Annex A. Methodology note on Survey of Career Guidance for Adults
This report uses data collected in the 2020 Survey of Career Guidance for Adults (SCGA). The SCGA was conducted to better understand adults’ experience with career guidance services and to improve international data on coverage.
Fieldwork was conducted by Cint1 in two phases using an online questionnaire developed by the OECD. The first phase of fieldwork took place from mid-June to early July 2020 in six countries: Chile, France, Germany, Italy, New Zealand and the United States. The second phase took place in November 2020 in Argentina, Brazil and Mexico.2 Among the four Latin American countries covered by this report, the Chilean data was thus collected five months earlier than the other three countries’ data. The sample was restricted to adults aged 25-64, in order to target those who had left initial education.
The survey was prepared in six languages (English, French, German, Italian, Portuguese and Spanish) and distributed in the country’s official language. Cint disseminated the online survey to a “pre-approved” panel of registered users using a stratified sample methodology, which imposed quotas on age, gender and region. This means that Cint drew a sub-sample from its panel that is representative of each country’s population in terms of age, gender and region. The age and gender quotas were based on UN World Population Prospects statistics (https://population.un.org/wpp), while the region quotas were based on Cint’s own data.
Education quotas were added in the second phase of data collection (i.e. Argentina, Brazil and Mexico). In the first phase, adults with higher levels of education were over-sampled (see Annex in OECD (2021[1]). This is expected to some extent because online surveys tend to over-represent the behaviour of people who are online, who tend to be individuals with higher levels of formal education (Van Der Heyden et al., 2017[2]). However, the oversampling of higher educated adults was much higher in Chile than in the other countries in the first phase of data collection. To avoid such bias with other Latin American countries in the second phase of data collection, quotas on education were applied. Education quotas were based on OECD Education at a Glance.
After data collection, two quality checks were applied. First, if a respondent completed the survey in two minutes or less, the respondent was excluded. This is based on the assumption that the survey takes more than two minutes to complete with appropriate consideration. Second, if a respondent did not answer the final question of the survey, they were also excluded. This was to ensure that only respondents who completed the full survey were captured in the final dataset.
To ensure adequate sample sizes and comparability, the data collection aimed at 1 000 observations per country. Annex Table 1 shows the final sample sizes by country, after sample restrictions, quotas and the quality checks had been applied.
Annex Figure 1 compares the composition of the country-level samples with the composition of the actual population in each country. Thanks to quotas, the sample is very close to the actual population on age and gender in all four countries. The education quotas also ensure that the education profile of the sample is close to that of the actual population in Argentina, Brazil and Mexico. To enable cross-country comparisons, the Chilean data were reweighted wherever the sample size permitted in order to account for over-representation of highly-educated adults. In charts where reweighting of the Chilean data was not possible, a reference can be found in the notes.
Adults in informal employment are under-represented in the SCGA. Adults who reported being “employed without a contract” – a proxy for informal employment – made up only 7% of employed adults in the sample, while 40% of employed adults find themselves in informal employment in the actual population (CEDLAS and The World Bank, n.d.[3]).
Note: “Highly educated” refers to the share of the population with a tertiary education (ISCED 4-6). The other education groups not shown here.
are below upper secondary education (ISCED 0-2), and upper secondary or post-secondary non-tertiary education (ISCED 3).
Source: The sample composition by age, gender and education is drawn from the Survey of Career Guidance for Adults (SCGA). The population.
distribution of age and gender were extracted from UN 2019 Revision of World Population Prospects (https://population.un.org/wpp/). For
population estimates by education group, OECD Education at a Glance 2020 was used.
References
[3] CEDLAS and The World Bank (n.d.), SEDLAC Statistics, https://www.cedlas.econo.unlp.edu.ar/wp/en/estadisticas/sedlac/estadisticas/#1496165509975-36a05fb8-428b (accessed on 3 December 2020).
[1] OECD (2021), Career Guidance for Adults in a Changing World of Work, Getting Skills Right, OECD Publishing, Paris, https://dx.doi.org/10.1787/9a94bfad-en.
[2] Van Der Heyden, J. et al. (2017), Additional weighting for education affects estimates from a National Health Interview Survey, http://dx.doi.org/10.1093/eurpub/ckx005.
Notes
← 1. Cint is a digital insights gathering platform (www.cint.com). The Cint platform and products comply with standards and certifications set out by various market research associations including ESOMAR, MRS, ARF, MRIA, AMA, AMSRO and Insights Association and ISO 20 252 quality standards.
← 2. The online survey was conducted in June-July and November 2020, in the middle of the COVID-19 crisis. One implication of this is that more people were able to respond to the survey because they were confined at home, were teleworking, and/or because they lost their job and had more time available. Cint noted that response rates were higher than expected as a result. Any impact this might have had on sample composition, however, was mitigated by the use of quotas. Countries were also at different stages of the pandemic when the survey was conducted. It is possible that policy measures adopted in different countries to cope with COVID-19 could have indirectly influenced the use of career guidance services. For instance, those countries that were more heavily affected by the pandemic at the time of the survey may have had more people out of work or at risk of losing their job as a result of the policy measures that were adopted (e.g. temporary business closures, travel restrictions). This could have affected the share of people who used career guidance services.