Is more vocational education the answer?

by Herman van de Werfhorst, Andrea Forster, Thijs Bol*

A few years ago, Eric Hanushek gave a keynote lecture at a conference at the Amsterdam Centre for Inequality Studies. The talk was entitled "Is more vocational education the answer?" and spoke to debates in the United States about whether or not to strengthen the vocational education and training sector. The U.S. education system is much more “general”  in nature than the German and Dutch education systems, which are more vocationally oriented. Is it sensible for the United States to adopt a German-style education system with a strong dual (work- and school-based) sector?

Vocational education and training can mean very different things to different people. In some countries, it refers to education and training provided by and in schools, with no or limited exposure to real work situations. In other countries, it designates systems where much of the training is provided in a work place by the employer. The latter is often called “dual system” or apprenticeship.

Hanushek's lecture warned against an overly optimistic interpretation of the vocational training sector. While strong vocational education and training systems may do well in integrating school leavers into the workforce (as has been documented before), vocational education is harmful in the later phases of work careers. More vocational education is definitely not the answer, according to Hanushek, Schwerdt, Woessmann and Zhang: Vocationally qualified workers are the first to be laid off after the age of 50 because their specific skills are likely to be outdated.

Looking at the impact of vocational education across a lifetime is fascinating, illuminating and highly relevant for policy debates on how to organise an education system. A true trade-off emerges between the short-term (early career) gains and the long-term (late career) losses, and tradeoffs like these
should be evident to policy makers when they think about changing the education system.

Micro-level and macro-level questions
We felt that it was premature to write off the vocational education and training system just yet. From Hanushek et al.'s paper, it wasn't clear whether the problem is a micro-level issue of workers educated in vocational schools relative to those educated in more general programmes, or a macro-level issue concerning the education system. It may be true that people with vocational qualifications are less likely to be employed later in their career, but maybe that pattern is unrelated to the way that vocational education is organised in a country. In fact, while the question "is more vocational education the answer?" is a system-level question, it was answered in a micro-level fashion by Hanushek and associates.

So we tested whether the lifetime employment profiles of adults with vocational versus general forms of education vary by the size of the dual system, using data on 22 countries from the Survey of Adult Skills (PIAAC). The warning of Hanushek to the proponents of a German-style vocational training system should imply that the late-career disadvantage of vocational degrees would be more pronounced in countries with a large dual system. However, we did not find evidence of that.

Two graphs from our paper tell most of the story. The figure below shows the age-employment profiles for people with vocational and general types of education. In line with Hanushek et al.'s micro-level hypothesis, we see higher probabilities of being employed for vocationally qualified workers at the start of their career, but lower probabilities late in their career, among both men and women. This is controlled for level of education, numeracy proficiency and parents’ education.  

                

                       Predicted probabilities of employment by type of education(vocational/general) 

Men

Women

Dual systems and workers
But is this pattern particularly strong in vocationally oriented systems? The figure below shows the effect of vocational education across a lifetime (relative to general education, again controlled for education level, numeracy proficiency and parents’ education), in two types of systems: systems with a strong dual sector and with a weak or no dual sector. It is clear that the early-career benefit of being educated in a vocational programme is strongest in countries with a large dual sector (Germany, Austria), while the advantage turns into a (slight) disadvantage later in the career (but mostly in countries with a weak vocational training sector). However, we do not see that the blue and red lines cross somewhere mid-career, which would be expected based on the thinking that German-style systems, in particular, result in later-career disadvantage for vocationally trained workers. In fact, strong dual systems are characterised by less disadvantage late in the careers of vocationally qualified workers; and the negative effect at the end of the career is not statistically significant (while it is negative and significant in societies that do not have dual systems, like the United States and Canada).

