An Analysis of User Attitudes to SNS

Social Networking Service (SNS) have become widely used in Japan in recent years with Facebook, mixi and Twitter being the most popular. These are used in various fields of life together with the convenient devices such as smart-phones. A questionnaire investigation was used to clarify the current usage condition, issues and desired function etc. Information for marketing purposes was then extracted. Fundamental Statistical Analysis, Multi Corresponding Analysis, Quantitative Analysis and Text Minig Analysis were then performed. Reviewing past research, there are some related papers, but they do not include new tools which are evolving rapidly. Moreover there has been little research conducted on this precise topic. Some interesting results were obtained.


Introduction
Social Networking Service (SNS) become widely used in Japan in recent years with Facebook, mixi and Twitter are the popular one.Big Disaster happened at 11/March/2011 in the east part of Japan.It is well known that Facebook played important role in communication under the condition that telephone and/or cellular phone connected with Internet could not make link.Google launched forth into SNS by the name Google+ at June 2011.Thus, it has become a hot business spot and it makes great influence upon society and economy.
In this paper, we make questionnaire investigation in order to clarify the current usage condition, issues, and desired functions etc. Marketing applications would then, be extracted.Reviewing past researches, Miyao et al. (2008) made questionnaire investigation concerning usage objective of SNS and its usage consciousness.They analyzed into 4 clusters by cluster analysis but it does not include Facebook and Google+, which are increasing among users recently.Yamamoto et al. (2011) made research on the intimacy among friends on SNS but it does not mention about sharing information.Thus, there are researches concerning SNS, but there has been few researches made on our above stated themes.
The rest of the paper is organized as follows.Current condition of SNS is stated in section 2. Investigating Former Researches is stated in section 3. Outline of questionnaire research is stated in section 4. Multivariate Analysis such as Multi Correspondence Analysis and Quantitative Analysis is executed in section 5.Text Mining Analysis is executed in section 6, 7. Section 8 is a summary.

Current Condition of SNS
SNS means the services to construct social network on the Internet.Friendster which has started in the year 2002 is said to be the father of SNS (Saito, 2010).Various typed SNS were born ever since.Japanese users have reached 42.89 million at the end of December 2011.It is reported that 45.1% of Internet users (95.1 million) use SNS.SNS users will increase as is shown in Figure 1.

Positioning in SNS
There are some related papers concerning SNS.Thomson Robert et al. (2012) made a study of 131 Japanese SNS users who use both Facebook and mixi.In that research, self-disclosure, numbers of contacts, in-group numbers, and levels of perceived commitment on mixi and Facebook were measured.The study found that such users showed a lower level of self-disclosure, connected with fewer people and had fewer categories of contacts, and felt a higher degree of commitment on mixi than on Facebook.It may have found the characteristics of the correlation and distinction among tools, but it lacks the viewpoints of organizational management theory.In this paper, focusing the past researches of Harada (2008), a new approach is executed based upon organizational management theory.Such research is quite a new one.
Harada arranged four business models as Figure 3.
The horizontal line shows the users' communication motive and the vertical line shows the solution to be captured.Former one is represented as "Construction of relation" vs. "Self-expression" and the latter one as "Realization of dialogue" vs. "Self-assertion".

SNS Users' Desire
Harada pointed out that SNS users want to have relation so as to satisfy following three desire.
・Desire for having relationship: Desire for having relationship by making communication with friends in having mail and/ or agreement behavior.
The expected role in using SNS is to build good relationship with friends and keep it long.Furthermore, it is used to keep contact with those who are busy or stay in the distance.
・Desire for sharing sympathy: Desire for sharing sympathy by enjoying hobbies with friends.
As for the second role, it is to talk with friends for about hobby freely.By attending the same hobby group, they seek to share the sympathy.
・Desire for making expression: Desire for making expression as actor/ actress and being fascinated.
As for the third role, it is to confirm his own value through the evaluation executed by other people.By letting others know himself and by expressing his skill, SNS can be utilized as a "Theater".

