Search

While there is increasing global attention to data privacy, most of their current theoretical understanding is based on research conducted in a few countries. Prior work argues that people's cultural backgrounds might shape their privacy concerns; thus, we could expect people from different world regions to conceptualize them in diverse ways. We collected and analyzed a large-scale dataset of tweets about the #CambridgeAnalytica scandal in Spanish and English to start exploring this hypothesis. We employed word embeddings and qualitative analysis to identify which information privacy concerns are present and characterize language and regional differences in emphasis on these concerns. Our results suggest that related concepts, such as regulations, can be added to current information privacy frameworks. We also observe a greater emphasis on data collection in English than in Spanish. Additionally, data from North America exhibits a narrower focus on awareness compared to other regions under study. Our results call for more diverse sources of data and nuanced analysis of data privacy concerns around the globe.

Online fanfiction communities enable young people to make connections through their shared interests. Prior research has found that informal learning takes place in fanfiction platforms through distributed mentoring, which includes the writing and receiving of story reviews to one another that have significant effects on young authors’ literacy skill development. The pandemic lockdown has limited young people’s access to forming social connections, and it also stimulated engagement in online informal learning. Therefore, it is crucial to study online fanfiction communities as a prototype for designing future informal learning platforms. This study focuses on reviewing, one of the main components of distributed mentoring, to explore whether reciprocal patterns exist in the sending and receiving of reviews among fanfiction authors. We used a social network analysis approach by modeling users’ interactions through reviews as directed graphs and performed correlation analysis to test the proposed hypotheses. The results provided significant evidence that supports the existence of reciprocal patterns among users’ interactions through reviews.

The recent rise in online education and the accompanying difficulties encountered by both students and educators demonstrate the value of better understanding how online environments can facilitate learning and community. Fanfiction websites contain an enormous amount of original creative writing, primarily written by young people, offering an opportunity to examine informal learning within an online community. Previous research has shown that fanfiction encourages this informal learning through “distributed mentoring,” which occurs when members of an online community quickly and asynchronously receive abundant feedback, while giving others feedback in turn [1]. However, not all interaction in online communities has such a positive nature, as a study of massive open online courses found students who received longer feedback became less likely to reciprocate with similar effort when leaving peer reviews [4]. This study looked at reciprocity between fanfiction reviewers and found a moderate, statistically significant correlation between the quality of reviews given and received by fanfiction authors. These findings are valuable in understanding the ways in which members of online communities interact and learn from each other, an area that would benefit from further research.

Data science provides powerful tools and methods. CSCW researchers have contributed insightfulstudies of conventional work-practices in data science - and particularly machine learning. However,recent research has shown that human skills and collaborative decision-making, play important rolesin defining data, acquiring data, curating data, designing data, and creating data. This workshopgathers researchers and practitioners together to take a collective and critical look at data sciencework-practices, and at how those work-practices make crucial and often invisible impacts on theformal work of data science. When we understand the human and social contributions to data sciencepipelines, we can constructively redesign both work and technologies for new insights, theories, andchallenges.

Social media platforms and social network sites generate a multitude of digital trace behavioral data, the scale of which often necessitates the use of computational data science methods. On the other hand, the socio-behavioral and often relational nature of the social media data requires the attention to context of user activity traditionally associated with qualitative analysis. Human-Centered Data Science (HCDS) attempts to bridge this gap by both harnessing the power of computational techniques and accounting for highly situated and nuanced nature of the social media activity. In this workshop we plan to consider the methods, pitfalls, and approaches of how to do HCDS effectively. Moreover, from collating and organizing these approaches we hope to progress to considering best (or at least common) practices in HCDS.

International audience; The Nearby Supernova Factory presents an interim data release of spectrophotometric timeseries of 210 SNe Ia. Two slightly different versions of the data are included, corresponding to the training data sets used for the SNEMO and SUGAR Type Ia models. The data has been shifted to the restframe and is blinded with respect to cosmology.

