2. Research Question #1: -Are the
differences between the gains made by the
MMT group and those made by the control
group statistically significant?
An independent-sample t-test was used to determine the effectiveness of the MMT
program by comparing the total number of points gained between the pre and post tests
for each student. The result was significant at the α=.05 level, t(346) = 19.41, p >.001,
two-tailed, d = 1.958. The mean of the MTT group (M = 84.1, SD = 20.6) was
significantly higher than the control group mean (M = 40.35, SD = 23.969), as shown
in figure 1. With a cohen's d of 1.957, the effect size was very large.
5. Research Question #2: Is overall
class rating equally distributed
across SES?
A chi-squared goodness of fit test was used to determine whether
overall class rating was equally distributed across SES in this sample
group. An analysis of the results suggest that ice cream preference was
not equal, χ2(4) = 55 , p >.001. This result does not support the
hypothesis that overall class rating is evenly distributed across SES in
the sample group. Overall class rating is not independent of SES. As
shown in Figure 2 while a majority of students fell within Benchmark
level, and in the free lunch category, 70% of the paid group of students
met the benchmarks, while only 50% and 56% of the free and reduced
categories respectively met the benchmarks. Only 25% of the students
in the paid category fell in the strategic category, while 33% and 37%
of the free and reduced students respectively fell into the strategic
category.
6. Research Question #3:
Tx, Ctrl, and State, on Composite
Score from All State Tests
- normally distributed, equally variant,
independent
Group ID
Mean
N
Std. Deviation
Skewness
MMT
165.84
262
40.429
-0.831
Control
134.43
202
46.134
-0.304
State
158.71
1136
41.821
-0.661
Total
156.81
1600
43.07
-0.65
7. Research Question #3:
Tx, Ctrl, and State, on Composite
Score from All State Tests (cont'd)
- One-way ANOVA
- F(2,1597) = 35.614, p<.001, η2 = .043,
almost-medium effect size
- Tukey's HSD found between all three
groups
- treatment group scored highest, then rest
of the state, then control group
8. Research Question #3:
Tx, Ctrl, and State, on Composite
Score from All State Tests (cont'd)
Score by Group ID
pity the
control
group!
9. Research Question #4: Tx Score on
final post-test x State Composite
Score
- normally distributed
Descriptives
Mean
Std.
Deviation
N
State Composite Score
156.81
43.07
1600
MMT Post KG2
66.08
24.941
479
10. Research Question #4: Tx Score on
final post-test x State Composite
Score (cont'd)
- two continuous variables
- Pearson correlation
- Pearson r = .734, p<.001, two-tailed
- r2 = .54, very large effect size!
11. Research Question #4: Tx Score on
final post-test x State Composite
Score (cont'd)
Post KG2 x State Composite
similar
content,
close test
occasions,
or effective
intervention?