Study On User Exodus Of Social Network Site Based On Stimuli-Organism-Response Theory

With the arrival of the era of Web2.0,users are seeking more pragmatic and real experience with increasing needs of involvement in information transmission and information interaction by combining online and offline activities(O2O).Then,Social Network Site(SNS)was born at the historic moment.Although many SNSs experienced great popularity in the past few years,it is also worth noting that most of them were confronted with massive and abrupt recession in a short period of time.It is clear that user exodus is an important issue for online service providers.However,the subsequent behavior research mainly concentrates on the adoption behavior and sustainable behavior,with less attention to user exodus behavior.Some researchers explored the reasons and factors connected with user exodus based on the using experience,but less research focused on about the importance of different factors,the relationship among these factors.As the core resources of SNS,the amount of active users and their engagement are playing an extremely important role in the success and sustainability of today's SNS.So it is particularly important to understand the reasons of user exodus for SNS and other Social Media.This dissertation investigates the factors of user exodus and their relationships through the open-ended questionnaire by merging Ethnographic Decision Tree Model(EDTM)and Means-End Chain(MEC).The integration model of user exodus's behavior is then constructed based on the literature review,interview and Stimuli-Organism-Response(S-O-R)theory and verified by using Structural Equation Model.At last,the study suggests that the design of SNSs and other Social Media need to follow the principle of "function first and interaction second" to slow down the rate of user exodus.This dissertation firstly introduces the research background and significance and clarifies the research objective.Then a detailed literature review is concluded and the research questions are proposed.At last,the paper states the research contents,methods and the road map of this study.The second chapter explores theoretical bases and methodology.The paper elaborately examines the S-O-R theory and its application,which is used to design the framework of user exodus's behavior.The paper also introduces the theoretical bases and implementation of EDTM,conducting the architecture of open-ended questionnaire.Moreover,this study discusses a qualitative method known as the means-end theory and its associated laddering technique,which can be used to design the open-ended questionnaire and investigate attitudes towards user exodus.The third chapter mainly detects the factors and their relationships from mass user loss of social networks.Firstly,the paper explores the feasibility and necessity of applying EDTM and MEC to the open-ended questionnaire and proposes the corresponding framework of open-ended questionnaire.Secondly,the interview architecture and specific interview items are proposed under the social networking environment.The kaixin001 is taken as an example to yield some implication for the critical influences to user exodus.Then the factors and their relationships are extracted with the form of Attributes-Consequence-Value(ACV)based on the open-ended questionnaire.At last,the paper constructs the Hierarchical Value Map(HVM)according to the relationship of ACV.The results are analyzed and discussed in the final section.The fourth chapter mainly builds the theoretical model of user exodus.Firstly,the paper refines the theories related to user exodus behavior from multiple disciplines(i.e.S-O-R)and influencing factors from related literature which are classified into three aspects based on the factors and its relationships from the third chapter.Then a theoretical model is proposed and analyzed.At last,the paper builds a research model and puts forward several hypotheses for empirical study.The fifth chapter firstly designs the variables scale based on the existing literature.The final questionnaire is conducted to collect the data from users based on the pretest questionnaire.Then the data is analyzed,including the statistical analysis of demographic characteristics,the data quality inspection of the validity and the reliability and the structural equation model analysis,etc.The sixth chapter mainly introduces the design-science research paradigm based on the research results.Social media's service providers need to put the function design in the first place and interaction design secondly according to the empirical results in the design philosophy.They also need to develop the social media product according to user requirements,market segmentation and positioning based on the results from A-C-V and S-O-R for the marketing strategy.The paper puts forward the development suggestions to the social media from four aspects:product design,privacy protection and attractive alternative and product promotion.The seventh chapter mainly illustrates the conclusions,limitations and outlook.The findings of this dissertation will contribute to the theoretical construction of user exodus and the anti-loss policies and also yield some implications,suggestions and guidelines for the server provider,designers,and cyber regulation department from social media in their real work.The dissertation has several major innovations as below.First,we design the architecture of open-ended questionnaire by the methods fusion of EDTM and MEC.Second,we obtain the elements and their relationship from open-ended questionnaire by laddering method.Third,we incorporate the S-O-R theory,the obtained factors and so on to build the integration model.However,the study also has some limitations.First,the moderating effects of demographic factors and perceived social capital as a multidimensional concept need to be further explored.Second,the influence of perceived privacy risk as an independent variable on social media remains to be further researched.Third,the study mainly focuses on the factors from user behavior and SNS.More attentions need to be paid to external variables in the future work,such as social culture and network environment.At last,the anti-loss design lacks empirical evaluation which needs to be tested and verified.