PAF Karachi Institute of Economics &

Technology

Report
On
Prime time viewership

Submitted to
Sir Rizwan Farooqui
Submitted by
Group: Kietians
Hassan Raza Haider (52847)

Muhammad Ali (52454)

Furqan Hussain Jaffrani (52509)

Furqan Ahmed (52434)

 Muhammad Kashif (52940)

S.no Contents p.g no

1. Title page 1
2. Acknowledgment 3
3. Introduction 4
4. Purpose of Study 5
5. Methodology 5
6. Information from Data 6
7. Descriptive working on Data 8
8. Hypothesis Testing for 8
9. Goodness of Fit Test 10
10. Correlation Coefficient 12
11. Regression Analysis 15
12. Conclusion 16

Statistics & Inference Page 2

 ACKNOWLEDGEMENT

I would like to thank everyone who helped me and guided me on this report. I would
specially like to express my warm gratitude towards the following persons:

Mr. Rizwan Farooqui,Professor at PAF KIET for giving us the concept and working,
and for helping and guiding us throughout the semester.

Last but not the least to my dear friends who provided me the latest knowledge and
information within a short span of time.

Statistics & Inference Page 3

INTRODUCTION:

PAKISTAN MEDIA INDUSTRY

In the worst situation regarding economy and terrorism in Pakistan, if there is an

institution which is progressing, is on the rise and capturing great heights, it is none
other than Pakistan fast growing electronic media. Starting with the on set of 21st
century our Electronic media is almost a decade old now with educated youngsters
joining media and we can witness information and knowledge spreading to all corners.
It is estimated that on an average day approximately 38 Million Pakistanis of age 10+
view cable and satellite TV. The estimated viewership of various genres of TV on an
average day is the following (because of multiple viewership the total will be greater
than the addition of various genre viewer ships):

Infotainment Channels 26 Million

News Channels 7 Million

Religious Channels 4 Million

Music Channels 500,000

Business Channels 10,000

The last few years have been very active and promising for the TV industry, the
number of channels, the size of audience, the variety and independence of media
content, have all seen progress and development. Thus one hopes that the current
recessionary trend in the global market that is affecting everyone will not damage the
general direction of our development, even though there might be temporary setbacks.
It is in this context that research can play a crucial role. Quoting from a recent
marketing conference, recessionary times call for the effective allocation of resources
even more than normal or booming times do. These are also the times to think out of
the box. Research is an important tool to help bring about ignored efficiencies and
new paradigms.

Rising Size of Audience:

Our latest research shows that TV audiences have risen dramatically since 2004. From
an estimated number of 63million viewers in 2004, audiences have risen to an
estimated number of 86 million by 2009. Most of this increase has happened in the
rural areas, many of which were previously excluded from TV viewing due to lack of
electricity. Despite the poor quality of supply, the electrification of rural areas has

Statistics & Inference Page 4

moved rapidly in the last five years. In the new scenario there are 52 million TV
viewers in the rural areas of Pakistan, 16million in small and medium size cities and18
million in large and metropolitan cities, adding up to a total of 86 million viewers of
age 10 years and above.

PURPOSE OF STUDY
This report is based upon Statistical and Inferential analysis of Prime Time
Viewership of different Pakistani Television Channels. The main purpose is to
find out is there a strong relationship between the prime time slot and viewership
share of different Television Channels. We will also be testing different hypothesis and
whether how much the obtained data is matched to our expected values. These data
and analysis has a direct bearing on the media and our industry. Such studies are very
important for market researchers and entrepreneurs because the majority of TV
audiences and potential buyers of goods and services advertised on TVs are in the
villages and towns as well as in small and medium sized cities of Pakistan. They
constitute nearly 80% of TV audience in the country. They also constitute nearly 70%
of the purchasing power of goods and services advertised on TV. The remaining 20%
of TV audience and around 30% of purchasing power lies in the metropolitan and
large cities of the country.

Methodology
Some of the statistical methods used are as follows:

Descriptive statistics

Hypothesis Testing for

Chi Square Goodness of Fit Test

Correlation

Regression

Sampling Technique:
By Secondary Data
www.medialogic.com.pk

Statistics & Inference Page 5

About Data Source:
The data has been taken from the Medialogic. The general survey is done by them
about the viewership of television duration and the out come was the primetime of the
viewership starts from 20: 00- 21:59. Medialogic was set up in late 2006 in order to
address a growing need for accurate, reliable and actionable TV audience data. It has
teamed up with one of the biggest names in Media Research Telecontrol GfK to bring
cutting edge technology in people meters to Pakistan. Telecontrol is a subsidiary of
global research powerhouse GfK that specializes in TV measurement. It has installed a
panel of approximately 600 meters across urban and rural areas of Pakistan to gauge
second-to-second viewership data for more than 4,000 individuals residing in
Pakistan.

