Professional Documents
Culture Documents
As one of our objective to find out the preference of the respondents, hence we
decided to carry out cross-tabulation on the basis of the satisfaction level of the
respondents and their frequency to watch a movie in the theatre in a month. The
area in which the preferences were looked after were
2) What kind of actors do they prefer to watch for particular script or director’s
philosophy?
About 92 % of the respondents watches movie for entertainment purpose and at the
same time they want that it should be more close to reality. It is also well clear from
the type of script they prefer to watch. From the graph below, we can see that
maximum number of respondents wants the script of the film should be close to
reality or current issues prevalent in the society.
It is not that other scripts or philosophy wil not work, for example Dhoom was a
Trendy and stylish kind of movie with all fast music numbers and comparatively
new face, but movie like 3 – Idiots, Taare Zameen Par etc were more preferred with
a repeat audience. The main USP of 3 – idiot was all the factors as preferred by the
respondents.
The above survey does not indicate that our TG group wants a real life story in a
serious manner. They want that this philosophy should be explained in a light way
or in comic way. This is clear from the graph below. Maximum number of
respondents prefered comedy movies over other genres like romance, action,
thriller etc.
On this front also, 3 – Idiots scores well. The movie brought about the problem of
graduating youth in a light humorous way which was well accepted by our TG.
From cross-tabulation, it is clear that maximum respondent prefered comedy.
However, there were many respondents who liked action, comedy, romance and so
on. Hence combining these genres, we found out that maximum respondents who
has multiple genre choices, majority of them prefered either Romantic Comedy or
action comedy. Few of them prefered Action Romance Comedy.
Multiple Regression to find relationship between Satisfaction level of the Respondent and
the main attributes of the film
Variables Entered/Removed(b)
Variables Variables
Model Entered Removed Method
1
Location,
Hero, Film
Title, Genre,
Director,
. Enter
Heroine,
Songs,
Prod.House
, Story(a)
Model Summary
Coefficients
Unstandardized Standardized
Coefficients Coefficients
The main aim of carrying out multiple regression was to find a correlation between the
satisfaction level of the respondent with the various attributes of the movie such as genre, title,
actor, director etc.
But as seen from the table, after running multiple regression in SPSS, it is clear that the change
in the satisfaction level is explained by only 11.6% change in the above mentioned independent
variable. Hence, there may be other variables affecting the satisfaction level of the respondents.
Checking out the multicollinearity, it was found that there was much any relation between the
independent variables hence clarifying that each variable is independent with respect to each
other.
In order to find out, whether the variables such as TV promo, Music, PR activities, Critics
Review, Word of Mouth etc have an equal impact on the respondents in influencing them to
watch a particular movie, we first applied one way ANNOVA test.
Alternative Hypothesis :- All the promotional effort have different impact on influencing
people to watch a movie.
ANOVA
Source of P-
Variation SS df MS F value F crit
Between 24.4065 2.2266
Groups 99.25 5 19.85 78 1E-22 49
0.81330
Within Groups 580.7 714 53
679.9
Total 5 719
As F(observed) > F Critical, our null hypothesis is rejected, and it is clear that all promotional effort
has different impact on influencing people to watch a movie.
To find out which of these promotional variables should be focused more, which can highly
influence the people. For that we carried out Factor Analysis.
Factor Analysis
Methodology
2. Using Principle component method for data extraction and saving factor scores as
variables to be formed basis for cluster analysis
The Output and Analysis
Comp Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings
onent Total % of Variance Cumulative % Total % of Variance Cumulative % Total % of Variance Cumulative %
Component
1 2
From the table of Total Variance Explained, we found out only two factors which explain about
53% cumulative variance of the variables given. This is mainly due to the fact that all the
variables are not having much difference. Hence we can say that all the promotional variables
can be grouped in to two factors namely Direct Promotion and Indirect Promotion
Direct Influence :- As seen from the Rotated Component Matrix, TV Promo, Music and PR
Activities are the variables which are related to the above factor.
Indirect Influence :- As seen from the Rotated Component Matrix, Critic review, Word of
mouth, Controversies are the variables which are related to the above factor.
Cluster Analysis
We carried out hierarchical clustering using agglomeration scheduling. We used Dendogram plot
to find out the clusters. Warde method of linkage was used and squared Euclidean distance used
as a basis to find out the clusters.
