Evidence hierarchy

An evidence hierarchy classifies evidence according to its rigour. ‘Gold standard’ evidence sits at the top. Less rigorous forms of evidence sit below it, in descending order.

The Centre for Education Statistics and Evaluation (CESE) has developed the following hierarchy for classifying evidence. It ranks evidence according to the rigour of the research that underpins it.

Gold standard evidence is the most reliable form of evidence, as the methodology used in the studies that underpin it is the most robust. The following types of reviews and studies make up this category:

  • Meta-analyses of randomised controlled trials: summaries and syntheses of multiple studies that involve randomised controlled trials (RCTs). Because these include multiple studies, and the studies use the most robust methodology, this is the best possible evidence.
  • Individual randomised controlled trials: single studies that involve randomised controlled trials. Although single studies are not as reliable as multiple studies, these use the most robust methodology and therefore provide the next best possible evidence.

Silver standard evidence is the next most reliable form of evidence. The methodology used in the studies that underpin this category is less rigorous than that used in studies in the gold standard category, however the evidence can still provide a solid indication of effectiveness. The following types of reviews and studies make up this category:

  • Meta-analyses of quasi-experiments:  summaries and syntheses of multiple studies that compare a group exposed to a treatment with a group not exposed to a treatment, but the assignment of subjects to groups is not random. Because these include multiple studies, the evidence is stronger than that arising from any one study in isolation, but the methodology is not as reliable as that used in randomised controlled trials.
  • Individual quasi-experiments:  single studies that compare a group exposed to a treatment with a group not exposed to a treatment, but the assignment of subjects to groups is not random. As these are single studies, they are less reliable than multiple studies, and the methodology used is not as reliable as that used in randomised controlled trials.

Evidence that falls into the ‘other evidence’ category should rarely be relied upon to make important decisions as it does not clearly show the effectiveness of programs. However, if no other studies are available, these can provide an indication of what future research might show. When the only evidence available on a topic arises from the types of studies in this category, it should be treated with caution. The following types of reviews and studies make up this category:

  • Pre-post comparisons:  studies in which subjects are observed before and after a treatment. Because these studies do not include control groups, they cannot usually rule out other explanations for any changes observed.
    Expert opinion:  the opinion of a person generally regarded as an expert in a particular field.
  • For many programs, such anecdotal evidence may be all that is available. However, these should not be relied upon if more robust pieces of evidence are available.

Category:

  • Evaluation evidence bank

Business Unit:

  • Centre for Education Statistics and Evaluation
Return to top of page Back to top