Kay 2016 - When-ish is My Bus
Routing Summary
Fully captures Kay, Kola, Hullman and Munson (2016), When (ish) is My Bus? User-centered Visualizations of Uncertainty in Everyday, Mobile Predictive Systems (CHI ‘16): survey and design requirements, the quantile dotplot, the 541-participant beta-regression experiment, and design implications. 5 notes.
- Need the paper on one page or the section-to-note map? → Kay 2016 - When-ish is My Bus - Overview
- Need the survey results or the five design elements? → Kay 2016 - Transit User Survey and Design Requirements
- Need how to build a quantile dotplot, or interval by counting? → Kay 2016 - Quantile Dotplot Construction
- Need the experiment, SDs, ratios or the beta regression? → Kay 2016 - Experiment Beta Regression and Results
- Need subitizing, trade-offs or limitations? → Kay 2016 - Discussion and Design Implications
Concept Map
| Concept | Note | Type | Depends On | Key Result |
|---|---|---|---|---|
| Design requirements | Kay 2016 - Transit User Survey and Design Requirements | definition | survey of 172 users | point + probabilistic + status + freshness |
| Quantile dotplot | Kay 2016 - Quantile Dotplot Construction | definition | quantile function | dots at |
| Interval by counting | Kay 2016 - Quantile Dotplot Construction | theorem | quantile dotplot | dots from tail gives |
| Error decomposition, beta regression | Kay 2016 - Experiment Beta Regression and Results | definition | logit error | bias vs variance in logit space |
| Variance results | Kay 2016 - Experiment Beta Regression and Results | example | model | dotplot-20 0.11 vs density 0.13 (1.15x) |
| Subitizing recommendation | Kay 2016 - Discussion and Design Implications | theorem | experiment | use small numbers of outcomes |
Notes
- Kay 2016 - When-ish is My Bus - Overview — CONTAINS: question, contributions, key result, motivating E16/E90, Source Map.
- Kay 2016 - Transit User Survey and Design Requirements — CONTAINS: background arguments, survey goals and unmet needs, five design elements, layout decisions.
- Kay 2016 - Quantile Dotplot Construction — CONTAINS: Def. quantile dotplot, counting property, encodings matrix, verified 20-dot example, ggdist/PyMC sketch.
- Kay 2016 - Experiment Beta Regression and Results — CONTAINS: method, participants, model spec and priors, Fig. 6-7 values, confidence and appeal.
- Kay 2016 - Discussion and Design Implications — CONTAINS: subitizing, attitudes, trade-offs, future work, risk slider.
Sources
- Kay Kola Hullman Munson 2016 - When (ish) is My Bus — CHI 2016, 12 pages, author-hosted at mjskay.com.