The assessment of student wellbeing has been often static and lagged behind for the intervention/diagnostic purpose. In this chapter, we aim to introduce an automated school wellbeing scoring dynamic real-time system, School Day Wellbeing Model. With Artificial Intelligence (AI)-based item sampling methods and answers scoring and reporting systems, the School Day Wellbeing Model can collect wellbeing data at low cognitive cost, track wellbeing real time at multiple levels (e.g., individual-, class-, school-level), and give immediate feedback. The model is constructed on the basis of the School Wellbeing Model, Study Demand-Resource Model, and OECD Social-Emotional Skill Model. In the book chapter, the wellbeing assessments, including AI-based assessments, are reviewed so that the strengths of the School Day Wellbeing Model are highlighted. User experiences are collected to show the utility of the model. During the COVID-19 pandemic, the need for such a model is imperatively high as students’ wellbeing has been largely dampened. As a result, the model has been appreciated by users and has served about 55,000 students so far in the globe. The future development of the model is also discussed.
- Students’ wellbeing
- Automated scoring system
- Social-emotional skills
- Artificial Intelligence
Wellbeing, the state of being well in physical, mental, and social aspects of life, has been a focus in research for the past two decades (Diener et al. 2017, 2018; Seligman 2011). People who have high wellbeing are likely to succeed in life (Lyubomirsky et al. 2005), to live longer (Diener and Chan 2011), and to conduct prosocial behaviors (Oishi et al. 2007). Students with high wellbeing are also the ones who have high academic achievement (Kiuru et al. 2020; Salmela-Aro 2020) and exhibit fewer problem behaviors (Arslan and Renshaw 2018). Given the important role of wellbeing, it is not surprising to see a surge of research on the assessment, antecedents, and outcomes of wellbeing.
The assessment of wellbeing has always been a significant theme in the society and research field. During the past few decades, national-level policy makers have tried to assess and track wellbeing to build a sustainable society (e.g., UK Office for National Statistics; Allin and Hand 2017). International comparison assessments (e.g., World Happiness Report; Helliwell et al. 2021) also collect wellbeing data to compare and understand wellbeing gaps between multiple countries across the world. Education sectors (e.g., education policy makers, schools, universities) are joining this endeavor to understand students’ wellbeing, with an aim of improving wellbeing to support learning gains (OECD 2013, 2019). Industries are also striving to provide applications to assess, track, and report wellbeing. With the development of Artificial Intelligence (AI) techniques, there are increasing applications and research on AI-based wellbeing assessments (Castro et al. 2018).
In this chapter, we aim to introduce a newly developed wellbeing assessment and enhancement system, School Day Wellbeing Model, as a joint product of researchers and industry practitioners. We first review the (traditional) assessments of wellbeing, and then review AI-based wellbeing assessments. After identifying some caveats in those assessments, the School Day Wellbeing Model is introduced to show its features and strengths as a novel AI-based wellbeing assessment application. The user experiences are also gathered to show its validity and the future directions of the Model are discussed.
2 The Assessments of Wellbeing
Measuring wellbeing has been a central task for the new science of wellbeing (Diener et al. 2018). Wellbeing has been assessed and indexed using objective measures (e.g., physiological data, life expectancy as for country-level wellbeing) and subjective measures (e.g., self-reported happiness, life satisfaction; for reviews, see Conceição and Bandura 2008; Ong et al. 2021). Though objective measures can provide some information on wellbeing, the majority of wellbeing assessments are subjective measures as wellbeing is largely idiographic (i.e., relating to an individual’s own experiences and interpretations; Rose et al. 2017; VanderWeele et al. 2020). To date, subjective wellbeing has been mainly examined from three approaches: evaluative, hedonic, and eudaimonic approaches. The evaluative approach portrays wellbeing as an individual’s view of satisfaction with life. The corresponding scales typically measure the overall life satisfaction or satisfaction in different domains of life (Diener et al. 1985; Pavot and Diener 1993). Hedonic approach examines wellbeing as positive affective experiences such as happiness or pleasure. Scales under this approach typically ask the participants to report their experiences of positive and negative emotions (e.g., Positive and Negative Affect Schedule – PANAS Scale; Watson et al. 1988). The last approach, eudaimonic approach, describes wellbeing from the perspective of meaning and purpose of life. Accordingly, scales from this approach typically measure the extent to which individuals live a purposeful life or fulfill their self-realization (Ryff 1989; Ryff and Singer 2008).
Recently, most wellbeing assessments acknowledge the multidimensional nature of wellbeing and typically include items from all three approaches (Diener et al. 2018; VanderWeele et al. 2020). For instance, Seligman’s model of Positive emotion, Engagement, positive Relationship, Meaning, and Accomplishment (PERMA) describes wellbeing as a compound concept (2011). According to the PERMA model, positive emotions denote hedonic experiences, such as feeling happy, joyful, and cheerful. Engagement represents positive experiences in activities, such as feeling absorbed and immersed in life. Positive relationships refer to the psychological connections with others (e.g., peers and parents). Meaning represents the feelings of being valuable and of being purposeful in life. Accomplishment refers to feeling capable to pursue goals and to finish tasks. A valid wellbeing assessment for adolescents on the basis of the PERMA model has also been established recently (Kern et al. 2015).
