Using Skills Profiling to Enable Badges and Micro-Credentials to be Incorporated into Higher Education Courses

Micro-credentials

The use of degree certificates to evidence learning achievements has not changed much over the last 200 years, with the credential’s credibility linked to the providing institution (Brown 2001). The potential for alternative digital credentials, to supplement traditional ones, has been promoted since the rise of the World Wide Web in the 1990s (Noam 1995). Indeed, for a while, Massive Open Online Courses (MOOCs) were the subject of much hype, such as the New York Times declaring 2012 as “The Year of the MOOC” (Pappano 2012). Whilst not meeting the original hype, the impact of MOOCs has been significant, with platforms such as Coursera, edX, FutureLearn, LinkedIn Learning, Udacity, etc., providing courses and resultant micro-credentials to many learners. Such development has not been without challenges in the form of bogus providers and credentials (Ezell & Bear 2005). One solution to this is to use micro-credentials to personalise professional development (Hunt et al. 2020) either alongside or within formal qualifications, especially where completing a university qualification is the foundation of professional formation (Raj et al. 2022) (e.g. Computing, Engineering, Medicine, Nursing, Teaching).

Skills-based hiring

Skills-based hiring is becoming an increasingly popular recruitment approach (Fuller, Langer & Sigelman 2022). Employers still ask for degrees, but increasingly they are specifying subject-specific and transferable skills (Gallagher 2018; Prospects 2021), including 21^st Century Skills. In this work 21^st Century Skills are defined as competencies gained by a learner that enable them to self-reflect, self-regulate and self-optimise their capabilities within highly emergent contexts. This definition is implicitly built on a personalised learning perspective (Barty et al. 2023).

A common theme in the 21^st Century Skills debate is sub-categories representing knowledge, skills and dispositions. For example, CC2020 considers knowledge, skills, and dispositions directly (Clear et al. 2020). ATC21S (2020) considers ten 21^st Century Skills within four main categories (ways of thinking, working and living, plus the tools to work). Other authors present slightly different but similar combinations of categories (Joynes, Rossignoli & Amonoo-Kuofi 2019).

By analysing job advertisements we can better understand these skills demands (Papoutsoglou et al. 2019). Indeed, such a research approach underpins reports by Harvard Business Review (Hershbein & Kahn 2017), Forbes (Bernick 2022) and the National Bureau of Economics (Hershbein & Kahn 2018). One commercial organisation undertaking such analysis is Lightcast (Lightcast 2022), who produce job skills mappings. From these, competency specifications for entry-level computing graduate roles (software engineer, data scientist, etc.) can be compared against skills gained within education.

Credential portability trends in Europe

The European Qualifications Framework (EQF) is a common European reference framework whose purpose is to make qualifications more readable and understandable across different countries and systems. Covering qualifications at all levels and in all sub-systems of education and training, the EQF provides a comprehensive overview over qualifications in the 38 European countries currently involved in its implementation (European Centre for the Development of Vocational Training n.d.). The European Skills, Competences and Occupations (ESCO) classification describes 13,485 skills and 2,942 occupations (ESCO 2022). ESCO aspires to align non-formal and informal learning through open badges and digital credentialing, providing a link between qualifications, skills, micro-credentialing frameworks and employment. Within the European Union, the European Skills Agenda (European Commission Directorate-General for Employment, Social Affairs and Inclusion 2022) highlights twelve actions, including (Action 10) developing a European Approach to Micro-credentials and (Action 11) creating a new Europass Platform (Europass 2022). The Europass platform provides online tools to help manage skills and support learning and career development. In doing so, it promotes transparency and understanding of formal, non-formal and informal qualifications and skills as identified, for example, through ESCO. Building upon this, in 2022, the Council of the European Union adopted a recommendation to develop a European approach to micro-credentials which seeks to support the development, implementation and recognition of micro-credentials across institutions, sectors and borders (European Commission 2022).

