The Science of Teaching Science: Evidence from PISA 2015

The Science of Teaching Science: Evidence from PISA 2015

Education experts have spent the last 50 years debating over a seemingly simple question: what’s the best way to teach science? On one side of the divide are those who support self-guided, enquiry-based approaches, under which students direct their own learning. On the other are proponents of teacher-directed instruction, who say this approach makes it easier for teachers to manage classrooms and cover a wider range of content. Complicating the debate even further is the increasing diversity of student populations, which has raised demands for science curricula to adapt to student needs through adaptive teaching approaches.

We take a closer look at each of these strategies in the latest issue of PISA in Focus. Using new evidence from PISA 2015, we found that each approach has advantages and drawbacks for learning – and that identifying the most effective strategy isn’t as clear cut a proposition as it may seem.

In almost all of the 68 countries and economies that participated in PISA, students in the least disciplined science classes perform worse when exposed to enquiry-based science teaching. But in 33 countries and economies, this negative association disappears when students are learning in a disciplined environment.

In Thailand, exposure to enquiry-based teaching accounted for a four-point increase in performance among students in the most disciplined science classes. But students exposed to enquiry-based teaching in the least disciplined classrooms, scored about 13 points lower than those in more disciplined environments. The benefits gained from attending disciplined science classes with enquiry-based teaching are largest in Georgia (+20 points), Kosovo (+15 points), Lebanon (+13 points), Malta (+14 points), and Slovenia (+13 points).

Our findings shed light on the real-world complexity of teaching.

In OECD countries, enquiry-based teaching seems like the most promising way to nurture positive attitudes toward science – including interest and enjoyment in science-related topics, and participation in science-related activities. We also found that all three teaching practices – enquiry-based, teacher-directed and adaptive teaching are associated with higher expectations among students to pursue a career in science. This association is particularly strong among girls who are exposed to enquiry-based teaching.

Teacher-directed science instruction, on the other hand, is associated with better science performance in almost all countries. This positive association is equally strong across all science sub-domains and proficiency levels, and does not vary with student and school characteristics (e.g. disciplinary climate, student composition, resources, etc.). Based on these findings, we can conclude that teacher-directed practices are likely to deliver good results regardless of environment.

Our findings also show that adapting science lessons to students’ needs is correlated with stronger science performance in the majority of countries, even after accounting for student and school characteristics. This relationship is particularly strong in the Nordic countries, which are known for their comprehensive education systems and their reliance on differentiated learning approaches.

So which strategy would be most effective for science teachers to deploy? Our findings suggest a combination of all three. For example, teachers with strong classroom-management skills and professional knowledge could guide student learning with explicit instruction of basic ideas, then ask them to carry out enquiry-based activities to consolidate their knowledge. At the same time, teachers could also adapt their science lessons to account for differences among students, and help those who have difficulty understand a particular topic.

This conclusion may not be satisfying for those who firmly support one approach over the other. But our findings shed light on the real-world complexity of teaching in various classroom environments – where teachers often have to find the right mix of different practices to achieve the best results for their students.

What makes for a satisfied science teacher?

What makes for a satisfied science teacher?

Teachers play a vital role in the lives of their students. They impart knowledge, provide pastoral care, act as role models and, above all, create an environment that’s conducive to learning. But teaching is fraught with numerous challenges that could lead to dissatisfaction and ultimately to drop-out from the profession. Science teachers are particularly vulnerable to quitting their jobs given the opportunities that exist outside the teaching profession.

So what makes a science teacher satisfied enough that he or she would want to keep teaching, despite the challenges they might face?

Data from PISA’s 2015 teacher questionnaire provide interesting evidence.

Science teachers who reported that pursuing a career in the teaching profession was their goal after finishing secondary school are far more satisfied with their jobs and with the profession as a whole. These teachers represent about 58% of all teachers on average across all countries. The relationship between these long-held ambitions and teacher satisfaction is strong across most countries and economies, and particularly in Beijing-Shanghai-Jiangsu-Guangdong (China), Brazil, the Dominican Republic, Peru and the United Arab Emirates.

