#Green space statistically significant portion of the variation#push out model summary as csv file for plottingtbl.basub <-tidy(modelgst_basub)tbl.basub <- tbl.basub %>%mutate(p.value =2* (1-pt(abs(statistic), df =df.residual(modelgst_basub))))tbl.basub
# A tibble: 8 × 7
effect group term estimate std.error statistic p.value
<chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl>
1 fixed <NA> (Intercept) -0.0891 0.231 -0.385 7.01e- 1
2 fixed <NA> GreenSpaceInstitutio… 1.15 0.296 3.88 1.96e- 4
3 fixed <NA> GreenSpacePark 2.16 0.317 6.80 9.89e-10
4 fixed <NA> GreenSpacePublic Rig… 1.81 0.284 6.36 7.40e- 9
5 fixed <NA> GreenSpaceResidential 1.17 0.284 4.12 8.18e- 5
6 fixed <NA> GreenSpaceVacant Lot 1.19 0.347 3.43 9.00e- 4
7 ran_pars Plot sd__(Intercept) 0.328 NA NA NA
8 ran_pars Residual sd__Observation 0.840 NA NA NA
#tukey test to check pairwise comparisons of species richness of each green space typePairwise.basub <-emmeans(modelgst_basub, pairwise ~ GreenSpace, type ="response")# Summary of pairwise comparisonssummary(Pairwise.basub$contrasts)
contrast ratio SE df null t.ratio p.value
Commercial / Institutional 0.317 0.0942 77.8 1 -3.869 0.0030
Commercial / Park 0.116 0.0369 81.4 1 -6.768 <.0001
Commercial / (Public Right-Of-Way) 0.164 0.0468 78.8 1 -6.343 <.0001
Commercial / Residential 0.310 0.0884 78.8 1 -4.108 0.0013
Commercial / Vacant Lot 0.304 0.1060 79.8 1 -3.416 0.0124
Institutional / Park 0.365 0.1120 82.8 1 -3.288 0.0180
Institutional / (Public Right-Of-Way) 0.517 0.1390 76.8 1 -2.452 0.1516
Institutional / Residential 0.977 0.2630 76.8 1 -0.085 1.0000
Institutional / Vacant Lot 0.957 0.3240 81.3 1 -0.131 1.0000
Park / (Public Right-Of-Way) 1.418 0.4130 79.1 1 1.199 0.8360
Park / Residential 2.679 0.7800 79.1 1 3.384 0.0137
Park / Vacant Lot 2.623 0.9410 84.4 1 2.687 0.0886
(Public Right-Of-Way) / Residential 1.890 0.4790 74.5 1 2.512 0.1338
(Public Right-Of-Way) / Vacant Lot 1.849 0.6060 81.9 1 1.875 0.4247
Residential / Vacant Lot 0.979 0.3210 81.9 1 -0.065 1.0000
Degrees-of-freedom method: kenward-roger
P value adjustment: tukey method for comparing a family of 6 estimates
Tests are performed on the log scale
#summarizing pairwise differences across green space typessummary(Pairwise.basub)
$emmeans
GreenSpace response SE df lower.CL upper.CL
Commercial 0.915 0.212 93.5 0.577 1.45
Institutional 2.882 0.612 91.9 1.890 4.39
Park 7.899 1.900 94.1 4.904 12.72
Public Right-Of-Way 5.571 1.070 89.1 3.801 8.16
Residential 2.948 0.567 89.1 2.012 4.32
Vacant Lot 3.012 0.854 95.0 1.716 5.29
Degrees-of-freedom method: kenward-roger
Confidence level used: 0.95
Intervals are back-transformed from the log scale
$contrasts
contrast ratio SE df null t.ratio p.value
Commercial / Institutional 0.317 0.0942 77.8 1 -3.869 0.0030
Commercial / Park 0.116 0.0369 81.4 1 -6.768 <.0001
Commercial / (Public Right-Of-Way) 0.164 0.0468 78.8 1 -6.343 <.0001