The European Commission also recognised that a growing number of European higher education institutions were working with micro-credentials and identified a need for a common definition and approaches for their application (Futures, Anderson & Larson 2020). The Micro-credential Higher Education Consultation Group provided a definition (Antonaci, Henderikx & Ubachs 2021), and micro-credentials linked to the Bologna (Joint declaration of the European Ministers of Education 1999) key commitments project (MICROBOL) specified an approach, requiring micro-credentials to: explicitly reference the European or National Quality Framework Levels; state the learning outcomes achieved; document related workload in terms of European Credit Transfer and Accumulation System (ECTS); and indicate the assessment approach adopted (Antonaci, Henderikx & Ubachs 2021). To implement these expectations, the European MOOC Consortium developed the Common Microcredentials Framework (CMF) (EMC Common Microcredential Framework. N.d.; European MOOC Consortium 2023). Other European Projects have similar aims, for example, MicroHE and OEPASS, European Short Learning Programmes Project (E-SLP) and Digital Credentials collaboration (CORSHIP 2019). Together, these initiatives support a rapidly maturing environment and promote the increased usage of micro-credentials.

Within the United Kingdom specifically, higher education micro-credential use has been debated in the Quality Assurance Agency (QAA) community of practice (QAA 2022a). In 2020, Professor Sue Reece published a discussion paper for QAA members regarding modular qualifications (focus on micro-credentials) and a report (QAA 2021), outlining, amongst other things, the opportunities for micro-credential approaches to enhance the student learning experience.

Micro-credentialing trends in computing and engineering professions

Employer micro-credential support is varied. One recent review suggested, experienced and aspiring computing professionals need to manage their qualifications according to current market needs. That includes certification achievement as well as formal education, experience, and licenses (Tannian & Coston 2021: 56). Amazon has committed $1.2 billion to provide free education and skills for their employees and hundreds of millions to provide free cloud computing-related training to the public (Amazon 2022). Google also recognises micro-credentials within its recruitment processes (Oliver 2020).

Together, this highlights a trend in defining micro-credentials and industry demand in a more systematic manner, recognising and employing credentials in the broadest sense with employment, careers and education/training. This is particularly true within computing and engineering professions, where reskilling and certification requirements are high, and opportunities to enter these professions through non-degree certificate routes are becoming increasingly popular.

In response to employer trends, there have been attempts to harmonise engineering graduate educational outcomes. The Washington, Sydney and Dublin Accords (International Engineering Alliance 2022a, 2022b, 2022c) advocate that engineering programme accreditation is foundational to engineering professional practice. The Seoul Accord extends this to computing (Seoul Accord Secretariat 2022). All have successfully promoted internationalisation of curricula and consistency and parity across higher education (Engineering Council n.d.). As a related example, the European Network for Accreditation of Engineering Education (ENAEE) EUR-ACE Framework Standards and Guidelines are intended to be widely applicable and inclusive so that they can be applied to all branches of engineering. They aim to reflect the diversity of engineering degree programmes which provides the education necessary for graduates to enter the engineering profession and to have their qualifications recognised throughout the area (EUR-ACE Framework Standards and Guidelines 2020). It should be noted that ENAEE does not accredit engineering degree programmes; using the standards, ENAEE evaluates the policies and procedures implemented by accreditation and quality assurance agencies which have applied for authorisation to award the EUR-ACE label to the engineering degree programmes which these agencies accredit.

Accreditation seeks to ensure the development of competent professionals. Competence applies to professions such as medicine, legal or teaching. The ACM and IEEE Computer Society suggest, “Competency = [Knowledge + Skills + Dispositions] in Task” (Clear et al. 2020: 47). In a higher education context, transferable skills have been embedded within degree programs (Kemp & Seagraves 1995) and formalised through transversal competencies, for example, in the computing discipline (Sicilia 2010). Alongside these developments, industry competency models have also developed, such as the Industry Structure Model (Johnson 1997). This digital skills model was refined into the Skills Framework for the Information Age (SFIA) in 2000 (SFIA Foundation 2022). SFIA describes 102 professional skill areas for the computing sector and specifies a range of generic skills. Related work in the UK includes initiatives by the Institute of Coding, a £40m+ strategic investment by the UK Government to “transform the digital skills profile of the country” aiming to “create a new way to develop the digital skills you’ll need at work and beyond” (Davenport et al. 2020). The IoC Accreditation Standard is a novel approach to accreditation of degree courses that focuses on the demonstration of competence in addition to academic knowledge, co-designed with industry to develop a new standard for “digital” graduates (Institute of Coding Accreditation Standard 2020).