                       Average marginal effect of VET on employment in countries with low and high dual system enrollment 

      Men

  Women

Based on these findings, we conclude that the pattern of late-career disadvantage is not typical for strong vocational systems; quite the contrary. In all countries, people with vocational degrees are more likely to lose their job late in their career, possibly because of a lack of adequate skills. But if anything, strong dual systems offer a safeguard for those with vocational qualifications. Such systems do not adversely affect employment either at the start or at the end of a career.

Links:
Forster, Andrea G., Thijs Bol, and Herman G. Van de Werfhorst. 2016. “Vocational Education and Employment over the Life Cycle.” Sociological Science 3:473–94.
Hanushek, Eric A., Guido Schwerdt, Ludger Woessmann, and Lei Zhang. 2016. “General Education, Vocational Education, and Labor-Market Outcomes over the Life-Cycle.” Journal of Human Resources, 10.3368/jhr.52.1.0415-7074R
Survey of Adult Skills (PIAAC)  

*Herman van de Werfhorst, Andrea Forster and Thijs Bol are affiliated to the University of Amsterdam and the Amsterdam Centre for Inequality Studies. Contact: H.G.vandeWerfhorst@uva.nl

Source figures: PIAAC 2012, release March 2015, calculations Forster, Andrea G., Thijs Bol, and Herman G. Van de Werfhorst.

Vocational education and training, depending on its design, can lead to very different outcomes. As argued above, systems with strong apprenticeship or dual systems are associated with better employment prospects than those relying heavily on school-based vocational education and training. The OECD forthcoming study, Striking the Right Balance: Costs and Benefits of Apprenticeships, provides further insights into this topic. Drawing on data from the Survey of Adult Skills, it compares outcomes from apprenticeships with outcomes from alternative education and training options. It also discusses different components of the apprenticeship system and conditions under which apprenticeships yield considerable benefits to employers and students.

A Brave New World: The new frontiers of technology and education

by Tracey Burns
Project Leader, Directorate for Education and Skills, OECD

“I don’t actually have an attention problem. I just take the pill when I need to be sharp”. Legal drugs such as Ritalin, used for treating attention deficit disorder, are increasingly being repurposed by healthy students to feel sharper on exam day.

"Smart drugs" allegedly improve memory and concentration. In addition to Ritalin other drugs are also taken to aid learning, such as modafinil, normally used to treat sleep disorders. University students can rely on them to pull all-nighters during exam weeks. The belief (true or not) that these drugs might boost academic performance has grown along with their availability – both through a marked increase in the number of prescriptions and  through more prevalent online markets where prescriptions are not as carefully scrutinised.

This raises a series of ethical and practical questions for education. Do smart drugs provide some students with an unfair advantage? Should tertiary institutions take a stand on the illegal use of cognitive performance-enhancing drugs? And what about younger users? Reports of teenagers and even pre-teens abusing smart drugs have raised concern about the lack of research on the impacts of these drugs on developing brains.

These questions highlight some of the more challenging aspects of the technological advances sweeping our classrooms and societies. Trends Shaping Education 2016 looks at how technology is transforming our lives – and asks whether education will be able to keep up.

When we think of technology and education, we usually think of information and communication technologies (ICTs). And indeed, ICTs have changed the way we live. Increasingly mobile technologies allow us to buy our groceries, pay our bills, watch films and attend meetings without ever leaving our homes. In fact, we increasingly do many of these things at once: Internet users perform seven activities at any one time on average, up from five just a few years previous and giving rise to worries of decreasing attention spans among today's youth.

However, technological advances are not exclusive to the Internet. Although it might seem like science fiction, biotechnology is used in medicine to combat disease, in agriculture to produce higher yields and more resistant crops and in the environment to develop cleaner energy. One example of how biotechnology is more integrated in our lives comes from genome sequencing, or the process of revealing the genetic make-up of cells. Once extremely expensive, technological advances have reduced the price exponentially in just a few years. Individuals can now afford to map their genes and identify whether they carry potentially life threatening mutations. Earlier this year scientists from the United Kingdom were given permission to edit the genes of human embryos for research purposes. Will designer babies (and designer students) be part of the future?