However, for most wellbeing assessments, the common collecting method is paper and pencil, which reduces data collection efficiency. In addition, much wellbeing information is collected only once per year, limiting the assessments’ ecological validity. One recent review (Ong et al. 2021) indicated that only 1.7% assessments ask for the reporting of wellbeing at the momentary level. As hedonic wellbeing (e.g., positive or negative emotions) is highly sensitive to situations, wellbeing assessments with high ecological features are imperatively needed. The School Day Wellbeing Model is a tool which measures subjective wellbeing in a timely manner, collects wellbeing data virtually, reports wellbeing automatically, and offers feedback correspondingly. We provide a detailed description of the School Day Wellbeing Model in the latter section.
3 Artificial Intelligence-Based Wellbeing Assessments and Enhancement
For several decades, educational assessments using artificial intelligence-based techniques and tools have been a research topic. To date, the most common AI-based assessments in the field of education are automated grading systems or adaptive assessment systems (Gardner et al. 2021; González-Calatayud et al. 2021). There has been also a great interest in collecting wellbeing information with the help of intelligent systems or devices in recent years. Nowadays, there is a wide use of intelligent devices (e.g., smartphones, smart watches, smart wristbands) that collect information on sleep patterns and physical exercise, which are essential parts of wellbeing (Castro et al. 2018). However, these measures are mostly for adults, rather than for school children or adolescent students. More importantly, the information collected by the intelligent devices mostly concern indicators of objective wellbeing rather than subjective wellbeing. Yet, as we stated, subjective wellbeing is a critical and indispensable part of one’s wellbeing. Some researchers even argue that we should focus mainly on subjective wellbeing as the interpretation process is critical for the final wellbeing status (Krueger and Stone 2014; OECD 2013). For instance, people may experience happiness even if their exercise is sparse or sleep only 6 h a night.
Researchers have attempted to examine associations between objective measures (e.g., heart rate variability, blood pressure, mobile log data) and subjective wellbeing (e.g., happiness, positive and negative emotions; Gordon and Mendes 2021; Jaques et al. 2015). Given the big data gathered through intelligent devices, researchers have utilized several machine learning algorithms to predict subjective wellbeing on the basis of data on objective measurements (Jaques et al. 2015; Taylor et al. 2020). The central idea is to see whether subjective wellbeing can be represented by merely looking at data on objective measures.
For instance, one study collected four types of data (physiological data, survey data, phone data, location data) with mobile sensors and smartphones (Jaques et al. 2015). University students participated in the study over two 1-month (30-day) experimental periods. Physiological data consisted of electrodermal activity (EDA; a measure of physiological stress), and three-axis accelerometer (a measure of steps and physical activity). The survey data consisted of questions about academic activity, sleep, drug and alcohol use, exercise, stress, and wellbeing measures such as health, energy, alertness, and happiness. The phone data included phone call, SMS, and usage patterns. The location data included the GPS coordinates throughout the day. The authors extracted and formulated features from each data source before they evaluated and reduced the number of features. After this step, multiple algorithms, such as Support Vector Machines (SVM), Random Forests (RF), Neural Networks (NN), Logistic Regression (LR), k-Nearest Neighbor (kNN), and AdaBoost, were applied to test the predictability of each algorithm for classifying subjective happiness. The results showed that an ensemble classifier they discovered can have about 70% accuracy rate in predicting the state of happiness. However, in this study and many other studies (e.g., Gordon and Mendes 2021; Taylor et al. 2020), the examinations of subjective wellbeing are very limited. In addition, the multidimensional nature of subjective wellbeing (including general and academic wellbeing) was unaddressed.
Besides the AI-based wellbeing assessments, there are also several intelligent applications that aim to improve wellbeing. The most typical application is conversational agents or chatbots (Dekker et al. 2020; Inkster et al. 2018). Chatbots or agents utilize natural language processing techniques with psychological counseling methods (e.g., dialectical behavior therapy, behavioral reinforcement, mindfulness) and can respond to users’ questions and requests and to reduce their health problems (e.g., anxiety, stress, sleeping problems). For instance, one conversational AI agent (Wysa App) used text-analysis techniques to converse with users who needed assistance for their wellbeing (Inkster et al. 2018). The authors revealed that the frequent use of this application improves the users’ wellbeing (by reducing their depressive symptoms) significantly.