Commercial / Residential 0.310 0.0884 78.8 1 -4.108 0.0013
Commercial / Vacant Lot 0.304 0.1060 79.8 1 -3.416 0.0124
Institutional / Park 0.365 0.1120 82.8 1 -3.288 0.0180
Institutional / (Public Right-Of-Way) 0.517 0.1390 76.8 1 -2.452 0.1516
Institutional / Residential 0.977 0.2630 76.8 1 -0.085 1.0000
Institutional / Vacant Lot 0.957 0.3240 81.3 1 -0.131 1.0000
Park / (Public Right-Of-Way) 1.418 0.4130 79.1 1 1.199 0.8360
Park / Residential 2.679 0.7800 79.1 1 3.384 0.0137
Park / Vacant Lot 2.623 0.9410 84.4 1 2.687 0.0886
(Public Right-Of-Way) / Residential 1.890 0.4790 74.5 1 2.512 0.1338
(Public Right-Of-Way) / Vacant Lot 1.849 0.6060 81.9 1 1.875 0.4247
Residential / Vacant Lot 0.979 0.3210 81.9 1 -0.065 1.0000
Degrees-of-freedom method: kenward-roger
P value adjustment: tukey method for comparing a family of 6 estimates
Tests are performed on the log scale
##OPTION 2: Gamma with log link, but I believe since we have so many small values, this is not the best option? I also get a warning when running the model (warning means the Gamma GLMM didn’t converge cleanly)
#Gamma with log linkmodelgst_basub1 <-glmer(basal_area_ha ~ GreenSpace + (1| Plot), data = subsite_table, family =Gamma(link ="log"))
Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
Model failed to converge with max|grad| = 0.0147709 (tol = 0.002, component 1)
summary(modelgst_basub1)
Generalized linear mixed model fit by maximum likelihood (Laplace
Approximation) [glmerMod]
Family: Gamma ( log )
Formula: basal_area_ha ~ GreenSpace + (1 | Plot)
Data: subsite_table
AIC BIC logLik -2*log(L) df.resid
500.3 521.2 -242.1 484.3 93
Scaled residuals:
Min 1Q Median 3Q Max
-1.2967 -0.6666 -0.1832 0.5650 4.3880
Random effects:
Groups Name Variance Std.Dev.
Plot (Intercept) 0.1234 0.3513
Residual 0.4981 0.7058
Number of obs: 101, groups: Plot, 22
Fixed effects:
Estimate Std. Error t value Pr(>|z|)
(Intercept) 0.240812 0.004912 49.03 <2e-16 ***
GreenSpaceInstitutional 1.226580 0.004909 249.88 <2e-16 ***
GreenSpacePark 2.083658 0.005081 410.13 <2e-16 ***
GreenSpacePublic Right-Of-Way 1.573543 0.004916 320.11 <2e-16 ***
GreenSpaceResidential 0.933804 0.005075 183.99 <2e-16 ***
GreenSpaceVacant Lot 1.394668 0.004920 283.47 <2e-16 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) GrnSpI GrnSpP GSPR-O GrnSpR
GrnSpcInstt -0.001
GreenSpcPrk 0.000 0.000
GrnSpPR-O-W 0.000 0.001 0.000
GrnSpcRsdnt 0.000 0.001 -0.250 0.000
GrnSpcVcntL 0.000 0.000 0.000 0.000 0.000
optimizer (Nelder_Mead) convergence code: 0 (OK)
Model failed to converge with max|grad| = 0.0147709 (tol = 0.002, component 1)
print(modelgst_basub1)
Generalized linear mixed model fit by maximum likelihood (Laplace
Approximation) [glmerMod]
Family: Gamma ( log )
Formula: basal_area_ha ~ GreenSpace + (1 | Plot)
Data: subsite_table
AIC BIC logLik -2*log(L) df.resid
500.2804 521.2014 -242.1402 484.2804 93
Random effects:
Groups Name Std.Dev.