Social Background for the Prevailing SNS
In Marketing 3.0 by P. Kotler (2010), consumers can gather information speedily by utilizing IT.He says that "cooperative marketing" utilizing SNS becomes an important keyword.Comparison of Marketing 1.0, 2.0 and 3.0 is shown in Table 3.   to collect/ pu " follow.
tweets" is the and acquainta .2) Why don't you use the SNS? (Q11) We can observe the following result from Figure 14.From the data, we can confirm that Dimension 1 axis can be interpreted as "Relationship" and Dimension 2 as "Indifference" and eigenvalue of Dimension 1 axis is 0.388 and those of Dimension 2 is 0.305.69.3% can be explained.We can extract four clusters.We can assume that the first cluster shows "Safety, anxiety about operation" ( ③ (Anxious about security concerning individual information), ⑨(Cannot continue because it is too bothering)).The second one would mean "Indifference" (① (Do not have interest), ②(Interesting but do not know how to use)).The third one "Anxious about fee" (④ (Anxious about fee), ⑩(Likely to increase the spam e-mail)).The fourth one "Anxiety for human relationship" (⑥(Become poor in human communication), ⑦(Will waist a lot of time)).We can observe the following result from Figure 15.From the data, we can confirm that Dimension 1 axis can be interpreted as "Function" and Dimension 2 as "Information".An eigenvalue of Dimension 1 axis is 0.471 and those of dimension 2 is 0.301.77.2% can be explained by these two major axes.We can extract three clusters.We can assume that the first cluster shows "relationship, participation and function" (①(Make full communication with friends and acquaintances), ②(Want to seek old friends, acquaintances and new friends), ⑥(Have a space/ field for exchanging opinion about goods, service and politics) , ⑫(Interconnection among SNS functions)).The second cluster would mean "Sharing・Diffusion and Function" (⑦(Gather interesting information), ⑩(Easiness in using)).The third one "Sympathy and Confirmation" (④(Provide the valuable information), ⑤(Enrich the collection of information)).We analyze Q12 by diving users and non-users.First of all, we see the users' case.From the data shown in Figure 16, ④(Provide the valuable information) has the largest score.This means that sharing information is valuable for them.Dimension 1 axis can be interpreted as "Sympathy" and Dimension 2 as "Society".This is the same with Figure 13.An eigenvalue of Dimension 1 axis is 0.327 and those of Dimension 2 is 0.265.59.2% can be explained.We can extract three clusters.The first cluster shows "Relationship and Function" (①(Make full communication with friends and acquaintances), ⑪(Restrict the writing in board by others), ⑫(Interconnection among SNS functions)).The second cluster would mean "Confirmation and Sharing・Diffusion" (⑤(Enrich the collection of information), ⑦(Gather interesting information)).The third one "Expression and Function" (⑧(To disclose the information of himself/herself), ⑨(To make perfect the security of individual information)).This is the one we could not find in Figure 15.Users want to share information or express themselves under the strengthened security system.

Remarks for Multi Correspondence Analysis
"Confirmation" of Q6⑤(Able to collect news and information efficiently), Q12⑤(Enrich the collection of information) and "Expression" of Q6⑧(Able to let others know about myself well), Q12⑧(To disclose the information of himself/herself) are located in the distant way and they have few relationship, which can be seen Figure 13 and Figure 14.Common theme is to share sympathy in society, which can be observed in Figure 13 and Figure 16.As for non-users, minus item can seen in Q11 (Why don't you use the SNS?) owing to the indifference.But the Figure 17 shows that the non-users have demands for relationship, attendance, Share & Spread.Therefore they have a probability to shift to users if anxiety items are resolved.