In the School Day Wellbeing Model, we choose an approach which combines wellbeing assessment and improvement simultaneously. As we value the complex and multidimensional nature of subjective wellbeing, we constructed a new wellbeing assessment model. More importantly, the way the data has been collected is largely different from traditional assessments. Experience sampling methods (Hektner et al. 2007), in which survey items are randomly repeatedly measured, have been used. More importantly, the randomization of the item sampling is driven by AI techniques (see the following section) to select the questions strategically and automatically. After the data has been collected, the wellbeing status will be reported automatically and the feedback for improvement will be delivered timely according to the status. A detailed description of the model is in the following sections.
4 School Day Wellbeing Model: A Model for Wellbeing Assessment and Enhancement
The School Day Wellbeing Model is constructed jointly by the researchers and practitioners as a response to the call for an ecologically valid measure of wellbeing and for an intelligent solution to detect and improve student wellbeing. A distinctive part of the School Day Wellbeing model, in comparison with other wellbeing assessments, is that it not only focuses on measuring wellbeing but also on improving wellbeing. In other words, it is a model for wellbeing assessment and enhancement simultaneously. The model intends to report, monitor, and track wellbeing live, so that it can provide timely feedback given the person’s current wellbeing status.
4.1 Theoretical Foundations for the School Day Wellbeing Model
The School Day Wellbeing Model is built by integrating three theoretical frameworks (see Fig. 1 for the latest model): School Wellbeing Model, Study Demands-Resources Model, and OECD Social Emotional Skills.
4.1.1 School Wellbeing Model
School wellbeing model (Konu et al. 2002; Konu and Rimpelä 2002) defined four broad indices to represent wellbeing and its supportive environment: school conditions, social relationships, means for self-fulfillment, and health status. School conditions include physical environment (e.g., ventilation is good; inappropriate desks), school organization (e.g., rules and regulations are sensible), and school services. Social relationships cover school climate (e.g., teachers treat pupils fairly), relationships with teachers and peers (e.g., I have friends in school; easy to get along with teachers), and bullying experiences (e.g., classmates intervene in bullying). Means for self-fulfillment includes autonomy support (e.g., pupils’ views are taken into account) and school engagement (e.g., I am able to follow teaching). Health status contains the evaluation of current physical health condition. The model has been recognized as a valid tool for assessing students wellbeing from grade 4 to 12 (Konu and Lintonen 2006).
4.1.2 Study Demands-Resources Model
Study Demands-Resources model (Salmela-Aro, Tang and Upadyaya, in press; Salmela-Aro and Upadyaya 2014) proposed that wellbeing (particularly school engagement and burnout) is based on the fit between demands and resources. Both demands and resources can be divided into school- and person-related factors. Demands are factors that cause exhaustion and burnout, such as school work load. Resources are factors that promote personal development, such as self-efficacy and social support. More importantly, the model proposes a synergistic role of demands and resources in determining wellbeing. In consequence, the assessment of wellbeing should consider the positive and negative side of environmental factors and of wellbeing itself. The model has been tested among students and shown its predictive validity in explaining well- and ill-being (Romano et al. 2020; Salmela-Aro et al. 2008; Salmela-Aro and Upadyaya 2014).
4.1.3 OECD Social Emotional Skills Framework
To understand the key factors that enhance wellbeing, the OECD social-emotional skill framework (Kankaraš and Suarez-Alvarez 2019) was adopted and included in the model. It defines social-emotional skills as: “individual capacities that (a) are manifested in consistent patterns of thoughts, feelings, and behaviors, (b) can be developed through formal and informal learning experiences, and (c) influence important socioeconomic outcomes throughout individual’s life” (OECD 2015, p. 35). The model proposed five broad skills: task performance, emotional regulation, collaboration, open-mindedness, and engaging with others. Task Performance refers to the ability to be self-disciplined, persistent, and dedicate effort in achieving goals and completing tasks. Emotional Regulation is the ability to control one’s emotional responses and moods, as well as to be positive and optimistic about self and life in general. Collaboration is the ability to maintain positive relations and to be sympathetic to others. Open-mindedness is the ability to engage with new ideas and generate novel ways to do or think. Lastly, Engaging with Others is the ability to engage with others, and to be energetic and assertive. The role of social-emotional skills in affecting students’ wellbeing and achievement has been established in the OECD international comparison study of social-emotional skills (OECD 2021) and other recent studies (Guo et al., 2022; Salmela-Aro et al. 2021; Salmela-Aro and Upadyaya 2020; Tang et al. 2019, 2021).
4.2 School Day Wellbeing Model
As an integrative model, the School Day Wellbeing Model has four broad domains: Learning, Social and Emotional Skills, Social Relationships, and Wellness (see Fig. 1). Learning is the domain that covers studying skills and environment factors, such as self-studying (e.g., I like studying on my own), study support (e.g., It is easy to get support from teachers), learning environment (e.g., I have a peaceful place to study), and learning material (e.g., I have the necessary school supplies). Social and Emotional Skills are five skills introduced above (i.e., task performance, emotional regulation, collaboration, open-mindedness, and engaging with others). Social Relationships is the domain related to the communication and interaction. It includes communication with teachers (e.g., It is easy to keep in touch with my teachers), communication with peers (e.g., I can get help from my classmates), communication outside school (e.g., I get support when studying at home), and student services (e.g., I can get help if I am overwhelmed). Wellness is the domain related to physical health, mental health, and academic wellbeing. It covers physical health (e.g., I am not concerned about my health), emotions (e.g., I feel happy; My anxiety is low), diet (e.g., My diet is healthy), psychological wellbeing (e.g., I like being at school), and academic wellbeing (e.g., Time flies when I am studying). Overall, the model has 64 items with each dimension having three to six items.