Machine learning (ML) has become increasingly influential to human society, yet the primary advancements and applications of ML are driven by research in only a few computational disciplines. Even applications that affect or analyze human behaviors and social structures are often developed with limited input from experts outside of computational fields. Social scientists—experts trained to examine and explain the complexity of human behavior and interactions in the world—have considerable expertise to contribute to the development of ML applications for human-generated data, and their analytic practices could benefit from more human-centered ML methods. Although a few researchers have highlighted some gaps between ML and social sciences [51, 57, 70], most discussions only focus on quantitative methods. Yet many social science disciplines rely heavily on qualitative methods to distill patterns that are challenging to discover through quantitative data. One common analysis method for qualitative data is qualitative coding . In this article, we highlight three challenges of applying ML to qualitative coding. Additionally, we utilize our experience of designing a visual analytics tool for collaborative qualitative coding to demonstrate the potential in using ML to support qualitative coding by shifting the focus to identifying ambiguity. We illustrate dimensions of ambiguity and discuss the relationship between disagreement and ambiguity. Finally, we propose three research directions to ground ML applications for social science as part of the progression toward human-centered machine learning.

Many successful digital interfaces employ visual metaphors to convey features or data properties to users, but the characteristics that make a visual metaphor effective are not well understood. We used a theoretical conception of metaphor from cognitive linguistics to design an interactive system for viewing the citation network of the corpora of literature in the JSTOR database, a highly connected compound graph of 2 million papers linked by 8 million citations. We created 4 variants of this system, manipulating 2 distinct properties of metaphor. We conducted a between-subjects experimental study with 80 participants to compare understanding and engagement when working with each version. We found that building on known image schemas improved response time on look-up tasks, while contextual detail predicted increases in persistence and the number of inferences drawn from the data. Schema-congruency combined with contextual detail produced the highest gains in comprehension. These findings provide concrete mechanisms by which designers presenting large data sets through metaphorical interfaces may improve their effectiveness and appeal with users.

The study of emoticon use in text communication is in its early stages (Aragon, Feldman, Chen & Kroll, 2014), with even less known about how emoticons function in multilingual environments. We describe a preliminary longitudinal analysis of text communication in an online bilingual scientific work environment and demonstrate how patterns of emoticon use constitute a novel yet systematic nonverbal aspect of communication. Specifically, coordination over bilingual speakers entails reductions in emoticon diversity over time that are greater for those who communicate in their L2 than in their L1. An analogous but weaker pattern is evident for lexical diversity in L2 but not L1. We hypothesize that reductions in emoticon diversity in the L2 are likely to reflect social contributions to alignment rather than purely proficiency.

The Cambridge Analytica scandal triggered a conversation on Twitter about data practices and their implications. Our research proposes to leverage this conversation to extend the understanding of how information privacy is framed by users worldwide. We collected tweets about the scandal written in Spanish and English between April and July 2018. We created a word embedding to create a reduced multi-dimensional representation of the tweets in each language. For each embedding, we conducted open coding to characterize the semantic contexts of key concepts: "information", "privacy", "company" and "users" (and their Spanish translations). Through a comparative analysis, we found a broader emphasis on privacy-related words associated with companies in English. We also identified more terms related to data collection in English and fewer associated with security mechanisms, control, and risks. Our findings hint at the potential of cross-language comparisons of text to extend the understanding of worldwide differences in information privacy perspectives., Comment: Proceeding CSCW '19: Conference Companion Publication of the 2019 on Computer Supported Cooperative Work and Social Computing

