Running head: TESTING HYPOTHESIS FOR MEANS 1
Testing Hypothesis for Means
Student’s Name
Institutional Affiliation
TESTING HYPOTHESIS FOR MEANS 2
Testing hypothesis for Means
Hypothesis-testing helps in establishing if two groups of interest have a relationship
(Laureate, 2016). One sample t-test is a method that determines if the researcher used a specific
mean value while obtaining a sample of observations for a particular experiment (Laureate,
2016). This approach is applicable in establishing if the current levels of democracy are
statistically different from the experts’ desired mean of 6.
Syntax:
EXAMINE VARIABLES=Q46A
/PLOT BOXPLOT HISTOGRAM NPPLOT
/COMPARE GROUPS
/STATISTICS DESCRIPTIVES
/CINTERVAL 95
/MISSING LISTWISE
/NOTOTAL.
The box plot indicates that there are no outliers.
TESTING HYPOTHESIS FOR MEANS 3
Tests of Normality
Kolmogorov-Smirnova
Statistic df Sig.
Level of democracy:
.111 46940 .000
today
a. Lilliefors Significance Correction
Kolmogorov-Smirnov test gives the results as D(46940)=.111, P=.000. This output indicates that
the data do not follow the normal distribution. There is a violation of the normality assumption.
Syntax:
T-TEST
/TESTVAL=6
/MISSING=ANALYSIS
/VARIABLES=Q46A
/CRITERIA=CI (.95).
Output:
Statistics for one sample
N Average Standard Std. Error
Deviation Mean
Q46a. Level of
46940 5.52 2.883 .013
democracy: today
Test for one sample
Test number = 6
t Degree Sig. (2- Mean 95% confidence gap
freedom tailed) Deviation
Lower Upper
TESTING HYPOTHESIS FOR MEANS 4
Level of
-35.924 46939 .000 -.478 -.50 -.45
democracy: today
The null hypothesis tested here is that Ho: μ=6 vs . Ha : μ≠ 6.
This output is a confirmation that the current democracy level differs from the value of 6 since
t (46939)=−35.924 , p< 0.05. These results are significant in the development of the society
because it shows the current status of governance. The leaders can, therefore, spot the sector that
needs improvements.
An independent sample t-test enables the researchers to compare the means of two different
types of data. Therefore, this technique is appropriate in determining if there is a significant
difference in people’s perception concerning the regions (North Africa and Southern Africa.)
The assumptions are the same as for one sample t-test except for the homogeneity of variance.
Syntax:
T-TEST GROUPS=COUNTRY.BY.REGION (3 4)
/MISSING=ANALYSIS
/VARIABLES=Q46A
/CRITERIA= CI (.95).
Output:
Statistics for Group
Country by N Average Standard Std. Error
region Deviation Mean
Today’s Level of South of Africa 15979 5.78 2.795 .022
democracy North of Africa 5418 4.90 3.092 .042
Samples Test for Independence
Test for Means Equality t-test
TESTING HYPOTHESIS FOR MEANS 5
equality of
variance using
Levene
statistics
Std. 95% Confidence
Mean Error gap
Sig. (2- Differe Differenc Lowe
F Sig. t df tailed) nce e r Upper
Q46a. Assumptio 130.64 .000 19.453 21395 .000 .879 .045 .790 .967
Level n of 9
of equality of
demo variance
No 18.510 8610.81 .000 .879 .047 .786 .972
cracy:
assumption 5
today
of the
equality of
variance
Levene’s test helps in determining the assumption of homogeneity of variance (Frankfort-
Nachmias & Leon-Guerrero, 2018). These results indicate that the variations of the two groups
are not the same since the p-value is less than .05. The researcher is 95% sure that the two
regions statistically differ since t (8610.815) =18.510, p<0.05. This depiction is essential because
it shows how governance varies from one place to another. Therefore, leaders from different
countries should borrow excellent leadership skills from each other.
A paired sample t-test is useful in case-control experiments (Laureate, 2016). This technique
is appropriate for case three. The assumptions are the same as those of a one-sample t-test.
TESTING HYPOTHESIS FOR MEANS 6
Syntax:
T-TEST PAIRS=X1MTHUTI WITH X2MTHUTI (PAIRED)
/CRITERIA=CI (.9500)
/MISSING=ANALYSIS
Output:
Statistics for Paired Samples
Average N Standard Std. Error
Deviation Mean
Freshman mathematics
-.0096 16021 .99040 .00782
utility
Set 1
Second-year
.0059 16021 1.00682 .00795
mathematics utility
Correlations for Paired Samples
N Correlation Sig.
Freshman mathematics
Set 1 utility & senior year 16021 .285 .000
mathematics utility
This table gives a correlation between T1 and T2, which is at .285. This figure also indicates the
effect size, and the researcher can use the results to make real-life decisions.
Test for Paired Samples
Paired Differences
Std. Standar 95% Confidence Sig.
Deviatio d Error gap (2-
Mean n Average Lower Upper T df tailed)
Se Freshman -.0155 1.19384 .00943 -.0340 .00293 -1.649 16020 .099
t1 mathematic 6 4
TESTING HYPOTHESIS FOR MEANS 7
s utility –
senior year
mathematic
s utility
This table affirms that mathematical utility does not change as the students proceed since t
(16020) = -1.649, p=.099 at a 95% confidence level. This outcome is essential to the education
sector since the teachers can understand how student's attitudes change towards some subjects
over time.
TESTING HYPOTHESIS FOR MEANS 8
TESTING HYPOTHESIS FOR MEANS 9
References
Frankfort-Nachmias, C., & Leon-Guerrero, A. (2018). Social statistics for a diverse society
(8thed.). Thousand Oaks, CA: Sage Publications.
Laureate Education (Producer). (2016l). The t-test for independent samples [Video file].
Baltimore, MD: Author
Laureate Education (Producer). (2016m). The t-test for related samples [Video file]. Baltimore,
MD: Author
