Empirical evidence contained in open digital badges has the capability to change educational curricula, assessments, and priorities. Because badge data in educational, social media, and workforce contexts is publicly available, questions of privacy and ethics should be scrutinized. Due to change driven by digital transparency, ethical questions at the intersection of learning analytics and the data contained in badges poses three distinct, yet related questions: within learning analytics systems, can the use of educational data in digital badges be used in a predictive manner to create a deterministic future for individual learners? Can badge data that is freely and openly accessible in social media be used against individuals if it exposes intellectual weaknesses? And, can the student data in badges be isolated to exploit particular skills for nefarious reasons, i.e. surveillance or hacking? These questions address ethical principles of human autonomy, freedom, and determinism.
The possible educational data contained in badges,
including assessments, validation, and demonstration of
skills, is of value as one aspect of a wider and growing
body of research in learning analytics [
Within learning analytics systems, can the use of educational data in digital badges be used in a predictive manner to create a deterministic future for individual learners? Can badge data that is freely and openly accessible in social media be used against individuals if it exposes intellectual weaknesses? Can the student data in badges be isolated to exploit particular skills for nefarious reasons, i.e. surveillance or hacking?
With the expansion of educational technology, some work is being done at the intersection of ethical theory and learning analytics [6]. Some propose ethical frameworks for development [7], while others appeal specifically to known problems in the legal use of student data in analytics systems [8]. This is a growing domain of research in the broader implications of how educational technology affects student growth and development.
Though some commentators have noted the potentiallyharmful aspects of having open data in social networks, to date there has been scant studies of ethical issues in open digital badges [9]. As research in the ethics of educational technology expands, a myriad of potential problems looms [10]. To bridge this gap, targeted ethical questions must be specific enough to demonstrate applicability, but also be generalizable enough to warrant attention outside of the broadly-considered technology.
Due to their connection with social networks, availability of meta-data, and transparency of learning, digital badges face important ethical questions best formulated within the larger context of learning analytics.
Learning analytics systems are becoming increasingly
powerful tools to help students utilize educational data to
achieve academic success [11]. Such data points are
everprecise, refined by multiple cohorts, experiments, and
digital developments. One of the drivers of such
development is statistical regression which helps predict
when students need help [12], what classes might be
beneficial in customized sequences [13], and how certain
interventions benefit different types of students [14]. The
robustness of learning analytics systems are built with data
refined over time and with evidence of successful student
outcomes [
Open digital badges contain multiple points of valuable educational data including assessments, specific skills development, and validation amongst others. Built within learning analytics systems, the evidence presented in badges can help detail a student’s educational strengths and weaknesses. Further, the data available in well-designed badge ecosystems could strengthen a learning analytics’ predictive strength. For example, with the completion of several badges in both curricular and extra-curricular activities, data points could be amalgamated to further bolster a student’s educational strengths and support possible weaknesses. The specificity of assessment data, where final grades in college courses could be aligned with performance data in badges, could provide extremely valuable information not only to the learner, but also the institution supporting the learner, as well as the businesses developing learning analytics systems.
As analytics systems increase their capability of predicting student outcomes, it may be difficult to distinguish between the strength of the predictive algorithm and the role of determinism as it affects students. This is not to say that the same would be true of digital badges, though. In this instance, determinism would entail the ability to either sequentially offer badges to purposely build a set of skills in a learner without his/her explicit knowledge or to suggest that a self-fulfilling prophecy of ability would be set forth with badge data. A student’s interaction with content leading to a badge may be examined for motivational aspects, perhaps even for so-called gamification reasons. However, could the data contained with badges be used to constrain a determined future? Meaning, if students are directly motivated to achieve certain badges, their interests may be piqued either with the content or with simply obtaining a badge; the question, then, is what effect the badge may have on future educational choices. Today’s use of digital badges is often to enhance learning and provide open and transparent evidence of learning. It is impossible to say if this will continue, and what possible effect badges may have on learners’ educational choices. A determined future, one shaped by an algorithm targeting content and ability, may not be suitable to human freedom and autonomy. Important to learning is the chance – and the eventuality – of failure. Learning and carrying on from failure is a hallmark of resilient students who become as selfactualized as possible. Badges can certainly help mitigate educational failure because they are used to supplement through micro-credentialing, build on skills to be successful in other educational contexts, and provide a lasting record of one’s accomplishments. Yet, human freedom and autonomy must be examined in light of these technological developments. If learning analytics in the systemic sense and digital badges in the individual sense are able to absolutely minimize failure, could learners suffer from not having to form resilience? The possibility of minimizing failure and preventing failure is determinism for individuals, though the data available in digital badges poses numerous ethical questions related to public disclosure of student data. Learning analytics systems used at schools are controlled by regulations (like FERPA), and are thus “closed” systems, whereas open badges make educational and learning data public.
