Note: Binomial logistic regression was used in all analyses to
predict differences in demographics across study condition (0 = control;
1 = intervention). Each demographic factor was compared across
conditions individually, in separate models. Any variable that differed
between groups at p < .10 was included as a potentially
confounding variable in follow-up models testing for the effect of the
intervention (highlighted in bold). Exp(B) is a odds ratio, with 1
meaning no difference, values under 1 meaning slightly less likely, and
values over 1 meaning more likely to occur in the intervention compared
to control group. English language proficiency scores ranged from 1 to
4, with 4 being completely English proficient and 1 being no English
proficiency. |
Note: Binomial logistic regression was used in
all analyses to predict differences in demographics across study
condition (0 = control; 1 = intervention). Each demographic factor was
compared across conditions individually, in separate models. Any
variable that differed between groups at p < .10 was included
as a potentially confounding variable in follow-up models testing for
the effect of the intervention (highlighted in bold). Exp(B) is a odds
ratio, with 1 meaning no difference, values under 1 meaning slightly
less likely, and values over 1 meaning more likely to occur in the
intervention compared to control group. English language proficiency
scores ranged from 1 to 4, with 4 being completely English proficient
and 1 being no English proficiency. |
Note: Binomial logistic
regression was used in all analyses to predict differences in
demographics across study condition (0 = control; 1 = intervention).
Each demographic factor was compared across conditions individually, in
separate models. Any variable that differed between groups at p
< .10 was included as a potentially confounding variable in
follow-up models testing for the effect of the intervention (highlighted
in bold). Exp(B) is a odds ratio, with 1 meaning no difference, values
under 1 meaning slightly less likely, and values over 1 meaning more
likely to occur in the intervention compared to control group. English
language proficiency scores ranged from 1 to 4, with 4 being completely
English proficient and 1 being no English proficiency. |
Note:
Binomial logistic regression was used in all analyses to predict
differences in demographics across study condition (0 = control; 1 =
intervention). Each demographic factor was compared across conditions
individually, in separate models. Any variable that differed between
groups at p < .10 was included as a potentially confounding
variable in follow-up models testing for the effect of the intervention
(highlighted in bold). Exp(B) is a odds ratio, with 1 meaning no
difference, values under 1 meaning slightly less likely, and values over
1 meaning more likely to occur in the intervention compared to control
group. English language proficiency scores ranged from 1 to 4, with 4
being completely English proficient and 1 being no English proficiency. |
Note: Binomial logistic regression was used in all analyses to
predict differences in demographics across study condition (0 = control;
1 = intervention). Each demographic factor was compared across
conditions individually, in separate models. Any variable that differed
between groups at p < .10 was included as a potentially
confounding variable in follow-up models testing for the effect of the
intervention (highlighted in bold). Exp(B) is a odds ratio, with 1
meaning no difference, values under 1 meaning slightly less likely, and
values over 1 meaning more likely to occur in the intervention compared
to control group. English language proficiency scores ranged from 1 to
4, with 4 being completely English proficient and 1 being no English
proficiency. |