Developing a skills taxonomy

This morning’s mailing from the Marchmont Employment and Skills Observatory reports that NESTA have launched an interesting new Tool – a UK skills taxonomy:

“Skill shortages are costly and can hamper growth, but we don’t currently measure these shortages in a detailed or timely way. To address this challenge, we have developed the first data-driven skills taxonomy for the UK that is publicly available. A skills taxonomy provides a consistent way of measuring the demand and supply of skills. It can also help workers and students learn more about the skills that they need, and the value of those skills.” NESTA

It should help with careers guidance and is ideal for people looking at the return to differing career choices and how you get there. NESTA began with a list of just over 10,500 unique skills that had been mentioned within the descriptions of 41 million UK job adverts, collected between 2012 and 2017 and provided by Burning Glass Technologies. Machine learning was used to hierarchically cluster the skills. The more frequently two skills appeared in the same advert, the more likely it is that they ended up in the same branch of the taxonomy. The taxonomy therefore captures ‘the clusters of skills that we need for our jobs’.

The final taxonomy can be seen here and has a tree-like structure with three layers. The first layer contains 6 broad clusters of skills; these split into 35 groups, and then split once more to give 143 clusters of specific skills. Each of the approximately 10,500 skills lives within one of these 143 skill groups.

The skills taxonomy provide a rich set of data although requiring some work in interpretation. The six broad clusters of skills are:

The ten clusters (at the third layer) containing the most demanded skills are:

  1. Social work and caregiving
  2. General sales
  3. Software development
  4. Office administration
  5. Driving and automotive maintenance
  6. Business management
  7. Accounting and financial management
  8. Business analysis and IT projects
  9. Accounting administration
  10. Retail

The five skill clusters at the third layer with the highest annual median salaries are:

  1. Data engineering
  2. Securities trading
  3. IT security operations
  4. IT security standards
  5. Mainframe programming

The five clusters with the lowest salaries are:

  1. Premises security
  2. Medical administration
  3. Dental assistance
  4. Office administration
  5. Logistics administration

While the taxonomy is based on web data collected between 2012 and 2017, the approach has teh potential to be developed on the basis of real time data. And it is likely to be only one of a number of tools produced in the next two years using machine learning to analyse large data sets. The use of real-time data from web vacancies is receiving a lot of attention right now.

There is also interest in the idea of skills clusters in the ongoing debate over the impact of Artificial Intelligence on jobs and employment. Rather than whole occupations disappearing (and others surviving) it is more likely that the different skills required within occupations may change

Four domains of learning

four development domaninspng

I came upon this text today when I was seeking to extend on an article I was writing that included the idea of learning in four domains. It was produced, I think, for the EmployID MOOC on the Changing World of Work and was probably written by Alan Brown and Jenny Bimrose.Sadly, I was so tied up with producing my own materials for the MOOC and didn’t get to read all of the other peoples. But at a time when there is a growing need to question to division between humanities and technical subjects, I think this offers a good way forward.

Relational development – learning with and from interacting with other people

A major route for relational development is learning through interactions at work, learning with and from others (in multiple contexts) and learning as participation in communities of practice (and communities of interest) while working with others. Socialisation at work, peer learning and identity work all contribute to individuals’ relational development. Many processes of relational development occur alongside other activities but more complex relationships requiring the use of influencing skills, engaging people for particular purposes, supporting the learning of others and exercising supervision, management or (team) leadership responsibilities may benefit from support through explicit education, training or development activities.

Jack from the UK had switched career and now who worked as a carer. From the outset Jack learned much about his work from engaging with residents in the care home as well as learning from other staff. He had received letters from residents expressing their gratitude, which had boosted his confidence. His manager encouraged him to become a trainer in the care home, and although nervous and unsure he delivered the training and his self-efficacy increased.

Cognitive development – acquiring knowledge and thinking skills

A major work-related route for cognitive development involves learning through mastery of an appropriate knowledge base and any subsequent technical updating. This form of development makes use of learning by acquisition and highlights the importance of subject or disciplinary knowledge and/or craft and technical knowledge, and it will be concerned with developing particular cognitive abilities, such as critical thinking; evaluating; synthesising etc.

Bernard, a Czech automotive worker, participated in a short internal company technical training programme which positively surprised him in terms of practical outcomes and motivated him to actively work on his vocational development. ‘You had to know your stuff, the trainer was extremely competent, he knew his field very well, but sometimes I had difficulties to follow him. Anyway, it was really done by professionals who knew their stuff, and I appreciated it very much. I was very satisfied. I learned lots of things that were later very useful for my work […] It was very interesting to meet people from a completely different and a rather specialised area. I learned a lot of things and I was proud of it. I think this was the moment that made me change my attitude towards learning. I became much more curious.’

