Poor, Good Enough or Even Better? Bridging the Gap between Acceptability and QoE of Mobile Broadband Data Services

Due to the accelerating uptake of mobile broadband services, Quality of Experience (QoE) assessment and optimization for wireless networks has become a topic of vital importance. In this field, subjective quality evaluation methods based on MOS (Mean Opinion Score) are the most widely used approaches for understanding and measuring perceived performance. However, while these methods have been proven to reliably quantify users\´ degrees of satisfaction, they do not sufficiently address the following important question: "Which quality levels are acceptable to end users and which ones not?". In this paper we investigate the application of acceptability to the domain of QoE assessment for data services. To this end, we perform a critical review of the term\´s utilization and operationalization in related fields and discuss the results of a series of mobile broadband QoE studies we performed in lab and field settings. Our results demonstrate that acceptability is a relevant and useful concept that also features a consistent mapping between subjects\´ binary acceptance and ordinal satisfaction ratings, even across application scenarios as different as web browsing and file downloads. However, it is also a highly context-dependent concept since users\´ rating behaviors and thresholds diverge between lab and field settings. Acceptability thus needs to be evaluated in situ in the field in order to guarantee generalizable, ecologically valid results.