This study has used causal research design to fulfil the objectives set and hypotheses developed after critical review of the relevant literature. The study has used quantitative data technique for collecting quantitative data by using questionnaire as a primary data collection tool. The researcher also has applied factor analysis in pilot study for the questionnaire for reducing the scale dimensionality and managing the items included in the original scale. The questionnaire was sent to the respondents through email followed by hard copy for better communication with the respondents.
One of the reasons for using google along with the hard copy was that the results obtained from google form could go directly into a shareable spreadsheet. The researcher could get email notification of results and being the mobile-friendly for quick updates simultaneously. The researcher shared questionnaire in google form with an email and a link. The researcher remained informed about all the updates regarding data collection and watched the responses appearing in real time.
3. 1 POPULATION
The population of the study comprised of the university students of public and private sector universities of Pakistan located in four major cities i.e. Karachi, Lahore, Rawalpindi and Islamabad. There were various reasons of selecting this population for the current study. One of the major cause was the availability of android mobile phone through which mobile Facebook advertisement can be seen. Firstly, in Pakistan most of the people did not have android phone and the school or college going students also
89
did not have android mobiles in which Facebook feature is incorporated. Secondly, majority of school or college students in Pakistan did not have email addresses on which they could be accessed. Thirdly, the university students had an idea of mobile Facebook advertisement and had enough power to purchase a product they like in the mobile Facebook advertisements. So, the prime purpose of the current study could only be fulfilled if the researcher had collected data from university students. Hence, the researcher selected university students as the population for this study.
As the population of the study was the university students, the researcher has taken three major categories of food, clothing and electronics products in this study.
There were the following reasons to choose these categories of the products:
i. Most of the companies who are the clients of facebook incorporation deal in food, clothing and electronics products (Harris and Dennis, 2011).
ii. Food, clothing and electronics products are frequently used by the majority of the masses in general and the respondents especially (Raju and Kumar, 2016).
iii. These products of food, clothing and electronics products are demanded by each and every person irrespective of their income level (Raju and Kumar, 2016).
iv. The population of this study consists of the university students, so the products like Car, Air Ticketing, Business/Industrial Goods etc. are irrelevant (Gao et al., 2016).
v. Food, clothing and electronics products are consumer goods with high demand among the masses (Ehrenberg and Smith, 2016).
90
After considering the factors mentioned above, the researcher took food, clothing and electronics products as his prime concern for this study.
3. 2 SAMPLE SIZE
The researcher took 1000 students as sample size of this study on the basis of statistical formulae provided in the table “Sample Size Determination for Given Population” by Krejcie and Morgan (1970). As the researcher did not know the exact size of population, he used statistical formulae to calculate sample size for this study as provided by Krejcie and Morgan (1970). While determining the sample size Margin of Error (Confidence Interval), Confidence Level, and Standard of Deviation were also kept in mind by the researcher.
3. 3 SAMPLING TECHNIQUE
Researcher has used cluster sampling technique for collecting the data from the respondents. The researcher made four (04) clusters of the population. Four cities in which the universities were located, were selected as independent clusters for gathering data and 250 respondents were selected as a sample size from each cluster.
In, cluster sampling, the population was divided into four (04) groups which were called “clusters”. The clusters of the current study were four major cities i.e. Karachi, Lahore, Rawalpindi and Islamabad. After making the clusters, the researcher randomly selected respondents form each cluster as sample.
91
3. 4 MOBILE FACEBOOK USAGE CHARACTERISTICS IN PAKISTAN There are many Facebook usage characteristics on mobile version. Mobile version also provides a wide range of activities just like the desktop version. The largest proportion of Facebook users are people who declare vocational, secondary and higher education – according to Gemius, they make up more than 60 per cent of users in total. The age group 15–24 is the second largest (22 per cent). So, it can be a good market for various corporations. In May, 2016, Pakistan deepens Facebook user base with over 40 million mobile subscribers in the country, 33% of Mobile users of Pakistan. From these, 25 million users use mobile version of Facebook.