But a lack of school educational and physical resources, and behavioural problems among students in school could undermine teachers’ satisfaction. For instance, teachers who perceive that the lack of teaching staff hinders instruction tend to be less satisfied with their profession and with their current job. The difference in satisfaction between the teachers who reported that these shortfalls hinder instruction to a great extent and those who reported little or no impact are the largest in Australia, Brazil, Chile, Germany, Macao (China) and the United Arab Emirates. The findings also show that in 10 out of 18 countries and economies, teachers’ satisfaction with their current job is positively associated with the disciplinary climate in science classes, as perceived by students. The associations are particularly strong in Brazil, the Dominican Republic, Germany, Peru and the United States.

The presence of a collaborative and collegial working environment could boost teacher satisfaction. In fact, teachers who reported frequent collaboration among their colleagues tend to be more satisfied with their job and with the profession as a whole. Collaborative activities are more common in Australia, Beijing-Shanghai-Jiangsu-Guangdong (China), Colombia, the Dominican Republic, Korea, Macao (China), Peru, Portugal and the United Arab Emirates, and less common in Brazil, Chile, the Czech Republic, Germany, Italy and the United States.

PISA 2015 also shows that science teachers who engaged in more than three types of professional-development activities during the preceding 12 months tend to be more satisfied with the teaching profession and with their current job. On average and across all countries, 52% of teachers undertook more than three different types of professional-development activities during the last 12 months. The proportions are particularly large in Beijing-Shanghai-Jiangsu-Guangdong China (82%), Brazil (65%), the Dominican Republic (76%), Peru (65%) and the United Arab Emirates (65%).

Last but not least, some factors usually associated with challenging learning environments, such as the presence of large proportions of immigrant students or of students who do not speak the language of the host country, are not linked to teachers’ dissatisfaction with their job or the profession. This finding is particularly interesting because it shows that teachers do not necessarily mind teaching in schools with more demanding student populations as long as the environment is conducive to learning, the school climate is positive, and adequate resources are available.

To sum up, teacher satisfaction is positively associated with factors known to improve students’ performance, such as collegial and positive school environments. In other words, teachers’ satisfaction is both an aspect and a consequence of the school environment. As such, one has to improve the learning experience for all students in order to boost teachers’ professional satisfaction.

The Programme for International Student Assessment (PISA)
PISA in Focus No. 81 – What do science teachers find most satisfying about their work?
Working Paper No. 168 – Science teachers’ satisfaction: Evidence from the PISA 2015 teacher survey

Follow the conversation on Twitter: #OECDPISA

Image source: @Shutterstock

The Trends in International Mathematics and Science Study 2015 National Report for England

I co-authored the Trends in International Mathematics and Science Study 2015 national report for England with colleagues at the UCL Institute of Education (Toby Greany, Iain Barnes, Nicola Pensiero and Christina Swenson).

The full report can be accessed through the following link

In what follows is a summary authored by Toby Greany for the UCL IOE blog


The Trends in International Mathematics and Science Survey (TIMSS) now provides 20 years-worth of internationally comparable data on the mathematics and science performance of primary and secondary pupils worldwide, and the contexts in which they learn. England has participated in the study, which is now in its sixth four-yearly cycle, since its inception in 1995.  The 2015 national report, which I and a team from the UCL Institute of Education authored for the Department of Education, can be found here: TIMSS 2015.

Over 8,800 pupils across 290 schools participated in England’s TIMSS assessments last spring. The year 5 and year 9 pupils that sat the assessments have experienced quite substantial curriculum and qualifications reforms during their time in school: the year 5 pupils sat new Key Stage 2 assessments this summer, while the year 9 pupils will sit the new GCSE English and maths assessments next summer.