Currently, there is a limited understanding of how data privacy concerns vary across the world. The Cambridge Analytica scandal triggered a wide-ranging discussion on social media about user data collection and use practices. We conducted an inter-language study of this online conversation to compare how people speaking different languages react to data privacy breaches. We collected tweets about the scandal written in Spanish and English between April and July 2018. We used the Meaning Extraction Method in both datasets to identify their main topics. They reveal a similar emphasis on Zuckerberg's hearing in the US Congress and the scandal's impact on political issues. However, our analysis also shows that while English speakers tend to attribute responsibilities to companies, Spanish speakers are more likely to connect them to people. These findings show the potential of inter-language comparisons of social media data to deepen the understanding of cultural differences in data privacy perspectives., Comment: 2019 World Wide Web Conference (WWW '19 Companion)

Raiding is a format in digital gaming that requires groups of people to collaborate and/or compete for a common goal. In 2017, the raiding format was introduced in the location-based mobile game Pokemon GO, which offers a mixed reality experience to friends and strangers coordinating for in-person raids. To understand this technology-mediated social phenomenon, we conducted over a year of participant observations, surveys with 510 players, and interviews with 25 players who raid in Pokemon GO. Using the analytical lens of Arrow, McGrath, and Berdahl's theory of small groups as complex systems, we identify global, local, and contextual dynamics in location-based raiding that support and challenge ad-hoc group formation in real life. Based on this empirical and theoretical understanding, we discuss implications to design for transparency, social affordances, and bridging gaps between global and contextual dynamics for increased positive and inclusive community interactions.

The process of qualitative coding often involves multiple coders coding the same data to ensure reliable codes and a consistent understanding of the codebook. One aspect of qualitative coding includes resolving disagreements, where coders discuss differences in coding to reach a consensus. We conduct a case study to evaluate four strategies of disagreement resolution and understand their impact on the coding process. We find that an open discussion and the n-ary tree metric lead coders to focus more on the disagreement of a particular data instance, whereas kappa values and Code Wizard direct coders to compare code definitions. We discuss opportunities for using different strategies at different stages of the coding process for more effective disagreement resolution.

As the era of ‘big data’ unfolds, researchers are increasingly engaging with large, complex data sets compiled from heterogeneous sources and distributed across networked technologies. The nature of these data sets makes it difficult to grasp and manipulate their materiality. We argue that moments of breakdown – points at which progress is stopped due to a material limitation – provide opportunities for researchers to develop new imaginations and configurations of their data sets' materiality, and serve as underappreciated resources for knowledge production. In our ethnographic study of data-intensive research in an academic setting, we emphasize the layers of repair work required to address breakdown, and highlight incremental innovations that stem from this work. We suggest that a focus on the breakdown–repair process can facilitate nuanced understandings of the relationships and labour involved in constituting data assemblages and constructing knowledge from them.

Inspired by the ACM SIGCHI Across Borders Initiative, this workshop focuses on ongoing CSCW research in, or about, Latin America (LATAM). We seek to position LATAM as the common context that unites students, academic and industry researchers who participate in the workshop. Our goals are: (1) to discuss the opportunities and challenges of doing CSCW research centered on LATAM, (2) to collaboratively mentor emerging projects focused on LATAM, and (3) to make LATAM research projects more visible to the international community. Senior CSCW researchers from LATAM and other regions will be invited to discuss the work presented at the event in order to facilitate a greater integration of Latin American CSCW into the CSCW community at large.

Communities focused on creating and sharing fanworks have existed since before the internet, but have thrived in the presence of online platforms. From young people learning literacy skills through writing fanfiction about their favorite media to complex infrastructural work in fans creating their own platforms to longstanding social norms that promote inclusivity, these communities showcase many core interests of CSCW-however, there has been little research focused on fandom within social computing and HCI. This panel brings together scholars who have conducted research related to online fandom to discuss their experiences, the challenges they have faced, and the vast opportunity for more work in this area from a range of perspectives.

