AD ALTA
JOURNAL OF INTERDISCIPLINARY RESEARCH
independent samples t–test with equality of variances. By non–
parametric Mann–Whitney U test, we test the variables that do
not meet the normal distribution condition.
The following tables provide an overview of only those variables
where have been demonstrated the statistical significance of the
examined differences (ISMS1 and ISMS2). Table 4 shows
Levene's test results that indicate the statistically significant
difference in group variances (F = 4,97; p = 0,028).
Subsequently, the t–test confirms the statistically significant
difference in the ISMS1 (the need for the ISMS in the
respondent's e–business) rating, t (49,70) = 2,23; p = 0,030,
among those enterprises which assess information security risks
(M = 66,64, SD = 25,13) and which do not assess the risks (M =
51,06, SD = 34,05).
Tab. 4 Testing differences of the ISMS1 variable depending on
the risk assessment
Source: Authors' own research
The results of the non–parametric test in Table 5 show that
enterprises dealing with information risks significantly differ in
the enterprise's ISMS practice (p < 0,05) from enterprises that do
not deal with the risks. The differences of the ISMS4 variable
are random (p > 0,05).
Tab. 5 Testing the differences of the ISMS2 variable depending
on the risk assessment
Variable
Mann–
Whitney U
Z
Asymp. Sig.
(2–tailed)
ISMS2
Practicing the ISMS
in the enterprise
328,00
-5,13
0,000
Source: Authors' own research
Figure 3 shows a summary of the surveyed variables with
average values for each group with a statistically significant
difference. Higher average values of variables are achieved in
enterprises that assess information security risks. Differences
between groups are evident and statistically significant according
to the test results.
Fig. 3
Average values of groups in points with a statistically
significant difference
Source: Authors' own research
Enterprises that assess information security risks achieve a
higher level of ISMS practice and emphasize the need for ISMS
in their e–business more than enterprises that do not assess such
risks.
4.2 Impact of practicing ISMS on the specified level of
information security
By linear regression analysis, we verify the hypothesis whether
the ISMS practice in the e–commerce enterprise (ISMS2)
significantly affect the specified level of information security in
that enterprise (ISMS4). We can observe a relationship of the
variables in Figure 4, where the systematic linear dependence is
not obvious, however, it cannot even be excluded.
Fig. 4 Correlation diagram of variables ISMS2 and ISMS4
Source: Authors' own research
The results of the regression are found in Table 6. The
correlation coefficient (R = 0,33) demonstrates the weak
interrelationship of the variables. The coefficient
of
determination (R
2
= 0,11) explains 11 % of variability of the
dependent variable ISMS4 affected by the independent variable
ISMS2. The remaining 89 % is influenced by other factors, such
as investments in the ISMS, security controls, security training,
incorporation of ISMS requirements into business processes,
supply contracts, etc.
Tab. 6 Regression model of the dependent variable ISMS4 and
the independent variable ISMS2
Model Summary (ISMS4)
R
R Square
Adjusted R
Square
Std. Error of
the Estimate
0,33
0,11
0,10
26,30
ANOVA (ISMS4)
Sum of
Squares
df
Mean
Square
F
Sig.
Regression 7397,66
1
7397,66
10,70
0,002
Residual 61543,95
89
691,51
–
–
Total
68941,60
90
–
–
–
Coefficients (ISMS4)
Unstandardized
Coefficients
Standardized
Coefficients
t
Sig.
95%
Confidence
Interval for B
B
Std.
Error
Beta
Lower
Bound
Upper
Bound
Constant 53,40
4,01
0,00
13,31 0,00 45,42 61,37
SY2
0,32
0,10
0,33
3,27 0,002 0,12 0,51
Source: Authors' own research
Analysis of variance (ANOVA) confirms a statistically
significant linear regression model with a good data description,
F (1, 89) = 10,70; p = 0,002. The regression coefficient b
1
= 0,32
significantly contributes to the prediction of the ISMS4 variable
(p = 0,002). Based on the results obtained, we reject the
hypothesis H
0
and at
the level of significance α = 0,05 we accept
66,64
40,95
51,34
10,03
0
20
40
60
80
100
ISMS1
ISMS2
Assessed information security risks
0
10
20
30
40
50
60
70
80
90
100
0 10 20 30 40 50 60 70 80 90 100
T
he
l
eve
l o
f
inf
o
rm
at
io
n s
ec
ur
it
y
(IS
M
S
4
)
Practicing the ISMS in the enterprise (ISMS2)
Levene's
Test for
Equality of
Variances
T–test for Equality of Means
95 %
Confidence
Interval of
the
Difference
E
q
u
al
v
ar
ian
ces
F
Sig.
t
df
S
ig
.
(2
–
ta
ile
d
)
M
ea
n
D
iffe
re
n
ce
S
td
. E
rro
r
D
iffe
re
n
ce
L
o
we
r
U
p
p
er
ISM
S1
A
ssu
m
ed
4,97 0,028 2,44 89,00 0,017 15,30 6,27 2,84 27,76
No
t
assu
m
ed
–
– 2,23 49,70 0,030 15,30 6,85 1,54 29,06
- 210 -
vol. 8
issue 1