In parallel, competency models developed via the Software Engineering Institute of Carnegie Mellon University evolved firstly into the software assurance competency model (Mead & Shoemaker 2013) and then the ACM and IEEE Computing Curricula 2020 Report (CC2020 report) model (Clear et al. 2020). This model presents competency as an individual-centred concept demanding the demonstration of technical knowledge, skills, and dispositions within a task or job context. Dispositions, such as collaboration or communication, parallel SFIA skill level qualities. Indeed, terminology aside, competency evidence is broadly comparable between the ACM/IEEE model and the SFIA model, though the SFIA model also links to specific job roles and expects repeated successful demonstration of competency. This work reflects the SFIA approach.

Data sample

Computing-related programmes at six different universities were used to explore opportunities for badging and micro-credentialing within UK higher education, covering the five approaches advocated by Professor Sue Reece (QAA 2021), as shown in Figure 1.

Figure 1

Modelling approach

Data was collected regarding existing course structures, learning outcomes and assessments. Learning outcomes were translated into 21st Century skills categories (Ward et al. 2021), with Figure 2 showing the six skills themes used to develop skills profiles. Translating in this way enabled learning to be combined differently to offer more flexible, efficient and personalised approaches to assessment, whilst maintaining the quality assurance and integrity of the degree programmes themselves.

Figure 2

The first stage involved calculating learning outcome hours based on total learning hours, assessment components and learning outcomes tested through each assessment. Skills hours were calculated pro-rata from learning hours, assessment weightings and learning outcomes. Two types of skills hours were calculated, subject-specific and transferable. Each was calculated based on the total learning hours of the degree. Subject-specific skills hours were calculated through a one-to-one mapping (learning outcome to subject-specific skills category). To illustrate this, in the example below, a 20 credit module with 200 learning hours is assumed to have two assessments weighted 60% and 40%. Skills hours associated with each assessment are 120 (60% of 200) and 80 (40% of 200). Assessment 1 addresses 4 learning outcomes, so 120/4 = 30 skills hours are allocated against each skill. Assignment 2 addresses 2 learning outcomes, so 80/2 = 40 skills hours are allocated against each skill. Skills hours are then collated (Figure 3).

Figure 3

A similar approach is used for transferable skills (Figure 4), however several transferable skill categories can be mapped to a single learning outcome. Skills hours allocation, for example, to learning outcome 4, is therefore split equally across T9D, T10D and T11D. Each of the three skills are therefore allocated 10 hours. Skills hours are then totalled to provide the profile.

Figure 4

Labour market analytics

Job analysis was undertaken using labour market analytics (Lightcast 2022). Relevant advertisements were searched for using degree specialism and skills quantified based on occurrence frequency. Skills were then collated and coded against skills categories and compared to courses by using the percentage distributions of skills by course and jobs. This approach identifies the “skills gap” between what employers seek and degree programmes offer.

Analysis approach

Results were analysed employing narrative analysis (Andrews, Squire & Tamboukou 2013), focusing on Big Stories i.e. more significant issues (Freeman 2011) since the modelling approach used is approximate and therefore minor differences may be the consequence of the modelling approach. Each participating institution compared results in their own context to develop their narrative. For example, Ulster used the mapping outcomes to help guide students in selecting activities contributing towards an extra-curricular award and how this will complement their formal learning. This facilitates micro-credentials being embedded within the curriculum in a non-credit-bearing way (Model D in Figure 1).

