Statistical database

Percentage/ confidence interval

VIN01211

Selected 1 of total 10

Selected 0 of total 1

Selected 0 of total 2

Selected 0 of total 3

Selected 0 of total 2

Number of selected data cells are:

(maximum number allowed is 100,000)

Presentation on screen is limited to 10,000 rows and 100 columns

Number of selected cells exceeds the maximum allowed 100,000

The results come from the European labor force survey. The calculations are based on the percentage of employees that have a union membership.
### Background

#### Immigrants

Immigrant is defined as an individual that is born abroad and has parents that both are born abroad and have a foreign background. That means that both grandparents are born abroad.
#### Locals

Local is used for everyone that cannot be defined as an immigrant. Local can be seen as a broad category that includes many sub-categories. For example a person can be seen as local if they have no foreign background, if they are born abroad but have an Icelandic background, individuals that are born in Iceland and have one parent that is an immigrant, an individual that is born abroad and has Icelandic grandparents, and an individual that has immigrant parents but is born in Iceland.
### Unit

#### Confidence interval, ±

The results come from the labor force survey that is based on a sample of the population and therefore there is some uncertainty surrounding the results. To estimate this uncertainty confidence intervals are calculated. The confidence interval estimates how exactly the sample value represents the true value of the population. The calculations of confidence interval are based on the standard error of the estimate. The confidence interval reaches above and below the percentage point, and is added and deducted from the percentage point. If an estimate is 10%, and the confidence interval is ±1,2, then the lower confidence interval is 8,8 and the upper confidence interval is 11,2. A difference between two estimates can be said to be statistically significant if the confidence interval of the estimates do not overlap. 95% confidence intervals are calculated by multiplying the standard error with 1,96. That confidence interval will contain the mean with 95% certainty. If a series of random samples of the same size are taken from the same population, then 95% of the confidence intervals would contain the population mean. The precision requirements for this publication are that an estimate does not have a larger standard error than 5%.