Astronomy and astrophysics 644, A176 (1-20) (2020). doi:10.1051/0004-6361/201730404, As part of an on-going effort to identify, understand and correct for astrophysics biases in the standardization of Type Ia supernovae (SNIa) for cosmology, we have statistically classified a large sample of nearby SNeIa into those located in predominantly younger or older environments. This classification is based on the specific star formation rate measured within a projected distance of 1kpc from each SN location (LsSFR). This is an important refinement compared to using the local star formation rate directly as it provides a normalization for relative numbers of available SN progenitors and is more robust against extinction by dust. We find that the SNeIa in predominantly younger environments are Δ$_Y$ = 0.163 ± 0.029 mag (5.7σ) fainter than those in predominantly older environments after conventional light-curve standardization. This is the strongest standardized SN Ia brightness systematic connected to host-galaxy environment measured to date. The well-established step in standardized brightnesses between SNeIa in hosts with lower or higher total stellar masses is smaller at Δ$_M$ = 0.119 ± 0.032 mag (4.5σ), for the same set of SNeIa. When fit simultaneously, the environment age offset remains very significant, with Δ$_Y$ = 0.129 ± 0.032 mag (4.0σ), while the global stellar mass step is reduced to Δ$_M$ = 0.064 ± 0.029 mag (2.2σ). Thus, approximately 70% of the variance from the stellar mass step is due to an underlying dependence on environment-based progenitor age. Standardization using only the SNeIa in younger environments reduces the total dispersion from 0.142 ± 0.008 mag to 0.120 ± 0.010 mag. We show that as environment ages evolve with redshift a strong bias on measurement of the dark energy equation of state parameters can develop. Fortunately, data to measure and correct for this effect is likely to be available for many next-generation experiments., Published by EDP Sciences, Les Ulis

Astronomy and astrophysics 615, A162 (2018). doi:10.1051/0004-6361/201832932, Context. Type Ia supernovae (SNe Ia) are widely used to measure the expansion of the Universe. To perform such measurements the luminosity and cosmological redshift (z) of the SNe Ia have to be determined. The uncertainty on z includes an unknown peculiar velocity, which can be very large for SNe Ia in the virialized cores of massive clusters.Aims. We determine which SNe Ia exploded in galaxy clusters using 145 SNe Ia from the Nearby Supernova Factory. We then study how the correction for peculiar velocities of host galaxies inside the clusters improves the Hubble residuals.Methods. We found 11 candidates for membership in clusters. We applied the biweight technique to estimate the redshift of a cluster. Then, we used the galaxy cluster redshift instead of the host galaxy redshift to construct the Hubble diagram.Results. For SNe Ia inside galaxy clusters, the dispersion around the Hubble diagram when peculiar velocities are taken into account is smaller compared with a case without peculiar velocity correction, which has a wRMS = 0.130 ± 0.038 mag instead of wRMS = 0.137 ± 0.036 mag. The significance of this improvement is 3.58σ. If we remove the very nearby Virgo cluster member SN2006X (z < 0.01) from the analysis, the significance decreases to 1.34σ. The peculiar velocity correction is found to be highest for the SNe Ia hosted by blue spiral galaxies. Those SNe Ia have high local specific star formation rates and smaller stellar masses, which is seemingly counter to what might be expected given the heavy concentration of old, massive elliptical galaxies in clusters.Conclusions. As expected, the Hubble residuals of SNe Ia associated with massive galaxy clusters improve when the cluster redshift is taken as the cosmological redshift of the supernova. This fact has to be taken into account in future cosmological analyses in order to achieve higher accuracy for cosmological redshift measurements. We provide an approach to do so., Published by EDP Sciences, Les Ulis

Collaborative qualitative coding often involves coders assign- ing different labels to the same instance, leading to ambiguity. We refer to such an instance of ambiguity as disagreement in coding. Analyzing reasons for such a disagreement is essential-- both for purposes of bolstering user understanding gained from coding and reinterpreting the data collaboratively, and for negotiating user-assigned labels for building effective machine learning models. We propose a conceptual definition of collective disagreement using diversity and divergence within the coding distributions. This perspective of disagreement translates to diverse coding contexts and groups of coders irrespective of discipline. We introduce two tree-based ranking metrics as standardized ways of comparing disagreements in how data instances have been coded. We empirically validate that, of the two tree-based metrics, coders' perceptions of dis- agreement match more closely with the n-ary tree metric than with the post-traversal tree metric.