Information Gathered from the Data

Time Slot 20:00 - 21:59

Channels
Ratings Share%
Ary News 0.10 4.00
Aaj TV 0.10 4.00
GEO News 1.00 40.00
Exp News 0.20 8.00
Samaa 0.00 0.00
Indus News 0.00 0.00
Dawn News 0.00 0.00
News1 0.10 4.00
ARY Digital 0.10 4.00
GEO Entertainment 0.20 8.00
Hum 0.40 16.00
TV one 0.10 4.00
Indus Vision 0.00 0.00
The Musik 0.00 0.00
MTV Pak 0.00 0.00
Aag 0.00 0.00
Play 0.00 0.00
HBO 0.10 4.00

Statistics &FTV Pak Page 6

Inference 0.00 0.00
QTV 0.10 4.00
STATISTICAL WORKING ON DATA

Serial Name of
No. Element Actual Results

01 Mean 05

02 Median 04

03 Mode 00

04 Range 40

05 Minimum 00

06 Maximum 40

Standard 9.7
Deviation
07

Coefficient of 183.30%
Variation
08 (or 1.833)

09 Skewness + 3.28

HYPOTHESIS TESTING FOR

Statistics & Inference Page 7

Consider the scenario in which there is the claim that the Average share in viewership
for any TV Channel in Prime Time Slot is less than 6 %. We selected the above
sample of 20 TV Channels along with their share % of viewership in primetime slot.
Average share per TV Channel is 5. The value of = 0.10 and the standard deviation
is 9.7 (calculated above).

The given data is:-

=6

=5

n = 20 = 0.10 s = 9.7

Step # 01: -

The hypothesis and Claim is: -

Ho: 6 And H1: < 6 (Claim)

Step # 02: -

The value of = 0.10; the test is left tailed test. The critical value is

t = -1.328.

Step # 03: -

The formula for test the value is: -

t cal = ( - ) / (s/n)

= (5 - 6) / (9.7/20)

= - 0.46

Step # 04: -

Statistics & Inference Page 8

As t calculated value lies in the acceptance region, the decision is to accept the
null hypothesis Ho.

Step # 05: -

There is not enough evidence to support the claim that average share in
viewership for any TV Channel in Prime Time Slot is less than 6 %.
Tv Channels Observed values
Ary News 0.10
GOODNESS OF FIT
Aaj TV 0.10
TEST
GEO News 1.00
ExpNews 0.20 From the above study the
Samaa 0.00 observed values are
IndusNews 0.00
reported. It is summarized
through table given below:
Dawn News 0.00
News1 0.10
ARYDigital 0.10
GEO Ent 0.20
Hum 0.40
TV one 0.10
IndusVision 0.00
TheMusik 0.00
MTV Pak 0.00
Aag 0.00
Play 0.00
Statistics & Inference Page 9
HBO 0.10
FTV Pak 0.00
QTV 0.10
At = 0.10, is there sufficient evidence that the proportion differ from those
stated in the given data.

Step # 01: - The hypothesis and the Claim is: -

Ho: proportion is distributed same as reported (Claim)

H1: proportion is not distributed same as reported.

Step # 02: -

The value of = 0.10. The degree of freedom is 20-1 = 19. The Chi tab value is
27.204

Step # 03: -
% of
Expecte (O - Chisq
observed d O-E E) / E
4.70
0.1 0.050 0.050 0.050
1.57
0.1 0.150 -0.050 0.017
25.08
1 0.600 0.400 0.267
9.40
0.2 0.100 0.100 0.100
4.70
0 0.050 -0.050 0.050
0 0.050 -0.050 0.050 4.70

Statistics & Inference Page 10

 4.70
0 0.050 -0.050 0.050
0.78
0.1 0.075 0.025 0.008
0.47
0.1 0.125 -0.025 0.005
1.57
0.2 0.150 0.050 0.017
0.00
0.4 0.400 0.000 0.000
4.70
0.1 0.050 0.050 0.050
4.70
0 0.050 -0.050 0.050
4.70
0 0.050 -0.050 0.050
4.70
0 0.050 -0.050 0.050
9.40
0 0.100 -0.100 0.100
4.70
0 0.050 -0.050 0.050
0.00
0.1 0.100 0.000 0.000
4.70
0 0.050 -0.050 0.050
4.70
0.1 0.200 -0.100 0.050
100.00
2.5 2.500 0.000 1.063

Chi square calculated value = 1.06

Step # 04: -

Statistics & Inference Page 11

As chi square value lies in the acceptance region, so accept the null hypothesis.

