Seven studies provided data on performance-based social and
daily living skills, usually collected through the use of role-play
tests of simulated social interactions. Four studies focused solely
on social skills, whereas one included performance-based measures
of activities of daily living only. The other two studies
included a combination of these measures. The weighted mean
effect size was moderate and significant (d 0.52; 95% CI
0.34, 0.71). There was also significant heterogeneity in overall
effect sizes between studies, suggesting that these studies did not
share a common effect. Analysis of client characteristics indicated
that SST had stronger effects on younger samples (Z3.78, p
.0005). Inpatient (compared with outpatient) sample studies had a
larger weighted mean effect size at a trend level (d1 0.82 vs.
d2 0.42; QB[1] 3.5, p .06). From a design perspective, those
studies that compared SST to a TAU control had significantly
larger effects than those comparing SST to an active control
condition (d1 0.87 vs. d2 0.09; QB[1] 16.4, p .0005), and
from a treatment perspective, paradoxically, SST programs with a
shorter duration in weeks and of less intensity produced a greater
weighted mean effect size than longer duration programs of greater
intensity (Z 4.31, p .0001; and Z 2.92, p .005,
respectively).
Seven studies provided data on performance-based social anddaily living skills, usually collected through the use of role-playtests of simulated social interactions. Four studies focused solelyon social skills, whereas one included performance-based measuresof activities of daily living only. The other two studiesincluded a combination of these measures. The weighted meaneffect size was moderate and significant (d 0.52; 95% CI 0.34, 0.71). There was also significant heterogeneity in overalleffect sizes between studies, suggesting that these studies did notshare a common effect. Analysis of client characteristics indicatedthat SST had stronger effects on younger samples (Z3.78, p .0005). Inpatient (compared with outpatient) sample studies had alarger weighted mean effect size at a trend level (d1 0.82 vs.d2 0.42; QB[1] 3.5, p .06). From a design perspective, thosestudies that compared SST to a TAU control had significantlylarger effects than those comparing SST to an active controlcondition (d1 0.87 vs. d2 0.09; QB[1] 16.4, p .0005), andfrom a treatment perspective, paradoxically, SST programs with ashorter duration in weeks and of less intensity produced a greaterweighted mean effect size than longer duration programs of greaterintensity (Z 4.31, p .0001; and Z 2.92, p .005,respectively).
การแปล กรุณารอสักครู่..