Context. Observations of Type Ia supernovae (SNe Ia) can be used to derive accurate cosmological distances through empirical standardization techniques. Despite this success neither the progenitors of SNe Ia nor the explosion process are fully understood. The U-band region has been less well observed for nearby SNe, due to technical challenges, but is the most readily accessible band for high-redshift SNe. Aims. Using spectrophotometry from the Nearby Supernova Factory, we study the origin and extent of U-band spectroscopic variations in SNe Ia and explore consequences for their standardization and the potential for providing new insights into the explosion process. Methods. We divide the U-band spectrum into four wavelength regions {\lambda}(uNi), {\lambda}(uTi), {\lambda}(uSi) and {\lambda}(uCa). Two of these span the Ca H&K {\lambda}{\lambda} 3934, 3969 complex. We employ spectral synthesis using SYNAPPS to associate the two bluer regions with Ni/Co and Ti. Results. (1) The flux of the uTi feature is an extremely sensitive temperature/luminosity indicator, standardizing the SN peak luminosity to 0.116 $\pm$ 0.011 mag RMS. A traditional SALT2.4 fit on the same sample yields a 0.135 mag RMS. Standardization using uTi also reduces the difference in corrected magnitude between SNe originating from different host galaxy environments. (2) Early U-band spectra can be used to probe the Ni+Co distribution in the ejecta, thus offering a rare window into the source of lightcurve power. (3) The uCa flux further improves standardization, yielding a 0.086 $\pm$ 0.010 mag RMS without the need to include an additional intrinsic dispersion to reach {\chi}$^2$/dof $\sim$ 1. This reduction in RMS is partially driven by an improved standardization of Shallow Silicon and 91T-like SNe., Comment: 19 pages, 17 figures, accepted for publication in A&A

With its roots dating to popular television shows of the 1960s such as Star Trek, fanfiction has blossomed into an extremely widespread form of creative expression. The transition from printed zines to online fanfiction repositories has facilitated this growth in popularity, with millions of fans writing stories and adding daily to sites such as Archive Of Our Own, Fanfiction.net, FIMfiction.net, and many others. Enthusiasts are sharing their writing, reading stories written by others, and helping each other to grow as writers. Yet, this domain is often undervalued by society and understudied by researchers. To facilitate the study of this large but often marginalized community, we present a fully anonymized data release (via differential privacy) of the metadata from a large fanfiction site (to protect author privacy, story, profile, and review text is excluded, and only metadata is provided). We use visual analytics techniques to draw several intriguing insights from the data and show the potential for future research. We hope other researchers can use this data to explore further questions related to online fanfiction communities.

Qualitative coding offers the potential to obtain deep insights into social media, but the technique can be inconsistent and hard to scale. Researchers using qualitative coding impose structure on unstructured data through “codes” that represent categories for analysis. Our visual analytics interface, Aeonium, supports human insight in collaborative coding through visual overviews of codes assigned by multiple researchers and distributions of important keywords and codes. The underlying machine learning model highlights ambiguity and inconsistency. Our goal was not to reduce qualitative coding to a machine-solvable problem, but rather to bolster human understanding gained from coding and reinterpreting the data collaboratively. We conducted an experimental study with 39 participants who coded tweets using our interface. In addition to increased understanding of the topic, participants reported that Aeonium's collaborative coding functionality helped them reflect on their own interpretations. Feedback from participants demonstrates that visual analytics can help facilitate rich qualitative analysis and suggests design implications for future exploration.

