25
McQuail’s (1983) classification of the following common reasons for media use:
Information
Finding out about relevant events and conditions in immediate surroundings, society and the world
Seeking advice on practical matters or opinion and decision choices
Satisfying curiosity and general interest
Learning; self-education
Gaining a sense of security through knowledge
Personality Identity
Finding reinforcement for personal values
Finding models of behavior
Identifying with valued others (in the media)
Gaining insight into oneself
Integration and Social Interaction
Having a substitute for real-life companionship
Helping to carry out social roles
Enabling one to connect with family, friends and society
Gaining insight into the circumstances of others; social empathy
Identifying with others and gaining a sense of belonging
Finding a basis for conversation and social interaction
Entertainment
Escaping, or being diverted, from problems
Relaxing
Getting intrinsic cultural or aesthetic enjoyment
Filling time
Emotional release
Sexual arousal
26
Katz, Gurevitch and Haas (1973) developed 35 needs taken from the social and psychological functions of the mass media and put them into five categories:
Cognitive needs,including acquiring information, knowledge and understanding
Affective needs, including emotion, pleasure and feelings
Personal integrative needs, including credibility, stability and status
Social integrative needs, including interacting with family and friends
Tension release needs, including escape and diversion.
2.3) Users’ motives for engaging in social media
Many studies on social media and user-generated media apply a uses and gratification approach (c.f. Shao, 2009; Park et al., 2009). This approach focuses on consumers’motives for using a certain media and on the consequences thatfollow from those motives (Blumler and Katz, 1974). Thegratification studies exploring social media show similarfindings. Stafford, Stafford and Schkade (2004) found that consumers have three main gratifications or motives for usingthe internet as a medium, namely, information, entertainment, and social aspects. This finding has been supported andextended by more recent research on user-generated media, which has identified information, entertainment, socialinteraction and community development, self-actualization, and self-expression as motives (Shao, 2009; Courtois et al., 2009).
Krishnamurthy and Dou (2008) summarized the motivations into two main groups:
rational motives, such asknowledge-sharing and advocacy, and emotional motives, such as social connection and self-expression. Park et al.(2009) found four motives for using social networking sites:socializing, entertainment, self-status seeking, and information.
27
Chapter 3 Methodology
Due to the scant research on social activities, an exploratory study design was chosen. A content analysis method was used to capture the contents which users post on their Facebook wall. This paper is not an experimental design so we use content analysis method to analyze and classify types of post on Facebook wall. Holsti (1969) offers a broad definition of content analysis as, "any technique for making inferences by objectively and systematically identifying specified characteristics of messages" (p. 14).
Content analysis has been defined as a systematic, replicable technique for compressing many words of text into fewer content categories based on explicit rules of coding (Berelson, 1952; GAO, 1996; Krippendorff, 1980; and Weber, 1990).Content analysis enables researchers to sift through large volumes of data with relative ease in a systematic fashion (GAO, 1996). It can be a useful technique for allowing us to discover and describe the focus of individual, group, institutional, or social attention (Weber, 1990). It also allows inferences to be made which can then be corroborated using other methods of data collection.
3.1) Content Analysis
The method of content analysis allows researchers to include huge amounts of textual information and systematically identify its properties, for example: the frequencies of most used keywords by locating the more important structures of its communication content. Content analysis has become an important tool in the measurement of success in the public relations programs and the assessment of media profiles and in these circumstances, content analysis is an element of media analysis ("Methods for Media Analysis". ReStore. Economic and Social Research Council.
Retrieved 13 June 2013.)
28
Uses of content analysis (Three basic categories by Holsti):
1. Make inferences about the antecedents of a communication
2. Describe and make inferences about characteristics of a communication 3. Make inferences about the effects of a communication
In qualitative data, content analysis can relate any kind of analysis where communication content such as speech, text, images etc. which is categories and classified. For reliability in content analysis, Neuendorf suggest that when people coders are used in content analysis two coders should be used. Reliability of people coding is often measured using a statistical measure of intercoder reliability or “the amount of agreement or correspondence two or more coders” (Neuendorf, Kimberly A. ,2002).
