Understanding Organizational Behavior with Wearable Sensing Technology

Waber, Benjamin N.; Olguı́n, D.; Kim, Taemie; Pentland, Alex

doi:10.2139/ssrn.1263992

Cited by 15 publications

(15 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The Sociometric badges have been used to predict organizationally relevant outcomes such as job attitudes and performance (Olguin-Olguin & Pentland, 2010b), job satisfaction (Olguin-Olguin, Waber, et al, 2009), workspace design (Orbach, Demko, Doyle, Waber, & Pentland, 2015), personal and group interaction satisfaction (Waber, Olguin-Olguin, Kim, & Pentland, 2008), network cohesion (Wu, Waber, Aral, Brynjolfsson, & Pentland, 2008), creativity (Tripathi & Burleson, 2012), personality traits (Olguin-Olguin, Gloor, & Pentland, 2009), group performance (Olguin-Olguin, Gloor, et al 2009;Olguin-Olguin & Pentland, 2010a), and group collaboration (Kim, Chang, Holland, & Pentland, 2008). A brief summary of these studies showing the types of sensors and metrics used in prior work is presented in Table 2.…”

Section: Introductionmentioning

confidence: 99%

How honest are the signals? A protocol for validating wearable sensors

et al. 2018

View full text Add to dashboard Cite

There is growing interest among organizational researchers in tapping into alternative sources of data beyond self-reports to provide a new avenue for measuring behavioral constructs. Use of alternative data sources such as wearable sensors is necessary for developing theory and enhancing organizational practice. Although wearable sensors are now commercially available, the veracity of the data they capture is largely unknown and mostly based on manufacturers' claims. The goal of this research is to test the validity and reliability of data captured by one such wearable badge (by Humanyze) in the context of structured meetings where all individuals wear a badge for the duration of the encounter. We developed a series of studies, each targeting a specific sensor of this badge that is relevant for structured meetings, and we make specific recommendations for badge data usage based on our validation results. We have incorporated the insights from our studies on a website that researchers can use to conduct validation tests for their badges, upload their data, and assess the validity of the data. We discuss this website in the corresponding studies.

show abstract

Section: Introductionmentioning

confidence: 99%

How honest are the signals? A protocol for validating wearable sensors

et al. 2018

View full text Add to dashboard Cite

show abstract

“…They provide ongoing and unobtrusive data that can be used to adapt technology to simulate real-world complex simulations while targeting emergent team processes (Kozlowski et al, 2015; Kozlowski and Chao, 2018). Furthermore, Waber et al (2008) discuss how team interactions sensors such as sociometric badges, a smart phone device, have been developed to accumulate data involving “bluetooth to detect people in proximity with one another, infrared to detect closer face-to-face interactions, accelerometers to assess movement, and microphones to detect vocalization” (Kozlowski and Chao, 2018, p. 581). These sociometric badges are unobtrusive, provided to large numbers of participants, and have the ability to obtain real-world data over long periods of time that can subsequently be incorporated as a source for advancing ABSs and computational modeling, avoiding multiple data collection points and ultimately minimizing the use of self-reported surveys.…”

Section: The Road Aheadmentioning

confidence: 99%

“…Luciano et al (2018) discuss how big data is generated through three general types of data streams: (a) behaviors, (b) words, and (c) physiological responses. Sociometric badges is a perfect example of behavior-related data streams due to its ability to measure proximity, movement, or interactions with other team members (Waber et al, 2008). When analyzing word-related data streams, Luciano et al (2018) discuss computer-aided text analysis (CATA) and Hidden Markov Model (HMM) (Pentland, 2007).…”