3. 5 DESIGN OF QUESTIONNAIRE BEFORE PILOT TESTING 3.5.1. Measurements of Variables
In my study, the related variables include mobile Facebook advertisement, purchase intentions, Ethnocentrism and perceived value. The variable of mobile Facebook advertisement was subdivided into three variable which were: social impression, homepage impression and organic impression. The variable of country of origin was subdivided into national products and foreign products.
Table 1: Measurement of Facebook Advertisement
Variable NO. Contents References
Social Impression
S 1 Products liked by your friends on mobile FB are attractive.
Nielsen’s (2012)
92
S 2 I am interested to fully see the advertisements which are liked by my friends on mobile FB
The book written by Gibs, J., &
Bruich, S.
(2010).
Advertising effectiveness:
understanding the value of a social media impression.
New York, NY: The Nielsen Company.
In 2012, the Nielsen company published a blog post “ Consumer trust in online, S 3 Mobile Facebook advertisements displayed with
likes by your FB friends contain detailed information about the features of the advertised products.
Homepage Impression
H 1 Mobile Facebook advertisements displayed
through pages/groups created by producers contain detailed information about the features of the advertised products.
H 2 Products advertised through mobile Facebook pages/groups created by producers you like are attractive.
H 3 I am interested to fully see the advertisements which are directly present on mobile Facebook pages/groups created by producers
Organic Impression
O 1 Mobile Facebook advertisements displayed with likes and ultimate purchases by your friends contain detailed information about all the features of the advertised products.
O 2 Products liked and purchased by your friends on mobile Facebook are attractive.
93
O 3
I am interested to fully see the advertisements which show that my mobile Facebook friends liked and purchased.
social and mobile advertising grows” http://www.ni elsen.com/us/e n/insights/new s/2012/consu mer-trust-in- online-social- and-mobile- advertising-grows.html
Respondents need to evaluate all items on a five-point Likert type scale, ranging from strongly disagree(1) to strongly agree(5).
Table 2: Measurement of Purchase Intentions
Variable NO. Contents References
Purchase Intentions
P 1 I Shall buy products that are advertised on mobile FB in the near future
Taylor & Hunter (2002)
94
P 2 I desire to buy products that are promoted on mobile FB
Taylor, S. A., &
Hunter, G. L.
(2002) published the research paper,
“The impact of loyalty with e-CRM software and e-services” in International Journal of Service Industry
Management, 13(5), 452-474.
P 3 Advertisements on mobile FB do not increase purchase intent of featured brands P 4 Advertisements on mobile FB have a
positive influence on my purchase decisions
P 5 I would buy the products that are advertised on mobile FB if I had the money
P 6 I do not intend to acquire products that are promoted on mobile FB
P 7 I am likely to buy some of the products that are promoted on mobile FB
P 8 I plan to purchase the products that are advertised on mobile FB
P 9 Advertisements on mobile FB have a negative influence on buying decisions
Respondents need to evaluate all items on a five-point Likert type scale, ranging from strongly disagree(1) to strongly agree(5).
95
Table 3: Measurement of Country of Origin
Variable NO. Contents References
Country of Origin
C 1 Pakistani people should always buy
Pakistani-made products instead of imports.
Shimp, T. and Sharma, S., (1987)
They published a research paper “ Consumer ethnocentrism:
construction and validation of the CETSCALE” in Journal of Marketing Research 24(3):
280–289.
C 2 Only products that are unavailable in Pakistan should be imported.
C 3 Buy Pakistani-made products. Keep Pakistan working
C 4 Pakistani products first, last and foremost.
C 5 Purchasing foreign-made products is un-Pakistani.
C 6 It is not right to purchase foreign products, because it puts Pakistanis out of jobs.
C 7 A real Pakistani should always buy Pakistani-made products.
C 8 We should purchase products manufactured in Pakistan instead of letting other countries get rich of us.
C 9 It is always best to purchase Pakistani products.
C 10 There should be very little trading or
purchasing of products from other countries
96
unless out of necessity.
C 11 Pakistanis should not buy foreign products, because this hurts Pakistani businesses and causes unemployment.
C 12 Curbs should be put on all imports.
