Social and Parasocial Relationships on Social Network Sites(2)

with psychological well-being

While parasocial SNS relationships between users and ce-lebrities resemble the mass media model,mutually interac-tive SNS friendships are akin to conversations via phones, e-mail exchange,or online chatting.13Since mutually inter-active online communication is interpersonal,like face to face interaction,a past studies have demonstrated that reciprocal interaction on the Internet strengthens existing friend-ships10,12,27and even facilitates of?ine meetings,which in turn lowers loneliness and forti?es social bonding.11Thus,we can anticipate that:

H2a:Reciprocal social relationships on SNSs are negatively related with loneliness but positively related with in-

terpersonal trust.

However,in terms of SNS addiction(i.e.,obsessed SNS use):

H2b:Dependency on reciprocal social relationships on SNSs could be positively correlated with SNS addiction.

This is because active interaction with SNS friends means users are more attentive on the given SNS,and users might be afraid that SNS abstinence might hurt the ongoing reciprocal relationship.

Methods

Sample

To test the hypotheses,national representative survey data were used.Survey respondents were recruited from the Ko-rean Ipsos panel,which is an online survey service based on national samples of Internet users in South Korea.There are about660,000people on the Ipsos panel.Out of these,a total of 404Korean adults who use at least one SNS(e.g.,Twitter, Facebook,me2day,Cyworld,or other services)were recruited to test our hypotheses.The survey was conducted in January 2012.On average,selected respondents utilized two SNSs (M=2.20,SD=1.10;median=2)and reported they used SNSs about six days per week(M=5.62,SD=2.28,median=6).

Types of SNS relationship

We distinguish social(i.e.,reciprocal)relationships from parasocial ones.To measure both types of relationship,we asked respondents for the number of their mutual SNS friendships and the number of their nonmutual ones with celebrities,such as movie stars,athletes,or politicians.In the case of Twitter,we distinguished mutual following from mere following.The distribution of social relationships was skewed,indicating that some respondents reported to have many relationships(M=51.03,SD=179.69,range0–3,000).

SOCIAL AND PARASOCIAL RELATIONSHIPS ON SOCIAL NETWORK SITES513

However,the median number of reciprocal relationships was 10,which corresponds to previous?ndings on the number of acquaintances per person.28In the case of parasocial rela-tionships,the problem of skewed distribution was also found (M=9.03,SD=12.77,median=4,range0–80).

To adjust for the skewed distribution,the number of each relationship was log transformed(M social=1.16,SD social= 0.65,range0–3.48;M parasocial=0.70,SD parasocial=0.55,range 0–1.91).The two types of SNS relationship are slightly cor-related with each other(r=0.25,p<0.001).b

Measures of psychological well-being

This study examined three constructs of psychological well-being:(a)loneliness,(b)interpersonal trust,and(c)SNS addiction.First,the UCLA loneliness scale29was modi?ed and shortened for the SNS context in Korea.There were?ve statements in the loneliness scale:‘‘When using SNS,I feel isolated from others’’;‘‘SNS friends are not‘with’me’’;‘‘When using SNS,I feel sometimes left out’’;‘‘No SNS friends know the‘real’me’’;and‘‘SNS relationships are super?cial.’’Based on a conventional5-point Likert scale(1=‘‘strongly disagree,’’5=‘‘strongly agree’’),respondents were asked to indicate their agreement with each statement(Cronbach’s a=0.89,M=2.70,SD=0.74).

Second,interpersonal trust was measured using Rotter’s interpersonal trust scale,30which was modi?ed for the SNS context in Korea.There were?ve statements in the inter-personal trust scale:‘‘Online users in general try to take advantage of others’’;‘‘Generally speaking,online users are sel?sh’’;‘‘You can’t be too careful when dealing with online users’’;‘‘Online users in general are fair’’;and‘‘Generally speaking,online users can be trusted.’’Respondents were asked to estimate their agreement with the?ve statements, based on a5-point Likert scale.After reversely coding the ?rst three statements,respondents’agreement on the?ve

statements was averaged(Cronbach’s a=0.84,M=3.55, SD=0.66).

Finally,SNS addiction was measured by modifying the Korean Internet addiction scale for the SNS context.31Basi-cally,SNS addiction aims to measure the degree of perceived obsession of SNS use or withdrawal symptoms from SNS abstinence.There were four statements on the SNS addiction scale:‘‘Whenever I have time,I log on to my SNS account’’;‘‘Frequently,I regret I consume too much time using SNS services’’;‘‘If I could not use SNS services,I would be de-pressed’’;and‘‘If I stopped using SNS,my social relationships would be ruined.’’Using the same scale as for the other constructs,respondents’agreement on the four statements was averaged(Cronbach’s a=0.83,M=2.94,SD=0.86). Statistical controls

Socio-demographical variables and SNS behaviors were additionally controlled for statistical adjustment.The socio-demographical variables were the respondents’gender,age (measured in years),monthly income,and educational achievement.Regarding SNS behaviors,a self-report of daily SNS use(measured by the hour)and number of SNS accounts was included.A dichotomous variable for whether a re-spondent used Twitter was also added because Twitter is the most popular SNS for forming parasocial interactions with celebrities.Descriptive statistics of socio-demographics and SNS behaviors can be found in Table1.

Results

To test the hypotheses,three outcome variables were re-gressed on two types of SNS relationships as well as seven statistical controls.Results of ordinary least square regression analyses are provided in Table2.A …… 此处隐藏：6298字，全部文档内容请下载后查看。喜欢就下载吧 ……