3.2) Concept Model of Social Media Activities
Due to the scant research on Facebook activities, it is difficult to find the theories or frameworks that completely suitable to this study. We found a conceptual model of social media activities (K. Heinonen, 2011) which based on Shao’s theory (2009). They propose that consumers’ social media activities can be based on two major dimensions: consumer motivation and consumer input. The separation of activities into three main types, i.e. consumption, participation, and contribution (Shao, 2009), provides general information about what consumers do, i.e. it illustrates consumers’
input. Examining consumer motivations in termsof entertainment, social connection, and information provides more details concerning the nature of three main activity types identified by Shao. In Figure 6, the two perspectives arecombined in a simple 3x3 matrix. Nine potential outcomes followfrom this categorization. Activities to the right involve a higher level of consumer input in terms of content contribution. The activities are based on motivation range from utilitarian to hedonic motives.
29
Figure 6. Model of Consumers’ social media activities (K.Heinonen,2011)
K. Heinonen (2011) applied this conceptual model to her study. The researcher collected data by using diary method. A diary method was usedto capture the voice of the customer (Griffin & Hauser, 1993;Bolger, Davis & Rafaeli, 2003). The diary method has been described as a method that captures life as it is lived where respondents are asked to report on daily events and experiences (Bolger, Davis
&Rafaeli, 2003).
3.3) Research Design
This study has used content analysis and empirical method for collect data. We selected two hundred people from Facebook friend’s list. We perceived their wall post and classified types of post such as Status update, Check-in, Selfie, Kid photo, Pet, Criticism, Relationship, Thought and feeling, (Re)Sharing information etc. We could classify it to more than fifty items, we gathered the same categories into one type. For example: there are many check-in such as home, restaurant, school, company, well-known place, gym etc., so named this category is Check-in. We can categorize these items into fifteen types. For the first time, we categorized to 17 items, but we decided to cut kid and pet out because these to items are special items. Kid and pet are limited item because there are only few people who have kid or pet. These two items are not the
Social surveillance Belonging and bonding Creating and Sharing and Being up-to-date Managing a social Experiencing with network others Staying in touch
Retrieving product Applying knowledge information Sharing and Collecting factual accessing opinions, Information reviews and rating
30 3.4) Sampling and Data Collection Procedure
After we categorized the types of post into fifteen types, we had two judges for collect the data. We selected two hundred people from our Facebook account and perceive their post on their Facebook wall. Two judges separated record the data for check the reliability. We selected 2014 all year (2014-01-01 to 2014-31-12) for collect data and two judges start to collect to data at the same period. For reliability, the alpha coefficient for the thirty items (fifteen pairs) is 0.903 ( 0.7), suggesting that the items have relatively high internal consistency. The intraclass correlation of average measures is 0.903 indicates that measures of type posting has a very high level of agreement.
Table 1 Reliability Statistics
Cronbach’s Alpha N of Items
0.903 30
Table 1.1. Intraclass Correlattion Coefficient
We used Factor Analysis for classify the type of posting. The KMO measure is 0.811 and Bartlett’s test of Sphericity is significant. That is, its associate probability is less than 0.05 (Table 2).
Table 2. KMO and Bartlett’s Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy 0.811 Bartlett’s Test Approx. Chi-Square 1157.863 Sphericity df 105 Sig. 0.00
Intraclass 95% Confidence Interval F Test with True Value 0 Correlation ᵇ Lower Bound Upper Bound Value df1 df2 Sig Single Measures .237ᵃ .201 .281 10.331 199 5771 .000 Average Measures .903ᶜ .883 .922 10.331 199 5771 .000
31
For all factors extractable from the analysis along with their eigenvalues, the percent of variance attribute to each factor, and the cumulative variance of the factor and the previous factors. Notice that the first factor accounts for 32.672% of the variance, the second 13.252%, the third 10.347%, and the forth 7.228%. All remaining factors are not significant (Table 3). For the scree plot, it can be seen that factor 5 has an eigenvalue of less than 1 so only four factors have been retained (Table3).