England’s pupils performed relatively badly in maths in 1995, coming below the international mean in both years 5 and 9. This prompted Michael Barber to argue for the introduction of a National Numeracy Strategy for primary schools a few years later.  Since then, performance in maths has improved significantly in both year groups, but particularly in year 5, where it is now significantly above the international mean.  2015 saw further small increases – compared to 2011 – in maths in both years 5 and 9, although neither increase was significant.

Performance in science was much better in 1995, and has remained significantly above the international mean in both year groups ever since. The worrying drop in year 5 science results in 2011 – which some attributed to the removal of universal science SATs for primary schools – appears to have been reversed between 2011 and 2015, with a significant improvement over the 20 year period. Performance in year 9 science has remained steady over the 20 years, with a small but not significant increase since 2011.

This picture of incremental improvement in maths, more volatile performance in year 5 science, and minimal change in year 9 science, leaves England firmly in the second highest-performing group of countries internationally in 2015. Meanwhile, the profile of countries above and below England has changed considerably over the two decades.  While the five East Asian countries (Singapore, Hong Kong, South Korea, Taiwan and Japan) have generally occupied the highest performing group of countries in recent rounds of TIMSS, Singapore performed at a similar level to England in year 5 science in 1995, while Hong Kong performed significantly lower. Several countries (such as Kazakhstan, Poland, Slovenia and the Czech Republic) have seen significant improvements in one or more subject in recent rounds, sometimes placing them in the highest performing group overall.  Others (such as Australia, Canada and the Netherlands) that performed above or at the same level as England in 1995 have now dropped below us.

In many respects, England’s schools compare favourably with their international comparators. For example, according to their head teachers and/or teachers, schools here have fewer challenges with a lack of resources, with poor conditions, and with pupil behaviour than schools in most other countries, although there are other areas, such as teacher recruitment, staff challenges and job satisfaction for teachers, where they perform notably less well.  England’s schools score highly for their focus on academic performance, and this appears to be a particularly important factor in student attainment.

Digging below the headline results, some further key findings for England in 2015 that are explored in more depth in the report include:

  • boys are significantly ahead of girls again in year 5 maths (having previously narrowed this gap);
  • whilst there has been an increase over the last 20 years in our pupils’ maths performance at both years 5 and 9, pupils here make relatively little progress in maths between years 5 and 9;
  • the performance of our lowest achievers in England has improved in recent years, however far higher proportions of pupils in the highest performing countries achieve the Advanced and High benchmark scores than in England;
  • England has wider gaps between our more and less advantaged pupils (determined according to the number of books pupils report having at home) than most other high performing countries;
  • three curriculum areas where England performs poorly compared to our overall results are Chemistry and Algebra in year 9, and Geometric Shapes and Measures in year 5;
  • pupil confidence in a subject appears to matter more to pupil performance than engaging teaching or whether or not pupils value it, so it is encouraging that pupils in England are relatively confident in their maths and science education; and
  • English as an Additional Language appears to be a barrier in science, but not in maths.

So England has improved its performance in TIMSS over the past 20 years, if not perhaps by as much as we might have hoped given the relentless pace and scale of the reforms we have seen in that period.

Both TIMSS and PISA (which will be published on 6th December) provide important sources for understanding trends in own national performance as well as starting points for exploring how and why different countries improve or decline on these international assessments.  TIMSS itself is not designed to answer such questions, but it does offer some tantalising clues that warrant further investigation.  For example, the East Asian group of countries performs phenomenally well in the assessments, but these countries often score less well in other areas; for example, with fewer pupils valuing or liking learning maths and science than in many other parts of the world, and with high levels of challenge for teachers and low levels of teacher job satisfaction.  It is also important to note the high levels of home tutoring in these countries, often involving more than 50% of pupils.  As ever with international comparisons the importance of cultural and contextual differences cannot be ignored here but, as England adopts ever more of its curriculum content and maths pedagogy from East Asia, it will be important to understand how these wider factors interact to secure high performance.

In a time of global uncertainty and complex social and economic challenges, education becomes more important than ever: the need for social justice, international understanding and engaged democracies has never been greater and education can contribute in all three areas. International studies such as TIMSS can support informed debate and thereby help build high quality education systems around the world, but in order for this to happen we must go beyond simplistic policy borrowing to enable genuine opportunities for system learning.