Practical development – learning by doing, by experience, by taking on challenges

For practical development the major developmental route is often learning on the job, particularly learning through challenging work. Learning a practice is also about relationships, identity and cognitive development but there is value in drawing attention to this idea, even if conceptually it is a different order to the other forms of development highlighted in this representation of learning as a process of identity development. Practical development can encompass the importance of critical inquiry, innovation, new ideas, changing ways of working and (critical) reflection on practice. It may be facilitated by learning through experience, project work and/or by use of particular approaches to practice, such as planning and preparation, implementation (including problem-solving) and evaluation. The ultimate goal may be vocational mastery, with progressive inculcation into particular ways of thinking and practising, including acceptance of appropriate standards, ethics and values, and the development of particular skill sets and capabilities associated with developing expertise.

Davide, an Italian carpenter, saw learning as a practice-based process driven by curiosity, a spirit of observation, and trial and error. A major role was played by his passion for the transformation of matter, which he perceived as an almost sacred event: ‘It really struck me to see that from a piece of wood one can create a piece of furniture’.

Emotional development – making sense of your own feelings and how others feel 

For emotional development, the major developmental routes are learning through engagement,  reflexiveness that leads to greater self-understanding, and the development of particular personal qualities. Much emotional development may occur outside work, but the search for meaning in work, developing particular mind-sets, and mindfulness may be components of an individual’s emotional development. Particular avenues of development could include understanding the perspectives of others, respect for the views of others, empathy, anticipating the impact of your own words and actions, and a general reflexiveness, which includes exploring feelings. Identity development at work may also be influenced by changing ideas individuals have about their own well-being and changing definitions of career success (Brown & Bimrose 2014).

Henrik from Denmark switched career, moving into caring and developed a new relationship with his work, which he found much more emotionally engaging. While studying for his skilled worker qualification, Henrik immersed himself in individual assignments of his own choice. In one assignment, he developed a ‘product’ to help improve a pupil’s ability to communicate, an ability which was being lost due to a rare disease. When Henrik talked about the assignment he was very engaged and showed insight into the syndrome. Because the assignment was closely related to his experience and practice, he saw meaning in undertaking it: ‘It was as though there was a circle I could complete on my own.’ He received a top grade for the assignment, and it is evident that positive learning experiences and the perception of entering into learning processes that are meaningful to his life and work situation are strong motivating factors in his engagement in further learning.

The future of work and changing occupational identities

The debate over the future of work, long running in research circles but kicked into public consciousness amongst others a Oxford University study titled ‘The Future of Employment: How susceptible are jobs to computerisation’ suggesting over 40 per cent of jobs are at threat in the next 11 years due to technology, emgineercontinues. In truth there is little agreement from economists and labour market specialists. Some claim techn0logy is leading to more jobs, some that it is destroying jobs and still other that it is neutral. Some claim technology is leading to jobs being deskilled, others the reverse.

I like a recent blog post entitled ‘More on digitalisation and skills: What happens within occupations?’, by Guillermo Montt on the OECD Skills and Work web site. The article says that “as technology enters the workplace, the tasks related to a job and an occupation change” citing  Alexandra Spitz-Oener (2006) who found that in Germany, occupations in the 2000s require more complex skills than in 1979 and that this change is more pronounced in occupations that adopted computers. Although something of a simplification, that finding is largely born out in analysis of the USA O*NET data. The article also draws attention to research by James Bessen published in his recent book ‘Learning by Doing: The Real Connection between Innovation, Wages and Wealth‘. “He follows the evolution of occupations over time and claims that accelerated technological change has implications for inequality within occupations with more and more occupations becoming winner-take-all markets.” Essentially, as new technology is introduced pay and opportunities in occupations bifurcate with a few taking high high, pay levels and more taking home lower pay. “In occupations requiring above-median computer use, the 90th to 50th percentile wage ratio has risen by 0.2% per year but has remained stagnant in occupations with below-median computer use. Workers who stay ahead of the curve, those who learn by doing, reap the wage benefits of technological change.”

This has major implication for training and continuing professional development. CPD has traditionally been organised through courses. But as we have already found in in the EmployID project working with employees in European Public Employment Services, traditional course delivery is both too slow to respond to change and even more problematic is unable to deliver the volume of training required. The approach adopted in EmployID is both to look at using new technologies for learning and for promoting informal learning in the workplace but also to center on changing occupational identities. For instance there is a very different occupational identity associated with a print graphic designer than todays web designer. But the ability to change occupational identities may be shaped by previous learning experiences and by motivation as well as the ability to reflect on both individual and group learning. Within EmployID we are exploring how Learning Analytics can bets be deployed to assets people in reflection (Reflection Analytics) and to assist in transforming identities to deal with such change. I am presenting this work next week at a LAKs pre conference workshop in Glasgow and will publish by slides on this blog.