4.3 How Does the School Day Wellbeing Model Work?
The School Day Wellbeing Model is driven by several automated techniques (Kylväjä et al. 2019) in sampling the items, cleaning the data, scaling the answers, reporting the results, and providing feedback (see Fig. 2). Information concerning subjective wellbeing is collected through a mobile, web, or an online platform (e.g., Microsoft Teams). The platform notifies students to answer questions once a week. Once a classroom takes School Day into use for the first time, the model asks all the 64 questions so that an immediate baseline can be formed in the classroom. After the initial 64 questions, the amount of questions to be answered is limited to 10 items per week per student to reduce cognitive burden. The question sampling procedure is not purely random. The questions are delivered by an Artificial Intelligence algorithm built by School Day that selects the items strategically from the item pool so that a balanced sample of student wellbeing can be formed at any particular time. The answers to the items are recorded on a Likert scale (5 = totally agree, 1 = totally disagree) and scaled to the point from 1 to 100 with scaling functions. The wellbeing reports are then generated automatically based on the answers. The reports can be read by teachers concerning their own class, by principals concerning their school, and by administrators concerning the region they are responsible for. The wellbeing reports also include trends of change so that the wellbeing status of each entity (classroom, school, region) can be compared weekly, monthly, or yearly. It is also possible to compare the wellbeing performance across classes and schools.
Once the wellbeing status has been recorded and reported, the feedback module will function to provide adaptive group level feedback according to the wellbeing status. The feedback is delivered to students, teachers, principals, and educational administrators. School Day AI module distributes weekly (e.g., cards) content highlighting what is going well and what needs attention and improvement. The weekly feedback content covers a broad series of wellbeing improvement practices (e.g., how to cope with stress, if there is a report on high level of stress). Additionally, monthly (e.g., lesson plans) contents are provided for teachers on broader topics in the School Day Wellbeing Model such as social skills, task performance, physical health etc. Moreover, social-emotional learning tools (Durlak et al. 2015) have been used to guide feedback provision.
4.3.1 Ethical Code When Implementing School Day Wellbeing Model
The School Day Wellbeing Model is operated following the General Data Protection Regulation (GDPRFootnote 1) and research ethics. The collected data is stored in secure Microsoft Azure storages hosted in respective regions where the users are using the platform in North America, Europe, or Asia. In most countries, for students who are under age 16, parental consents have been collected prior to the participation in the data collection. The participation in the data collection is voluntary, students can quit the data collection at any time they prefer. The answers are fully anonymized and only analyzed on group/classroom level with a minimum of five respondents. Individual students and responses are not identified and only an answer distribution chart will be shown to teachers and administrators.
5 Features of the School Day Wellbeing Model
As a whole, besides the rigorous theoretical foundations, the School Day Wellbeing Model has several features that are distinctive from other wellbeing models.
One strength of the model is that it has a broad scope on wellbeing. As we have indicated, the model focuses on wellbeing assessment and enhancement together. Moreover, both general wellbeing and academic wellbeing are measured in the model. Consequently, the School Day Wellbeing Model can provide an overview of the student’s daily life and school life.
By asking students to respond to survey questions once a week, the School Day Wellbeing Model measures wellbeing at the momentary level regularly. The momentary assessment can have high ecological validity in reflecting the authentic phenomena of wellbeing. The automated reporting procedure can track and present wellbeing continuously. The visualization of wellbeing status can show the trends of change and reflect the dynamic nature of wellbeing (see Fig. 3).
Once the data has been gathered, wellbeing can be reported automatically. More importantly, wellbeing is layered for different audiences. Students will receive class-level wellbeing status. Teachers can oversee class-level wellbeing status. Principals can additionally see school-level wellbeing status. The wellbeing information can also be seen at the district- or city-level when it is needed. The multilayered wellbeing reports can have important practical implications, so that each stakeholder receives corresponding feedback and can use the most appropriate strategies to improve wellbeing (see Fig. 4 for a teacher’s view).
Timely Feedback and Intervention for Wellbeing Improvement
Given the dynamic nature of the model, feedback that is delivered to each stakeholder is highly time appropriate (see Fig. 5). This feature allows the School Day Wellbeing Model to provide timely intervention to the stakeholders when some mental or physical health problems have been reported frequently. This feature also makes the model distinct from traditional wellbeing assessment systems where wellbeing is measured only once or twice per year.