Table 3. Total Variance Explained and scree plot
Initial Eigenvalues Extraction Sums of Squared Loading Rotation Sums of Squared Loading Component Total % of Variance Cumulative% Total % of Variance Cumulative % Total % of Variance Cumulative%
1 4.901 32.672 32.672 4.901 32.672 32.672 3.842 25.613 25.613 2 1.988 13.252 45.924 1.988 13.252 45.924 2.055 13.698 39.312 3 1.552 10.347 56.272 1.552 10.347 56.272 1.922 12.811 52.123 4 1.084 7.228 63.499 1.084 7.228 63.499 1.706 11.377 63.499 5 .920 6.134 69.633
6 .757 5.045 74.678 7 .717 4.783 79.461 8 .589 3.923 83.384 9 .559 3.730 87.114 10 .466 3.109 90.222 11 .367 2.445 92.667 12 .335 2.235 94.902 13 .303 2.022 96.924 14 .236 1.572 98.496 15 .226 1.504 100.00
32
At the table of rotated component matrix, we can see that check-in, show off, activities interest, friend and family gathering, selfie, self-report and thought and feeling expression are substantially loaded on Factor (Component) 1 as self-expression group.
For criticism, complaint and caviling are substantially loaded on Factor (Component) 2 as criticism group while recommendation (WOM), ask for help and spread information are substantially loaded on Factor (Component) 3 as announcement group. All the remaining variables, (re)sharing informationis substantially loaded on Factor4 as (re)sharing post group (Table 4).
Table 4. Rotated Component Matrix
After we had four components, we used cluster analysis method for classify our samples. At table 5, we can see samples were divided to 4 groups: group of self-expression was classified to cluster 1, group of announcement was classified to cluster 2, group of share link/post was classified to cluster 3 and group of criticism was classified to cluster 4 (Table 5).
33 3.5) Questionnaire Design
We design online questionnaire for study individual’s motive on Facebook wall post. The questionnaire consists of three motivations: Entertainment Motives, Social Connection Motives, and Information Motives. We use five-point scales indicates frequency, amount and size of idea (1= strongly disagree and 5= strongly agree). The questions of each motivation were mixed together. Each motivation has five questions so we designed twelve questions in total.
3.6) Measures
Entertainment: According to Figure 6. Model of Consumers’ social media activities (K.Heinonen, 2011), we assume that entertainment motive may effect to people post about self-expression and criticism contents on their wall-posts. When people feel relax or want to escape from real world for a while, we assume that they tend to more post about yourself activities, their emotions and their opinions etc.
Social Connection: Based on the model, we also assume that social motive may effect to people post about announcement and sharing information posting such as spread information, recommend some products or services to their friends and share their own experiences. This type of activities, we assume that people tend to more post about announcement some news to their friends so this type of people would like to connect with others.
Information: According the model of Consumers’ social media activities (K.Heinonen, 2011), the researcher used interview method for collect the data. This part, participants were asked about what kind of social media or links that participants use to search for gain information? For our thesis, we used the empirical method to investigate samples’
posts so we could not know what kind of information that they would like to search for gain information or how often do they search? We only saw their post about the links that they shared so we assume that they saw the links that their friends post or they searched by themselves, then they like it or they thought that it is useful or interesting so they would like to (re)sharing information to others.
34 Table 6. Measure Items
Entertainment
Q1. When I post on Facebook, I am looking for inspiration and encouragement.
Q4. When I post on Facebook, I can escape for a while and release my tension.
Q7. When I post on Facebook, I love to express about myself.
Q10. When I post on Facebook, I fell enjoy and relaxing.
Social Connection
Q2. When I post on Facebook, I feel can connect with my friends.
Q5. When I post on Facebook, my friends can share their opinions or experiences with me.
Q8. When I post about my groups or clubs (activities), I feel belong to that community.