Does technology destroy jobs

Infoposter_V1The argument over whether technology creates or destroys jobs has been going on for as long as I can remember.

Only yesterday John Naughton, in an article entitled “We are ignoring the new machine age at our peril“, worried about the impact of self driving cars and other technology on the future of employment. Naughton argued that there are “radical discontinuities that nobody could have anticipated”, driven by “combinatorial” effects of different technology trends coming together. These, he siad, include: “the near-infinite computing power provided by Moore’s law; precise digital mapping; GPS; developments in laser and infrared sensor technology; and machine-learning algorithms plus the availability of massive data-sets on which to train them.”

He warned the outcome could be “that vast swaths of human activity – and employment – which were hitherto regarded as beyond the reach of “intelligent” machines may now be susceptible to automation.” he went on to quote a studyby  Dr Carl Benedikt Frey and Michael Osborne, two researchers at the Martin School in Oxford,T heir report, The Future of Employment: How Susceptible Are Jobs to Computerisation?,  estimates the probability of computerisation for 702 detailed occupations, based on US government classifications of those occupations.  About 47% of total US employment, they conclude, is at risk from technologies now operational in laboratories and in the field.

However a study entitled ‘Are ICT Displacing Workers? Evidence from Seven European Countries‘ by Smaranda Pantea, Federico Biagi and Anna Sabadash from the Institute of Prospective Technologies in Seville comes up with a different answer. Looking at micro data ins even European countries for companies in the manufacturing, ICT producing and service sector the study found “a non-significant relationship between employment growth and ICT intensity among ICT-using firms.: The authors say: “Since our estimates mainly capture the “substitution” effects of ICT on employment (i.e. those due to ICT substituting for some type of labour and to ICT increasing productivity and hence reducing demand for inputs, for constant values of output), our results indicate that these effects are statistically insignificant.”

Of course this study and the American study are not directly comparable. They looked at different things and used different methodologies. One conclusion might be that whilst technology is not being directly substituted for overall employment, it is changing the nature of jobs available. Some labour market studies (for instance based on the US O*Net surveys) have suggested that what is happening is a bifurcation of labour, with an increasing number of high qualified jobs and of low skilled (and consequently low paid) service sector jobs. And of course another impact may be on the ;content’ and different skills required in different jobs. For instance our work in the construction industry through the Learning layers project suggests increasing adoption of technology is leading to the need for new (and higher) skills levels within what was traditionally seen as a lower skills sector. This has considerable implications for vocational education and training. ather than training for presents skills demands VET systems need to be looking at future skills. And by providing those future orein3eteds kills this could provide a workforce and society with the abilities and motivation to shape our use of technology in society, rather than as John Naughton fears that “we’re bound to lose this race against the machine” and in the course “enrich the corporations that own it.”

Developing a response to youth unemployment

Since I wrote my last article on ‘What is the answer to youth unemployment?‘, elections in Greece, France and Germany have seen a decisive rejection of European austerity politics. This is hardly surprising. It doesn’t take a genius to work out that ever deeper cuts and austerity, whilst ultimately cutting the real cost of labour and thus boosting corporate profits, are unlikely to boost growth, jobs or individual prosperity in any way.

The EU reaction has been to call for a new strategy for growth, although details of what that might entail are pretty hazy.

As I wrote in the previous article, one of the main results of the recession has been a massive increase in youth unemployment and, in particular, a substantial increase in graduate unemployment. At the same time companies are increasingly requiring work experience prior to employment resulting in increasing pressure for new graduates to undertake low paid of unpaid internships. Pretty clearly new policies are needed for education and training but there seems little public discussion of this, let alone of what such policies might be. The prevailing EU policy is more of the same and try harder.

To rethink policies for education and training requires looking back at how we got where we are now. And it requires looking at more than just education and training policy – we need to examine the relationship between education and training, labour market policy and economic policy. here I am going to look at just a few aspects of such policies and hope to develop this a little more in the next week or so.

For the last decade – or even longer – economic policy has been driven by a liberal free market approach. In turn labour market policy has similarly been based on deregulating labour markets and removing protection for workers (interestingly, Germany, the one country in Europe where the economy is growing, has probably one of the highest levels of labour market regulation). At a European level, education and training policy has been dominated by a drive to make qualifications more transparent and thus comparable in order to promote the mobility of labour. Employers have been given a greater role in determining the content and form of qualifications. Employability has become a key theme, with individuals being made responsible for keeping their knowledge and skills up to date, often as considerable personal expense. A number of countries have tried to liberalise education and training systems by reducing subsidies for public education and introducing individual voucher schemes.