Social-Emotional Skills as Key Enhancers
While multiple feedbacks and interventions can be suggested, social-emotional skills play a key role in improving wellbeing. In modern society, students may often face unexpected environmental changes (e.g., transitioning to an unfamiliar school, or moving to a new city/country) in their life. When interventions targeting environmental factors are difficult to manage or too slow to see the actual effects, equipping students with necessary skills is a central task. Those skills are transferable so that students can cope with any situation to maintain their wellbeing. Thus, the School Day Wellbeing Model emphasizes social-emotional skills and aims to build those transferable skills for students.
Cognitive Cost Efficiency
Although the item pool is comparatively large, students are not required to answer all of them each time when they receive the notifications. The model has an AI-driven question analytics system so that a balanced sample of student wellbeing can be formed without continuously having answers from all the students in the group. This feature also significantly reduces the cognitive demands of question answering.
6 User Experiences
The School Day Wellbeing Model was launched in January 2019 and has served approximately 55,000 students in 26 countries (e.g., UK, USA, Finland) in the world so far. We also contact users to collect their experiences and to give feedback on using the model. In general, the feedback is positive, and many users have reported that the use of the School Day Wellbeing Model improves their wellbeing. Below are some examples of the feedback we have received from students, teachers, and school staff.
One eighth grader from Finland expressed that “Personally, I think that it is a very helpful and handy app to use. Mainly because you do not have to expose your name, which, of course, gives honest feedback. It really improved the mood in school and helped us feel better and learn more.” Similarly, one sixth grader from the UK said that “Answering the questions and going through the data together with the whole class has made me realize I am not the only one who has felt a certain way.” Even a younger student in the third grade from Finland expressed that “It’s great when I can tell how I feel without fear of being judged or causing a disappointment.”
A school teacher from Finland said that “We have been able to teach students about wellbeing factors and how they can observe their emotions. This has helped me to reflect my own work broadly and to apply tools promoting wellbeing in my class. It has been easier to keep track of students’ experiences of wellbeing, as well as the atmosphere and learning process of the class. We’ve had good discussions, even on the more difficult themes.” An educational department head from Finland also said that “The data has clarified and deepened our understanding of existing wellbeing issues. Based on the shared data we have discussed together with students how to maintain the positive development and deal with the challenges.”
Teachers, school leaders, and administrators from other countries have also expressed their appreciation of the model. One UK teacher who also serves as the head of school wellbeing said that “As many teachers do not feel fully confident in discussing and handling mental health and wellbeing, the app has proved very useful for developing their abilities in the said areas.” Another teacher from the USA expressed that “School Day gives me a way to check in with the students without face-to-face checking in with them. This has been helpful in quarantine, but also when students are uncomfortable about what’s on their mind and find it difficult to share their feelings. Now they can reflect their emotions at their own pace and talk to me or other adults when they feel ready.”
7 The Future Directions
Despite of many strengths the School Day Wellbeing Model can be improved for the future iterations. We suggest several future directions for the model’s development.
First, the current model only measures students’ wellbeing, however, teachers’ and principals’ wellbeing has not been measured. Teachers’ wellbeing, as it has been discovered (Zee and Koomen 2016), is important to be maintained to improve students’ wellbeing. Consequently, teachers’ and other staffs’ wellbeing is a critical component for building a comprehensive high wellness school environment. In the future, the School Day Wellbeing Model will have wellbeing assessments and enhancements for teachers and principals. Thus, both teachers and principals can receive feedback in order to maintain a good level of wellbeing.
Second, in the current model, the involvement of parents is only at a minimal level. That is, though parents provide the consent for children’s participation, they receive little information about their children’s wellbeing status. It is possible in the future the model can share a weekly or monthly summary report for parents, and to provide some feedback to parents concerning the children’s wellbeing status.
Third, the current model only focuses on the school children, from grade 1 to 12. Students beyond that level are not included. In the future, the School Day Wellbeing Model plans to have a version for higher education institutes. Thus, university students’, teachers’, and staffs’ wellbeing will also be measured to serve the stakeholders in higher education.
Fourth, in the future, the model can integrate the school grades system so that students’ academic performance can be combined with wellbeing datasets. Consequently, on the one hand, students’ academic performance can be traced and recorded at multiple levels. On the other hand, the relationships among wellbeing, social-emotional skills, and academic performances can be examined. These are imperatively needed to understand, for instance, the role of wellbeing in students’ learning outcomes or vice versa, and how to promote academic development by enhancing wellbeing and building social-emotional skills.
Finally, although it is indispensable that subjective wellbeing is measured, future development of the School Day Wellbeing Model may include some objective measures (e.g., footsteps per day, sleeping hours, heart rate variability) to make it a hybrid assessment model. Combining subjective and objective wellbeing measures can possibly yield a stronger model on wellbeing assessments. Nowadays, there are several smartphones and applications (Gordon and Mendes 2021; MyBPLab 2021) that are pursuing in this direction, though their measures of subjective wellbeing are limited.