The Centre for Longitudinal Studies Missing Data Strategy

To impute or not to impute? A practical guide for handling missing data in longitudinal surveys. George Ploubidis.

Maximising the plausibility of the Missing At Random Assumption: Results from the 1958 British birth cohort. Tarek Mostafa.

How well do methods that assume data are Missing At Random perform when the missing data generating mechanism is not ignorable?  Evidence from the 1958 British birth cohort. Brian Dodgeon.

Variation within households in consent to link survey data to administrative records: Evidence from the UK Millennium Cohort Study

photo_consentTarek Mostafa

This article was first published in the NCRM newsletter.

Longitudinal surveys face significant challenges due to the rise in survey costs, attrition over time, and non-coverage of the target population. A promising solution to some of these problems is survey and administrative data linkage. Administrative data linkage leads to shorter interviews, less respondent burden and an overall reduction in costs (1) in addition to the gain of valuable information on respondents. However, access to administrative data is restricted by consent. Non-consent occurs when respondents refuse permission to link their administrative records to their survey data. It results in smaller samples and possibly in sample bias if the likelihood of consent is related to the characteristics of respondents.

This study aims to advance our knowledge about consent by analysing adult respondents’ behaviour when consenting to link their own administrative records in contrast to their behaviour when consenting to link someone else’s records (i.e. the cohort member in the Millennium Cohort Study). These variations in consent behaviour have not been explored in the past. All previous studies focused on respondents consenting to link their own records but not those of other members of their household (2, 3). The paper uses data from the UK Millennium Cohort Study (MCS) and focuses on consent to link the cohort member’s health and educational records and the main respondent’s health and economic records (all consents are sought in wave 4 of MCS). The study attempts to answer the following research questions:

Do respondents behave differently when consenting to link their own administrative records in comparison to consenting to link those of their children?

Does respondents’ consent behaviour vary according to the domain of consent, e.g. health, economic, education records?

What is the impact of interviewers on consent outcomes and can interviewer effects be separated from the impact of an interviewer’s geographical assignment?

In summary, the findings show that main respondents behave differently when consenting to link their own records and when consenting on behalf of the cohort members. For instance, parents of children with high cognitive skills are more likely to consent on linking their children’s educational records. In contrast, the child’s cognitive skills do not affect the parents’ likelihood to link their own health and economic records. Moreover, being a private person has a more significant effect on the MRs outcomes than those of the CM. When it comes to loyalty to the survey, respondents who have missed a wave in the past are found to be less likely to consent irrespective of the outcome. In contrast, partial evidence was found in support of the impact of past relationship with the agency holding the administrative data. Among the sociodemographic characteristics of respondents, ethnicity was found to have the strongest impact irrespective of the outcome. Nonwhite respondents are less likely to consent. The cross-equation correlations measured through the multivariate probit models showed that the highest level of association is between outcomes sought for the same respondent (i.e. MRs consenting for linking their own records vs. MRs consenting for linking the CMs records).When interviewers’ effects were included through the use of fixed effects models, the explanatory power of the models increased by 3 to 4 times. This indicates that the interviewers’ characteristics and behaviour have a large effect on consent.

In terms of fieldwork practices, the findings suggest that it is possible to identify the respondents who are less likely to consent (ethnic minorities, respondents with higher privacy concerns, and respondents who have dropped out from the survey in the past). Interviewers have a strong impact on consent, therefore in the case of low consent rates, the matching of interviewers and respondents and the allocation of interviewers, with more survey experience, to difficult cases might improve consent rates. The findings also indicate that the linked administrative data is likely to suffer from sample composition bias due to non-consent. This is of a particular interest for the MCS data users. For instance the linked MCS and educational records are likely to lose children with lower cognitive skills. Similarly the high and significant impact of ethnicity means that samples are likely to lose non-white minorities. Since ethnicity is highly correlated with educational, health and economic outcomes, the data contained in the linked administrative records will be affected by non-consent. However, the total level of bias depends on non-consent and on the extent of non-linkage (the failure to link data even if consent was given) which might alleviate or exacerbate the initial non-consent bias.