Agglomeration Schedule
1 55 113 .000 0 0 67
2 77 110 .000 0 0 7
5 93 94 .000 0 0 75
6 20 92 .000 0 0 15
7 18 77 .000 0 2 32
8 34 68 .000 0 0 79
9 57 66 .000 0 0 19
10 44 56 .000 0 0 46
11 3 47 .000 0 0 78
12 31 33 .000 0 0 40
13 28 29 .000 0 0 90
14 12 27 .000 0 0 45
15 19 20 .000 0 6 43
16 1 4 .000 0 0 65
17 52 114 .000 0 0 39
18 39 116 .001 0 0 50
19 57 62 .002 9 0 86
20 6 8 .004 0 0 56
21 43 72 .007 0 0 73
22 79 87 .010 0 0 44
23 40 83 .013 0 0 27
24 46 112 .015 0 0 76
25 64 89 .018 0 0 58
26 9 80 .021 0 0 48
27 7 40 .025 0 23 52
28 96 120 .029 0 0 41
29 23 100 .032 0 4 47
30 98 103 .036 0 0 68
31 10 76 .040 0 0 39
Dendrogram using Ward Method
C A S E 0 5 10 15 20 25
Label Num +---------+---------+---------+---------+---------+
55 ─┐
113 ─┤
39 ─┤
116 ─┤
11 ─┤
34 ─┤
68 ─┼─┐
16 ─┤ │
118 ─┤ │
13 ─┤ │
25 ─┤ │
3 ─┤ │
47 ─┤ ├─────┐
9 ─┤ │ │
80 ─┤ │ │
85 ─┘ │ │
100 ─┐ │ │
101 ─┤ │ │
23 ─┤ │ ├─────────────────────┐
50 ─┼─┘ │ │
95 ─┘ │ │
97 ─┬─┐ │ │
108 ─┘ │ │ │
86 ─┐ │ │ │
109 ─┤ │ │ │
1 ─┤ ├─────┘ │
4 ─┤ │ │
78 ─┤ │ │
107 ─┤ │ │
40 ─┼─┘ │
83 ─┤ │
7 ─┤ │
14 ─┤ │
117 ─┤ │
57 ─┤ │
66 ─┤ │
62 ─┤ │
44 ─┤ │
56 ─┤ │
59 ─┤ │
58 ─┘ │
98 ─┐ │
103 ─┤ ├─────────────────┐
61 ─┤ │ │
48 ─┤ │ │
6 ─┼─┐ │ │
8 ─┤ │ │ │
67 ─┤ │ │ │
102 ─┤ │ │ │
91 ─┤ │ │ │
46 ─┤ │ │ │
112 ─┤ │ │ │
2 ─┤ │ │ │
36 ─┘ │ │ │
96 ─┐ │ │ │
120 ─┤ ├─────────┐ │ │
63 ─┤ │ │ │ │
71 ─┤ │ │ │ │
77 ─┤ │ │ │ │
110 ─┤ │ │ │ │
18 ─┤ │ │ │ │
88 ─┤ │ │ │ │
74 ─┤ │ │ │ │
79 ─┤ │ │ │ │
87 ─┤ │ │ │ │
37 ─┼─┘ │ │ │
12 ─┤ │ │ │
27 ─┤ ├─────────────────┘ │
5 ─┤ │ │
64 ─┤ │ │
89 ─┤ │ │
75 ─┤ │ │
82 ─┤ │ │
99 ─┘ │ │
41 ─┬─────┐ │ │
54 ─┘ │ │ │
31 ─┐ │ │ │
33 ─┤ │ │ │
21 ─┤ │ │ │
115 ─┤ │ │ │
24 ─┤ ├─────┘ │
119 ─┤ │ │
70 ─┤ │ │
43 ─┤ │ │
72 ─┤ │ │
26 ─┤ │ │
52 ─┼─────┘ │
114 ─┤ │
10 ─┤ │
76 ─┤ │
53 ─┤ │
65 ─┤ │
32 ─┤ │
111 ─┤ │
38 ─┤ │
106 ─┤ │
49 ─┤ │
22 ─┘ │
93 ─┐ │
94 ─┤ │
104 ─┤ │
30 ─┤ │
81 ─┼─────────────┐ │
84 ─┤ │ │
60 ─┤ │ │
45 ─┤ │ │
42 ─┘ │ │
35 ─┐ │ │
51 ─┤ ├─────────────────────────────────┘
73 ─┤ │
28 ─┼─┐ │
29 ─┤ │ │
20 ─┤ │ │
92 ─┤ │ │
19 ─┤ ├───────────┘
105 ─┤ │
17 ─┘ │
15 ─┐ │
69 ─┼─┘
90 ─┘
Based on the coefficients and looking at the Dendogram, we found out that there are two clusters
with different characteristics.