In conclusion, the data obtained via the School Day Wellbeing application provides researchers and users real-time dynamic information concerning students’ wellbeing. The information on wellbeing is described on multiple levels (e.g., class/group-, school-, regional-level), which provides users and researchers a more holistic picture of students’ current wellbeing. The anonymity of the users provides students better security that their answers will not be analyzed individually, which helps in giving less socially desirable and more honest answers. When decreases or room for improvement is recognized in wellbeing, the feedback module of the School Day Wellbeing application provides users information on enhancement. Being able to see the graphs for the whole classroom’s wellbeing may also enhance students’ sense of belonging, and reduce anxiety of being alone in the situation. The journey of the School Day Wellbeing Model is in its beginning stage; however, the results and feedback from the users are promising. The model is constantly developed further, and information concerning multiple levels of school societies’ wellbeing will become more detailed in the future when the model will target all levels of the school society. Similarly, the possibilities for collecting objective data on wellbeing through physical measures will give new possibilities for a more detailed wellbeing profile of the whole school.
Allin, P., & Hand, D. J. (2017). New statistics for old?-measuring the wellbeing of the UK. Journal of the Royal Statistical Society: Series A (Statistics in Society), 180(1), 3–43. https://doi.org/10.1111/rssa.12188
Arslan, G., & Renshaw, T. L. (2018). Student Subjective Wellbeing as a Predictor of Adolescent Problem Behaviors: a Comparison of First-Order and Second-Order Factor Effects. Child Indicators Research, 11(2), 507–521. https://doi.org/10.1007/s12187-017-9444-0
Castro, L. A., Rodríguez, M. D., Martínez, F., Rodríguez, L.-F., Andrade, Á. G., & Cornejo, R. (2018). Charting the Past, Present, and Future in Mobile Sensing Research and Development. In Intelligent Data Sensing and Processing for Health and Well-Being Applications (pp. 3–21). Elsevier. https://doi.org/10.1016/B978-0-12-812130-6.00001-9
Conceição, P., & Bandura, R. (2008). Measuring subjective wellbeing: A summary review of the literature.
Dekker, I., De Jong, E. M., Schippers, M. C., De Bruijn-Smolders, M., Alexiou, A., & Giesbers, B. (2020). Optimizing Students’ Mental Health and Academic Performance: AI-Enhanced Life Crafting. Frontiers in Psychology, 11. https://doi.org/10.3389/fpsyg.2020.01063
Diener, E., & Chan, M. Y. (2011). Happy People Live Longer: Subjective Well-Being Contributes to Health and Longevity. Applied Psychology: Health and Well-Being, 3(1), 1–43. https://doi.org/10.1111/j.1758-0854.2010.01045.x
Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The Satisfaction With Life Scale. Journal of Personality Assessment, 49(1), 71–75. https://doi.org/10.1207/s15327752jpa4901_13
Diener, E., Heintzelman, S. J., Kushlev, K., Tay, L., Wirtz, D., Lutes, L. D., & Oishi, S. (2017). Findings all psychologists should know from the new science on subjective well-being. Canadian Psychology/Psychologie canadienne, 58(2), 87–104. https://doi.org/10.1037/cap0000063
Diener, E., Oishi, S., & Tay, L. (2018). Advances in subjective well-being research. Nature Human Behaviour, 2(4), 253–260. https://doi.org/10.1038/s41562-018-0307-6
Durlak, J. A., Domitrovich, C. E., Weissberg, R. P., & Gullotta, T. P. (Eds.). (2015). Handbook of social and emotional learning: Research and practice. Guilford Publications.
Gardner, J., O’Leary, M., & Yuan, L. (2021). Artificial intelligence in educational assessment: ‘Breakthrough? Or buncombe and ballyhoo?’ Journal of Computer Assisted Learning, 37(5), 1207–1216. https://doi.org/10.1111/jcal.12577
González-Calatayud, V., Prendes-Espinosa, P., & Roig-Vila, R. (2021). Artificial Intelligence for Student Assessment: A Systematic Review. Applied Sciences, 11(12), 5467. https://doi.org/10.3390/app11125467
Gordon, A. M., & Mendes, W. B. (2021). A large-scale study of stress, emotions, and blood pressure in daily life using a digital platform. Proceedings of the National Academy of Sciences, 118(31), e2105573118. https://doi.org/10.1073/pnas.2105573118
Guo, J., Tang, X., Marsh, H., Parker, P., Basarkod, G., Baljinder, S., Ranta, M. & Salmela-Aro, K. (2022). The roles of social-emotional skills in students’ academic and life success: A multi-informant and multicohort perspective. Journal of Personality and Social Psychology. http://dx.doi.org/10.1037/pspp0000426
Hektner, J. M., Schmidt, J. A., & Csikszentmihalyi, M. (2007). Experience Sampling Method: Measuring the Quality of Everyday Life. Thousand Oaks: SAGE Publications.
