Habitual physical exercise is almost universally accepted as being a p
rotective behavior for physical and mental health. Despite the heterog
eneity of research results, reviewers conclude, that mental health ben
efits of physical activity include reductions in state and trait anxie
ty. These effects are believed to be independent of age, gender and fu
rther personal and situational characteristics. To clarify the inconsi
stent results a meta-analysis was conducted. Twenty studies published
between 1980 and 1990 were found that provided 22 independent effect-s
izes based on 1,306 subjects. Using the Schmidt-Hunter-procedure, a po
pulation effect-size of r(w) = - .15 resulted. This coefficient is com
paratively small and not significantly different from zero. According
to various tests the data set has to be considered as heterogeneous. S
ubsequent meta-analyses aimed at identifying possible moderators such
as age, sex, etc. There is evidence for a marginal anxiolytic effect o
f physical exercise in middle-aged persons (30 to 50 years).