Sandberg et al. (2014) – The added value of gaming context and intelligent adaption for a mobile learning application for vocabulary learning Mobiles allow learners the freedom to learn where
and whenever they want. The learning content can be adapted to the location of the learner, enriching the learner experience. MEL2 explores the added value of intelligent
adaption combined with game elements. Public education faces the challenge of incorporating the digital competence of youngsters into the school curriculum, with more
informal ways of learning outside schools → serious games could offer support in dealing with these challenges. Game characteristics: ‘Fantasy’ is related to the group concepts
of immersion and engagement, ‘Rules and goals’ are related to challenge and underlying learning objectives, ‘Sensory stimuli’ are needed to fully engage the learner, ‘Challenge’
refers to the level of difficulty of the goals set of the learner, ‘Control’ addresses the learner's influence on the flow of the game, ‘Mystery’ refers to introducing uncertainty and
surprise to a game, ‘Rewards’ in games are often scoring points, winning medals etc., ‘Competition/cooperation’ refers to possible social interaction the player may have during
a game. Student modelling: refers to the representation of the developing knowledge and skill of a student. Contextual learning: where the learner is expected to infer the
meaning of new words from the context, better facilitating the transfer of acquired vocabulary to new contexts. Requires a higher mental effort than translation learning.
METHOD: Participants: 51 MEL-original; 55 MEL-enhanced. Study based on a quasi-experimental pre- and post- test design. Themes: Zoo animals & neighbourhood. MEL original;
application built around 25 animals, distributed over 5 continents; multiple choice quiz, yes or no quiz, picture picker quiz, spelling quiz, jigsaw puzzle → learning the target
words (goal). MEL-enhanced; They wanted to know whether a mobile game could enhance performance on the skill tests in English proficiency, they compared the new
application with MEL-original. CONCLUSIONS: Children who worked with MEL-enhanced progressed significantly more from pre-test to post-test on the active items. They didn’t
spend significantly more time working with the learning material. They learned significantly more, but do not appear to be more motivated as measured in terms of time spent
learning. Even though the children who used MEL-enhanced didn't work longer with the material, the gaming aspect may have contributed to their concentration and effort to
understand and retain the material presented. They conclude that the gaming aspect doesn't motivate the children to spend more time on the learning material.
Wouters et al. (2013) - A meta-analysis of the cognitive and motivational effects of serious games (SG) Goal: to statistically summarise the research on the effects of SG on learning and
motivation. Their meta-analysis adopted the media comparison approach, which investigates whether people learn better from SG than from conventional media ( lectures,
reading etc.) Computer games: being interactive, based on a set of agreed rules and constraints and directed toward a clear goal that is often set by a challenge. Games may
influence learning in two ways; by changing the cognitive processes and by affecting the motivation. The most important factors that make playing a computer game intrinsically
motivating are challenge, curiosity and fantasy. They investigate how situational/contextual variables may moderate learning with SG: passive vs active instruction, SG combined
with other instructional methods, Number of training sessions, Group size. Other investigated variables; instructional domain, age, level of realism, narrative. METHOD:
Knowledge was used when a test involved knowledge of concepts, principles, definitions, symbols or facts. Studies in which the learners had to solve problems, make decisions,
or apply rules to a situation where coded as cognitive skills. Retention was coded when a delayed measure for learning was available. Questionnaire to measure motivation.
Comparison group: active instruction refers to instruction methods that explicitly prompt learners to learning activities (exercises, hypertext training) passive instruction includes
listening to lectures, reading textbooks etc. 39 studies were identified. CONCLUSIONS: Results on knowledge/cognitive skills suggest that training with SG is more effective than
training with conventional instruction methods. The retention outcome shows that the cognitive gains persist in the long term. Supports what teachers/instructors deem
important: that SG lead to well-structured prior knowledge on which learners can build on during their learning career. SG are not more effective than passive instruction. SG are
more effective when supplemented with other instructional methods. SG are more effective when played in groups. More basic designs (schematical/textual/cartoonlike) are
equally/more effective. SG with a narrative are not more effective, can cause too much use of the cognitive capacity for processing narrative info that isn’t directly related to the
learning content.SG aren't more motivating than the instructional methods. Conditions that limit the sense of control or freedom of action may undermine intrinsic motivation.
It's relevant to investigate whether variations in the level of control that SG offer moderate intrinsic motivation. The question can be raised whether it makes sense to measure
affective states such as motivation and enjoyment with questionnaires and surveys after gameplay (perhaps eye tracking/skin conductance is better).