Section: The Road Aheadmentioning

confidence: 99%

Advancing Teams Research: What, When, and How to Measure Team Dynamics Over Time

2019

View full text Add to dashboard Cite

Teams are complex and dynamic entities that face constant changes to their team structures and must simultaneously work to meet and adapt to the varying situational demands of their environment ( Kozlowski and Ilgen, 2006 ). Agencies, industries, and government institutions are currently placing greater attention to the influence on team dynamics and teamwork as they are important to key organizational outcomes. Due to increased emphasis being placed upon the understanding the maturation of team dynamics, the incorporation of efficient methodological tools to understand how teams are being measured over time becomes critical. Thus, the purpose of this paper is to present a review of relevant academic articles detailing the science behind methodological tools and general approaches to study team dynamics over time. We provide an overview of the methodological tools used to understand team dynamics with accordance to specific temporal elements. Drawing from Kozlowski et al. (1999) process model of team development, we highlight relevant emergent team constructs within each stage. As well, for each stage, we discuss the what and how to measure team dynamics. Our analyses bring to light relevant, novel and complex approaches being used by researchers to examine specific constructs within different team developmental phases (e.g., agent-based simulations, computational modeling) and the importance of transitioning from a single source methodology approach. Implications and future research are also discussed.

show abstract

“…For example, one of the newest iterations of the sociometer, the Sociometric Badge (Olguín‐Olguín, ), has the capacity to collect information on (a) speech features such as volume, tone of voice, and speaking time; (b) body movement features such as energy and consistency; (c) information regarding people nearby who are also wearing a Sociometric Badge; (d) the proximity of Bluetooth‐enabled devices; and (e) approximate location information. To date, sociometers have been used in research examining interpersonal behavior and performance across a wide variety of organizational settings, including healthcare (Olguín‐Olguín, Gloor, & Pentland, ; Olguín‐Olguín & Pentland, ), marketing (Waber et al, ), sales (Olguín‐Olguín & Pentland, ), and information technology (Wu, Waber, Aral, Brynjolfsson, & Pentland, ).…”

Section: Inferring Noncognitive Skills Through Nonverbal Behavior: a mentioning

confidence: 99%

“…Some of these tools, such as the automated feature extraction technology and sociometers, can be implemented without necessitating that a third‐party judge or rater assess and evaluate nonverbal data, further increasing their practical value within organizational settings. Not surprisingly, organizations have already begun to implement some of these tools, the sociometer, in particular, as part of their employee performance evaluation system (Waber, Olguín‐Olguín, Kim, & Pentland, ). In this sense, nonverbal measures constitute a practical addition to nearly any assessment context, as they do not detract from nor compromise test takers' ability to complete an assessment but still provide administrators with additional information that may be useful in determining test takers' noncognitive skills.…”

mentioning

confidence: 99%

We Are What We Repeatedly Do—But in What Context? The Role of Situational Factors in Assessing Personality via Nonverbal Behavior

Klafehn

Kell

Andrews

et al. 2016

ETS Research Report Series

View full text Add to dashboard Cite

While the vast majority of assessments designed to measure noncognitive skills rely heavily on self‐report, concerns surrounding faking and response bias have led to an increased demand for the development of new and innovative methods by which to measure such traits, particularly in high stakes contexts. The focus of this report is to address one such method—the inference of noncognitive skills, namely personality, through nonverbal behavior—and the role contextual factors may play in influencing the validity of conclusions drawn from its application. Specifically, this report discusses the various ways in which characteristics of the task (e.g., level of cooperation, physical effort) may facilitate or hinder the assessment of personality when personality is being inferred through observable behavior. This discussion is dovetailed by a review of research from the nonverbal assessment and interpersonal task literatures, as well as a synthesis of these literatures to highlight the degree to which situational factors (as manifested through different tasks) influence the assessment of personality via nonverbal behavior. Additionally, the potential value of using noninvasive tools, such as the sociometer, to collect nonverbal behavioral data during these tasks is discussed. The report concludes with a brief commentary on the feasibility of using nonverbal methods to assess noncognitive skills, with a specific focus on the extent to which structuring the context to elicit certain behaviors may influence the validity and robustness of nonverbal responses as a measurement source.

show abstract

Understanding Organizational Behavior with Wearable Sensing Technology

Abstract: Abstract

Cited by 15 publications

References 42 publications

How honest are the signals? A protocol for validating wearable sensors

How honest are the signals? A protocol for validating wearable sensors

Advancing Teams Research: What, When, and How to Measure Team Dynamics Over Time

We Are What We Repeatedly Do—But in What Context? The Role of Situational Factors in Assessing Personality via Nonverbal Behavior

Contact Info

Product

Resources

About