Incorporating non-formal learning through the EDGE Award (University of Ulster, Model A)

Incorporating non-formal learning through iDEA digital badges (Manchester Metropolitan University, MMU, Model A)

Reflecting on intended and specified learning outcomes (Abertay University, Model B)

Incorporating micro-credentials for credit (Northumbria University, Model C)

Reflecting on learning outcome definitions and institutional approaches (University of Bath, Model D)

Course alignment example (University of Huddersfield, Model E)

Contextual narrative – What was the outcome?

An embedded stackable credit-bearing approach (Figure 5) was applied to BSc Software Engineering, where subject-specific skills were focussed in three key areas – Process and Production (32%), Technical Writing (26%) and Theory (23%). Comparing these to Lightcast data (Figure 6), there is a much higher requirement for Process and Production skills and an increased requirement for Business Requirements and Applications, reflecting, in part, different education and employment expectations regarding breadth and depth, though it is interesting to note a lack of requirements for Innovation, Self-Reflection and Technical Writing despite these being in high demand amongst more rounded graduates (Succi & Canovi 2020). The Lightcast data highlights career entry expectations though later expectations could be different.

Figure 5

Figure 6

Figure 7 shows transferable skills were focussed in four key areas – Information Literacy (18%), Numeracy (14%), Problem Solving (14%) and Leadership (30%) all contributing significantly to the overall skills profile. Here the pattern is somewhat different to the labour market information, as shown in Figure 8. Leadership, Problem Solving and Information Literacy are all more developed within the programme than is sought in job roles (though if we combine leadership and management the first of these becomes much less different). However, there are large numbers of transferable skills sought by employers that are not clearly developed within the courses including, in particular, Social Learning, Sustainability and Communication. Similar comparisons were made across each institution and course under consideration. They demonstrated reasonably good agreement between programmes and labour market needs and where variation occurs it is typically indicative of differences in priorities between capabilities (within degree programmes) and competencies (within job roles).

Figure 7

Figure 8

Personalised learning example (University of Huddersfield, Model E)

Higher Education Institutions

Reflecting on learning outcome definitions and institutional approaches, it is clear that there were different approaches between universities in how learning outcomes are expressed. Learning outcomes tend to focus on

• what is taught?

• what is to be learnt?

• how will they subsequently be applied? (although this is very rare)

Which focus has been adopted may help explain the confusion amongst employers. There appear to be two common challenges: learning outcomes descriptions and constraints. Policies can compound this, as some providers limit the number of learning outcomes and therefore generate multi-faceted learning outcomes with several embedded competencies. Assessment drives the language used in many learning outcomes; thus, items that are straightforward to evidence via assessment can predominate. This may mean critical learning, that is more challenging to assess directly, is overlooked. Clear et al. (2020) suggest a number of dispositions which are relevant characteristics to develop within computing education but are not commonly assessed (passion, conviction, responsiveness and respect). The career relevance of such dispositions is apparent; however, the assessment approach is less obvious.

Different approaches to expressing expectations could be partly responsible for course alignment differences (Huddersfield, Model E) and mismatches between competencies desired by employers and learning outcomes. Some disciplines have developed competency-focused criteria (e.g. Nursing, Medicine, Education) (Raj et al. 2022), though not the computing discipline. Indeed, a recent UK governmental review made several recommendations regarding improving computing graduate work-readiness (Shadbolt 2016). This suggests a need to further explore how competency and 21st Century Skills can be more formally embedded into programme specifications with a great emphasis on learners employing knowledge, skills and dispositions in a real context. For example, being an effective collaborator is often the desired outcome of T18F – Collaboration rather than solely an in-depth knowledge related to the theories of collaboration. Skill frameworks, such as SFIA, can help define and explain these required competencies.

For education and employment policies

The skills profiling approach can reassure stakeholders that degree programmes are providing employable graduates, as well as identifying where programmes are strongly and weakly aligned to industry needs. Differences in meeting industrial needs may be down to the type of course (generalist or specialist), the topic (arts-based or science-based) or the way learning is defined or prioritised.

Job postings generally do not contain the subject-specific skills of Innovation, Self-Reflection and Technical Writing (Figure 5); however, these are embedded in learning outcomes. It is perhaps unsurprising that degree programmes seek to provide broader preparation as a graduate than simply an alignment to a specific job role, but when employers often voice concerns about employees’ creativity, self-regulation and communication it is interesting to note that these areas are not explicitly sought in job postings.

