Group comparisons were performed using student's t-test for
continuous variables, and the chi-square test for categorical variables.
The pre-post test difference was compared using a paired t-test for
each group. The strength of the intervention for each group was
illustrated by using the effect size, termed as “Cohen's d”, in which
values greater than 0.2, 0.5, and 0.8 was defined as small, medium,
and large effect, respectively (Cohen, 1988). The effect of intervention
was examined by one-way analysis of covariance (ANCOVA), using
general linear models with confounding factors included as covariates.
For each outcome variable, ANCOVA with the pretest score and
age as the covariate (control variables) were used to compare the
adjusted post-test score between the 2 groups. Before using ANCOVA,
the assumption of homogeneity of the regression slope was checked
by adding an interaction term of group by pretest score in the model.
In addition, age was chosen as a covariate in ANCOVA because the 2
groups differed both statistically and substantially. The SPSS software
package 15.0 for Windows was used for all analyses. Considered
significant was p b 0.05.
Group comparisons were performed using student's t-test forcontinuous variables, and the chi-square test for categorical variables.The pre-post test difference was compared using a paired t-test foreach group. The strength of the intervention for each group wasillustrated by using the effect size, termed as “Cohen's d”, in whichvalues greater than 0.2, 0.5, and 0.8 was defined as small, medium,and large effect, respectively (Cohen, 1988). The effect of interventionwas examined by one-way analysis of covariance (ANCOVA), usinggeneral linear models with confounding factors included as covariates.For each outcome variable, ANCOVA with the pretest score andage as the covariate (control variables) were used to compare theadjusted post-test score between the 2 groups. Before using ANCOVA,the assumption of homogeneity of the regression slope was checkedby adding an interaction term of group by pretest score in the model.In addition, age was chosen as a covariate in ANCOVA because the 2groups differed both statistically and substantially. The SPSS softwarepackage 15.0 for Windows was used for all analyses. Consideredsignificant was p b 0.05.
การแปล กรุณารอสักครู่..

Group comparisons were performed using student's t-test for
continuous variables, and the chi-square test for categorical variables.
The pre-post test difference was compared using a paired t-test for
each group. The strength of the intervention for each group was
illustrated by using the effect size, termed as “Cohen's d”, in which
values greater than 0.2, 0.5, and 0.8 was defined as small, medium,
and large effect, respectively (Cohen, 1988). The effect of intervention
was examined by one-way analysis of covariance (ANCOVA), using
general linear models with confounding factors included as covariates.
For each outcome variable, ANCOVA with the pretest score and
age as the covariate (control variables) were used to compare the
adjusted post-test score between the 2 groups. Before using ANCOVA,
the assumption of homogeneity of the regression slope was checked
by adding an interaction term of group by pretest score in the model.
In addition, age was chosen as a covariate in ANCOVA because the 2
groups differed both statistically and substantially. The SPSS software
package 15.0 for Windows was used for all analyses. Considered
significant was p b 0.05.
การแปล กรุณารอสักครู่..
