Skip to content

Data quality

What are you looking for?

Information or data quality is the term used to describe the relevance and correctness of information. It provides clues as to how well the data describes reality or actual situations. The quality of the collected data is crucial for an impact assessment to be able to supply exact results. There are two criteria for quality that the collected data needs to satisfy according to scientific data collection methodology:

  • Reliability
    The term “reliability” refers to the relevance and correctness of information. A data collection method is deemed reliable if a rerun of the data collection or measurement in the same conditions leads to the same results.
  • Validity
    Data collection is valid if it measures what it was intended to measure. A measurement or survey is valid if the data collected provide fitting figures for the question under investigation.

Checking data and data sources

Data and data sources should be checked for reliability and validity. That is especially necessary when there are outside data sources or if new data sources are being used. New collection and processing methods should also be checked. It can be worthwhile doing a test run for data collection. It should also be checked whether the surveys deliver the desired information.

Identify and minimise sources of error

There are various sources of error that should be avoided when one is collecting or recording qualitative and quantitative data. If an organisation carries out its own data collection using qualitative and quantitative collection tools, then these tools (questionnaires, conversation guidelines, etc.) should where possible be pre-tested. This involves checking the collection tools on test individual or test cases. These should, where possible, be similar to the target group in the survey or the cases to be analysed. In addition, the pre-test should be carried out under conditions that are as similar as possible to the planned survey. Depending on the results of the pre-test, these collection tools might need to be revised or adjusted. It is therefore important that the time this takes is taken into account at the planning stage. Sources of error often come from the selection of the unit of analysis.