Execution of Analysis
Factor Analysis is executed so as to extract viewpoints or axes for the appraisal concerning Q6 (What are the SNS's interesting and fascinating points?),Q11 (Why don't you use the SNS?) and Q12 (What do you expect the SNS in the future?).As for the extraction method of the factor, "the principal axis factoring" is adopted, and as for the rolling-method, "the varimax rotation which is accompanied by the normalization of kaiser as orthogonal rotation" is adopted.Also, we confirmed the validity of the model in the KMO (Kaisel-Meyer-Olkin) specimen validity measure.When the value is more than 0.8, it is said meritorious, more than 0.7 middling, more than 0.6, mediocre.Bartlett sphericity test is executed as well.If the score is less than 0.05, then it is said that there is a correlation among observation variables.
1) What are the SNS's interesting and fascinating points?(Q6) KMO measure is 0.798 and Bartlett Score is 0.000.We can confirm a rather appropriate common factor.From the Factor Matrix after rotation, we can extract 3 meaningful axes.We can see that the first axis is the factor about "Sympathy", "Participation" and "Share & Spread" as the score for ③(Able to find new friend who has the same hobby and interest via the Net), ⑥(Able to share sympathy by joining a group) and ⑦(Able to share hobby and interests with friends and acquaintances) is high.Second axis is the factor about "Relationship" as the score for ①(Able to communicate with each other by diary and tweets) and ②(Obtained much opportunities to contact with friends and acquaintances who were under rare contact) is high, Third axis is the factor about "Function" as the score for ⑪(Rich online game) is high.
2) Why don't you use the SNS? (Q11) KMO measure is 0.690 and Bartlett sore is 0.000.We can confirm an ordinary common factor.From the Factor Matrix after rotation, we can extract 3 meaningful axes.We can see that the first axis is the factor about "Anxiety for communication tool" (④(Anxious about fee)).Second axis is the factor about "Fatigue in soul and anxiety for human relationship" (⑦(Will waist a lot of time), ⑧(Feel uneasy how friends and acquaintances make response)).Third axis is the factor about "Anxiety for operation" (⑨(Cannot continue because it is too bothering)).
3) What do you expect the SNS in the future?(Q12) KMO measure is 0.814 and Bartlett score is 0.000.We can confirm an appropriate common factor.From the Factor Matrix after rotation, we can extract 4 meaningful axes.We can see that the first axis is the factor about "Sympathy", "Identify" and "Share & Spread" (④(Provide the valuable information), ⑤(Enrich the collection of information), ⑦(Gather interesting information)).Second axis is the factor about "Function" (⑩(Easiness in using), ⑪(Restrict the writing in board by others)).Third axis is the factor about "Relationship" (①(Make full communication with friends and acquaintances), ②(Want to seek old friends, acquaintances and new friends)).Forth axis is the factor about "Participation" (⑥(Have a space/ field for exchanging opinion about goods, service and politics)).
4) Difference between users and non-users in Q12 KMO measure is 0.758 and Bartlett score is 0.000 in Table 11.We can confirm a rather appropriate common factor.From the Factor Matrix after rotation, we can extract 4 meaningful axes.We can see that the first axis is the factor about "Sympathy" and "Identify" (④(Provide the valuable information), ⑤(Enrich the collection of information)).Second axis is the factor about "Function" (⑪(Restrict the writing in board by others)).Third axis is the factor about "Relationship" (①(Make full communication with friends and acquaintances), ②(Want to seek old friends, acquaintances and new friends)).Fourth axis is the factor about "Participation" (⑥(Have a space/ field for exchanging opinion about goods, service and politics)).In Table 10, KMO measure is 0.885 and Bartlett score is 0.000.We can confirm an appropriate common factor.First axis would mean "Function" (⑨(To make perfect the security of individual information)).Second axis would mean "Anxiety for relationship" (②(Want to seek old friends, acquaintances and new friends)).

Remarks for Factor Analysis
Factors for Q6 (What are the SNS's interesting and fascinating points?) and Q12 (What do you expect the SNS in the future?) are arranged in Table 11.As is shown in Table 11, we can confirm the combination of common factors between Q6 and Q12.One is "Relationship" and another one is "Function"."Participation" which is the first factor in Q6 is replaced into "Identify" in Q12.This implies that after confirming the profile of friend and goods, they proceed to the "Participation" stage.Thus, effective approach is considered by these combinations in demand and action.

5.3
The Analysis by the Quantification Method Ⅱ

Execution of Analysis
Quantification Method Ⅱ is executed in order to grasp the contribution to the usage of SNS.Outer criteria is set "Use or do not use SNS".Explanation variables are set for each question number of Q12.The discriminate hitting ratio is 66.9%.It is rather good in this case.Utilizing the result of coefficient of canonical discrimination function, discrimination equation of outer criteria can be stated as follows.
Where question number is added to Q12**, for example, Q1209 means that question number is 9 in Q12.From Figure 18, we can observe that big contribution is achieved by the following items.First, ①(Make full communication with friends and acquaintances).Second, ⑩(Easiness in using).Third, ④(Provide the valuable information).Fourth, ⑨(To make perfect the security of individual information).Fifth, ⑫(Interconnection among SNS functions).From these, we can see that "Relationship", "Function", "Sympathy", "Identify" and "Expression" are important factors in utilizing SNS."Relationship", "Sympathy" and "Expression", which are listed in the former study, are all included.

