Models Testing Division of Labor
Chapter 4 evaluated the presence of a division of labor in the maintenance of the urban commons through two tests. The first set of analyses found that both typical and exemplar custodians contributed to the likelihood of a neighborhood reporting a broken sidewalk (Table D.1 ) or streetlight outage (Table D.2 ). The second set of analyses revealed distinctions in the types of situations and contexts in which typical and exemplar custodians reported issues in the public space. They also found that city employees and constituents differed in their patterns of reporting (Table D.3 ).
|
TABLE D.1 Parameter estimates from models using the relative prevalence of typical and exemplar custodians and demographic characteristics to predict a census tract’s likelihood to report a broken sidewalk |
|||||||||||||||||||||||||||||
|
Model 1 |
Model 2 |
Model 3 |
Model 4 |
Model 5 |
Model 6 |
Model 7 |
|||||||||||||||||||||||
|
Beta (SE) |
Stand. Beta |
|
Beta (SE) |
Stand. Beta |
|
Beta (SE) |
Stand. Beta |
|
|
Beta (SE) |
Stand. Beta |
|
Beta (SE) |
Stand. Beta |
|
Beta (SE) |
Stand. Beta |
|
Beta (SE) |
Stand. Beta |
|||||||||
|
Typical custodians / mi2 a |
.0.21*** (0.03) |
.47 |
— |
— |
(0.17*** (0.04) |
.38 |
(0.17*** (0.03) |
.38 |
— |
— |
— |
— |
(0.17*** (0.03) |
.38 |
|||||||||||||||
|
Exemplarsa |
— |
— |
.0.24*** (0.04) |
.41 |
(0.17*** (0.04) |
.29 |
0.17*** (0.04) |
.29 |
— |
— |
— |
— |
0.17*** (0.05) |
.30 |
|||||||||||||||
|
Typical * exemplars |
— |
— |
— |
— |
— |
— |
0.05* (0.03) |
.13 |
— |
— |
— |
— |
0.05* (0.03) |
.14 |
|||||||||||||||
|
Total custodians / mi2 |
— |
— |
— |
— |
— |
— |
— |
— |
.0.22*** (0.03) |
.48 |
— |
— |
— |
— |
|||||||||||||||
|
Total custodians |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
.0.20*** (0.04) |
.35 |
— |
— |
|||||||||||||||
|
% Immigrant |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
(0.16 (0.03) |
.06 |
|||||||||||||||
|
% Hispanica |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
(0.52 (0.03) |
.16 |
|||||||||||||||
|
% Blacka |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
(0.02 (0.01) |
.13 |
|||||||||||||||
|
Median income |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
(0.03 (0.15) |
.02 |
|||||||||||||||
|
Adjusted R 2 |
.21 |
.16 |
.29 |
.30 |
.22 |
.12 |
.31 |
||||||||||||||||||||||
|
Note: n = 155 census tracts. |
|||||||||||||||||||||||||||||
|
a. log-transformed to adjust for skew. |
|||||||||||||||||||||||||||||
|
* p < .05, ** p < .01, *** p < .001. |
|||||||||||||||||||||||||||||
|
TABLE D.2 Parameter estimates from models using the relative prevalence of typical and exemplar custodians and demographic characteristics to predict a census tract’s likelihood to report a streetlight outage |
||||||||||||||||||||||||
|
Model 1 |
Model 2 |
Model 3 |
Model 4 |
Model 5 |
Model 6 |
|||||||||||||||||||
|
Beta (SE) |
Stand. Beta |
|
Beta (SE) |
Stand. Beta |
|
Beta (SE) |
Stand. Beta |
|
Beta (SE) |
Stand. Beta |
|
Beta (SE) |
Stand. Beta |
|
Beta (SE) |
Stand. Beta |
||||||||
|
Typical custodians / mi2 a,b |
.0.17* (0.08) |
.28 |
— |
— |
+0.14+ (0.08) |
.23 |
— |
— |
— |
— |
— |
— |
||||||||||||