Alsonso-Fernandez et al. (2019) – Lessons learned applying learning analytics to assess serious games New techniques (Learning Analytics (LA)) are trying to provide insight about the
educational processes and improve the common educational scenarios benefitting from data-driven approaches. LA aims to measure, collect, analyse and report data from
learning to extract useful information about how students learn, with the purpose of understanding and optimising their learning processes and contexts. LA techniques can be
applied to game environments where their interactive nature is adequate to the data capturing process. They aim to fill the gap by providing data-based evidence of the possible
applications of Game Learning Analytics (GLA) data for SG. Conectado → tool to be used by teachers in classroom to address (cyber)bullying. GLA was used to improve validation
and deployment in schools. Aims to raise awareness + promote empathy with victims of (cyber)bullying. (12-17 years). Statistically significant difference in cyberbullying
awareness. Second version, GLA data showed that younger players required more time to complete the game. SG Downtown → designed for promoting independence in users
with intellectual disabilities (age 18-45). Train them to use the subway in Madrid. GLA illustrates how to validate a game design in situations where information can’t be directly
gathered from the users. Aims to train skills. Errors decreased over time, as did inactivity time. GLA allowed developers to correct an error. First Aid Game → designed to teach
first-aid manoeuvres to teenagers (12-17 years). GLA focuses on improving the evaluation and deployment of games by applying data mining models to predict students'
knowledge after playing based on interaction data, proving that games can help to accurately assess students' knowledge. Aims to improve students' knowledge. Game-like
simulation with different scenarios (chest pain, choking, unconsciousness). xAPI-SG profile, a standard collection model for tracking interaction data for SG. Game analytics
provide information about the appropriateness of the game design and its mechanics. GLA combines LA+GA in order to understand how students learn using games and to
validate the actual educational and game designs. GLA have the potential to shorten development time while improving their impact, thus increasing their use. CONCLUSIONS:
Statistically significant difference in learning (pre- post-test). Constructed prediction models based on GLA data: used to predict post-test scores with Exact scores and Pass/Fail.
Could be possible to eliminate pre- post-tests and simply rely on game performance to assess learning, because GLA had highly accurate results. GLA may simplify the
deployment of SG and the evaluation of students + lower costs. GLA can be useful to: Iteratively improve the design of the learning game (e.g. gameplay time, level difficulty),
Understand, in real time, how students are using the game, Better understand students' learning
in-game (with the potential to avoid out of game assessments). Standardised GLA has simplified
collection as they could easily define and match the interactions to be captured for each game
with the specific verbs and activity types of the xAPI-SG profile.
Chen. (2020) – A visual learning analytics (VLA) approach to video-based teacher professional development:
impact on teachers’ beliefs, self-efficacy and classroom talk practice Visual learning analytics (VLA)
combines LA + VA in order to understand educational problems and support educational decision
making. VLA: the employment of visual analytics for improving educational decision-making and
emphasises the combination of visual and automated analysis for solving complex educational problems. Teachers often
find it challenging to integrate effective talk into classroom practices. Videos can be useful in helping teachers reflect on
their teaching practices. Problem is the sheer amount of info contained in a video (potential overload). We conceptualise
an alternative solution, arguing that this problem can be mitigated through VLA approach to supporting video-based
teacher learning and Teacher Professional Development (TPD). Academically Productive Talk (APT) is a whole-class
discussion framework concerning how learning and cognition take place in the contexts of productive and meaningful talk,
tasks and tools through social interactions. There’s an increase in the videos to facilitate teacher reflection in TPD
programs. METHOD: Classroom Discourse Analyzer 2.0 (CDA) (web-based): allows teachers to visualise video data of their
teaching, provides different views, visual information seeking mantra: "overview first, zoom and filter, then details-on
demand", automatic extraction of low-inference discourse info. CDA leverages data analytics and information visualisation
techniques to provide teachers with 1) data access 2) data navigation 3) evidence extraction of teacher-students classroom
discourse. CONCLUSIONS: Significant changes in teacher beliefs about the effects of classroom talk and self-efficacy across
pre-, post, and delayed-post-tests for all teachers. Only teachers in the treatment group sustained the change to the
delayed post-test (for belief and self-efficacy).We found that, relative to the control teachers, the treatment teachers
significantly improved in their use of APT strategies in the classroom for facilitating students' elaboration, reasoning, and thinking with others. A visual learning analytics tool
(CDA 2.0) to support video-based teacher professional development is effective: It leads to sustained changes in teacher beliefs and self-efficacy, It leads to the greater use of
effective talk in the classroom.