One of the benefits of open digital badges is that they can
publically demonstrate a set of skills; downloadable into
various social media, badges can evidence real learning
[
Questions about what data points are useful in social media contexts are open for debate, but generally arguments seem to tentatively hinge in favoring the position of the individual learner’s choice to expose or withhold information. This matter becomes an issue of privacy within digital badge ecosystems, however, when students may or may not fully comprehend the possible outcomes of publishing particular data points. Further, if such algorithms can threaten learning weaknesses in individual learners, the same may be true for the entire badge ecosystems because a badge’s validity, no matter the expertise involved in the content development, deployment, and verification, could be threatened. The possibilities highlighted within learning analytics systems and in individual learning may well point to more sinister uses of data.
The increasing specificity of learning data within digital badges may lead to nefarious uses of that data in aggregate or individual cases. The use of predictive models in learning analytics coupled with evidence of skills development in digital badges disseminated across social media could help companies, governments, and perhaps even disreputable organizations recruit for nefarious purposes like hacking or surveillance. Furthermore, if learners are completing badges “recommended” to them for the ulterior motives of developing certain skills, could such information be used against them if they later refuse to participate in questionable activities? This may sound rather extreme, but when educational data is used across social media and within predictive systems, it is impossible to state how such data might be used. Additionally, while this may not come to fruition, the logical extremity is useful to examine the possible uses of future data. Ethically, the question of learners knowingly or unknowingly participating in skills development for nefarious purposes is a question of human freedom, and ultimately it rests with the actions of the learner. However, the possibility of future manipulation certainly conflicts with how popular notions of badges today include that of supplementing learning, offering opportunity for job creation and advancement, and branching learning into social spheres.
Open digital badges are changing the educational opportunities for learners of today and tomorrow. The intersection of learning analytics systems that may incorporate badges, as well as the possibility of learners to disseminate evidence of learning across social media platforms, creates unique ethical questions that fit within the larger discussion of machine learning. The uses of data for prediction, isolation of strengths and weaknesses, and potential manipulation have direct consequences for questions of human autonomy, freedom, and determinism. Such scenarios described herein may appear to be exceedingly negative or dystopic to some readers. While true that some of the scenarios may or may not come to fruition, they function here not only as a thought experiment, but also as a model of what may go awry as increasing amounts of personalized data are distributed, shared, and examined online. It is also a model for how data crawlers may use student data with present technologies. While not a warning in the formal sense, such thought experiments are useful to describe how the ethics of technology, broadly understood, ought to enter into development discussions.
Like other ethical discussions concerning digital learning and educational technology, solutions cannot be prescriptive. The speed of development may render such prescriptions neutral. Similarly, ethical discussions cannot be reactive, either. Once technology exists, especially for public use, it cannot be retracted; thus, ethical discussions belong in the foreground of development, if even for thought experiments.
As an ethics of educational technology pivots between the prescriptive and reactive, so too they must occupy this third space of ethical reflection. Working through the potential outcomes of development does not stymie innovation, but rather acts as a partner to innovate the autonomy of the individual, the role of the community, and the potential intersection of the two. This third space, then, is occupied with lists of potential questions, perhaps operating somewhere between utopic and dystopic, that drive purposeful and responsible innovation. Exploring the ethical implications at stake is often enough not only to aid in responsible innovation, but also to avoid potential abuses of such systems. Digital badges are no different: if they are to become ubiquitous in the digital age as key components and measures of learning and as effective credentials, then ethical innovation and use of data is ever-more important. [7] Willis, III, J.E. 2014. Learning analytics and ethics: A framework beyond utilitarianism. EDUCAUSE Review Online. [8] Pardo, A. and Siemens, G. 2014. Ethical and privacy principles for learning analytics. British Journal of Educational Technology, 45, 3, 438-450.
DOI=10.1111/bjet.12152. [9] Halavais, A.M.C. 2012. A genealogy of badges.
Information, Communication & Society. 15, 3, 354373, DOI= 10.1080/1369118X.2011.641992. [10] Introna, L.D. 2007. Maintaining the reversibility of foldings: Making the ethics (politics) of information technology visible. Ethics and Information Technology, 9, 11-25. DOI=10.1007/s10676-0069133-z. [11] Siemens, G. and Long, P. 2011. Penetrating the fog: Analytics in learning and education. Educause Review. 46, 5. 30–32. [12] Baker, R. and Siemens, G. 2014. Educational data mining and learning analytics. in Sawyer, R.K. ed. Cambridge Handbook of the Learning Sciences. Cambridge University Press, New York, 2014, 353372. [13] Denley, T. 2014. How predictive analytics and choice architecture can improve student success. Research & Practice in Assessment. 9, 2. 61-69. [14] Smith, V.C., Lange, A., Huston, D.R. 2012. Predictive modeling to forecast student outcomes and drive effective interventions in online community college courses. Journal of Asynchronous Learning Networks. 16, 3, 51-61.