The above table shows the mean of the values for the given variables for each cluster. It is seen
from table that clusters – 2 which has maximum respondents music has the high mean value as
compared to cluster -1. Also there is not much significant difference in the mean values for TV
promo and PR activities.
From the cluster Analysis, we can interpret that Direct Influencing variables like TV
Promo, Music, PR Activities should be given more importance while designing promotion
campaign in order to influence people to watch movies.
Analysis of Q – 5 - Factors influencing the respondents to watch a movie
Genre
Film Title
Story
Lead Hero
Lead
Heroine
Director
Production
House
Songs
Sets/Locatio
n
have an equal impact on the respondents in influencing them to watch a particular movie, we
first applied one way ANNOVA test.
Null Hypothesis :- All the variables mentioned above create an equal impact in influencing
people to watch a movie.
Alternative Hypothesis :- All the promotional effort have different impact on influencing
people to watch a movie.
ANOVA
Source of
Variation SS df MS F P-value F crit
5.0650 3.36792E- 1.9470
Between Groups 55.56296296 8 6.94537037 22 06 33
Within Groups 1468.6 1071 1.37124183
As F(observed) > F Critical, our null hypothesis is rejected, and it is clear that all the different factor
has different impact on influencing people to watch a movie.
To find out which of these promotional variables should be focused more, which can highly
influence the people. For that we carried out Factor Analysis.
Factor Analysis
Methodology
1. Different kind of variables such as Genre, Story, Lead Hero etc as attributes on basis of
which factoring is to be done.
2. Using Principle component method for data extraction and saving factor scores as
variables to be formed basis for cluster analysis
Cast, Crew& Content: - As seen from the Rotated Component Matrix, Story, Lead Hero,
Lead Heroine and Director are the variables which are related to the above factor.
Film Identity: - As seen from the Rotated Component Matrix, Film Title and Production
House are the variables which are related to the above factor.
X-factor:- As seen from the Rotated Component Matrix, Songs and Sets/Location are the
variables which are related to the above factor.
Cluster Analysis
We carried out hierarchical clustering using agglomeration scheduling. We used Dendogram plot
to find out the clusters. Warde method of linkage was used and squared Euclidean distance used
as a basis to find out the clusters.
Agglomeration Schedule
2 48 49 .000 0 0 57
3 43 46 .000 0 0 62
4 28 29 .000 0 0 107
5 67 84 .004 0 0 72
6 15 117 .016 0 0 21
7 61 64 .029 0 0 102
8 80 94 .046 0 0 88
9 74 114 .066 0 0 32
10 12 16 .090 0 0 82
11 7 23 .116 0 0 45
12 13 95 .145 0 0 16
13 2 31 .175 0 0 22
14 17 33 .212 0 0 56
15 86 112 .255 0 0 32
16 4 13 .303 0 12 43
17 5 11 .350 0 0 69
18 14 90 .399 0 0 67
