Data Analytics: What To Use and What To Lose

Jonathan Hill

November 6, 2017

Gestalt Principles

  • Gestalt principles describe how the human eye perceives visual elements.

Cont.

Proximity
Similarity
Enclosure
Closure
Continuity
Connection
  • Complex scenes are reduced to simple shapes.
  • The eyes perceive shapes as a single, united form rather than the separate simpler elements involved.

Tables vs. Pictures

Best of Both Worlds: The Web

Friends don’t let friends…

  • Clearly indicates the nature of the relationship? (Yes)
  • Represents the quantities accurately? (No)
  • Makes it easy to compare the quantities? (No)
  • Makes it easy to see the ranked order of values? (No)
  • Makes obvious how people should use the information? (Partially)

  • Clearly indicates the nature of the relationship? (Yes)
  • Represents the quantities accurately? (Yes)
  • Makes it easy to compare the quantities? (Yes)
  • Makes it easy to see the ranked order of values? (Yes)
  • Makes obvious how people should use the information? (Yes)

The most misleading charts of 2015, Fixed

Skewing the y-axis

Fixed

No context

Fixed

No Perspective

The only #climatechange chart you need to see. http://natl.re/wPKpro - National Review

Fixed

Misleading story

Fixed

Lies, plain and simple

Fixed

DatasauRus

Getting Started

# Install the software
install.packages("datasauRus")
require(datasauRus)
head(
  datasauRus::datasaurus_dozen
)
## # A tibble: 6 x 3
##   dataset       x       y
##     <chr>   <dbl>   <dbl>
## 1    dino 55.3846 97.1795
## 2    dino 51.5385 96.0256
## 3    dino 46.1538 94.4872
## 4    dino 42.8205 91.4103
## 5    dino 40.7692 88.3333
## 6    dino 38.7179 84.8718

Data

dataset mean_x mean_y std_dev_x std_dev_y corr_x_y
away 54.3 47.8 16.8 26.9 -0.064
bullseye 54.3 47.8 16.8 26.9 -0.069
circle 54.3 47.8 16.8 26.9 -0.068
dino 54.3 47.8 16.8 26.9 -0.064
dots 54.3 47.8 16.8 26.9 -0.060
h_lines 54.3 47.8 16.8 26.9 -0.062
high_lines 54.3 47.8 16.8 26.9 -0.069
slant_down 54.3 47.8 16.8 26.9 -0.069
slant_up 54.3 47.8 16.8 26.9 -0.069
star 54.3 47.8 16.8 26.9 -0.063
v_lines 54.3 47.8 16.8 26.9 -0.069
wide_lines 54.3 47.8 16.8 26.9 -0.067
x_shape 54.3 47.8 16.8 26.9 -0.066

Graphs

Thank you!

Jonathan Hill, MPA

Senior Consultant, Data Science DevOps

hill@ficonsulting.com

585-967-5007