CHAPTER IV

(Introductory paragraph here)

1. What is the profile of the respondents that can be drawn in terms of:

1.1. Age:

Table 1. Respondents’ profile in terms of age.

Table 1 shows that there are twenty-eight (28) respondents ages 15 – 16 years
old corresponding to a 24.35% of the total respondents. Moreover, eighty – one (81) age
17 – 18 years old matching to 70.43% and six (6) ages above eighteen years old which
contributes to a 5.22%. (You may rephrase this for the other discussions)

1.2. Sex:

Table 2. Respondents’ profile in terms of sex.

Table 2 shows that there were seventy-seven female respondents and thirty-
eight male respondents matching 66.96% and 33.04% of the total respondents,
respectively. (You may rephrase this for the other discussions)

2. Is there a significant difference on the variables when grouped according to the profiled
variables? (For this question, I will only present two profiled variables: age and sex; do
the same for the rest of the profiled variables you have and take note on the number of
groups each profiled variable have. Use Wilcoxon rank sum test for two groups and
Kruskal-Wallis for three or more groups)

Figure 1: Graphical illustrations for the different assumptions

The figure above shows that linearity, homogeneity of variance and normality of
residuals assumptions failed to satisfy. With this and since data is of ordinal level of
measurement, we should use non-parametric test.
Table 2. Normality test for the factors (variables).
Factors Statistical Test p-value Decision Interpretation
Concentration of the Study Shapiro – Wilk 0.0001129<0.05 Reject Ho Not Normally Distributed
Behavior of the Students Shapiro – Wilk 0.0006382<0.05 Reject Ho Not Normally Distributed
Self-Confidence Shapiro – Wilk 0.0005707<0.05 Reject Ho Not Normally Distributed

Table 2 shows that using Shapiro – Wilk test for normality, all p-values for each
factor is less than the level of significance ( ∝) which is 0.05, indicating that normality
assumption fails to satisfy. These results suggest that we should use non-parametric
tests i.e., Wilcoxon rank sum test and Kruskal-Wallis test.

Table 3. Descriptive analysis on the factors when grouped according to age.

Factors Age Median Mean SD SE
Concentration of the Study 15 – 16 years old 3 2.89 0.683 0.13
Concentration of the Study 17 – 18 years old 3 2.90 0.669 0.07
Concentration of the Study Above 18 years old 2.8 2.87 0.484 0.20
Behavior of the Students 15 – 16 years old 3.2 3.13 0.656 0.12
Behavior of the Students 17 – 18 years old 3 3.07 0.555 0.06
Behavior of the Students Above 18 years old 3 2.87 0.755 0.31
Self-Confidence 15 – 16 years old 3.2 3.26 0.530 0.10
Self-Confidence 17 – 18 years old 3 3.06 0.568 0.06
Self-Confidence Above 18 years old 3 2.77 0.731 0.30

The table above (Table 3) shows the descriptive analysis on the factors when
grouped according to their age. Two of the measures of central tendency such as
median and mean provides an exact representation of the entire collected data which
can be seen on the said table. Furthermore, the standard deviation (SD) with the
standard error (SE) shows how dispersed is the data and again showcased in Table 3.
Lastly, with negligible difference, the response of the students ages above eighteen
years old on the impact of bullying towards behavior of the students is the most
dispersed and according to students ages 15 – 16 years old, bullying has most
significant impact towards behavior of the students when based on the mean.

For the graphical representation (box – plot) of Table 3, see the immediate figure below.
 Figure 2: Box – plot of the factors grouped according to age.

Table 4. Analysis on significant differences on each factors grouped according to age.

Factors Statistical Test p-value Decision Interpretation
Concentration of the Study Kruskal-Wallis 0.927>0.05 Fail to reject Ho Not Significant
Behavior of the Students Kruskal-Wallis 0.604>0.05 Fail to reject Ho Not Significant
Self-Confidence Kruskal-Wallis 0.172>0.05 Fail to reject H o Not Significant

Table 4 shows that using Kruskal-Wallis test, there is no significant differences

on the level of influence of the three factors (concentration of the study, behavior of the
students, and self-confidence) between the identified age groups (15 – 16 years old, 17
– 18 years old, and above 18 years old) since all p-values seen is greater than the level
of significance (α ) of 0.05. (In this case, since all p-values are greater than 0.05, we will
not proceed with pairwise comparison)

(Suppose one of the variables in Table 4 has a p-value which is less than 0.05, proceed
with pairwise comparison or the post – hoc test. Note that only the variable(s) or factor(s)
with a p-value which is less than 0.05 will proceed to the pairwise comparison or post
hoc test)

For this example, assuming that the variable concentration of the study has a p-value
less than 0.05.