A full paper on this topic was published in the International Journal of Social Research Methodology: Mostafa, T. (2015). Variation within Households in Consent to Link Survey Data to Administrative Records: Evidence from the UK Millennium Cohort Study. International Journal of Social Research Methodology. 579.2015.1019264


1 Sakshaug, J., Couper, M., Ofstedal, M., & Weir, D. (2012). Linking survey and administrative records: Mechanisms of consent. Sociological Methods and Research, 41, 535–569

2 Korbmacher, J., & Schroeder, M. (2013). Consent when linking survey data with administrative records: The role of the interviewer. Survey Research Methods, 7, 115–131

3 Sala, E., Burton, J., & Knies, G. (2012). Correlates of obtaining informed consent to data linkage: Respondent, interview, and interviewer characteristics. Sociological Methods and Research, 41, 414–439

How consistent is respondent behaviour to allow linkage to health administrative data over time?

Tarek Mostafa and Richard D Wiggins, University College London.

This article was first published in the NCRM newsletter (Summer 2015).

Household surveys are increasingly being linked to administrative records with the potential of greatly enriching survey content on subjects such as health, education and income. One major challenge to data linkage is nonconsent. Non-consent occurs when respondents refuse permission to link their administrative records to their survey data. This problem inevitably leads to information loss and possibly bias if consent is correlated with key characteristics of the respondents.

Despite the recent developments in the analysis of consent [2, 3, 4, 5] the evidence is still scarce. The existing literature focused on the patterns of consent arising in cross-sectional surveys and very little is known about consent behaviour over time. Our study is one of the first longitudinal explorations of consent in the context of attempts to link survey and administrative data. It relies on a theoretical framework which distinguishes between passive, active, consistent and inconsistent consent behaviour and uses a series of models including multivariate probit analyses in order to identify the nature of consent as a latent trait and linear probability models which include interviewers as fixed effects.

The study exploits three consent questions to link health records held by the National Health Service (NHS) to the Millennium Cohort Study (MCS). The questions correspond to waves 1 (age 9 months), 2 (age 3 years) and 4 (age 7 years) of MCS. The questions are: To what extent is consent behaviour consistent over time? and Can consent behaviour be described as active or passive?

In order to answer the questions Cialdini’s [1] framework on compliance in survey research is developed by testing four scenarios which sub-divide consistency/ inconsistency in consent behaviour along the lines of activeness/passiveness. Active consistency is the case where respondents are aware of their previous choices and are committed to make the same choices on future occasions because of stable beliefs or personality traits (e.g. belief in the importance of scientific research, being a private person, etc). Passive consistency is the case where respondents make consistent choices over time even though the decision making process is passive. This means that consent decisions do not reflect an active adherence to well-defined beliefs but rather external influences such as the respondents’ circumstances at the time of the interview and the impact of the interviewers. Active inconsistency is the case where respondents are aware of their previous choices and intentionally behave in inconsistent ways. This change in behaviour could be the result of a change in convictions. For instance, a past consenter might actively decide to withhold consent after a breach to data confidentiality. Passive inconsistency is the case where respondents switch from consenters to non-consenters or vice versa. This switch is not the result of changes in convictions but rather the result of changes to the circumstances of the respondent (e.g. divorce, acute health problems), changes to the interviewers over time (e.g. persistence in pursuing consent), and the fact that respondents could have forgotten what they did in the past. In all cases, the respondent has a passive role and the changes in consent behaviour are caused by extrinsic factors.

