2 Sampling and stratification
Norway´s National Population Register is used as a sample frame. Gross samples are drawn from the registry after aquiring necessary permissions.
The table below outlines a historic overview of the main samples drawn from the population registry. The main samples represents the Norwegian population between the age of 16 and 84.
| Round | Size |
|---|---|
| 1 | 63 000 |
| 2 | 63 000 |
In order to correct for systematic biases in response rates, the sample is stratified. Sub-populations known to be less inclined to participate are oversampled. Sub-populations known to be more inclined to participate are undersampled. The stratums are as follows:
- Gender: Male, female
- Age: 16-29, 30-59, 59-84
- Geography: Norway´s 11 counties
The sample is drawn randomly within each of the stratums.
The first round of National Crime Survey, and other population surveys in Norway such as the Norwegian Citizen Panel, clearly shows that men, and especially young men are less inclined to participate. With the exception of men above 59 years who has a higher response rate than their female counterpart. It is especially men between 18-29 years that has a relatively lower response rate, both compared to women in the same age group and the overall response rate.
Age, gender and geography are demographic characteristics available in the population registry. Therefore, it is possible to correct for biases along these parameters as described above. But equally, or more important as a predictor for participation, is the respondents level of education. Experience from both the National Crime Survey and the Norwegian Citizen Panel is that there is a systematic underrepresentation of individuals with lower education levels. In order to counter this issue, we use survey weights as a method of post-stratification (see dedicated chapter on weighting).
In addition, following sub-samples have been targeted. The sub-samples are designed to increase representation of sub-populations known to be 1) less inclined to participate in surveys, or 2) especially crime-prone, or 3) both. The sub-samples are random without stratification
| Round | Size | Sub-sample |
|---|---|---|
| 1 | 5 000 | Immigrants |
| 2 | 5 000 | Immigrants |
2.1 Definitions
- Unit of analysis: Individuals
- Universe: The Norwegian population above the age of 16.
- Immigrants: Minimum one parent born in a country outside Europe, Oceania or North America
2.2 Weighting
To compensate for the observed bias, a set of weights has been calculated. The weights equal the relation between a given strata in the population and the total population, divided by the relation between a given strata in the net sample and the total net sample. This procedure returns values around 1, but above 0. Respondents who are underrepresented will receive a weight above 1 and respondents who are overrepresented a weight below 1. The weights of the different stratums are listed in the technical report.
When calculating the weights, the information regarding the respondent’s geographical location, gender and age are based on registry data. These attributes were included in the sample file we received from the Norwegian Population Register.
Information regarding the level of education is provided by the respondents when answering the questionnaire.
Two different weights have been calculated:
- Weight 1 based on demographic variables (age, gender and geography)
- Weight 2 combining the demographic variables with education.
Respondents with missing data on the education variable are only weighted on demography (the education component of the weight is set to 1 in these cases).
When applied, both weights will provide a weighted N equal to the number of cases in the dataset.