Table 5. Post – hoc test for the different age groups in terms of the perceived impact of
bullying towards concentration of the study
Factors Group 1 Group 2 p-value Interpretation
Concentration of the Study 15 – 16 yrs. old 17 – 18 yrs. old 0.913 Not Significant
Concentration of the Study 15 – 16 yrs. old Above 18 yrs. old 0.758 Not Significant
Concentration of the Study 17 – 18 yrs. old Above 18 yrs. old 0.701 Not Significant

Table 5 shows all p-values are greater than the level of significance of 0.05. This
suggests that pairwise, there is no significant difference on the impact of bullying
towards students’ concentration on their studies between age groups.

Table 6. Descriptive analysis on the factors when grouped according to sex.

Factors Sex Median Mean SD SE
Concentration of the Study Male 3.2 3.10 0.584 0.10
Concentration of the Study Female 3 2.79 0.674 0.08
Behavior of the Students Male 3.1 3.10 0.523 0.17
Behavior of the Students Female 3 3.06 0.621 0.14
Self-Confidence Male 3.4 3.19 0.573 0.09
Self-Confidence Female 3 3.05 0.572 0.07

The table above (Table 6) shows the descriptive analysis on the factors when
grouped according to their age. Two of the measures of central tendency such as
median and mean provides an exact representation of the entire collected data which
can be seen on the said table. Furthermore, the standard deviation with the standard
error shows how dispersed is the data and again showcased in Table 6. Lastly, with
negligible difference, the response of the female students on the impact of bullying
towards their concentration to their studies is the most dispersed and according to male
students, bullying has most significant impact towards self-confidence when based on
the mean. (If necessary, rephrase.)
For the graphical representation (box – plot) of Table 6, see the immediate figure below.

Figure 3: Box – plot of the factors grouped according to sex.

Table 7. Analysis on significant differences on each factors grouped according to sex.

Factors Statistical Test p-value Decision Interpretation
Concentration of the Study Wilcoxon Rank 0.027<0.05 Reject Ho Significant
Behavior of the Students Wilcoxon Rank 0.969>0.05 Fail to reject Ho Not Significant
Self-Confidence Wilcoxon Rank 0.166>0.05 Fail to reject Ho Not Significant

Table 7 shows that using Wilcoxon rank sum test, there is no significant
differences on the level of influence of the factors behavior of the students and self-
confidence between male and female since the two p-values (0.969 and 0.166) seen is
greater than the level of significance ( α ) of 0.05. However, there is a difference on the
perceived level of influence of bullying towards concentration of the study between male
and female since the p-value of 0.027 is less than 0.05 level of significance. Additionally,
males were greatly influenced by bullying in terms on their concentration to their studies
since it can be seen that the mean for males under the said factor in Table 6 is bigger
compared to that of females in the same factor (variable).

(Do the same for the rest of the profiled variables)

3. Is there a significant difference between the factors (or variables) found in question
number two in terms of the level of influenced of bullying?

Table 8. Descriptive analysis on the factors (or variables)

 Factors Min Max Median Mean SD SE
Concentration of the Study 1 4 3 2.89 0.659 0.06
Behavior of the students 1.4 4 3 3.07 0.589 0.06
Self-Confidence 1.6 4 3.2 3.09 0.574 0.05

(Try to follow the discussions made for tables 3 and 6)

For the graphical representation (box – plot) of Table 3, see the immediate figure below.

Figure 4: Box – plot of the factors.

Table 9. Analysis on significant differences between the variables or factors.

Statistical Test p-value Decision Interpretation
Factors Kruskal-Wallis 0.08 > 0.05 Fail to reject H o Not Significant

Table 9 shows that using Kruskal-Wallis test for normality, the p-value of 0.08 is
obviously greater than the level of significance (∝) which is 0.05, indicating that there is
no significant difference between the factors or variables. (Suppose the p-value is less
than the level of significance of 0.05, proceed with post – hoc test or commonly known
as pairwise comparison.)

4. Which of the variables does bullying has the most significant impact?

Based on Table 9, with negligible difference (with negligible difference if p-value

is greater than 0.05), the variable with highest mean is self-confidence. Hence, bullying
has the most significant impact towards self-confidence.