Analysis Using Key Graph
Key Graph Method is to clarify the data structure by using key words.Data occurrences at the same period of time are exhibited by the link of Key Graph.
1) Used SNS and their using frequency (Q2-Q3) The Key Graph Analysis is executed by utilizing Q2 (How often do you use the SNS?) and Q3 (What kind of the SNS do you use?).The occurrence rate and co-occurrence rate are important measures.From Figure 19, we can see that "More than 5 times a day" has the high co-occurrence rate with "Facebook", "mixi", "Twitter", "You Tube" and "Niconico β", while "Around 3-4 times a day" with "Facebook", "mixi" and "You Tube", furthermore "Around 1-2 times a day " with "Facebook", "Twitter" and "You Tube".Among them, Integrated Genre and Moving Picture Genre were used well.
They use "Facebook", "mixi", "Twitter" and "You Tube" with the commencement reason "Identify" factor.Integrated genre and Moving Picture Genre have the similar commencement reason.For the second cluster, "-60" and "Part-timers" have the high co-occurrence rate.They use "Ameba" and "Ameba pico".Housewives mainly use them as a diary."-60" has the high co-occurrence rate with "-50", "School teacher/staff" and "Independents".They use "Niconico β" and "Ustream" with the commencement reason, "Utilize in Business" and "To apply for the campaign".We can extract keywords which have high co-occurrence rate by utilizing the Key Graph Analysis Method.The co-occurrence condition is exhibited by Key Graph and we can classify the data by the Key Graph Clusters.Now, we make analysis utilizing "Text Analytics for Surveys" by focusing important keywords found in Key Graph.
Two clusters, which is analyzed in 6.3) Attributes and the Reason for Commencement of SNS (Q13-14-15, Q3, Q7), are analyzed as follows.
In Figure 22, we make focus on "Confirmation" Factor for the 1 st cluster, which is in the question item "To collect information".We can confirm from this Figure that keyword which has strong correlation with "To collect information" is "Male", "-40" and "Company employee" and the site used is "Facebook", "mixi" and "You Tube".In Figure 23, we make focus on "Utilize in Business" Factor for the 2 nd cluster.We can confirm from this Figure that keyword which has strong correlation with "To collect information" is "-50" and "Independents" and the site used is "Ameba", "Ameba pico" , "Niconico β" and "Ustream".

Figure 1 .
Figure 1.Users number of SNS in Japan

Figure 3 .
Figure 3. Positioning of SNS Source: Harada Tamotsu/ Formation of Tribes by Cooperative Experiences-Revised by the writer.

Figure 14 .
Figure 14.Why don't you use the SNS? (Q11) 3) What do you expect the SNS in the future?(Q12)

Figure 15 .
Figure 15.What do you expect the SNS in the future?(Q12)

Figure 16 .
Figure 16.Users' expecting items for SNS in the future (Q12)Now we see non-users' case(Fig 17).Dimension 1 axis can be interpreted as "Function" and Dimension 2 as "Relationship".⑧(To disclose information of himself/herself) is located in the small value part, which shows their attitude toward SNS.An eigenvalue of Dimension 1 axis is 0.482 and those of Dimension 2 is 0.267.74.8% can be explained.We can extract three clusters.The first cluster shows "Participation and Function" (⑥(Have a field for exchanging opinion about goods, service and politics), ⑫(Interconnection among SNS functions)).The second cluster would mean "Relationship and Sharing ・ Diffusion" (①(Make full communication with friends and acquaintances), ⑦(Gather interesting information)).The third one "Function" (⑩(Easiness in using), ⑪(Restrict the writing in board by others)).

Figure 17 .
Figure 17.Non-users' expecting items for SNS in the future (Q12)

Figure 18 .
Figure 18.Influence of demand item for SNS 5.3.2Remarks for Quantification Method Ⅱ Analysis

Figure 21 .
Figure 21.Attributes and the reason for commencement of SNS 7. Analysis by "Text Analytics for Surveys"

Figure 22 .
Figure 22.Attributes and the reason for commencement of SNS (1 st cluster)

Figure 23 .
Figure 23.Attributes and the reason for commencement of SNS (the 2 nd cluster)

Table 3 .
Comparison of marketing 1.0, 2.0, and 3.0 4. Outline of the Qestionnaire Research and Examinees4.1 Outline of the Qestionnaire Research

Table 6 .
What are the SNS's interesting and fascinating points?(Q6)(factor matrix after rotation)

Table 7 .
Why don't you use the SNS? (Q11) (factor matrix after rotation)

Table 8 .
What do you expect the SNS in the future?(Q12) (factor matrix after rotation)

Table 9 .
Users' expecting items for SNS in the future (factor matrix after rotation)

Table 10 .
Non-users' expecting items for SNS in the future (factor matrix after rotation)

Table 11 .
Factors for Q6 (What are the SNS's interesting and fascinating points?) and Q12 (What do you expect the SNS in the future?)Q6: What are the SNS's interesting and fascinating points?Q12.What do you expect the SNS in the future?