Data quality and validation

One of the most critical aspects of building trust in data quality is abiding by pre-defined ways of collecting and sharing data.

Certification/accreditation and associated definitions

  • Best Practice: An agreed way of doing something.
  • Standard method: An agreed method of doing something; bringing together expertise of wide range of users and end users to establish a best practice method/approach to doing something.
  • Compliance: The act of following a method or process
  • Quality Assurance: Assessing the whole method or process to prevent any errors or mistakes
  • Quality Control: Assessing the possible errors and mistakes within a process or method.
  • Accreditation/Certification: Official recognition of individuals’ or organisations’ compliance to a standard method. This can range from local certification following training e.g. ARMI to professional level accreditation process against a national/international standard (BSI, CEN, ISO) used by the Environment Agency and water company labs.

Quality Assurance (QA) versus Quality Control (QC)

Quality Assurance (QA)

Quality Assurance (QA) is a ‘proactive process’ that assesses the whole method or process to prevent any errors or mistakes.

  • Focus is on the prevention of mistakes and errors
  • Process-based approach
  • Manages quality

Sampling example: a QA process would be calibrating your Hanna checker to test for phosphates using a sample blank to ‘zero’ your sample between each sample.

Data example: a QA process would only allow a volunteer to enter readings within the possible range. The user could not enter a pH above 14 or a phosphorus reading above the detection limit of their equipment.

Quality Control (QC)

Quality Control (QC) is a ‘reactive process’ that assesses the possible errors and mistakes within a process or method.

  • Focus on the identification of mistakes and errors
  • Product-based approach
  • Verifies the quality

Sampling example: a QC process would be if you noticed a volunteer was always recording a zero value on their Kyoritsu tests for phosphate-phosphorus and so you meet with them to ensure they’re doing the test correctly. You then replace equipment as it appears sample tubes are out of date.

Data example: a QC process in water monitoring would be to have an expert review results to check for results that don’t make sense for a specific waterbody. They may catch an error, such as the reading for phosphorus being recorded where the pH would be recorded, or note that results are in the wrong units.

Related Articles

We've just published the CaSTCo Roadmap. Learn more

X