19 51 91 .449 0 0 27
20 27 89 .501 0 0 60
21 3 15 .554 0 6 30
22 2 75 .608 13 0 56
23 9 120 .662 0 0 42
24 98 105 .722 0 0 89
25 38 42 .783 0 0 70
26 93 99 .845 0 0 65
27 19 51 .921 0 19 60
28 81 115 .996 0 0 64
29 10 56 1.072 0 0 46
30 3 72 1.156 21 0 85
31 34 96 1.241 0 0 39
Dendrogram
* * * * * * * * * * * * * * * * * * * H I E R A R C H I C A L C L U S T E R A N A L
Y S I S * * * * * * * * * * * * * * * * * * *
C A S E 0 5 10 15 20 25
Label Num +---------+---------+---------+---------+---------+
100 ─┐
101 ─┤
83 ─┤
12 ─┤
16 ─┼─┐
24 ─┤ │
70 ─┤ │
107 ─┘ │
77 ─┐ │
118 ─┤ │
102 ─┤ ├─────────┐
81 ─┤ │ │
115 ─┤ │ │
27 ─┤ │ │
89 ─┤ │ │
51 ─┤ │ │
91 ─┼─┘ │
19 ─┤ │
15 ─┤ │
117 ─┤ │
3 ─┤ │
72 ─┤ │
74 ─┤ ├─────────────────────────┐
114 ─┤ │ │
86 ─┤ │ │
112 ─┤ │ │
26 ─┘ │ │
52 ─┐ │ │
116 ─┤ │ │
108 ─┤ │ │
82 ─┤ │ │
9 ─┼─┐ │ │
120 ─┤ │ │ │
111 ─┤ │ │ │
53 ─┘ ├─────────┘ │
93 ─┐ │ ├───────┐
99 ─┤ │ │ │
25 ─┼─┘ │ │
79 ─┤ │ │
97 ─┤ │ │
110 ─┘ │ │
73 ─┐ │ │
104 ─┼─┐ │ │
85 ─┘ │ │ │
88 ─┐ ├───────┐ │ │
109 ─┤ │ │ │ │
6 ─┼─┘ │ │ │
5 ─┤ │ │ │
11 ─┤ │ │ │
78 ─┘ ├───────────────────────────┘ │
98 ─┐ │ │
105 ─┼─────┐ │ │
30 ─┤ │ │ │
76 ─┤ │ │ │
8 ─┘ │ │ │
34 ─┐ ├───┘ │
96 ─┤ │ │
66 ─┼─┐ │ │
22 ─┤ │ │ │
92 ─┘ │ │ ├─┐
17 ─┐ ├───┘ │ │
33 ─┤ │ │ │
2 ─┤ │ │ │
31 ─┤ │ │ │
75 ─┼─┘ │ │
80 ─┤ │ │
94 ─┤ │ │
13 ─┤ │ │
95 ─┤ │ │
4 ─┤ │ │
119 ─┤ │ │
67 ─┤ │ │
84 ─┤ │ │
65 ─┤ │ │
103 ─┘ │ │
28 ─┬───┐ │ │
29 ─┘ │ │ │
14 ─┐ ├─────────────┐ │ │
90 ─┤ │ │ │ │
71 ─┤ │ │ │ │
113 ─┼───┘ │ │ │
7 ─┤ │ │ │
23 ─┤ │ │ │
18 ─┤ │ │ │
87 ─┤ ├───────────────────────────┘ │
1 ─┘ │ │
48 ─┐ │ │
49 ─┤ │ │
39 ─┤ │ │
50 ─┼─────┐ │ │
35 ─┘ │ │ │
61 ─┬─┐ ├───────────┘ │
64 ─┘ │ │ │
55 ─┐ ├───┘ │
62 ─┤ │ │
106 ─┼─┘ │
60 ─┘ │
10 ─┐ │
56 ─┼─┐ │
69 ─┘ │ │
38 ─┐ ├───┐ │
42 ─┤ │ │ │
47 ─┼─┘ │ │
68 ─┤ │ │
37 ─┤ │ │
40 ─┘ │ │
43 ─┐ ├─────────────────────────────────────────┘
46 ─┤ │
41 ─┼─┐ │
63 ─┘ │ │
45 ─┐ │ │
54 ─┤ │ │
36 ─┤ │ │
58 ─┼─┼───┘
44 ─┤ │
57 ─┘ │
59 ───┘
Based on the coefficients and looking at the Dendogram, we found out that there are two clusters
with different characteristics.
The above table shows the mean of the values for the given variables for each cluster. It is seen
from table that clusters – 1 which has maximum respondents Genre, Story, Lead Hero, Lead
Heroine, Director and Production House has the high mean value as compared to cluster -2
and3.
From the cluster Analysis, we can interpret that Cast, Crew and Content factor should be
given more importance while making a movie in order to influence people to watch movies.
Que-11-Satisfaction Level with the Movie
Here we used one sample z-test to find out about the satisfaction level of the respondent
with the movie released in the recent past.
So we will reject the Null Hpothesis and will accept the Alternative Hypothesis.
It means the most of the people in sample believe that they are not satisfied with the kind of
movie in the recent past.