Helliwell, J. F., Layard, R., Sachs, J., & Neve, J.-E. De. (2021). World Happiness Report 2021. New York.
Inkster, B., Sarda, S., & Subramanian, V. (2018). An Empathy-Driven, Conversational Artificial Intelligence Agent (Wysa) for Digital Mental Well-Being: Real-World Data Evaluation Mixed-Methods Study. JMIR mHealth and uHealth, 6(11), e12106. https://doi.org/10.2196/12106
Jaques, N., Taylor, S., Azaria, A., Ghandeharioun, A., Sano, A., & Picard, R. (2015). Predicting students’ happiness from physiology, phone, mobility, and behavioral data. In 2015 International Conference on Affective Computing and Intelligent Interaction (ACII) (pp. 222–228). IEEE. https://doi.org/10.1109/ACII.2015.7344575
Kankaraš, M., & Suarez-Alvarez, J. (2019). Assessment framework of the OECD Study on Social and Emotional Skills. https://doi.org/10.1787/5007adef-en
Kern, M. L., Waters, L. E., Adler, A., & White, M. A. (2015). A multidimensional approach to measuring well-being in students: Application of the PERMA framework. The Journal of Positive Psychology, 10(3), 262–271. https://doi.org/10.1080/17439760.2014.936962
Kiuru, N., Wang, M.-T., Salmela-Aro, K., Kannas, L., Ahonen, T., & Hirvonen, R. (2020). Associations between Adolescents’ Interpersonal Relationships, School Well-being, and Academic Achievement during Educational Transitions. Journal of Youth and Adolescence, 49(5), 1057–1072. https://doi.org/10.1007/s10964-019-01184-y
Konu, A., & Lintonen, T. P. (2006). School well-being in Grades 4-12. Health Education Research, 21(5), 633–642. https://doi.org/10.1093/her/cyl032
Konu, A., Lintonen, T., & Rimpelä, M. (2002). Factor structure of the School Well-being Model. Health Education Research, 17(6), 732–742. https://doi.org/10.1093/her/17.6.732
Konu, A., & Rimpelä, M. (2002). Well-being in schools: a conceptual model. Health Promotion International, 17(1), 79–87. https://doi.org/10.1093/heapro/17.1.79
Krueger, A. B., & Stone, A. A. (2014). Progress in measuring subjective well-being. Science, 346(6205), 42–43. https://doi.org/10.1126/science.1256392
Kylväjä, M., Kumpulainen, P., & Konu, A. (2019). Application of data clustering for automated feedback generation about student well-being. In Proceedings of the 1st ACM SIGSOFT International Workshop on Education through Advanced Software Engineering and Artificial Intelligence - EASEAI 2019 (pp. 21–26). New York, New York, USA: ACM Press. https://doi.org/10.1145/3340435.3342720
Lyubomirsky, S., King, L., & Diener, E. (2005). The Benefits of Frequent Positive Affect: Does Happiness Lead to Success? Psychological Bulletin, 131(6), 803–855. https://doi.org/10.1037/0033-2909.131.6.803
MyBPLab. (2021). MyBPLab APP. https://mybplab.com/. Accessed 9 October 2021
OECD. (2013). OECD Guidelines on Measuring Subjective Well-being. OECD. https://doi.org/10.1787/9789264191655-en
OECD. (2015). Skills for Social Progress: The Power of Social and Emotional Skills. OECD Publishing. https://doi.org/10.1787/9789264226159-en
OECD. (2019). PISA 2018 results (Volume III): What school life means for students’ lives. OECD. https://doi.org/10.1787/acd78851-en
OECD. (2021). Beyond Academic Learning: First Results from the Survey of Social and Emotional Skills. Paris: OECD. https://doi.org/10.1787/92a11084-en
Oishi, S., Diener, E., & Lucas, R. E. (2007). The Optimum Level of Well-Being: Can People Be Too Happy? Perspectives on Psychological Science, 2(4), 346–360. https://doi.org/10.1111/j.1745-6916.2007.00048.x
Ong, Z. X., Dowthwaite, L., Perez Vallejos, E., Rawsthorne, M., & Long, Y. (2021). Measuring Online Wellbeing: A Scoping Review of Subjective Wellbeing Measures. Frontiers in Psychology, 12. https://doi.org/10.3389/fpsyg.2021.616637
Pavot, W., & Diener, E. (1993). Review of the Satisfaction With Life Scale. Psychological Assessment, 5(2), 164–172. https://doi.org/10.1037/1040-35220.127.116.11
Romano, L., Tang, X., Hietajärvi, L., Salmela-Aro, K., & Fiorilli, C. (2020). Students’ Trait Emotional Intelligence and Perceived Teacher Emotional Support in Preventing Burnout: The Moderating Role of Academic Anxiety. International Journal of Environmental Research and Public Health, 17(13), 4771. https://doi.org/10.3390/ijerph17134771
Rose, T., Joe, S., Williams, A., Harris, R., Betz, G., & Stewart-Brown, S. (2017). Measuring Mental Wellbeing Among Adolescents: A Systematic Review of Instruments. Journal of Child and Family Studies, 26(9), 2349–2362. https://doi.org/10.1007/s10826-017-0754-0
Ryff, C. D. (1989). Happiness is everything, or is it? Explorations on the meaning of psychological well-being. Journal of personality and social psychology, 57(6), 1069.
