EU Adult learning statistics Complementary methodological information

version 24/04/2017 home page

Section 3 – Participation rate in education and training – LFS data

Annual data on participation of adults aged 25-64 in education and training (referred to as "LFS/AL") are based on the EU Labour Force Survey (LFS). The reference period for the participation in education and training is the four weeks prior to the interview. The series are generally available from 1992 (according to EU membership).

From 2000 onwards, the indicator “participation rate in education and training (last 4 weeks)” is calculated as an annual average of quarterly LFS data. See more information on this source in the following statistics explained articles:

- the EU Labour Force Survey – methodology (in particular, user guide for the coding)

- the EU Labour Force Survey – main features and legal basis

- the EU Labour Force Survey – data and publication (in particular, publication guidelines)

Two major changes are worth being mentioned as regards the improvement of the EU-LFS for the LFS/AL series:

- The transition from spring results to annual averages of quarterly data.

- The improvement of the variables on participation in education and training.

3.1. Implementation of variables on participation in education and training

·         From spring data to annual averages of quarterly data

LFS/AL data are available from 1992. Between 1992 and 1997, the EU Labour Force Survey was conducted only in spring (quarter 1 or 2 depending on the country). The data for remaining quarters started to become progressively available from 1998 onwards but more significantly from 2000. The LFS/AL series are therefore based on spring data until 1999 and on annual averages of quarterly data from 2000 onwards.

For some countries however, the transition to a quarterly survey was not possible until 2005. The spring series are therefore used until the transition is complete in these countries.

In addition to this process, the EU-LFS regulations requested the transition to a continuous survey (where the reference weeks are spread uniformly throughout the year). This was generally implemented by Member States at the same time as they introduced the quarterly survey. However, some countries first introduced a continuous annual survey (meaning the reference weeks were uniformly distributed throughout the reference spring quarter) and then switched to a quarterly collection, whereas few others moved first to a quarterly survey and after to a continuous quarterly survey (with other changes in the questionnaire introduced at the same time). This had some impact on the series in some countries like Austria (2004).

·         Improvement of the variables on participation in education and training

The following improvements of the LFS/AL data took place in 2003 and 2006:

- 2003: the variable EDUC4WN (participation in any education and training – for the coding, see below and the LFS user guide available from the EU Labour Force Survey – methodology) was divided into two variables: participation in regular learning activities (“formal education”, EDUCSTAT) and participation in other taught learning activities (“non-formal education”, COURATT) as shown below.

Participation in regular (formal) education (EDUCSTAT=1)	No education (EDUCSTAT= COURATT=2)
Participation in non-formal education (COURATT=1)

 

- 2006: a code was added to the variable EDUCSTAT to single out persons in regular (formal) education but on holidays. This additional code applies to students enrolled in a formal programme but on school holidays and to those who will continue their studies after holidays. The code was introduced in order to minimise doubts about being in formal education during holiday periods.

Participation in formal education (EDUCSTAT=1)	Students (formal education) on holidays (EDUCSTAT=3)	No education (EDUCSTAT= COURATT=2)
Participation in non-formal education (COURATT=1)

 

·         Variables available for indicators on participation in education and training

As explained above, the collection of the variable(s) on participation in education and training changed in 2003 (from EDUC4WN to EDUCSTAT and COURATT). The following variables are therefore available in the EU-LFS database.

EDUC4WN
Q		Education or training received during previous four weeks (1992-2002, derived from 2003 onwards)
	0	Received no education or training
	1	Received some education or training
	9	Not applicable (child less than 15 years)
	blank	No answer
EDUCSTAT
Q		Student or apprentice in regular education during the last 4 weeks (from 2003 onwards)
	1	Has been a student or an apprentice
	3	Person in regular education but on holidays (from 2006)
	2	Has not been a student or apprentice
	9	Not applicable (child less than 15 years)
	Blank	No answer
COURATT
Q		Did you attend any courses, seminars, conferences or received private lessons or instructions outside the regular education system (hereafter mentioned as taught learning activities) within the last 4 weeks (from 2003 onwards)
	1	Yes
	2	No
	9	Not applicable (child less than 15 years)
	Blank	No answer

 

EDUC4WN remains the reference for the calculation of the LFS/AL series for participation in education and training in the last 4 weeks.

Accordingly, the variable EDUC4WN has been derived from 2003 onwards – or 2004 in some countries as explained below – as follows (SAS codes):

IF      EDUCSTAT='1' or EDUCSTAT='3' (from 2006) or COURATT='1' then EDUC4WN='1';

ELSE IF EDUCSTAT='2' and                            COURATT='2' then EDUC4WN='0';

ELSE IF EDUCSTAT='9' and                            COURATT='9' then EDUC4WN='9';

ELSE IF EDUCSTAT=' ' or                             COURATT=' ' then EDUC4WN=' ';

3.2 Comparability across countries

Although the changes introduced in 2003 and 2006 have had a general impact on the EU-LFS results, the following aspects need to be considered when analysing the series:

- The changeover to two variables EDUCSTAT and COURATT could not take place in all countries at the same time in 2003 (IT and IE from 2003 Q2, UK from 2003 Q3). BE, IT, LT, MT, PL, PT and HR implemented this in 2004 only.

- The questions to collect the EU-LFS variables are not the same in all countries. The EU-LFS regulations define the principles to be respected when providing data to Eurostat (as reflected in the definitions of the codes). Countries are free to choose the questions translating these principles in their national LFS. Therefore, a given country might already have two or several questions between 1992 and 2003 and others having only one. This implies that the changes implemented in 2003 introduce a break only in the latter since the former were deriving the variable EDUC4WN from their national questions already in the past (i.e. they did not change their questionnaire in 2003 but only the way they derived the data for the EU-LFS database).

- The introduction of the extra category for students on holidays could not take place in all countries at the same time in 2006. Prior to this introduction, some countries already included students on holidays with students while others did not. For details on the year of introduction and inclusion/exclusion before that date, please see table A2 in annex A.

Further information on the comparability of national LFS questionnaires as regards participation in education and training is presented in the annex A. National questionnaires are available from the statistics explained article the EU labour force survey – methodology.

Explanations for the main breaks in series (flags) are listed in a separate document.

3.3 Calculation of the LFS series on participation in education and training

The time series for the LFS indicator “adult participation in learning” is based on the EU-LFS variable EDUC4WN which is, where necessary and possible, adjusted in order to:

(a)   correct for main breaks in the LFS series,

(b)   estimate the missing values, i.e. in case of missing quarters until 2005, annual results and EU aggregates are estimated using:

§  the average of the quarterly information available,

§  interpolations of the EU-LFS data with reference to the available quarter(s).

Corrections and estimations generally apply to data prior to 2005.

Annex B presents the exceptions applied to the calculation of the indicator LFS/AL when using the EU-LFS microdata.

3.4 Ad-hoc modules

Four ad-hoc modules to the LFS are dedicated to education and training:

- 2016: Young people on the labour market

- 2009: Entry of young people into the labour market

- 2003: Lifelong learning

- 2000: Transition from school to working life

The regulations, evaluation reports and list of variables are presented on the statistics explained article the EU labour force survey – ad hoc modules.