Identifying and Measuring Urban
Design Qualities Related to
Walkability
Funding by the Active Living Research program
of The Robert Wood Johnson Foundation
Reid Ewing, Otto Clemente,
Susan Handy, Ross Brownson, and
Emily Winston
Growing Body of Evidence
Mostly Gross Qualities
Even Audit Instruments
Not a Lot to Do with Experience
of Walking Down a Street
Genesis of Project
• Christopher Alexander, A Pattern Language -
Towns Buildings Construction
• Richard Hedman, Fundamentals of Urban
Design
• Allan Jacobs, Great Streets
• Kevin Lynch, The Image of the City
• Amos Rapoport, History and Precedent in
Environmental Design
• William H. Whyte, City—Rediscovering the
Center
Conceptual Framework
Physical
features
-Sidewalk width
-Street width
-Traffic volumes Perceptual Overall
Walking
-Tree canopy qualities walking
behavior
-Building height quality
-Number of people -Imageability
-Weather -Legibility
-Etc. -Complexity Individual
-Coherence reactions
-Enclosure
-Human scale -Sense of Safety
-Linkage -Sense of Comfort
-Transparency -Level of Interest
More objective More subjective
Work Plan
• Expert Panel
• Literature Review with Definitions
• Library of Video Clips
• Visual Assessment Survey
• Quantitative and Qualitative Analyses
• Instrument Development, Testing, and
Training
• Illustrated Field Manual
National Expert Panel
• Victor Dover - urban designer
• Rob Lane - urban designer
• Geoffrey Ferrell - urban designer/code expert
• Tony Nelessen - urban designer
• Anne Vernez Moudon - urban designer/planner
• Mark Francis - landscape architect
• Michael Southworth - urban designer
• Michael Kwartler - architect/simulations expert
• John Peponis - architect/space syntax expert
• Dan Stokols - social ecologist
Perceptual Qualities
41 qualities • Imageability
• Enclosure
• Human Scale
• Transparency
8 qualities • Linkage
• Complexity
• Coherence
• Legibility
Textbook Definitions
Imageability is the quality of a place
that makes it recognizable and
memorable. A place has high
imageability when specific physical
elements and their arrangement
evoke distinct images or positive
feelings.
Video Library
• Urban Streetscapes
• 22 Cities
• 205 Video Clips
• 86 Clips Rated by Research Team
• 48 Clips Rated by Panel
Fractional Factorial Design
full factorial design =
28 design = 256 clips
1/16 fractional design =
28-4 design = 16 clips
Best Match Run
High Values of All Eight Qualities
Best Match Run
High Values of Imageability, Human
Scale, Linkage, and Tidiness
Best Match Run
High Values of Enclosure, Transparency,
Complexity, and Coherence
Visual Assessment Survey Form
Individual Ratings
Walking Quality
Transparency
Human Scale
Imageability
Complexity
Coherence
Enclosure
Legibility
Tidiness
Linkage
Peponis 4 4 5 3 3 4 4 2 4 4
Southworth 5 5 5 4 4 3 4 3 5 5
Lane 5 5 5 5 5 5 5 4 5 5
Ferrell 5 4 5 4 4 4 5 4 5 4.8
Nelessen 4 4 4 4 4 4 4 3 4 4
Dover 5 4 5 5 5 5 4 5 5 5.0
Regression Model for Overall
Walkability
Standardized
Variable Coefficient Coefficient t-statistic p-value
constant -0.226 -1.503 0.140
human scale 0.411 0.420 5.814 0.000
transparency 0.137 0.149 2.366 0.023
tidiness 0.070 0.059 1.598 0.117
enclosure 0.140 0.157 2.504 0.016
imageability 0.307 0.310 5.153 0.000
N 48
R-square .959
Adjusted R-
.954
square
Inter-rater Reliability
Ratings of Urban Design
Qualities
Intra-class
Correlation 95% Confidence Cronbach’s
Coefficient Interval of ICC Alpha
imageability .494 .385-.618 .930
legibility .380 .276-.509 .895
enclosure .584 .478-.697 .945
human scale .508 .399-.630 .928
transparency .499 .390-.622 .926
linkage .344 .169-.621 .896
complexity .508 .398-.632 .926
coherence .374 .271-.504 .880
Tidiness .421 .314-.550 .915
N 48
Content Analysis of Sampled
Scenes
• 48 Clips
• More than 100 Physical Features
• Operational Definitions for All
Feature
• 1 Hour-plus per Clip
• Gold Standard
• Inter-rater Reliability Test
Selection of Physical Features
• Logic/Common Sense
• Literature
• Interviews with Expert Panel
• Operational Definitions from Expert
Panel
Operational Definitions
Inter-rater Reliability
Estimates of Physical Features
Variable ICC Alpha Variable ICC Alpha
proportion
number of
street wall –
courtyards .471 .611 .588 .737
opposite
etc.
