Final Exam Write Up
By
Lauren Barker
Final Exam
Requirements for EDUC 812
Liberty University
2019
FINDINGS
Research Question
The research question for this study was:
RQ1: Is there a significant difference among the attitude toward science fairs of students’
who either won 1st place, 2nd place, 3rd place, or no award at all at the local science fair?
Null Hypothesis
The null hypothesis for this study was:
H01: There is no significant difference among the attitude toward science fairs of
students’ who either won 1st place, 2nd place, 3rd place, or no award at all at the local science fair.
Descriptive Statistics
Data obtained for the dependent variable (attitude scores) toward science fairs of
students’ who either won (1st place, 2nd place, 3rd place, or 4th= no award at all) is specified in
Table 1.
Table 1
Descriptive Statistics
Dependent Variable: Attitude Scores
Award Mean Std. Deviation N
1.00 8.1429 .89974 7
2.00 8.0000 .81650 7
3.00 8.0000 .81650 7
4.00 3.8571 .89974 7
Total 7.0000 2.01843 28
Results
Data screening
Data screening was conducted on each group’s (1st place, 2nd place, 3rd place, or 4th= no
award at all) dependent variable (attitude scores) to determine data inconsistencies and outliers.
A representation of data is presented in a box and whisker plot to detect outliers on each
dependent variable. No data errors, outliers, or inconsistencies were identified. See Figure 1 for
box and whisker plot.
Figure 1. Box and Whisker Plot for 1st place, 2nd place, 3rd place, or 4th= no award at all.
Assumptions
An analysis of variance (ANOVA) looked at the differences among the attitude toward
science fairs of students’ who won 1st place, 2nd place, 3rd place, or 4th= no award at all at the local
science fair to test the null hypothesis. A Levene’s test was used to examine the assumption of
homogeneity of variance in the data. The ANOVA required that the assumptions of normality and
homogeneity of variance were met. A Shapiro-Wilk test was conducted because of the small
sample sizes (N < 50). No violations of normality were found. See Table 2 for the results of the
Shapiro-Wilk test.
Table 2
Tests of Normality
Award
Kolmogorov-SmirnovaShapiro-Wilk
Statistic df Sig. Statistic df Sig.
Attitude Scores 1.00 .258 7 .174 .818 7 .062
2.00 .214 7 .200*.858 7 .144
3.00 .214 7 .200*.858 7 .144
4.00 .258 7 .174 .818 7 .062
*. This is a lower bound of the true significance.
a. Lilliefors Significance Correction
The groups were also tested for the assumption of homogeneity of variances. The Levene’s
test provides an F value that shows no violation of this assumption occurred (p = .85), as seen
below. This concludes that the variances were equal among the groups. See Table 3 for Levene’s
Test.
Table 3
Levene's Test of Equality of Error Variancesa
Dependent Variable: AttitudeScores
F df1 df2 Sig.
.267 3 24 .848
a. Design: Intercept + Award
Results for Null Hypothesis
A one-way ANOVA test was conducted to examine the attitude toward science fairs of
students’ who either won 1st place, 2nd place, 3rd place, or no award at all at the local science fair.
The means and standard deviation for differences among attitude scores was presented in Table
1. A preliminary screening using Shapiro-Wilk test indicated the data was normal. The variances
were determined to be equal using Levene’s test, (p = .85). (p = .062, .14, .14, and .062
respectively). The variances were determined to be equal using Levene’s test, (p = .85), which
indicated no homogeneity of variance. See Table 4 for the Tests Between-Subjects Effects.
Tests of Between-Subjects Effects
Dependent Variable: Attitude Scores
Source
Type III Sum
of Squares df
Mean
Square F Sig.
Partial Eta
Squared
Corrected Model 92.286a3 30.762 41.677 .000 .839
Intercept 1372.000 1 1372.000 1858.839 .000 .987
Award 92.286 3 30.762 41.677 .000 .839
Error 17.714 24 .738
Total 1482.000 28
Corrected Total 110.000 27
a. R Squared = .839 (Adjusted R Squared = .819)
Findings for the ANOVA suggest a significant difference among the attitude toward
science fairs of students’ who either won 1st place, 2nd place, 3rd place, or no award at all at the
local science fair, where F (3, 24) = 41.68, p < .001, using partial 2 = .84, which is large,
causing the researcher to reject the null hypothesis at a 95% confidence level. A post hoc Tukey
HSD test was required for the null hypothesis revealing a significant difference among attitude
toward 1st place (M = 8.14, S.D. = .90), 2nd place (M = 8.0, S.D. = .82), 3rd place (M = 8.0, S.D.
= .82), and I did not when an award (M = 3.90, S.D. = .90) See Table 5 for the results of the
Tukey HSD test.
Table 5
Multiple Comparisons
Dependent Variable: Attitude Scores
Tukey HSD
(I) Award (J) Award
Mean
Difference
(I-J)
Std.
Error Sig.
95% Confidence
Interval
Lower
Bound
Upper
Bound
1st Place 2nd Place .1429 .45922 .989 -1.1240 1.4097
3rd Place .1429 .45922 .989 -1.1240 1.4097
I did not win an
award
4.2857*.45922 .000 3.0189 5.5525
2nd Place 1st Place -.1429 .45922 .989 -1.4097 1.1240
3rd Place .0000 .45922 1.000 -1.2668 1.2668
I did not win an
award
4.1429*.45922 .000 2.8760 5.4097
3rd Place 1st Place -.1429 .45922 .989 -1.4097 1.1240
2nd Place .0000 .45922 1.000 -1.2668 1.2668
I did not win an
award
4.1429*.45922 .000 2.8760 5.4097
I did not win an
award
1st Place -4.2857*.45922 .000 -5.5525 -3.0189
2nd Place -4.1429*.45922 .000 -5.4097 -2.8760
3rd Place -4.1429*.45922 .000 -5.4097 -2.8760
Based on observed means.
The error term is Mean Square(Error) = .738.
*. The mean difference is significant at the .05 level.
Powered by TCPDF (www.tcpdf.org)