C 13 It may cost me in the long-run, but I prefer to support Pakistani products.
C 14 Foreigners should not be allowed to put their products on our markets.
C 15 Foreign products should be taxed heavily to reduce their entry into the Pakistani market.
C 16 We should buy from foreign countries only those products that we cannot obtain within our own country.
C 17 Pakistani consumers who purchase products made in other countries are responsible for putting their fellow Pakistanis out of work.
Respondents need to evaluate all items on a five-point Likert type scale, ranging from strongly disagree(1) to strongly agree(5).
97
Table 4: Measurement of Perceived Value
Variable NO. Contents References
Perceived Value
PV 1 The price of the brand will be very economical for me.
Chowdhury, H. K.,
& Abe, S. (2002).
They published a research paper “ Consumer Perceptions of Value: a Multi-Item Scale For Its Measurement” in ACR Asia-Pacific Advances.
PV 2 The price of the brand will be unacceptable for me.*
PV 3 The price of the brand will appear as bargain.
PV 4 The price to acquire the brand will be reasonable.
PV 5 The brand will have a longer product life.
PV 6 I will not feel superiority by owing this brand.*
PV 7 From my experience of this brand, I feel that I will get quality attributes.
PV 8 I do not feel that the brand for the price will be of good quality.*
PV 9 I do believe that the brand for the price will outperform other brands.
PV 10 I think the brand will contain unique features for the price.
98
PV 11 The purchase of the brand will be a good deal but I have to search for the related information before making the purchase decision.
PV 12 Acquisition of this brand will not be worthwhile if I do not spend time to compare with others.*
PV 13 The purchase of the brand will have a good value, for that I would have to employ a lot of effort.
Respondents need to evaluate all items on a five-point Likert type scale, ranging from strongly disagree(1) to strongly agree(5). The questions with asterisk marks are reverse coded which means that response for these questions will be 1=5, 2=4, 3=3, 4=2, 5=1.
3. 6 PILOT TESTING FOR ITEMS REDUCTION IN QUESTIONNAIRE For decreasing the number of items/questions in each variable, the researcher would a pilot study in which data from 30 respondents will be collected as a pilot test.
The researcher will use factor analysis for data reduction and on the basis of factor loading the researcher will decide to delete items used in questionnaire. The items with low factor loading will be deleted from the original scale.
99
3. 7 QUESTIONNAIRE AFTER PILOT TESTING 3.7.1. Measurements of Variables
After conducting pilot testing, the questionnaire was revised as per the results of Principal Component Factor Analysis (PCA). The questions in the scales of variables of Facebook advertisement, purchase intentions, Ethnocentrism and perceived value were reduced. The variable of mobile Facebook advertisement was subdivided into three variable which are: social impression, homepage impression and organic impression. The variable of country of origin was subdivided into national products and foreign products. The revised version of variables and their questions were as follow:
Table 5: Measurement of Facebook Advertisement
Variable NO. Contents References
Social Impression
S 1 Products liked by your friends on mobile FB are attractive.
Nielsen’s (2012) S 2 I am interested to fully see the advertisements
which are liked by my friends on mobile FB S 3 Mobile Facebook advertisements displayed with
likes by your FB friends contain detailed information about the features of the advertised products.
100
Homepage Impression
H 1 Mobile Facebook advertisements displayed
through pages/groups created by producers contain detailed information about the features of the advertised products.
H 2 Products advertised through mobile Facebook pages/groups created by producers you like are attractive.
H 3 I am interested to fully see the advertisements which are directly present on mobile Facebook pages/groups created by producers
Organic Impression
O 1 Mobile Facebook advertisements displayed with likes and ultimate purchases by your friends contain detailed information about all the features of the advertised products.
O 2 Products liked and purchased by your friends on mobile Facebook are attractive.
O O 3
I am interested to fully see the advertisements which show that my mobile Facebook friends liked and purchased.
Respondents need to evaluate all items on a five-point Likert type scale, ranging from strongly disagree(1) to strongly agree(5).