There may also be issues with describing learning with learning outcomes. Leadership and Management could perhaps have been combined (Figure 7 shows one scores highly for course, the other for job postings). Social Learning and Communication should be part of graduate expectations, but may not be strongly addressed in specific programmes especially Masters. Innovation, Self-Reflection and Technical Writing, Professionalism, Ethics, Risk Analysis and Sustainability are curricula areas promoted by professional bodies, higher education institutions and employers alike and hence may be assumed knowledge by employers. However, if assumed, their lack of visibility and promotion in job postings means that signals of their importance are missed, limiting incentives for these competencies to be developed further in programmes.

There is a difference between education and training programmes. It is possible to argue that higher education programmes are from a liberal education tradition, whereas training courses are more vocationally focused. Liberal education can be argued to have three central discourses (Collini 2012; Newman 1886):

• Knowledge for its own end;

• Knowledge viewed in relation to learning;

• Knowledge used in relation to professional skill.

This may explain some of the observed differences. The degree is broader than the professional skills embedded within or the explicitly stated job posting requirements. This highlights an ongoing debate in terms of education programmes that follow a liberal education model i.e. should this still be the approach that universities follow or should they focus more on professional skills? This paper cannot resolve these questions but it can provide a helpful way of considering and reflecting on such issues. Education and employment agencies may wish to explore such models and approaches to promote the more formal embedding of competencies within programmes. It also points to further work in terms of exploring the skills expectations of mid-career or senior professionals, the role of competency frameworks in clarifying skills progression requirements and how a skills profiling approach could be used to demonstrate how skills are developed throughout learner-earner journeys. Senior positions are likely to require a different skills profile from entry level roles, and specific aspects of study may well support this development for senior positions that are not immediately evident in entry level positions, such as leadership and management skills developed through studying for a Masters in Business Adminstration course, for example.

For assessment

A key consideration when seeking to express learning through skills, badges and micro-credentials is how does it demonstrate authenticity for learners whilst also enabling authenticity for earners i.e. how does it support both capability and competency measures? Fully aligning the skills profile mappings in this work would require a move towards assessing competency. Traditional approaches for assessment can find it challenging to evidence competencies (Raj et al. 2021). Authentic assessment approaches are required to enable learners to evidence expected knowledge, skills and dispositions. This would require changes to programme specifications and more formal consideration of skills, capabilities and competencies within assessed activities; moving away from focusing on learning that is straightforward to assess (knowledge and, to a slighter extent, skills), and greater alignment with employer needs at a more granular skills-based level. This is part of the reason for both current micro-credentialing trends and why this skills profiling approach is so important. It provides many more opportunities for authentic assessments in ways that clearly differentiate individual learner-earners.

For personalised learning

Two examples of learner-earners differentiating their capabilities and competencies are provided in this work (personalised learning (Huddersfield, Model E) and incorporating micro-credentials for credit (Northumbria, Model C)). It is possible to conceive of an extension to these examples whereby a learner chooses to complete some learning via micro-credentials and some via a traditional learning route. The Lifelong Loan Entitlement (UK Government 2022) is predicated on opening up such opportunities. The approach outlined in this work would facilitate such opportunities, promoting both lifelong and life-wide learning. It also provides a mechanism to help identify learning equivalencies between formal, informal and non-formal contexts. Doing so helps encourage personalised pathways through programmes, where some learning can be completed as credentials or micro-credentials, and additional learning from a formal education setting added such that the learning can be combined into a formal qualification. This would facilitate a learner-earner’s understanding of their own development and enable better-informed learning choices that then enable better alignment between education and employment and between skills, capabilities and competencies development and what is sought by employers. Opening up personalised learning pathways in higher education is not without its challenges. Nonetheless, it represents a vast opportunity to enhance the learner-earner experience and improve both the efficiency and effectiveness of education and employment pathways thus improving their outcomes.