Research On The Influencing Factors Of Algorithmic Avoidance Behavior Of Users In Online Health Communities

Online health communities are an important part of the online healthcare industry,providing users and patients with access to health knowledge and medical services in the form of online communities.To address information overload,improve user efficiency,and maximize the value of community content,online health communities have introduced recommendation algorithms for content distribution.However,recommendation algorithms may pose privacy risks,false content,and data misuse,which can lead to algorithm avoidance behavior by users.Algorithmic avoidance behavior can prevent recommendation algorithms from performing their intended function,so it is necessary to conduct empirical research to understand the factors and mechanisms that influence algorithm avoidance behavior,and provide reference for improving online health community algorithm recommendation services and user experience.Firstly,this study constructed a model of factors influencing algorithmic avoidance behavior among users of online health communities based on the cognitive load theory and the CAC pattern.Drawing on the technology threat avoidance theory,as well as concepts such as techno-stress,algorithm anxiety,and trust,relevant variables from existing research were summarized through literature review,and four cognitive factors were proposed: privacy concern,perceived intrusion,perceived threat,and system function overload,as well as the emotional factor of algorithm anxiety.Additionally,perceived trust was introduced to explore whether the impact of algorithm anxiety on algorithmic avoidance behavior can be weakened.Secondly,based on existing mature scales,a questionnaire was developed and data was collected through online surveys.The SEM-ANN-NCA three-stage data analysis method was used to analyze data from the perspectives of sufficiency,importance ranking,and necessity.The research results showed that privacy concerns,perceived intrusiveness,perceived threat,and system overload have a significant positive impact on algorithm anxiety,and all of them are necessary conditions for algorithm anxiety.At the same time,algorithm anxiety also has a significant positive impact on algorithm avoidance behavior and is a necessary condition for algorithm avoidance behavior.The relative importance ranking of the four cognitive factors on algorithm anxiety is perceived intrusiveness 100%,privacy concerns 85%,system feature overload 36%,and perceived threat 32%.In addition,perceived trust cannot moderate the effect of algorithm anxiety on algorithm avoidance behavior.Based on the above results,this study proposes suggestions and strategies from three aspects: enhancing users’ autonomy in data and privacy security,cultivating and improving users’ algorithm literacy,and introducing a people-centric concept into algorithm design,to help online health community algorithm recommendation service operators,designers,and regulatory agencies better fulfill their respective roles and enhance user experience,promoting the sustainable development of online health communities.