Netherlands Working Conditions Survey (NEA)

What does the survey entail?

Purpose

The purpose of the Netherlands Working Conditions Survey (NEA) is to collect data on working conditions, accidents at work, work content, workplace relations and the employment conditions of workers.

Partnerships

The Netherlands Working Conditions Survey (NEA) is conducted by Statistics Netherlands (CBS) and the Netherlands Organisation for Applied Scientific Research (TNO) together with the Ministry of Social Affairs and Employment. Since 2018, additional responses among employees in primary and secondary education have been collected at the request of the Ministry of Education, Culture and Science (OCW) in order to compile statistics on their working conditions, with a focus on pressure and workload. The statistical reports are published on the Netherlands Organisation for Applied Scientific Research (TNO) website.

Target population

All employed persons aged 15 to 74 years working in the Netherlands.

Statistical unit

Employed persons.

Date/year survey started

The Netherlands Working Conditions Survey (NEA) has been carried out since 2005.

Frequency

Every year.

Publication strategy

Definite figures.

How is the survey conducted?

Survey type

Sample survey.

Survey method

Respondents to the Netherlands Working Conditions Survey (NEA) fill in an online survey. Data is collected between 1 October and 31 December.

Respondents

Employed persons.

Scope of sample survey

In 2023, the gross sample size was over 175 thousand employed persons. The nett sample size is 62 thousand employed persons. This includes the additional responses collected from employees in primary and secondary education.

Checking and correction methods

The sample survey is checked every year for plausibility, internal consistency and completeness. Overrepresentation and underrepresentation of certain groups in the response is corrected using weighting.

Weighting

Since 2022, the sample survey has been divided into two random parts. The survey is set up in such a way that some of the questions are shown to the first half of the respondents, while the others are shown to the second half of the respondents. This method is called split-half reliability. In order to make statements on both halves of respondents and the population as a whole, Statistics Netherlands (CBS) weights the data collected from both halves. The two halves are then combined using proportional weighting. The following model is used for both halves of the weighting: gender, age, sector, origin, type of household, type of contract, primary income, country of origin. Linear weighting is carried out for both halves, taking into account unequal inclusion probabilities on the one hand, and potentially selective non-response on the other.

Quality of the results

Accuracy

Because the survey is based on a sample, the results of the Netherlands Working Conditions Survey (NEA) are subject to random fluctuations. The figures are based on data collected from at least 100 respondents. Both the value and the confidence interval for the figures are published on StatLine. The actual value lies in this interval with a probability of 95 percent. The size of the sample survey partially determines the range of the confidence interval. Larger sample surveys lead to narrower confidence intervals and more accurate estimates.

Sequential comparability

Since reporting year 2022, the way in which data for the Netherlands Working Conditions Survey (NEA) has been collected and processed has diverged from previous reporting years in some respects. As a result, the figures from 2022 onwards may not be comparable with the figures from before 2022 in all cases.

Even though a large number of changes were implemented in 2022 and the figures on some topics cannot be compared on a sequential basis, there are also topics for which it is likely that the figures can still be compared sequentially. Statistics Netherlands (CBS) only does this in cases where the survey questions have not changed, the use of the previous weighting model for the 2022 figures does not lead to new conclusions on differences between results of 2021 and 2022, and there are no other ‘plausible reasons’ to assume a method break.

In addition to this, the design of the Netherlands Working Conditions Survey (NEA) also changed in 2018. The gross sample size was increased to over 172 thousand employees and the printed version of the survey was no longer available. According to exploratory analyses, the discontinuation of paper-based responses caused a slight change in the composition of the response, but had little or no impact on the results on a number of core indicators.

The design of the Netherlands Working Conditions Survey (NEA) changed for the first time in 2014, when the gross sample size was enlarged significantly from 80,000 to over 140,000 workers, and their age range was also expanded from 15 – 65 years to 15 – 75 years. Also, Computer-Assisted Web Interviewing (CAWI) became the main survey method instead of Paper and Pencil Interviewing (PAPI) and several question modules were modified. Finally, since 2014, data collection has been managed by CBS. Due to these changes, not all results of the Netherlands Working Conditions Survey (NEA) from 2014 and onwards are sequentially comparable with 2013 and before.

There were no changes in the design of the Netherlands Working Conditions Survey (NEA) between 2005 and 2013. However, the content of the survey was slightly different each year. In 2007, the survey was modified in such a way that, for a limited number of variables, the trends found between 2006 and 2007 may (in part) be attributable to methodology.

Quality strategy

ll data is checked for plausibility, internal consistency and completeness.