In summary the analyses provide evidence in support of passive consistent behaviour. Firstly, 77% of respondents behaved consistently over time by either consenting or refusing to consent in all waves. Secondly, the cross-equation correlations from the multivariate probit models showed that the unobserved parts of the consent outcomes are weakly associated over time, and therefore, cannot really be held to indicate the existence of strongly held latent convictions about consent. Thirdly, the likelihood of consent and the likelihood of switching behaviour over time are related to the respondents’ circumstances, and to the variation the impact interviewers have on the respondents’ willingness to consent. These three findings indicate that, for the majority of respondents, consent is not driven by personal convictions but rather depends on the circumstances of the respondent at the time of the interview and on the potential influence of the interviewers.

In sum when it comes to using linked survey and administrative data, users need to take into account the potential sample bias resulting from the correlations between the respondents’ characteristics and the agreement to consent. On a practical level the longitudinal evidence based on our project suggest that it is important to brief interviewers about what to expect and encourage them to adapt to the respondent’s circumstances when attempting to gain their cooperation to consent to record linkage.


1 Cialdini, R. Wosinska, W. Barrett, D. Butner, J. and Gornik-Durose, M. 1999. Compliance with a Request in Two Cultures: The Differential Influence of Social Proof and Commitment/ Consistency on Collectivists and Individualists. Personality and Social Psychology Bulletin 25, 1242-1253.

2 Korbmacher, J. and Schroeder, M. 2013. Consent When Linking Survey Data with Administrative Records: The Role of the Interviewer. Survey Research Methods 7, 115–131.

3 Mostafa, T. 2015. Variation within Households in Consent to Link Survey Data to Administrative Records: Evidence from the UK Millennium Cohort Study. International Journal of Social Research Methodology, published online on: 23 March 2015.

4 Sakshaug, J. Couper, M. Ofstedal, M. and Weir, D. 2012. Linking survey and administrative records mechanisms of consent. Sociological Methods and Research 41, 535-569.

5 Sala, E. Burton, J. and Knies, G. 2012. Correlates of obtaining informed consent to data linkage: respondent, interview and interviewer characteristics. Sociological Methods and Research 41, 414–439.

Providing Education to the Syrian Refugees: ‘No Lost Generation’ a Promising Initiative

Syrian refugee children sit inside a makeshift school tent in Lebanon

Since the start of the conflict in Syria in March 2011, more than 6.5 million Syrians were displaced and another 3 million became refugees in neighbouring countries such as Turkey, Jordan, Lebanon and Iraq (figures from October 2014). According to UNHCR 3.2 million Syrians are now registered with the UN refugee agency; half of them are children below the age of 17. In some countries the influx of refugees has raised serious challenges to the abilities of governments, NGOs and international organizations, to provide the needed services such as food, shelter, education, and health.

In Lebanon, according to UNHCR, the number of Syrian school-aged children will exceed the number of Lebanese children who were enrolled in the public system in 2012. Eighty percent of those children were not in school as of September 2013. In addition to this, drop-out among Syrian refugees is twice as high as the national average. In Jordan, the situation is not much better with 56% of Syrian children not enrolled in school.

Aerial view of the Zaatari refugee camp in Jordan

The inability of the educational infrastructure to cope with the influx of refugees is compounded by the lack of adequate funding, the lack of trained teachers, the need for transportation, and the need for specialized services for children with disabilities. Other challenges also include the differences between the Syrian and Lebanese or Jordanian curriculums. In particular, in Lebanon, Syrian children have to study science and mathematics in either English or French while in Syria teaching is done exclusively in Arabic. Furthermore, children who have missed school for more than two years (usually older than 12) are more difficult to reintegrate into the education system since they are older than their peers. In addition to this, a large number of children are obliged to work in order to sustain their families.

Zaatari refugee camp, Jordan

Facing all these challenges a number of initiatives are being developed with the objective of alleviating the impact of the conflict on young Syrians. The ‘No Lost Generation Initiative’ is the largest among them. The initiative was launched in October 2013 by a consortium of UN agencies, governments, NGOs, and donors. Since then, numerous countries and organizations have pledged funding and the results are promising. Over the last year, 770,000 children received some form of education and more than 660,000 received psychological support. Enrolment of Syrian children in neighbouring countries has increased from 169,500 in 2013 to 489,000 in 2014. Now the major challenge facing the initiative is ensuring the sustainability of its funding.

