Contingency tables
Contingency tables
Privacy Read vs Gender
Chi-Square Tests
Crosstab
Gender
Female
Male
Total
Count
% within Gender
% of Total
Count
% within Gender
% of Total
Count
% within Gender
% of Total
Privacy _read
Yes
No
26
82
24,1%
75,9%
13,5%
42,5%
32
53
37,6%
62,4%
16,6%
27,5%
58
135
30,1%
69,9%
30,1%
69,9%
The V Cramer coefficient is
Significant ; its value (0,147)
signals some statistical dependence
between the considered variables
Total
108
100,0%
56,0%
85
100,0%
44,0%
193
100,0%
100,0%
Pearson Chi-Square
Continuity Correction a
Likelihood Ratio
Fis her's Exact Test
Linear-by-Linear
As sociation
N of Valid Cases
Value
4,169b
3,548
4,153
df
4,147
1
1
1
As ymp. Sig.
(2-sided)
,041
,060
,042
1
Exact Sig.
(2-sided)
Exact Sig.
(1-sided)
,057
,030
,042
193
a. Computed only for a 2x2 table
b. 0 cells (,0%) have expected count less than 5. The minimum expected count is
25,54.
Symmetric Measures
Nominal by
Nominal
Phi
Cramer's V
N of Valid Cases
Value
,147
,147
193
Approx. Sig.
,041
,041
a. Not ass umi ng the null hypothesis.
We can then assume that there is
a difference in terms of behavior
between genders.
Males are more privacy
concerned
b. Us ing the asymptotic standard error as sum ing the null
hypothesis.
25
Contingency tables
Privacy Share vs Region
With regards to privacy, both the Phi squared and the Cramer’s index are acceptable and
the test remains significant:
• We can therefore assume that there are different degree of concerns with privacy both
within Italy as outside
•In particular users from the north of Italy seems to be the most privacy conscious ones.
Users from the south and foreigners seems to be far less privacy conscious.
Chi-Square Te sts
Crosstab
Region
Total
North Italy
Count
% within Region
% of Total
Center Italy Count
% within Region
% of Total
South Italy
Count
% within Region
% of Total
Outside Italy Count
% within Region
% of Total
Count
% within Region
% of Total
Privacy_Share
No
Yes
34
57
37,4%
62,6%
17,6%
29,5%
10
11
47,6%
52,4%
5,2%
5,7%
21
9
70,0%
30,0%
10,9%
4,7%
33
18
64,7%
35,3%
17,1%
9,3%
98
95
50,8%
49,2%
50,8%
49,2%
Total
91
100,0%
47,2%
21
100,0%
10,9%
30
100,0%
15,5%
51
100,0%
26,4%
193
100,0%
100,0%
Pearson Chi-Square
Lik elihood Ratio
Linear-by-Linear
As soc iation
N of Valid Cases
Value
15,030 a
15,292
13,033
3
3
As ymp. Sig.
(2-sided)
,002
,002
1
,000
df
193
a. 0 c ells (,0% ) have expected count less than 5. The
minimum expected count is 10, 34.
Symmetric Measures
Nominal by
Nominal
Phi
Cramer's V
N of Valid Cases
Value
,279
,279
193
Approx. Sig.
,002
,002
a. Not ass uming the null hypothesis.
b. Us ing the asymptotic standard error as suming the null
hypothesis.
28
Contingency tables
time to adoption
0
sesso

maggioranza
anticipatrice
early adopters
maggioranza
ritardataria
ritardatari
0
donna
0
14
20
22
22
78
uomo
2
14
33
25
25
99
2
28
53
47
47
177
Total

Total
FIRST TABLE:
At a first sight we
notice that the time
that customers take
to adopt a new
product does not
depend on the
gender.
SECOND TABLE:
We see that that the “gender” variable is not
significant as .537 is higher than 0,05.
Value
Pearson Chi-Square
Likelihood Ratio
Linear-by-Linear Association
df
Asymp. Sig. (2-sided)
3,124(a)
4
,537
3,880
4
,423
,311
1
,577
N of Valid Cases
177
Linear Correlation
2.2 Data elaboration through SPSS
Univariate analysis
Correlation between AGE and FREQUENCY TO CINEMA IN A MONTH
The sample presents a positive linear correlation
between the age and the frequency they go to cinema
in a month (the Pearson correlation is positive). This
is a quite weak correlation and there are some
outliers (as it can be seen from the scatterplot).
Anyway this result is significant for our objectives:
the older the customers are, the more often they go
to cinema, while the youngs tend to be rare users.
Correl ations
frequency to
cinema in a month
age
The Pearson Correlation is positive and the PValue is lower than 0.05: the test is significant
and there is a positive correlation
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
frequency
to cinema
in a month
1
194
.175*
.015
193
*. Correlation is s ignificant at the 0.05 level (2-tailed).
age
.175*
.015
193
1
199
Linear Correlation
Correlation between AGE and how many
movies are seen in a certain kind of cinema
Correlations
Correlations
A multisala
cinema
1
-.408**
.000
199
192
-.408**
1
.000
192
192
age
A multisala cinema
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
**. Correlation is s ignificant at the 0.01 level (2-tailed).
The age is indirectly correlated to
the percentage of movies seen in a
multiplex cinema:
the younger
tend to prefer this kind of cinema
that is quite different from Apollo
in terms of offering and business
model. This is the basis to
understand the needs of the
customers
in
this
specific
demographic segment
Another ess ay
cinema
1
.347**
.000
199
189
.347**
1
.000
189
189
age
age
age
Another ess ay cinema
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
**. Correlation is significant at the 0.01 level (2-tailed).
On the other side, the percentage
of movie seen in an essay cinema
is directly correlated to the age.
The eldest respondents are the
ones who frequent more essay
cinemas. Even in this case the
result is relevant for specific
managerial implications.
Comparison between the means
Comparison between the means
There is significance, even if Eta Squared is low.
We examined also other variables, but we didn’t find
any interesting value in significance and eta
squared.