|
Exemplarsa |
— |
— |
.0.27* (0.12) |
.30 |
+0.23+ (0.12) |
.26 |
— |
— |
— |
— |
— |
— |
||||||||||||
|
Typical * exemplars |
— |
— |
— |
— |
— |
— |
0.21** (0.08) |
.36 |
— |
— |
— |
— |
||||||||||||
|
Total custodians / mi2 |
— |
— |
— |
— |
— |
— |
— |
— |
.0.17* (0.08) |
.29 |
— |
— |
||||||||||||
|
Total custodians |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
.0.26* (0.12) |
.28 |
||||||||||||
|
% Immigrant |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
||||||||||||
|
% Hispanica |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
||||||||||||
|
% Blacka |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
||||||||||||
|
Median income |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
||||||||||||
|
Adjusted R 2 |
.08 |
.08 |
.11 |
.12 |
.06 |
.06 |
||||||||||||||||||
|
Note: n = 53 census tracts with streetlight outages. A final model found no demographic factors to be significant for which reason they are not reported. |
||||||||||||||||||||||||
|
a. log-transformed to adjust for skew. |
||||||||||||||||||||||||
|
b. During the three months in advance of the streetlight outage audits. |
||||||||||||||||||||||||
|
* p < .05, ** p < .01, *** p < .001. |
||||||||||||||||||||||||
|
TABLE D.3 Parameter estimates from multilevel models predicting whether a report was made by an exemplar or typical custodian or by an employee or constituent based on characteristics of the report, street, and census tract |
||||||||
|
Exemplar vs. typical |
City employee vs. constituent |
|||||||
|
Beta (std. error) |
Odds ratio |
Beta (std. error) |
Odds ratio |
|||||
|
Report characteristics |
||||||||
|
Man-made incivilitya |
0.54*** (0.01) |
1.72 |
−1.10*** (0.01) |
0.33 |
||||
|
Weekenda |
0.09*** (0.02) |
1.10 |
−0.33*** (0.02) |
0.72 |
||||
|
Springa |
0.05** (0.02) |
1.05 |
−0.11*** (0.01) |
0.90 |
||||
|
Summera |
−0.06*** (0.02) |
0.94 |
−0.10*** (0.01) |
0.90 |
||||
|
Falla |
−0.18*** (0.02) |
0.83 |
0.17*** (0.01) |
1.18 |
||||
|
Snow removala |
−0.03 (0.02) |
0.97 |
−2.31*** (0.03) |
0.10 |
||||
|
Street characteristics |
||||||||
|
Lengthb |
−0.06*** (0.01) |
0.94 |
−0.06*** (0.01) |
0.94 |
||||
|
Main street |
0.26*** (0.02) |
1.29 |
0.33*** (0.02) |
1.39 |
||||
|
Commercial |
0.22*** (0.04) |
1.25 |
0.42*** (0.03) |
1.52 |
||||
|
Industrial |
0.34*** (0.06) |
1.40 |
0.57*** (0.05) |
1.77 |
||||
|
Exempt |
0.33*** (0.04) |
1.39 |
0.46*** (0.04) |
1.58 |
||||
|
No zoning |
0.50*** (0.04) |
1.65 |
0.46*** (0.03) |
1.59 |
||||
|
Neighborhood characteristics |
||||||||
|
Population densityc |
0.01*** (0.001) |
1.01 |
−0.01** (0.002) |
0.99 |
||||
|
Downtown |
0.14 (0.10) |
1.15 |
−0.42** (0.15) |
0.66 |
||||
|
Institutional |
0.20** (0.06) |
1.22 |
−0.29** (0.10) |
0.75 |
||||
|
Park |
−0.03 (0.10) |
0.97 |
−0.09 (0.15) |
0.92 |
||||
|
Second-level variance |
.60*** |
.52*** |
||||||
|
Third-level variance |
.07*** |
.21*** |
||||||
|
N (roads / tracts) |
152,556 (13,163 / 178) |
265,243 (12,124 / 178) |
||||||
|
a. Dichotomous variable with 1 equal to the variable’s name. |
||||||||
|
b. 100s of meters. |
||||||||
|
c. 1,000s / mile2 . |
||||||||
|
** p < .01, *** p < .001. |
||||||||