At them same time the rather ridiculous EU Lisbon declaration, declared the aim to make the EU “the most competitive and dynamic knowledge-based economy in the world capable of sustainable economic growth with more and better jobs and greater social cohesion”, by 2010. Obviously this failed. But in line with such thinking most countries in Europe saw the way forward as moving from old fashioned vocational training to mass university education to cater for the demand for the thousands of new knowledge jobs. These jobs never materialised (except in countries such as the UK in the deregulated financial services sector which ultimately triggered the economic meltdown). As Wikipedia notes:

Much of the initial theorizing about the advent of a fundamentally new era in which economic activity is increasingly ‘abstract’, i.e., disconnected from land, labour, and physical capital (machines and industrial infrastructure) was associated with the ‘business management’ literature of the ‘new economy’ NASDAQ bubble, which collapsed in 2001 (but slowly recovered, albeit, in a leaner format, throughout the 2000s). This literature was initially known more for its hyperbole and faddishness than for its academic/empirical integrity.

In reality, many of the new degree courses were vocational in orientation – such as in the new Universities in the UK or in the Fachshule in Germany. These courses were either for new occupations – for instance in computing or simply replaced traditional vocational qualifications. It is arguable whether such a policy was financially sustainable or even desirable. It is certainly arguable whether an academic programme of learning is more effective for such subjects than traditional forms of work related learning.

To further policies associated with the obsession with the knowledge economy were the raising of the school leaving age and the so called lifelong learning policy. Longer schooling was needed, it was argued, to cope with the needs for higher levels of knowledge and skills for the knowledge rich jobs of the future. And lifelong learning was needed for the learning economies in which knowledge is the crucial resource and learning is the most important process.

At them same time the EU and national governments identified a number of key sectors which were felt to be crucial and which were then promoted through he education systems. In the late 1990s, there were dire predications of a massive shortage of computer programmers which never came to pass. And in the last five years or so EU and national governments have promoted the importance of STEM subjects – Science, Technology, Engineering and Maths as key to the future of employment and economies. Such priorities were based on a business driven policy of skills-matching promoting the “involvement of businesses in forecasting skills needs, through an employers’ survey tool and qualitative studies on the skills needs of business” (EU New Skills, New Jobs policy).

It is clear such policies have failed  and exhorting governments and agencies to try harder will go nowhere. What is needed is a fundamental rethink. As Professor Phillip Brown points out, the Lisbon Strategy was based on the idea that the technological lead then enjoyed by advanced industrial economies would be maintained with an increasing polarisation between highly skilled and well paid jobs in those countries and low paid low skilled manufacturing jobs being undertaken in developing countries. For a variety of reasons, including rapid technology transfer and a massive expansion of public education systems in countries like China and India, this hasn’t happened.

Indeed it may be the very manufacturing sector which was downgraded by EU policy which is the future for jobs in Europe especially in Small and Medium enterprises. For all the talk of high tech, knowledge based jobs. The construction industry is the biggest industrial employer in Europe with 13,9 million operatives making up 6,6% of the total employment in EU27. In addition it has a substantial influence on other industries represented by a multiplier effect. According to a study by the European Commission, 1 person working in the construction industry is responsible for 2 further persons working in other sectors. Therefore, it is estimated that 41,7 million workers in the EU depend, directly or indirectly, on the construction sector. Out of the 3,1 million enterprises 95% are SMEs with fewer than 20 and 93% with fewer than 10 operatives (pdf file). And manufacturing makes up almost 25 percent of the German economy, as opposed to only 11 percent in the United States. German mittelstands – small, family-owned and mid-size manufacturing companies – are key to the manufacturing sector. Rather than relying on university graduates for skills and knowledge, the mittelsands tend to employ graduates from the Dual apprenticeship system.

Indeed, many countries are promoting apprenticeships as one way out of the present mess. The present English coalition government boasts of the increase in the number of apprenticeship places. But in truth most of these places are apprenticeships only in name. The supermarket chain, Morrisons is the largest apprenticeship provider in the UK with many apprenticeship consisting of short induction training courses. To deliver the skills and knowledge for workers in a manufacturing economy through apprenticeship requires high quality training and the active involvement of employers and train unions alike. Moreover it requires social (and financial) recognition fo the value of apprenticeships. that seems a long way away.

To overcome the present crisis of youth unemployment requires a series of radical and interlinked policy initiative involving economic and labour market policies rather than just tinkering with education and training curricula. At a macro econ0omic level it means developing manufacturing industry rather than merely relying on financial services and the high tech knowledge industry sector. It means making sure companies provide high quality training, rather than forcing individuals to be responsible for their own employability. It means making sure that those who have gained vocational qualifications have opportunities to use those skills and knowledge and are properly rewarded for their learning. It means freeing up capital for starting small companies. It means proper financing for vocational schools and providing alternatives to young people rather than just more school and expensive university courses. It means abandoning skills matching and planning for future societal skills needs.

In other words we have to abandon liberalisation and free market ideologies and to recognise that economies and employment are a social function. As such society has to plan for the future of employment and the provision of jobs for young people. Is this too much to ask?