A distance cartogram (DC) is a technique that alters distances between a user-specified origin and the other locations in a map with respect to travel time. With DC, users can weigh the relative travel time costs between the origin and potential destinations at a glance because travel times are projected in a linearly interpolated time space from the origin. Such glance-ability is known to be useful for travelers who are mindful of travel time when finding their travel destinations. When constructing DC, however, uneven urban traffic conditions introduce excessive distortion and challenge user intuition. In addition, there has been little research focusing on DC's user interaction design. To tackle these challenges and realize the potential of DC as an interactive decision-making support tool, we derive a set of useful interactions through two formative studies and devise two novel techniques called Geo-contextual Anchoring Projection and Scalable Road-network Construction. We develop an interactive map system using these techniques and evaluate this system by comparing it against an equidistant map (EM), a widely used conventional layout that preserves the geographical reality. Based on the analysis of user behavior and qualitative feedback, we identify several benefits of using DC itself and of the interaction techniques we derived. We also analyze the specific reasons behind these identified benefits.

From Harry Potter to American Horror Story, fanfiction is extremely popular among young people. Sites such as Fanfiction.net host millions of stories, with thousands more posted each day. Enthusiasts are sharing their writing and reading stories written by others. Exactly how does a generation known more for videogame expertise than long-form writing become so engaged in reading and writing in these communities? Via a nine-month ethnographic investigation of fanfiction communities that included participant observation, interviews, a thematic analysis of 4,500 reader reviews and an in-depth case study of a discussion group, we found that members of fanfiction communities spontaneously mentor each other in open forums, and that this mentoring builds upon previous interactions in a way that is distinct from traditional forms of mentoring and made possible by the affordances of networked publics. This work extends and develops the theory of distributed mentoring. Our findings illustrate how distributed mentoring supports fanfiction authors as they work to develop their writing skills. We believe distributed mentoring holds potential for supporting learning in a variety of formal and informal learning environments.

Social media has become a fruitful platform on which to study human behavior and social phenomena. However, social media data are usually messy, disorganized, and noisy, which makes finding patterns in such data a challenging task. Visualization can help with the exploration of such massive data. Researchers studying social media often begin by reviewing related research. In this paper, we consider the idea that information from related research can be incorporated into social media visualization tools in order to spark creativity and guide exploration. To develop an effective overview of social media research with which to seed our tool, we conducted a content analysis of social media related papers and designed SparQs, a visual analytics tool to spark creativity in social media exploration. We conducted a pilot evaluation with three social media researchers as well as a participatory design workshop to explore further directions.

We draw parallels between emoticons in textual communication and gesture in signed language with respect to the interdependence of codes by describing two contexts under which the behavior of emoticons in textual communication resembles that of gesture in speech. Generalizing from those findings, we propose that gesture is likely characterized by a nuanced interdependence with language whether signed, spoken or texted.

Background Validated methods of objective assessments of surgical skills are resource intensive. We sought to test a web-based grading tool using crowdsourcing called Crowd-Sourced Assessment of Technical Skill. Materials and methods Institutional Review Board approval was granted to test the accuracy of Amazon.com's Mechanical Turk and Facebook crowdworkers compared with experienced surgical faculty grading a recorded dry-laboratory robotic surgical suturing performance using three performance domains from a validated assessment tool. Assessor free-text comments describing their rating rationale were used to explore a relationship between the language used by the crowd and grading accuracy. Results Of a total possible global performance score of 3-15, 10 experienced surgeons graded the suturing video at a mean score of 12.11 (95% confidence interval [CI], 11.11-13.11). Mechanical Turk and Facebook graders rated the video at mean scores of 12.21 (95% CI, 11.98-12.43) and 12.06 (95% CI, 11.57-12.55), respectively. It took 24 h to obtain responses from 501 Mechanical Turk subjects, whereas it took 24 d for 10 faculty surgeons to complete the 3-min survey. Facebook subjects (110) responded within 25 d. Language analysis indicated that crowdworkers who used negation words (i.e., "but," "although," and so forth) scored the performance more equivalently to experienced surgeons than crowdworkers who did not (P < 0.00001). Conclusions For a robotic suturing performance, we have shown that surgery-naive crowdworkers can rapidly assess skill equivalent to experienced faculty surgeons using Crowd-Sourced Assessment of Technical Skill. It remains to be seen whether crowds can discriminate different levels of skill and can accurately assess human surgery performances. © 2014 Elsevier Inc. All rights reserved.

