Table 2.
A framework for assessing research quality (Gorard et al., 2019).
| Design | Scale | Completeness of data | Data quality | Rating |
|---|---|---|---|---|
| Strong design for research question | Large number of cases per comparison group | Minimal missing data, no evidence of impact on findings | Standardised, independent, pre-specified, accurate | 4  |
| Good design for research question | Medium number of cases per comparison group | Some missing data, possible impact on findings | Standardised, independent, not pre-specified, some errors | 3  |
| Weak design for research question | Small number of cases per comparison group | Moderate missing data, likely impact on findings | Not standardised, independent, or pre-specified, some errors | 2  |
| Very weak design for research question | Very small number of cases per comparison group | High level of missing data, clear impact on findings | Weak measures, high level of error, too many outcomes | 1  |
| No consideration of design | A trivial scale of study, or number is unclear | Huge amount of missing data, or not reported | Very weak measures, or accuracy not addressed | 0  |