How to use Surveybanken

What kind of data can be found in Surveybanken?

Surveybanken contains several thousand surveys conducted since 1957, primarily in Norway.

Each of these surveys typically includes several hundred questions on various topics. Examples of question you can search for:

• Is climate change caused by humans?
• During the past week, have you been troubled by any of the following: Felling unhappy, sad or depressed?
• Think about an average day. How much time do you spend on the following: Social media?
• Did you vote in the election this autumn?

In total, there are about 750 000 of these kinds of questions available for search.

The unique thing about Surveybanken is that you can search for questions across all the surveys in the database.

Your search goes through 750 000 questions/variables and several thousand surveys. For instance:

• trust – returns results for all questions/variables and data sets containing this word.
• trust politicians– returns results for all questions/variables or data sets containing both of these words.
• election survey 2017 – returns results for surveys or variables containing both the word and the number.
• trust election survey 2017 – returns results for questions about trust in the 2017 election survey.

For those familiar with advanced searches: The search does not support the use of asterisk, apostrophe, plus, minus, AND, OR, etc. When multiple words are included in the search, the search is performed as if using AND.

Your search can return results for questions/variables, in the data set itself, or in a data collection.

The search phrase trust politicians can result in 55 variable matches and 31 dataset matches. You can choose whether to go directly to a variable or access a data set to view its documentation.

Questions may be phrased in different ways. Therefore, try to think of words that might have been used in the questions. For example, searching for trust politicians will return a match for the question: "How great trust in municipality politicians" from the Norwegian Citizen Panel.

When you click on a variable in the search results list, you will see default graphical representations, typically broken down by gender, age and education, and sometimes including maps. Not all data sets are processed to show all the analyses, so the number of graphs may vary.

To switch to another variable, you can scroll or search through the list of questions/variables. You can also filter questions by topic.

In addition to the automatic visualizations in Analysis, you can choose which variables you want to analyze. For example, you can explore how trust in politicians varies in relation to all the other questions/variables in the survey.

Click on “Add question/variable” to see how one variable is distributed across other variables. Here we examine the relationship between trust in politicians and political interest.

You can analyze up to three variables simultaneously.

You must log in to order data. Some data sets can be downloaded after you’ve logged in, while others need to be ordered.

Log in, add data sets to your cart, and go to the cart to complete your order.

Read more in our guide to ordering data.

Interpreting survey data involves analyzing the responses to questions so you can draw meaningful conclusions. The data is based on a sample, and interpretations depend on how well the sample represents the entire population.

Here are some steps to follow when interpreting survey data:

1. Check that data for completeness and accuracy: Before you analyze the data, you must make sure that all responses are complete and accurate. You should also check for missing data or deviations.
2. Identify patterns and trends: Look for patterns and trends in the data. For example, are there any common themes or standout responses?
3. Examine frequency distributions, percentages and averages, to summarize the data and identify key findings.
4. If your survey includes different groups of respondents (e.g., age, gender, income), compare their responses to identify significant differences or similarities.
5. Consider the context: It’s important to consider the context in which the survey was conducted. For example, results may be influenced by factors such as the timing of the survey or the phrasing of the questions. In the “Study documentation” you can find information about the background of the survey, the sampling method, the target population (who was surveyed) and more.

Remember that survey data is a sample of respondents intended to represent the entire population. This requires, for instance, a sufficient number of respondents to ensure that the results from the sample can provide insights about the whole population.