101
Table 6: Measurement of Purchase Intentions
Variable NO. Contents References
Purchase Intentions
P 1 I Shall buy products that are advertised on mobile FB in the near future
Taylor & Hunter (2002)
P 2 I desire to buy products that are promoted on mobile FB
P 3 Advertisements on mobile FB have a positive influence on my purchase decisions
P 4 I plan to purchase the products that are advertised on mobile FB
Respondents evaluated all items on a five-point Likert type scale, ranging from strongly disagree (1) to strongly agree (5) for the main study.
Table 7: Measurement of Country of Origin
Variable NO. Contents References
Country of Origin
C 1 Pakistani people should always buy
Pakistani-made products instead of imports.
Shimp, T. and Sharma, S., (1987) C 2 Buy Pakistani-made products. Keep
Pakistan working
C 3 Pakistani products first, last and foremost.
102
C 4 Pakistanis should not buy foreign products, because this hurts Pakistani businesses and causes unemployment.
C 5 Curbs should be put on all imports.
C 6 Foreign products should be taxed heavily to reduce their entry into the Pakistani market.
Respondents evaluated all items on a five-point Likert type scale, ranging from strongly disagree (1) to strongly agree (5) for the main study.
Table 8: Measurement of Perceived Value
Variable NO. Contents References
Perceived Value
PV 1 The price of the brand will be very economical for me.
Chowdhury, H. K.,
& Abe, S. (2002).
PV 2 The price to acquire the brand will be reasonable.
PV 3 I do believe that the brand for the price will outperform other brands.
PV 4 The purchase of the brand will be a good deal but I have to search for the related information before making the purchase decision.
103
Respondents evaluated all items on a five-point Likert type scale, ranging from strongly disagree (1) to strongly agree (5) for the main study.
3. 8 RELIABILITY TESTS VALUES
To test the reliability of questionnaire’s items for each scale, the test of reliability Cronbach’s alpha (Cronbach, 1951) was run. Cronbach’s alpha coefficient of reliability usually varies between 0 and 1. But there is no such lower limit to the coefficient. For greater consistency Cronbach’s alpha value should be close to 1. Basic formula for Cronbach’s alpha is _ = rk / [1 + (k -1)r] where the number of items considered are denoted by k is and represents is the mean of the inter-item correlations.
The following rules of thumb were provided by George and Mallery (2010).
1. _ > .9 – Excellent, 2. _ > .8 – Good, 3. _ > .7 – Acceptable, 4. _ > .6 – Questionable, 5. _ > .5 – Poor,
6. _ < .5 – Unacceptable” (p. 231).
In pilot study, data was collected from 42 respondents from Karachi, Lahore, Rawalpindi and Islamabad. The following table showed the results:
Reliability Statistics Cronbach's
Alpha
N of Items
.895 45
104
Overall Reliability Statistics (Cronbach's Alpha) for the study was 0.895 which showed a greater reliability before running the factor analysis.
3. 9 RELIABILITY TESTS FOR PILOT STUDY
To test the reliability of questionnaire’s items for each scale in the main study, the test of reliability Cronbach’s alpha (Cronbach, 1951) was run. In main study, data was collected from 1015respondents from Karachi, Lahore, Rawalpindi and Islamabad. The following table showed the results:
Reliability Statistics Cronbach's
Alpha
N of Items
.891 20
3. 10 SOFTWARE AND STATISTICAL TESTS
The researcher gathered the data from the respondents and used advanced statistical packages like Statistical Package for Social Sciences (SPSS 21.0 version).
Inferential statistics [Correlation and Regression Analysis for measuring effect, Independent Sample T-Test for measuring differences between national and foreign brands and descriptive statistics were used to draw valuable results and conclusions therefrom.
3. 11 SUMMARY OF THE CHAPTER
This chapter discussed the research design which was causal research design in order to fulfil the objectives set and hypotheses developed by the researcher. The
105
researcher has taken university students of public and private sector universities of Pakistan located in four major cities i.e. Karachi, Lahore, Rawalpindi and Islamabad as population of this study. The researcher took 1000 students as sample size of this study.
The researcher also conducted pilot test and measured its reliability which was good enough and the researcher moved forward for the main study.
106