Taking study quality into account is important in meta-analyses. Two approaches for factoring
in quality are (1) to include only studies deemed to be of acceptably high quality or (2) to include
studies varying in their quality and then examine the effect of coded (aspects of) quality in moderator
analyses. Among the challenges to implementing these goals are the numerous scales and instruments
that have been developed to assess quality, the lack of consensus on what factors should be focal
and the difficulty in accurately coding quality, given variation in the completeness and clarity of
study reporting. Only factors related to ‘internal validity’ (the level of confidence one has that
intervention/comparison status caused the outcome difference) or level of bias in effect sizes are
useful for weighting or adjusting effect sizes. However, separate assessments of external validity
(generalisability), adherence to ethical standards, and completeness and cogency of reporting can
help to qualify the conclusions of investigations and can be used to describe and compare different
areas of study (Moyer & Finney, 2005).