side
number of
arcades --- --- enclosed .389 .640
sides
average
number of building
.763 .878 .215 .338
landmarks setback –
same side
Relate Urban Design
Qualities Ratings to Physical
Features
Cross-Classified Random
Effects Model
• New Class of Model
• Only a Handful of Applications
• Hierarchical in Nature
• Partitions Variance
Partitioning Total Variance
Viewer Measurement
Scene Variance Variance Error Total Variance
0.67 0.16 0.50 1.33
imageability
(50) (12) (38)
0.46 0.17 0.55 1.18
legibility
(39) (14) (47)
0.83 0.10 0.48 1.41
enclosure
(59) (7) (34)
0.68 0.11 0.50 1.29
human scale
(53) (8) (39)
0.77 0.13 0.62 1.52
transparency
(51) (8) (41)
0.51 0.26 0.74 1.51
linkage
(34) (17) (39)
0.6 0.09 0.47 1.16
complexity
(52) (8) (40)
Cross-Classified Random Effects
Models (fixed effects too)
actual score = predicted score +
measurement error
predicted score = constant + viewer
random effect + scene random
effect + a*viewer variables +
b*scene variables
Best-Fit Transparency Model
Variable Coefficient t-statistic p-value
constant 1.709
proportion first floor 1.219 3.13 0.002
with windows
proportion active uses 0.533 2.96 0.004
proportion street wall 0.666 2.57 0.011
– same side
Proportion of Scene 0.62
Variance Explained
Proportion of Total 0.32
Variance Explained
Performance of Urban Design
Qualities
Relationship to Inter-rater Portion of Scene Inter-rater Criteria Met
Walkability in reliability Variance/Total Reliability of
Best-Fit Model (ICC) Variance Significant
(p-value) Explained by Variables
Best-Fit Models (number with
ICC>0.4)
Imageability 0.000 0.494 0.72/0.37 7 of 7 5 of 5
(1 missing)
Legibility --- 0.380 0.54/0.21 5 of 5 1 of 5
(1 missing)
Enclosure 0.016 0.584 0.72/0.43 5 of 5 5 of 5
Human scale 0.000 0.508 0.62/0.35 7 of 7 5 of 5
Transparency 0.023 0.499 0.62/0.32 3 of 3 5 of 5
Linkage --- 0.344 0.61/0.21 4 of 5 1 of 5
Complexity --- 0.508 0.73/0.38 5 of 6 3 of 5
Coherence --- 0.374 0.67/0.25 3 of 4 1 of 5
Tidiness 0.117 0.421 0.70/0.30 2 of 3 3 of 5
(1 missing)
Develop Draft Field Manual
Focuses on Urban Design Qualities
that Meet Performance Criteria
• Imageability
• Enclosure
• Transparency
• Human scale
• Complexity
• Tidiness
Qualitative Introduction to
Urban Design Quality
Detailed Illustrated Steps
Scoring Sheet
Next steps
• Test and Refine Instrument
• Provide Lab and Field Training to
Lay Observers
• Finalize Instrument