The impact of technological trends on education is clear. A great deal of work has already been done to identify how and where education can better use technology in the classroom. And there is interesting new research on emerging opportunities for education and work that could develop from human enhancement and biotechnologies.

In contrast to many trends that are relatively gradual and often linear, the pace of technological development is exponential and its impact much less predictable. One of the most difficult issues will be staying abreast of the evolution of technology and human behaviour: the use of smart drugs is one example. Another is the delicate terrain of human emotion and large online audiences, which has given rise to new risks such as cyber bullying and revenge porn.

In education, schools and teachers are increasingly asked to guide students through the advantages and disadvantages of the virtual world without always having the necessary skills themselves. Difficult questions will evolve as quickly as the technology. For example, how does "textbook learning" interact with the easy answers available at the simple push of a button? Whose voice counts if there is competing information? And what should we do, if anything, about smart drugs and other biotech advances?

The key is adaptability. Worries about decreasing attention spans, digital withdrawal disorder and “fear of missing out” syndrome illustrate the shifting landscape of the future. Advances in biotechnology and smart drugs will continue to raise difficult technical and ethical questions as well as provide new opportunities. All of these issues need to be part of a long-term strategy to help education keep pace with modern society. When Aldous Huxley wrote A Brave New World in 1931 he was worried about the fast paced world of the future. That time has now come, and it is up to us – and our education systems - to make the most of it.

Links:
Trends Shaping Education 2016
Students, Computers and Learning: Making the Connection
Trends Shaping Education 2014 Spotlight 5, Infinite Connections: Education and new technologies
Measuring the Digital Economy: A New Perspective
Centre for Educational Research and Innovation (CERI)
Photo credit: Scientist examining samples with plants @Shutterstock

Can analogue skills bridge the digital divide?

by Marilyn Achiron
Editor, Directorate for Education and Skills

The digital divide has shifted. Instead of (and in some places, in addition to) separating people with Internet access from those without access, it now cuts a wide chasm between those who know how to get the most out of the Internet and those who don’t. It’s no longer a matter of getting the tool into people’s hands; it’s a matter of getting people to understand how the tool can work for them.

This month’s issue of PISA in Focus reveals that the fault line at the bottom of this digital divide is socio-economic status. In recent years, there has been great progress in expanding access to the Internet for rich and poor alike. In Denmark, Finland, Hong Kong-China, Iceland, the Netherlands, Norway, Sweden and Switzerland, for example, more than 98% of disadvantaged students have access to the Internet at home. In some countries and economies where disparities in home Internet access persist, schools try to compensate. For example, among the most disadvantaged students, 50% of students in Turkey and 45% in Mexico have access to the Internet at school. PISA results show that, given the wide availability of Internet access, disadvantaged students now spend about the same amount of time on line during the weekend as advantaged students do.

But as with any tool, the Internet is most useful when you know how to use it. Results from PISA 2012 show that just because students have access to an Internet connection, it doesn’t mean that they know how to use it for learning. And differences in how students use the Internet seem to be linked to socio-economic status, although the strength of that link varies widely across countries. For example, PISA finds that while disadvantaged students play videogames on line as much as advantaged students do, they are far less likely to read the news or search for practical information on the Internet than their more advantaged peers.

These differences also seem to mirror disparities in more traditional academic abilities – to the extent that once differences in the ability to read and understand printed texts are taken into account, students’ socio-economic status has only a weak, and often insignificant, relationship with students’ performance in the PISA test of reading on line. In other words, rich or poor, students who can read well are better-equipped to make the most of the Internet’s considerable assets.

So the best way to narrow this digital divide is to be sure that all students are given the same opportunities to acquire solid reading and Internet navigation skills – the equivalent of a user’s manual (and a driving permit) for what has become an indispensable tool.

Links

PISA in Focus No. 64: Are there differences in how advantaged and disadvantaged students use the Internet?, by Francesco Avvisati.