Q11. When I post on Facebook, I feel up-to-date with others.
Information
Q3. When I post on Facebook, I would like to share information and experience.
Q6. When I post on Facebook, I would like to share knowledge with others.
Q9. When I post on Facebook, I would like to share and spread something that I trust and valuable.
Q12. When I post on Facebook, I would like to share information that important to me.
Source: Adapt from K.Heinonen,2011
35
We did the reliability test for this survey. For Entertainment factor, Cronbach’s Alpha = 0.697 that lower than 0.7 so we decided to cut off question no.1. After we cut out Q.1, Cronbach’s Alpha is higher than 0.7 (α= 0.748). Social Connection factor (α = 0.711), we also decided to cut Q.11 out then Cronbach’s Alpha changes to 0.724 and Information factor, Cronbach’s Alpha is 0.824 (see table 7).
Table 7. Item total-statistic
3.7) Model of Users’ motives on Facebook wall posting
Figure 7. Model of Users’ motives on Facebook wall posting Motivations
36 3.7) Hypothesis
Based on model of consumer’s social media activities (K. Heinonen, 2011), we can classify Facebook post activities by use this concept. This framework based on Shao’s theories (2009) that use consumer activities in social media approaches consumer’s motivation and behavior. The hypotheses were recommended follow:
H1a: Entertainment has significant effect on total post H1b: Social connection has significant effect on total post H1c: Information has significant effect on total post
Figure 8. Hypothesis Model 1 (Total post)
H2a: Entertainment has significant effect on Self-expression H2b: Entertainment has significant effect on Criticism
Figure 9. Entertainment Motivation Entertainment
Social connection
Information
Total post
H1a
H1b
H1c
Entertainment
Self-expression
Criticism
37
H3a: Social connection has significant effect on Announcement H3b: Social connection has significant effect on (re) Sharing post
Figure 10. Social connection Motivation
H4a: Information has significant effect on (re) Sharing post H4b: Information has significant effect on Criticism
Figure 11. Information Motivation Social connection
Announcement
(re)Sharing post
Information
(re)Sharing post
Criticism
38
Chapter 4 Results
This chapter presents the data gathering of the study, interpretation of the results from the conducted survey and the software product analysis.
4.1) Descriptive statistic for Age, Gender and Education
According to descriptive statistic, age factor was divided to four groups: 20-29 years old has 51 people, 30-39 years old has 103 people while 40-49 years old has 37 people and more than 50 years old has only 9 people. It indicates that most of samples have age between 30-39 years old.For gender factor, male has 63 people that is fewer than Female has 137 people from 200 people.Education factor, we divided to four groups: High School, Bachelor’s Degree, Master’s Degree and Ph.d..Bachelor’s Degree has 105 people that is a great number when compare with other education groups (table 8).
39 4.2) Hypothesis results
H1a: Entertainment has significant effect on total post H1b: Social connection has significant effect on total post H1c: Information has significant effect on total post
In the total post model, the study results of regression analyses are presented.
Hypotheses (H1a, H1b and H1c) were tested by using ANOVA and regression analysis.
Results indicated that Model (total post) is considered significant (R=0.499, R² = 0.249, P< 0.05). Beta coefficients of entertainment and social connection are quite the same (β=0.283, β=0.278), both of them are positive correlation. It indicates that entertainment and social connection have effect to total post. Information has significant effect to total post but negative correlation (β = -0.365). It indicates that when total post score is high, information score is low. Entertainment, social connection and information contribute statically significant to the model. From the results, we assume that when individuals like to post for information, the frequency of post for information may be lower than the frequency of post for entertainment and social connection.
Individuals who post for entertainment and social connection tend to post more frequently.
40
Model Summary
ANOVAᵃ
Coefficients ᵃ
Table 9. Result of Total post model
41
For this part, we used Confirmatory Factor Analysis (CFA) for test our measures.
Goodness of Fit Statistic of this model is good, Chi-Square (χ²= 141.24), df = 51, Normed Chi-Square = 2.769, P<0.01, RMSEA= 0.094, CFI= 0.907, GFI= 0.893 and AGFI= 0.836.
