Limitations
We could only include just over a quarter of the identified
studies in our meta-analysis; the remainder of studies did not
control for relevant stressors. There was a particular dearth of
longitudinal studies, and only two included data that we
could extract for analysis. A number of coping subscales could
not be extracted from otherwise relevant studies, because the
coping subscales did not fit into one of our three pre-defined
categories, but we have no reason to think that this
introduces a particular bias to our findings.