Affect has been identified as an important component of the communication practices of distributed teams. Our emerging theory of distributed affect moves beyond the individual as the primary unit of analysis, focusing instead on affect as a dynamic group process. Drawing upon a data set of over four years of chat logs from a distributed scientific collaboration relying on text-based communication to coordinate their work, we expand upon the framework of distributed affect and characterize the concept through five features: transference, resonance, pervasiveness, persistence, and representation. These features provide a set of descriptive components for interactions between people and their environment, their tools, and their present and historical references as part of a dynamical system of affect. We examine specific events in the group's history which highlight the dynamic way affect is operating in this context, and how it influences factors such as creative problem solving. The framework we describe offers a unique analytic lens for the study of computer-supported group work, and a useful tool for framing questions about the continued study of affect in collaborative teams.

The study and analysis of large and complex data sets offer a wealth of insights in a variety of applications. Computational approaches provide researchers access to broad assemblages of data, but the insights extracted may lack the rich detail that qualitative approaches have brought to the understanding of sociotechnical phenomena. How do we preserve the richness associated with traditional qualitative methods while utilizing the power of large data sets? How do we uncover social nuances or consider ethics and values in data use? These and other questions are explored by human-centered data science, an emerging field at the intersection of human-computer interaction (HCI), computer-supported cooperative work (CSCW), human computation, and the statistical and computational techniques of data science. This workshop, the first of its kind at CSCW, seeks to bring together researchers interested in human-centered approaches to data science to collaborate, define a research agenda, and form a community.

Residential thermostats control a substantial portion of both fuel and electrical energy—9% of the total energy consumption in the U.S. Consumers install programmable thermostats to save energy, yet numerous recent studies found that homes with programmable thermostats can use more energy than those controlled manually depending on how—or if—they are used. At the same time, thermostats are undergoing a dramatic increase in capability and features, including control of ventilation, responding to electricity price signals, and interacting with a home area network. These issues warrant a review of the current state of thermostats, evaluating their effectiveness in providing thermal comfort and energy savings, and identifying areas for further improvement or research. This review covers the evolution in technologies of residential thermostats; we found few standards and many features. We discuss studies of how people currently use thermostats, finding that nearly half do not use the programming features. The review covers the complications associated with using a thermostat. Finally, we suggest research needed to design—and especially test with users—thermostats that can provide more comfortable and economical indoor environments.

Residential thermostats control 9% of the total energy use in the United States and similar amounts in most developed countries; however, the details of how people use them have been largely ignored. Five parallel investigations related to the usability of residential thermostats were undertaken. No single investigation was representative of the whole population, but each gave insights into different groups or usage patterns. Personal interviews revealed widespread misunderstanding of thermostat operation. The on-line surveys found that most thermostats were selected by previous residents, landlords, or other agents. The majority of occupants operated thermostats manually, rather than relying on their programmable features and almost 90% of respondents reported that they rarely or never adjusted the thermostat to set a weekend or weekday program. Photographs of thermostats were collected in one on-line survey, which revealed that about 20% of the thermostats displayed the wrong time and that about 50% of the respondents set their programmable thermostats on “long term hold” (or its equivalent). Low-income families were visited and their thermostats photographed. Even though 85% of the respondents declared that they use programming features to automatically raise or lower the temperature, the photos indicated that 45% were in hold. Laboratory tests were undertaken to measure usability of thermostats. A measurement protocol was developed and a metric was created that could quantitatively distinguish usability among five thermostats. This metric could be used to establish minimum levels of usability in programmable thermostats and other energy-using devices with complex controls.