Figure 12 provides the results model that we can see that factor loading of Q.1 is 0.37 (λ less than 0.6 is not acceptable) and Q.11 is 0.42. From the results, we decided to cut out these two factors (Q.1 and Q.11). From this Model, we found that the correlation of Entertainment and Information has negative correlation (-0.20), and the correlation between Entertainment and Social connection has small correlation (0.06), however, Social connection and Information has rather high correlation (0.53). From this result, we assume that entertainment has low correlation to social connection may caused by entertainment motive activities is relate to relaxing or releasing tension and enjoying oneself online which rather differ from social connection motive activities that relate to connecting with people and sharing experience with others ( K. Heinonen, 2011)
Figure 12. CFA of Measurement Model
42
H2a: Entertainment has significant effect on self-expression H2b: Entertainment has significant effect on criticism
The model was fitted to the empirical data after removing question 1 (0.37). The result indicated that H2a and H2b were supported. As shown in table 10, goodness-of-fit to the data, with Normed Chi-square=2.915, p<0.001, SRMR=0.0945, RMSEA=0.098, GFI=0.903, AGFI=0.847, CFI=0.90
(Figure 13) The model suggested that entertainment motive has significant effect to self-expression (β=0.36) and entertainment also has significant effect to criticism but in small value (β=0.19) when compare with self-expression. This model was cut off Q.1 before analyze data that we stated on previous part. From this result, we assume that people who has entertainment motive will more post about self-expression and people who has entertainment motive might more post about criticism. These results quite same as previous research. Based on previous research, K. Heinonen provided the description of entertainment activities are escaping the real world and relaxing, mood management, entertaining one-self and self-expression. We employ these dimensions for mapping our activities. We supposed that criticism is kind of mood management and entertaining one-self. Criticism is a kind of thought expression but stronger because criticism is judgment based on individual’s feeling or experiences. When individual read or saw something that made them feel uncomfortable or unreasonable, they will expression their thoughts instantly. This is a way of release their tension or mood managing.
43
Table 10.Fit indices of Entertainment Motivation Model
Figure 13. Result of Entertainment Motivation Model (SEM)
Fit indices Criteria (Perfect) Results
χ² - 122.433
44
H3a: Social connection has significant effect on announcement H3b: Social connection has significant effect on (re) sharing post
The result indicated that H3a and H3b were supported. As shown in table 11, goodness-of-fit to the data, with Normed Chi-square=2.209, p<.005, SRMR=0.0826, RMSEA=0.078, GFI=0.953, AGFI=0.907, CFI=0.916 (Table 11). The model suggested that social connection motive has significant effect to announcement (β=0.36) and social connection also has significant to (re)sharing post (β=0.20) but the value is smaller than announcement (Figure 14). This model was cut off Q.11 before analyze data that we stated on previous part. From this result, we assume that people who have high social connection motive tend to more post announcement posting. On the other hand, people who have social connection motive also tend to (re) sharing post. Base on conceptual model of consumers’ social media activities (K. Heinonen, 2011 p.359), the activities of social connection motive are being up-to-date, staying in touch and sharing
The result indicated that H3a and H3b were supported. As shown in table 11, goodness-of-fit to the data, with Normed Chi-square=2.209, p<.005, SRMR=0.0826, RMSEA=0.078, GFI=0.953, AGFI=0.907, CFI=0.916 (Table 11). The model suggested that social connection motive has significant effect to announcement (β=0.36) and social connection also has significant to (re)sharing post (β=0.20) but the value is smaller than announcement (Figure 14). This model was cut off Q.11 before analyze data that we stated on previous part. From this result, we assume that people who have high social connection motive tend to more post announcement posting. On the other hand, people who have social connection motive also tend to (re) sharing post. Base on conceptual model of consumers’ social media activities (K. Heinonen, 2011 p.359), the activities of social connection motive are being up-to-date, staying in touch and sharing