and whenever they want. The learning content can be adapted to the location of the learner, enriching the learner experience. MEL2 explores the added value of intelligent
adaption combined with game elements. Public education faces the challenge of incorporating the digital competence of youngsters into the school curriculum, with more
informal ways of learning outside schools → serious games could offer support in dealing with these challenges. Game characteristics: ‘Fantasy’ is related to the group concepts
of immersion and engagement, ‘Rules and goals’ are related to challenge and underlying learning objectives, ‘Sensory stimuli’ are needed to fully engage the learner, ‘Challenge’
refers to the level of difficulty of the goals set of the learner, ‘Control’ addresses the learner's influence on the flow of the game, ‘Mystery’ refers to introducing uncertainty and
surprise to a game, ‘Rewards’ in games are often scoring points, winning medals etc., ‘Competition/cooperation’ refers to possible social interaction the player may have during
a game. Student modelling: refers to the representation of the developing knowledge and skill of a student. Contextual learning: where the learner is expected to infer the
meaning of new words from the context, better facilitating the transfer of acquired vocabulary to new contexts. Requires a higher mental effort than translation learning.
METHOD: Participants: 51 MEL-original; 55 MEL-enhanced. Study based on a quasi-experimental pre- and post- test design. Themes: Zoo animals & neighbourhood. MEL original;
application built around 25 animals, distributed over 5 continents; multiple choice quiz, yes or no quiz, picture picker quiz, spelling quiz, jigsaw puzzle → learning the target
words (goal). MEL-enhanced; They wanted to know whether a mobile game could enhance performance on the skill tests in English proficiency, they compared the new
application with MEL-original. CONCLUSIONS: Children who worked with MEL-enhanced progressed significantly more from pre-test to post-test on the active items. They didn’t
spend significantly more time working with the learning material. They learned significantly more, but do not appear to be more motivated as measured in terms of time spent
learning. Even though the children who used MEL-enhanced didn't work longer with the material, the gaming aspect may have contributed to their concentration and effort to
understand and retain the material presented. They conclude that the gaming aspect doesn't motivate the children to spend more time on the learning material.
Wouters et al. (2013) - A meta-analysis of the cognitive and motivational effects of serious games (SG) Goal: to statistically summarise the research on the effects of SG on learning and
motivation. Their meta-analysis adopted the media comparison approach, which investigates whether people learn better from SG than from conventional media ( lectures,
reading etc.) Computer games: being interactive, based on a set of agreed rules and constraints and directed toward a clear goal that is often set by a challenge. Games may
influence learning in two ways; by changing the cognitive processes and by affecting the motivation. The most important factors that make playing a computer game intrinsically
motivating are challenge, curiosity and fantasy. They investigate how situational/contextual variables may moderate learning with SG: passive vs active instruction, SG combined
with other instructional methods, Number of training sessions, Group size. Other investigated variables; instructional domain, age, level of realism, narrative. METHOD:
Knowledge was used when a test involved knowledge of concepts, principles, definitions, symbols or facts. Studies in which the learners had to solve problems, make decisions,
or apply rules to a situation where coded as cognitive skills. Retention was coded when a delayed measure for learning was available. Questionnaire to measure motivation.
Comparison group: active instruction refers to instruction methods that explicitly prompt learners to learning activities (exercises, hypertext training) passive instruction includes
listening to lectures, reading textbooks etc. 39 studies were identified. CONCLUSIONS: Results on knowledge/cognitive skills suggest that training with SG is more effective than
training with conventional instruction methods. The retention outcome shows that the cognitive gains persist in the long term. Supports what teachers/instructors deem
important: that SG lead to well-structured prior knowledge on which learners can build on during their learning career. SG are not more effective than passive instruction. SG are
more effective when supplemented with other instructional methods. SG are more effective when played in groups. More basic designs (schematical/textual/cartoonlike) are
equally/more effective. SG with a narrative are not more effective, can cause too much use of the cognitive capacity for processing narrative info that isn’t directly related to the
learning content.SG aren't more motivating than the instructional methods. Conditions that limit the sense of control or freedom of action may undermine intrinsic motivation.
It's relevant to investigate whether variations in the level of control that SG offer moderate intrinsic motivation. The question can be raised whether it makes sense to measure
affective states such as motivation and enjoyment with questionnaires and surveys after gameplay (perhaps eye tracking/skin conductance is better).
