate a dataset that contains information of daily step counts and social intervention for 83 people. InspiRE successfully uses the observations from this dataset, and proposes recommendations that are both effective and feasible. We also show that InspiRE can generalize to other dimensions of well being: we demonstrate this through a dataset that contains the snacking patterns of 73 people, who receive message-based interventions. We observe that InspiRE's recommendation strategy is in line with theories of behavior change.

Ensuring that the growing mobility demand is met in a sustainable manner is important for our climate goals, and this would require changes to our current mobility behaviours. Behaviour and behaviour change theories have an important role in informing the mobility behaviour research and practices that seek to achieve these changes, and this paper discusses the application of these theories in both research and practice. Newer and more comprehensive theories have been developed in this area in the last decade, improving the quality of theories available. However, two key challenges of using theories are identified. The great number of theories today makes it challenging when deciding what and when to use them, and the appropriateness of many theories are either too simplistic or too complex. Thus, there is a need to critically review the state of theories in this area and identify ways to advance our application and development of behavioural theories. Here, two suggestions are put forward on how we can address these challenges while advancing the use of theories in mobility behaviour research and practice. First, integrate and summarise multiple, individual theories into a comprehensive and accessible framework for researchers and practitioners. Second, expand this framework beyond behaviour and behaviour-change theories by mapping out their interconnections with theories from other paradigms and disciplines using a systems approach to facilitate a more holistic theorisation.

Physical activity (PA) and exercise programmes are recommended for the management of fibromyalgia. Despite positive effects on symptoms and function, PA promotion remains a significant clinical challenge. Behaviour change theories and techniques are recommended as part of complex health interventions; their integration into interventions aimed at PA behaviour in people with fibromyalgia is not known. This review explored behaviour change interventions targeting PA in adults with fibromyalgia. A systematic review was conducted; randomized and quasi-randomized controlled trials with at least one behaviour change intervention targeting PA were included. MEDLINE/OVID, EMBASE, PEDro, PsychINFO, CINAHL, Scopus, Web of Science, the Cochrane Central Register of Controlled Trials and relevant conference abstracts were searched. Two authors independently screened studies for inclusion and performed risk of bias assessments. Articles were reviewed for their use of behaviour change theory and behaviour change techniques (BCTs). The search identified 2491 records, from which eight studies (1416 participants) were included. PA and exercise behaviours were the primary focus of four interventions and were components of broader interventions in four studies. Behaviour change theories informed four interventions. Thirty-two different BCTs were used across studies. Five studies reported improvements in PA either post-intervention or at follow-up. Two studies used objective PA measures and seven studies used self-report measures. Short-term benefits in pain, quality of life, and physical fitness were also observed. Behaviour change interventions targeting PA in people with fibromyalgia have had limited success to date. With significant variations in intervention designs, the optimal intervention remains unknown.

The rising use of the Internet and information technology has made computerized interventions an attractive channel for providing advice and support for behaviour change. Health behaviour and behaviour change theories are a family of theories which aim to explain the mechanisms by which human behaviours change and use that knowledge to promote change. Among the best-known of these theories are the Social Learning and Social Cognitive theories, the Health Belief Model, the Theory of Reasoned Action and its successors the Theory of Planned Behaviour and the Reasoned Action Approach, and the Transtheoretical model. We discuss three examples of how behaviour change theories have been applied in computer-based interventions: a system to aid users to quit smoking, a decision aid for choice of breast cancer therapy, and an internet-based exercise program for reducing cardiovascular risk. We also discuss misapplication of theory, and reflect on how these theories can best be used. Behaviour change theory can be applied in health informatics interventions in several ways; for example, to select participants for a particular intervention, to shape the content of the intervention to effectively influence behaviour, or to tailor content to individual needs. Application of these theories to provide personalized advice ("decision support") is a young but promising area of research, and could inform other decision support interventions, including those that provide support for clinicians.

This entry provides an overview of human‐centered geospatial data science, highlighting the gaps it aims to bridge, its significance, and its key topics and research. Geospatial data science, which derives geographic knowledge and insights from large volumes of geospatial big data using advanced geospatial artificial intelligence, has been widely used to tackle a wide range of geographic problems. However, it often overlooks the subjective human experiences that fundamentally influence human–environment interactions, and few strategies have been developed to ensure that these technologies follow ethical guidelines and prioritize human values. Human‐centered geospatial data science advocates for two primary focuses. First, it advances our understanding of human–environment interactions by leveraging geospatial data science to measure and analyze human subjective experiences at place including emotion, perception, cognition, and creativity. Second, it advocates for the development of responsible and ethical geospatial data science methods that protect geoprivacy, enhance fairness, and reduce bias, and improve the explainability and transparency of geospatial technologies. With these two missions, human‐centered geospatial data science brings a fresh perspective to develop and utilize geospatial technologies that positively impact society and benefit human wellbeing and the humanities.

Although the HPV vaccine is highly safe and effective, its uptake is sub-optimal in many countries, including Ireland. There is therefore a need to identify appropriate interventions that will increase HPV vaccine acceptance by parents. In this study, we took a systematic approach to understand the factors that influence HPV vaccine uptake by parents of adolescent girls in Ireland to define suitable behaviour change interventions that would support positive vaccine decision-making in the future. Specifically, we conducted semi-structured interviews, used a Theoretical Domains Framework (TDF)-based topic guide, to gain insight into the knowledge, beliefs, attitudes and current behaviours of parents with respect to their HPV vaccine decision. Transcripts were analysed using the TDF. The Behaviour Change Wheel (BCW) was used to identify relevant intervention functions and the Behaviour Change Technique Taxonomy version 1 (BCTTv1), to identify relevant intervention techniques. All parents discussed the essential role of healthcare providers in vaccine decision-making. Complacency and confidence were important factors in decision-making by vaccine hesitant parents. Five BCW intervention functions were identified as appropriate, namely; education; persuasion; environmental restructuring; modelling and enablement. To our knowledge, this is the first study to systematically evaluate HPV vaccine decision-making using behaviour change theory and identify suitable intervention strategies to promote positive vaccine decision-making using this approach.