In: Statistics and Probability
In an effort to evaluate different methods of teaching a foreign
language, a researcher examined
the performance of students learning Spanish under four different
teaching methods. Group 1 learns in a
traditional lecture/lab format; group 2 is taught from a programmed
text; group 3 is taught from podcasts; and
group 4 is taught from videos of people and life in Spain. After
several weeks of instruction, all students were
given a Spanish quiz; these scores were recorded:
Group 1 Group 2 Group 3 Group 4
17 14 18 27
18 22 27 30
20 16. 24 25
17 15 18 26
16 16 20 32
21 20 22 27
18 18 19 31
15 14 25 28
a. Describe the statistical test to be done.
b. State the null and alternative hypotheses.
c. What is the critical value of the test statistic? What is the
decision rule for rejecting Ho?
CV: ______________________ Reject Ho if F ≥ _________________
d. Paste in SPSS output below, circling the observed value of the
test statistic. Also show your work for eta2
, and a
Tukey test, if necessary.
e. Write out an APA Style conclusion about the research project, using the context of the study.
a) Since there are four groups we have to do comparison between the means of the data and hence we perform one factor Analysis of Variance (ANOVA).
b) Null Hypothesis
Ho :
H1 : Atleast on mean is different.
c)
the critical value of the test statistic? What is the decision
rule for rejecting Ho?
CV: 2.946 Reject Ho if F ≥ Critical value.
d)
One-way ANOVA: Group 1, Group 2, Group 3, Group 4
Method
Null hypothesis | All means are equal |
Alternative hypothesis | Not all means are equal |
Significance level | α = 0.05 |
Equal variances were assumed for the analysis.
Factor Information
Factor | Levels | Values |
Factor | 4 | Group 1, Group 2, Group 3, Group 4 |
Analysis of Variance
Source | DF | Adj SS | Adj MS | F-Value | P-Value |
Factor | 3 | 643.7 | 214.583 | 28.37 | 0.000 |
Error | 28 | 211.8 | 7.563 | ||
Total | 31 | 855.5 |
Model Summary
S | R-sq | R-sq(adj) | R-sq(pred) |
2.75 | 75.25% | 72.60% | 67.67% |
Means
Factor | N | Mean | StDev | 95% CI |
Group 1 | 8 | 17.750 | 1.982 | (15.758, 19.742) |
Group 2 | 8 | 16.88 | 2.90 | (14.88, 18.87) |
Group 3 | 8 | 21.63 | 3.42 | (19.63, 23.62) |
Group 4 | 8 | 28.250 | 2.493 | (26.258, 30.242) |
Pooled StDev = 2.75
Tukey Pairwise Comparisons
Grouping Information Using the Tukey Method and 95% Confidence
Factor | N | Mean | Grouping | ||
Group 4 | 8 | 28.250 | A | ||
Group 3 | 8 | 21.63 | B | ||
Group 1 | 8 | 17.750 | C | ||
Group 2 | 8 | 16.88 | C |
Means that do not share a letter are significantly different.
Tukey Simultaneous Tests for Differences of Means
Difference of Levels |
Difference of Means |
SE of Difference |
95% CI | T-Value |
Adjusted P-Value |
Group 2 - Group 1 | -0.88 | 1.38 | (-4.63, 2.88) | -0.64 | 0.919 |
Group 3 - Group 1 | 3.88 | 1.38 | (0.12, 7.63) | 2.82 | 0.041 |
Group 4 - Group 1 | 10.50 | 1.38 | (6.75, 14.25) | 7.64 | 0.000 |
Group 3 - Group 2 | 4.75 | 1.38 | (1.00, 8.50) | 3.45 | 0.009 |
Group 4 - Group 2 | 11.38 | 1.38 | (7.62, 15.13) | 8.27 | 0.000 |
Group 4 - Group 3 | 6.63 | 1.38 | (2.87, 10.38) | 4.82 | 0.000 |
e) Since CV: 2.946 is less than the calculated test statistic F (28.37) Reject Ho , hence we conclude that all the means are not equal.
From tukey's results we can see that all the pairs are significantly different except group 2 and group1 because they share zero in their confidence interval.