Step # 05: -

There is enough evidence to support the claim that the proportion is

distributed as reported.

The Correlation Coefficient:

It is computed from the sample data measures the strength and directions of a
linear relationship between two variables.

I used the Regression and Correlation Analysis to study and measure the
statistical relationship that exists between two variables (Ratings and
Viewership Share)

Solution: To find out the values of variables;

20:00 - 21:59
Time Slot
Ratings (x) Share%(y) (xy) x y

AryOneWor 0.10 4.00 0.4 0.01 16

Statistics & Inference Page 12

Aaj TV 0.10 4.00 0.4 0.01 16

GEO News 1.00 40.00 40 1 1,600

ExpNews 0.20 8.00 1.6 0.04 64

Samaa 0.00 0.00 0 0 0

IndusNews 0.00 0.00 0 0 0

Dawn News 0.00 0.00 0 0 0

News1 0.10 4.00 0.4 0.01 16

ARYDigital 0.10 4.00 0.4 0.01 16

GEO Ent 0.20 8.00 1.6 0.04 64

Hum 0.40 16.00 6.4 0.16 1,600

TV one 0.10 4.00 0.4 0.01 16

IndusVision 0.00 0.00 0 0 0

TheMusik 0.00 0.00 0 0 0

MTV Pak 0.00 0.00 0 0 0

Aag 0.00 0.00 0 0 0

Play 0.00 0.00 0 0 0

HBO 0.10 4.00 0.4 0.01 16

FTV Pak 0.00 0.00 0 0 0

QTV 0.10 4.00 0.4 0.01 16

Total 2.50 100.00 52.4 1.31 3440

Now the Data is; n = 20

(xy) = 52.4

(x) = 2.50

Statistics & Inference Page 13

 (y) = 100

(x2) = 1.31

(y2) = 3440

By putting the values in the Correlation Formula;

r = n (xy) ( x) ( y) / [n (x2) ( x)2] [n (y2) ( y)2]

r =20(52.4)-(2.50)(100) /

[20(1.31)-(2.50)2] [20(3440)-(100)2]

r = 798/ 19.95 58800

r = 798 /1083.07

r = 0.736

Result:
There is positive correlation exists between the Prime Time and the viewership
shares.

Regression Analysis for Viewership Share

In Regression Analysis, an estimating or predicting equation is developed to describe
the pattern or functional nature of the relationship that exist between the variables.

We will compute an estimating or regression equation to describe the relationship

between the variables. My interest here is limited to an analysis of simple linear
regression of y on x. The formula or equation that we will use to predict the value of
Inflation is;

y = a + bx

Statistics & Inference Page 14

Here y = a computed estimate of the viewership share

a = the y intercept or the value of y when x is equal to zero

b = the slope of the regression line, or the increase or decrease in y for

each change of one unit of x

x = a given value of independent variable (prime time rating)

Solution:
The values need for equation are

n= 20, x= 2.50, y= 100.00,

xy = 52.4, x = 1.31, y = 3440

The values of a and b in the regression equation are

computed with the following formulas;

b = n (xy) ( x) ( y) / n (x2) ( x)2

a = ( y / n) - b ( x / n) or a = - b

By putting the data in the formulas;

b = 40

a=0

Hence the equation of the regression line y = a+ bx

is y = 40x

Graph Plotted :

Statistics & Inference Page 15

Conclusion:
The above statistical analysis has clearly shown that there is strong relationship
between Prime time slot and viewership. The rating of the channels actually depends
quite heavily on the viewership of Prime time. Hence not only the concerned parties
from TV channels side must work upon it but also the marketers and advertisers can
pay attention to this most important time slot in order to promote their advertising
campaigns and increase the sale of their products.

Statistics & Inference Page 16