Ryff, C. D., & Singer, B. H. (2008). Know Thyself and Become What You Are: A Eudaimonic Approach to Psychological Well-Being. Journal of Happiness Studies, 9(1), 13–39. https://doi.org/10.1007/s10902-006-9019-0
Salmela-Aro, K. (2020). The Role of Motivation and Academic Wellbeing – the Transition from Secondary to Further Education in STEM in Finland. European Review, 28(S1), S121–S134. https://doi.org/10.1017/S1062798720000952
Salmela-Aro, K., Kiuru, N., Pietikäinen, M., & Jokela, J. (2008). Does School Matter? The role of school context in adolescents’ school-related burnout. European Psychologist, 13(1), 12–23. https://doi.org/10.1027/1016-9040.13.1.12
Salmela-Aro, K., Tang, X., & Upadyaya, K. (in press). Study Demands-Resources Model of Student Engagement and Burnout. In Reschly, A. L. & Christenson, S. (eds.), Handbook of Research on Student Engagement, second edition. Springer. https://doi.org/10.1007/978-3-031-07853-8_4
Salmela-Aro, K., & Upadyaya, K. (2014). School burnout and engagement in the context of demands-resources model. British Journal of Educational Psychology, 84(1), 137–151. https://doi.org/10.1111/bjep.12018
Salmela-Aro, K., & Upadyaya, K. (2020). School engagement and school burnout profiles during high school – The role of socio-emotional skills. European Journal of Developmental Psychology, 17(6), 1–22. https://doi.org/10.1080/17405629.2020.1785860
Salmela-Aro, K., Upadyaya, K., Vinni-Laakso, J., & Hietajärvi, L. (2021). Adolescents’ Longitudinal School Engagement and Burnout Before and During COVID-19—The Role of Socio-Emotional Skills. Journal of Research on Adolescence, 31(3), 796–807. https://doi.org/10.1111/jora.12654
Seligman, M. E. P. (2011). Flourish. New York, NY: Simon & Schuster.
Tang, X., Upadyaya, K., & Salmela-Aro, K. (2021). School burnout and psychosocial problems among adolescents: Grit as a resilience factor. Journal of Adolescence, 86, 77–89. https://doi.org/10.1016/j.adolescence.2020.12.002
Tang, X., Wang, M.-T., Guo, J., & Salmela-Aro, K. (2019). Building Grit: The Longitudinal Pathways between Mindset, Commitment, Grit, and Academic Outcomes. Journal of Youth and Adolescence, 48(5), 850–863. https://doi.org/10.1007/s10964-019-00998-0
Taylor, S., Jaques, N., Nosakhare, E., Sano, A., & Picard, R. (2020). Personalized Multitask Learning for Predicting Tomorrow’s Mood, Stress, and Health. IEEE Transactions on Affective Computing, 11(2), 200–213. https://doi.org/10.1109/TAFFC.2017.2784832
VanderWeele, T. J., Trudel-Fitzgerald, C., Allin, P., Farrelly, C., Fletcher, G., Frederick, D. E., et al. (2020). Current recommendations on the selection of measures for well-being. Preventive Medicine, 133, 106004. https://doi.org/10.1016/j.ypmed.2020.106004
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063–1070. https://doi.org/10.1037/0022-3518.104.22.1683
Zee, M., & Koomen, H. M. Y. (2016). Teacher Self-Efficacy and Its Effects on Classroom Processes, Student Academic Adjustment, and Teacher Well-Being. Review of Educational Research, 86(4), 981–1015. https://doi.org/10.3102/0034654315626801
The study has been supported by the Academy of Finland Grants 1336138, 308351 and 345117, Strategic Research Council 345264, which were awarded to Katariina Salmela-Aro. The study has been supported by Business Finland, AI in learning project.
Mika Kasanen holds the shares of School Day Oy. None of the other authors (i.e., XT, KU, HT, KSA) received the financial support from School Day Oy.
Editors and Affiliations
Rights and permissions
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
© 2023 The Author(s)
About this chapter
Cite this chapter
Tang, X., Upadyaya, K., Toyama, H., Kasanen, M., Salmela-Aro, K. (2023). Assessing and Tracking Students’ Wellbeing Through an Automated Scoring System: School Day Wellbeing Model. In: Niemi, H., Pea, R.D., Lu, Y. (eds) AI in Learning: Designing the Future. Springer, Cham. https://doi.org/10.1007/978-3-031-09687-7_4
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-09686-0
Online ISBN: 978-3-031-09687-7
eBook Packages: Behavioral Science and PsychologyBehavioral Science and Psychology (R0)