Sources: The No Lost Generation Initiative.

For more information on the situation of Syrian refugees refer to:

The World Bank’s report on Syrian refugees in Lebanon.

The UNHCR’s report on the countries hosting Syrian refugees.

The different reports of the No Lost Generation Initiative.

Oxfam’s Syria, a children’s crisis?

Save the Children’s Untold Atrocities: The Stories of Syria’s Children.

Consent to admin data linkage and bias

1Surveys are increasingly being linked to administrative records with the potential of greatly enriching survey content on subjects such as health and education at a limited extra cost. One major challenge to data linkage is when respondents refuse permission to link their administrative records to a survey. This problem leads to a reduction in sample size and to sample bias if consent is correlated with respondents’ characteristics.

At the Centre for Longitudinal Studies we are carrying out an ESRC-funded project to analyse consent to admin data linkage in the Millennium Cohort Study (MCS). MCS follows 19,000 children born in the UK in 2000-01 and, so far, data has been collected when the cohort members were nine months, three, five, and seven years old. One of the important features of MCS is that the same consent questions are sought from the main respondent (i.e. the child’s parent, in most cases the mother) and the partner, and some of the questions are asked at each wave of data collection. The aim of this project is to assess the correlates of consent and to ascertain whether a latent propensity to consent exists across domains, at the household level and over time.

When we analysed multiple consent questions from MCS wave 4 for the main respondent, we found that ethnicity, gender, age, the respondent being a reserved person (as reported by the respondent), and the number of siblings in the household had a significant effect on the likelihood to consent. We also found a highly significant latent propensity to consent between all consent questions irrespective of the domain since all questions were answered by the main respondent.

At the household level four we analysed consent outcomes for two domains – health and economic records – two for the main respondent and two for the partner. The variables with the highest and most significant effects were: religion, the respondent being reserved, respondent’s age and whether the respondent received benefits. The analyses also showed that even if different respondents in the same household answered these questions, a strong and significant latent propensity to consent did exist; the strongest being for the combinations of questions answered by the same respondent.

In general, we found that respondents from ethnic minorities, older respondents and those who are more reserved were less likely to consent to data linkage. In contrast, non-religious respondents, white respondents, those with larger families, and those receiving benefits were more likely to consent. However, the impact of these variables varied according to the subject of consent and who was asked for consent.

When consent was analysed over time for the same respondent and the same domain (i.e. health), only ethnicity and whether the respondent is a reserved person had a significant effect. Further, the results showed that a weak latent propensity to consent exists between consent outcomes over time. This particular finding indicates that the likelihood to consent depends on two major groups of factors, those that are fixed over time (e.g. ethnicity) and those that reflect the circumstances surrounding the interview. The first group is likely to influence consent over time because it is intrinsic to the respondent. In contrast the second group of variables represent extrinsic circumstances which are time specific. Further, the weak latent propensity to consent over time is probably an indication that people forget what they did on the last survey and hence only their intrinsic characteristics affect their choice.

Beyond MCS, the findings of this study suggest that if consenters were very different in terms of their characteristics from the entire sample of respondents and if non-consent is high, then the significant impact of some of the correlates (e.g. ethnicity, openness, age,) requires adjustment for consent bias. Weights and multiple imputations can be used for this purpose. However, the efficiency of these techniques depends on the ability of the researcher to identify the variables which correlate with consent.

This article was first published in the December issue of Research Matters.

Learning: a cost effective intervention for a healthier old age

This post was first written for the blog of the Institute of Education by Dr. Andrew Jenkins.

It is based on research undertaken by Andrew and I for the Department of Business, Innovation and Skills. Please follow the link to download the report.

We live in an ageing society. On current estimates the European population aged over 60 will continue to grow by about 2 million people a year over the next couple of decades and by 2060 over 65s will make up some 30% of the European population. Finding ways to minimise the resulting strain on pension and healthcare systems is a major long-term challenge for policy-makers.