Alsonso-Fernandez et al. (2019) – Lessons learned applying learning analytics to assess serious games New techniques (Learning Analytics (LA)) are trying to provide insight about the
educational processes and improve the common educational scenarios benefitting from data-driven approaches. LA aims to measure, collect, analyse and report data from
learning to extract useful information about how students learn, with the purpose of understanding and optimising their learning processes and contexts. LA techniques can be
applied to game environments where their interactive nature is adequate to the data capturing process. They aim to fill the gap by providing data-based evidence of the possible
applications of Game Learning Analytics (GLA) data for SG. Conectado → tool to be used by teachers in classroom to address (cyber)bullying. GLA was used to improve validation
and deployment in schools. Aims to raise awareness + promote empathy with victims of (cyber)bullying. (12-17 years). Statistically significant difference in cyberbullying
awareness. Second version, GLA data showed that younger players required more time to complete the game. SG Downtown → designed for promoting independence in users
with intellectual disabilities (age 18-45). Train them to use the subway in Madrid. GLA illustrates how to validate a game design in situations where information can’t be directly
gathered from the users. Aims to train skills. Errors decreased over time, as did inactivity time. GLA allowed developers to correct an error. First Aid Game → designed to teach
first-aid manoeuvres to teenagers (12-17 years). GLA focuses on improving the evaluation and deployment of games by applying data mining models to predict students'
knowledge after playing based on interaction data, proving that games can help to accurately assess students' knowledge. Aims to improve students' knowledge. Game-like
simulation with different scenarios (chest pain, choking, unconsciousness). xAPI-SG profile, a standard collection model for tracking interaction data for SG. Game analytics
provide information about the appropriateness of the game design and its mechanics. GLA combines LA+GA in order to understand how students learn using games and to
validate the actual educational and game designs. GLA have the potential to shorten development time while improving their impact, thus increasing their use. CONCLUSIONS:
Statistically significant difference in learning (pre- post-test). Constructed prediction models based on GLA data: used to predict post-test scores with Exact scores and Pass/Fail.
Could be possible to eliminate pre- post-tests and simply rely on game performance to assess learning, because GLA had highly accurate results. GLA may simplify the
deployment of SG and the evaluation of students + lower costs. GLA can be useful to: Iteratively improve the design of the learning game (e.g. gameplay time, level difficulty),
Understand, in real time, how students are using the game, Better understand students' learning
in-game (with the potential to avoid out of game assessments). Standardised GLA has simplified
collection as they could easily define and match the interactions to be captured for each game
with the specific verbs and activity types of the xAPI-SG profile.
Chen. (2020) – A visual learning analytics (VLA) approach to video-based teacher professional development:
impact on teachers’ beliefs, self-efficacy and classroom talk practice Visual learning analytics (VLA)
combines LA + VA in order to understand educational problems and support educational decision
making. VLA: the employment of visual analytics for improving educational decision-making and
emphasises the combination of visual and automated analysis for solving complex educational problems. Teachers often
find it challenging to integrate effective talk into classroom practices. Videos can be useful in helping teachers reflect on
their teaching practices. Problem is the sheer amount of info contained in a video (potential overload). We conceptualise
an alternative solution, arguing that this problem can be mitigated through VLA approach to supporting video-based
teacher learning and Teacher Professional Development (TPD). Academically Productive Talk (APT) is a whole-class
discussion framework concerning how learning and cognition take place in the contexts of productive and meaningful talk,
tasks and tools through social interactions. There’s an increase in the videos to facilitate teacher reflection in TPD
programs. METHOD: Classroom Discourse Analyzer 2.0 (CDA) (web-based): allows teachers to visualise video data of their
teaching, provides different views, visual information seeking mantra: "overview first, zoom and filter, then details-on
demand", automatic extraction of low-inference discourse info. CDA leverages data analytics and information visualisation
techniques to provide teachers with 1) data access 2) data navigation 3) evidence extraction of teacher-students classroom
discourse. CONCLUSIONS: Significant changes in teacher beliefs about the effects of classroom talk and self-efficacy across
pre-, post, and delayed-post-tests for all teachers. Only teachers in the treatment group sustained the change to the
delayed post-test (for belief and self-efficacy).We found that, relative to the control teachers, the treatment teachers
significantly improved in their use of APT strategies in the classroom for facilitating students' elaboration, reasoning, and thinking with others. A visual learning analytics tool
(CDA 2.0) to support video-based teacher professional development is effective: It leads to sustained changes in teacher beliefs and self-efficacy, It leads to the greater use of
effective talk in the classroom.