Students, Computers and Learning: Making the Connection

PISA 2012 Results
Photo credit: Concept: the digital divide @Shutterstock

What does a country average actually mean?

by Dirk Van Damme
Head of the Innovation and Measuring Progress Division, Directorate for Education and Skills

The institutional framework of the international community was created in the period following the Second World War. The building blocks for international organisations, including the OECD, were and are the nation-states of the post-World War and post-colonial order. However, nation-states are not fixed entities, but historical constructions. Hence, they take many different forms and change as a consequence of socio-political transformations. Few states correspond to the ideal form of a nation – identified by a common history, language and religion – or state. In a complex and diverse world, national identities change and become less homogeneous. Today, many states are confronted with political pressures originating from regional aspirations for more autonomy. Sometimes such pressures lead to a separation of political entities and the creation of new states, as was the case in the former Yugoslavia, the former Czechoslovakia and the republics of the former Soviet Union. No one can predict the future, but it would be illusory to expect that the current global order will not continue to evolve during the 21st century.

The international statistical system, one of the great achievements of international organisations, has mirrored the evolution of the nation-state. International statistics – and those related to education are no exception – were tuned towards comparing and benchmarking countries against each other. National averages thus became the dominant data. Most of the data points in Education at a Glance, for example, are national averages. However, the expansion and increased sophistication of data collection and data processing have allowed for the development of many more measures than just national averages. Indeed, averages without more detailed measures of how indicators are distributed across various subpopulations offer little added value when it comes to understanding the real world.
Through its “New Approaches to Economic Challenges” initiative, the OECD is working to highlight distributional measures in its statistical apparatus. In Education at a Glance, for example, our analyses increasingly focus on the distribution of education indicators by gender, age, socio-economic status and immigrant background around the national average.

So far, little effort has gone into exploring regional variations within countries. Technical shortfalls, such as the lack of regional data in existing data collections, but also political sensitivities, have hindered the analysis of regional variations. After a few years of hard work, a pilot project under the auspices of the INES Working Party has gathered a range of interesting regional data on some key education indicators. The most recent edition of Education Indicators in Focus (EDIF) explores subnational variations in educational attainment and labour market outcomes.

The chart above shows clearly the relevance of subnational variations. For one of the key measures of a country’s human capital, the tertiary attainment rate in the adult population, the subnational variation in some countries is almost as wide as between-country variations. This is true, obviously, for large countries, such as Canada, the Russian Federation and the United States, but also for Germany, Spain and Sweden. Smaller countries, such as Belgium, Ireland and Slovenia, show less variation, but differences are still significant.

In all countries, the capital region, which attracts a large share of the nation’s human capital for the government and the industries and services concentrated around it, has a larger population of tertiary-educated adults than most other regions. This observation in itself is relevant for education policy: the civil servants and advisors designing those policies often live in environments that bear no resemblance to other parts of the country.

A better understanding of the magnitude of subnational variations in education indicators prompts a range of policy-relevant questions. Huge disparities in human capital between regions call into question the validity of uniform nation-wide education and skills strategies. Regional variation calls for policies that are adapted to the regions’ specific contexts and realities. But nation-states might also have an interest in promoting educational inclusion in the country by taking the steps necessary to help regions at the bottom of the distribution move closer to the average. Significant regional variation might also signal the need for continuing involvement of the central state to ensure that regions have similar capacity and resources to support skills development.

From a statistical point of view, exploring subnational variations raises doubts about the meaningfulness of national averages in international statistics. It is necessary to understand what the country average is and the magnitude of the regional variation around it. After all, an average is just an average, a statistical construct, not a reality.

Links: 

Education Indicators in Focus No. 43 Subnational variations in educational attainment and labour market outcomes, by Markus Schwabe, Rachel Dinkes, Tom Snyder.
Chart source: OECD/NCES (2015), Education at a Glance Subnational Supplement, http://nces.ed.gov/surveys/AnnualReports/oecd/.