But older adults can also be considered a resource to society so the idea of active ageing must be central in addressing demographic change. Active ageing means growing old in good health and as a full member of society: having the opportunity to continue to participate in paid or voluntary work, remaining independent in daily life and involved as citizens. Older people have much to contribute to society and in turn will enjoy a better quality of life if they are able to do so. The EU’s designation of 2012 as the European Year of Active Ageing and Solidarity between Generations underscores this point.

So, if the objective is to maintain the wellbeing of individuals as they age, what is the contribution that participation in learning can make? Research has been growing on this topic but has been skewed towards small scale qualitative studies which, while of much interest and value, are not readily generalisable. Our study “Learning and Wellbeing Trajectories Among Older Adults in England” (pdf) aimed to strengthen the evidence base by drawing on quantitative data.

Using data from the English Longitudinal Study of Ageing (ELSA), a large-scale, nationally-representative survey of older adults, we focused on people in their 50s and 60s, and related a measure of their wellbeing to participation in several types of learning.

The most striking finding was that it was consistently the non-vocational and relatively informal types of learning (such as music/arts groups and evening classes) which were associated with increases in wellbeing, rather than formal, more vocationally-oriented education and training courses.

Quantitative studies also have the advantage of yielding precise estimates of the magnitude of effects. As people grew older, their wellbeing gradually declined. But the measured impact of participation in (non-vocational) learning was at least sufficient to offset this gradual decline in wellbeing as people became older. Another way of expressing these results is that the boost to wellbeing delivered by engagement in learning was about one-quarter of the size of moving from the bottom to the middle of the wealth distribution. These estimates, then, show very clearly that learning participation has a useful role to play as a contributor to the wellbeing of older adults.

Of course, quantitative studies of this type do not tell us very much about the reasons why participation in non-vocational learning affected the wellbeing of older adults while vocational courses did not. But previous, qualitative evidence, can help to fill that gap. It seems plausible that vocational courses would only have benefits in the longer term and only when they led on to more satisfying work or promotion. Participation in non-vocational learning activities such as music or arts groups or evening classes, on the other hand, would be more likely to be undertaken because of their intrinsic enjoyment or possibly because of opportunities for getting out and socialising. These are important reasons for learning at older ages. Older adults often appreciated learning because it helped them to be receptive to new ideas, to improve understanding and maintain a positive outlook. Opportunities for increased social participation, for meeting up and studying with friends and the forming of new networks, were also important factors.

In general, it is not at all easy to think of policy instruments which can make an effective contribution to active ageing. If learning can play even a small part in contributing to good health and wellbeing, or helping people to live independent lives for longer, then providing relevant and interesting courses for older adults is a remarkably cheap and cost-effective intervention.

Despite the accumulating evidence of the benefits of learning for older adults in England, rates of participation in adult education have tended to decline in recent years. Some of this decline may be blamed on the recession – people having less money to spend on learning. But the decline has occurred primarily because of deliberate changes in adult education policy. Since 2004 – that is under both Labour and Conservative/Liberal governments – the policy has been to reduce the amount of money available for short courses and other unaccredited types of learning in order to concentate funding on longer-term, qualification-bearing courses aimed at 16-24 year olds.

It was no great surprise then, that when the Department for Business, Innovation and Skills recently, and rather belatedly, published the results of the 2010 National Adult Learning Survey (pdf) it showed some sharp declines in both non-formal learning (taught classes not leading to qualifications) and informal learning (involving self-study to improve knowledge). Indeed it was found that participation in learning had declined across almost all age groups with the exception of 16-19 year olds, with the decline being most noticeable among those aged 60 and over. This was a significant reversal of rising participation rates shown by earlier surveys.

Policies channelling public funds towards accredited and vocational learning carries the risk that other forms of learning, and any benefits which derive from them, will be neglected. Yet, in our ageing society, if adult learning can play a role in maintaining the health and wellbeing of older citizens then there must be a strong case for the state to invest in it.