Unobtrusive Continuous Stress Detection in Knowledge Work—Statistical Analysis on User Acceptance

Johanna Kallio, Elena Vildjiounaite, Julia Kantorovitch, Atte Kinnula, Miguel Bordallo LópezSustainability 2021 · DOI

TL;DR

This study investigates knowledge workers' acceptance of unobtrusive, continuous stress monitoring technologies and their willingness to share this data for well-being. The findings suggest that while privacy concerns exist, knowledge workers are generally interested in using these technologies and sharing data, especially when environmental sensors are involved.

What Problem Does This Paper Address?

Modern knowledge work is characterized by high intensity and demands, leading to significant psychosocial stress and fatigue. While stress detection technologies offer a promising avenue for continuous monitoring and intervention, there's a critical gap in understanding how employees perceive these technologies and their willingness to share sensitive personal data. This lack of insight hinders the adoption and effective implementation of such systems, despite their potential to improve individual, team, and organizational well-being. The paper aims to bridge this gap by exploring user acceptance and privacy perceptions of various unobtrusive stress monitoring methods in the knowledge work context.

Methods

The researchers developed a theoretical model based on the Theory of Planned Behavior (TPB), the Technology Acceptance Model (TAM), and privacy concern theories. This model predicts knowledge workers' intentions to share stress-indicative data by considering their interest in employing different stress monitoring technologies and their privacy concerns. They then empirically tested this model using Structural Equation Modeling (SEM) on data collected from a survey of 181 European knowledge workers. The survey employed a four-point Likert scale for attitudinal statements and multiple-choice questions to gather data on perceptions of various stress monitoring methods and privacy sensitivity.

Datasets

The study utilized a dataset collected through an anonymous online survey conducted in spring 2020. The survey received 181 responses from knowledge workers across 12 European countries. The data is available as a public dataset via Zenodo (doi: 10.5281/zenodo.4457393).

Key Insights and Ideas

  • Environmental sensors are a privacy win: The study found that privacy concerns did not significantly hinder the acceptance of environmental sensors (air quality, sound level, motion) for stress monitoring. This suggests these less intrusive methods are highly acceptable.
  • Tracking software is less of a deterrent than expected: Despite privacy concerns associated with tracking personal device usage (e.g., keyboard, mouse, smartphone), these methods did not prevent user acceptance or the intent to share data. This implies that the perceived benefits of stress monitoring outweigh privacy worries for these technologies.
  • Interest in well-being information is high: A significant majority (81.8%) of knowledge workers expressed strong interest in receiving information about their well-being, with stress levels and concentration being the most sought-after data points.
  • Indirect mediation for environmental sensors: While direct interest in sharing data from environmental sensors wasn't strongly linked, their acceptance was mediated by an interest in tracking software, suggesting a potential pathway for adoption.
  • Video cameras are a no-go: Unsurprisingly, video cameras were identified as the most privacy-sensitive and least acceptable method for stress monitoring by a large margin.

Key Results

  • High overall interest: 81.8% of respondents were interested in receiving well-being information measured by sensors.
  • Top well-being interests: Stress level (75.1%) and concentration level (66.9%) were the most desired information types.
  • Preferred monitoring methods: Air quality sensors (mean=3.47) and wearables (mean=3.46) were the most preferred methods for monitoring stress.
  • Most privacy-sensitive: Video cameras were considered the most privacy-sensitive by 82.3% of respondents.
  • SEM Model Fit: The amended SEM model showed an acceptable fit, with RMSEA=0.048, CFI=0.965, TLI=0.950, and SRMR=0.057, indicating the model's validity in explaining knowledge workers' interest in sharing data.
  • Hypothesis H1 (CON -> SHA): Privacy concerns negatively influenced the interest in sharing personal data.
  • Hypothesis H2 (TRA -> SHA): Interest in employing tracking software positively influenced the interest in sharing personal data.
  • Hypothesis H3 (ENV -> SHA): Interest in employing environmental sensors did not directly influence the interest in sharing personal data, but had an indirect positive effect mediated by TRA.
  • Hypothesis H4a (CON -> TRA): Privacy concerns negatively influenced the interest in employing tracking software.
  • Hypothesis H4b (CON -> ENV): Privacy concerns did not negatively influence the interest in employing environmental sensors.

Limitations and Future Work

The authors acknowledge several limitations:

  • Sample Size and Geography: The study's sample of 181 respondents was predominantly from Finland and Spain (89.9%), with only 10.1% from other countries. While gender, age, and nationality did not show significant differences in this study, future work should aim for broader geographical and national representation.
  • Model Simplification: The SEM model represents a simplification of a complex reality and could be extended to include more influencing factors at individual and organizational levels.

Future work could focus on:

  • Broader Model Coverage: Incorporating additional factors that influence data sharing and technology adoption.
  • Organizational Context: Further exploring how organizational culture, leadership, and policies impact the acceptance and implementation of stress monitoring technologies.
  • Longitudinal Studies: Investigating the long-term effects and sustained acceptance of these technologies.
  • Privacy-Preserving Design: Developing more robust privacy-preserving mechanisms to further enhance user trust and data security.

Summary updated

Abstract

Modern knowledge work is highly intense and demanding, exposing workers to long-term psychosocial stress. In order to address the problem, stress detection technologies have been developed, enabling the continuous assessment of personal stress based on multimodal sensor data. However, stakeholders lack insights into how employees perceive different monitoring technologies and whether they are willing to share stress-indicative data in order to sustain well-being at the individual, team, and organizational levels in the knowledge work context. To fill this research gap, we developed a theoretical model for knowledge workers’ interest in sharing their stress-indicative data collected with unobtrusive sensors and examined it empirically using structural equation modeling (SEM) with a survey of 181 European knowledge workers. The results did not show statistically significant privacy concerns regarding environmental sensors such as air quality, sound level, and motion sensors. On the other hand, concerns about more privacy-sensitive methods such as tracking personal device usage patterns did not prevent user acceptance nor intent to share data. Overall, knowledge workers were highly interested in employing stress monitoring technologies to measure their stress levels and receive information about their personal well-being. The results validate the willingness to accept the unobtrusive, continuous stress detection in the context of knowledge work.
BibTeX
@inproceedings{kallio2021unobtrusive,
  author = {Johanna Kallio and Elena Vildjiounaite and Julia Kantorovitch and Atte Kinnula and Miguel Bordallo López},
  title = {{Unobtrusive Continuous Stress Detection in Knowledge Work—Statistical Analysis on User Acceptance}},
  booktitle = {Sustainability},
  year = {2021},
  doi = {10.3390/su13042003},
}
Figure