Plot (Intercept) 0.3513
Residual 0.7058
Number of obs: 101, groups: Plot, 22
Fixed Effects:
(Intercept) GreenSpaceInstitutional
0.2408 1.2266
GreenSpacePark GreenSpacePublic Right-Of-Way
2.0837 1.5735
GreenSpaceResidential GreenSpaceVacant Lot
0.9338 1.3947
optimizer (Nelder_Mead) convergence code: 0 (OK) ; 0 optimizer warnings; 1 lme4 warnings
#Residualsresid_gamma1 <-residuals(modelgst_basub1, type ="pearson")qqnorm(resid_gamma1); qqline(resid_gamma1)
hist(resid_gamma1)
plot(resid_gamma1)
#check effect of green space type on species richness using model outputsAnova(modelgst_basub1)
#Green space statistically significant portion of the variation#push out model summary as csv file for plottingtbl.basub1 <-tidy(modelgst_basub1)tbl.basub1 <- tbl.basub1 %>%mutate(p.value =ifelse(p.value <0.0001, "<0.0001", round(p.value, 4)))tbl.basub1
#tukey test to check pairwise comparisons of species richness of each green space typePairwise.basub1 <-emmeans(modelgst_basub1, pairwise ~ GreenSpace, type ="response")# Summary of pairwise comparisonssummary(Pairwise.basub1$contrasts)
contrast ratio SE df null z.ratio p.value
Commercial / Institutional 0.293 0.001440 Inf 1 -249.875 <.0001
Commercial / Park 0.124 0.000632 Inf 1 -410.127 <.0001
Commercial / (Public Right-Of-Way) 0.207 0.001020 Inf 1 -320.114 <.0001
Commercial / Residential 0.393 0.001990 Inf 1 -183.989 <.0001
Commercial / Vacant Lot 0.248 0.001220 Inf 1 -283.473 <.0001
Institutional / Park 0.424 0.003000 Inf 1 -121.306 <.0001
Institutional / (Public Right-Of-Way) 0.707 0.004910 Inf 1 -49.961 <.0001
Institutional / Residential 1.340 0.009460 Inf 1 41.484 <.0001
Institutional / Vacant Lot 0.845 0.005870 Inf 1 -24.185 <.0001
Park / (Public Right-Of-Way) 1.665 0.011800 Inf 1 72.146 <.0001
Park / Residential 3.158 0.025400 Inf 1 143.221 <.0001
Park / Vacant Lot 1.992 0.014100 Inf 1 97.427 <.0001
(Public Right-Of-Way) / Residential 1.896 0.013400 Inf 1 90.533 <.0001
(Public Right-Of-Way) / Vacant Lot 1.196 0.008320 Inf 1 25.722 <.0001
Residential / Vacant Lot 0.631 0.004460 Inf 1 -65.203 <.0001
P value adjustment: tukey method for comparing a family of 6 estimates
Tests are performed on the log scale
#summarizing pairwise differences across green space typessummary(Pairwise.basub1)
$emmeans
GreenSpace response SE df asymp.LCL asymp.UCL
Commercial 1.27 0.00625 Inf 1.26 1.28
Institutional 4.34 0.03010 Inf 4.28 4.40
Park 10.22 0.07220 Inf 10.08 10.36
Public Right-Of-Way 6.14 0.04260 Inf 6.05 6.22
Residential 3.24 0.02290 Inf 3.19 3.28
Vacant Lot 5.13 0.03570 Inf 5.06 5.20
Confidence level used: 0.95
Intervals are back-transformed from the log scale
$contrasts
contrast ratio SE df null z.ratio p.value
Commercial / Institutional 0.293 0.001440 Inf 1 -249.875 <.0001
Commercial / Park 0.124 0.000632 Inf 1 -410.127 <.0001
Commercial / (Public Right-Of-Way) 0.207 0.001020 Inf 1 -320.114 <.0001
Commercial / Residential 0.393 0.001990 Inf 1 -183.989 <.0001
Commercial / Vacant Lot 0.248 0.001220 Inf 1 -283.473 <.0001
Institutional / Park 0.424 0.003000 Inf 1 -121.306 <.0001
Institutional / (Public Right-Of-Way) 0.707 0.004910 Inf 1 -49.961 <.0001
Institutional / Residential 1.340 0.009460 Inf 1 41.484 <.0001
Institutional / Vacant Lot 0.845 0.005870 Inf 1 -24.185 <.0001
Park / (Public Right-Of-Way) 1.665 0.011800 Inf 1 72.146 <.0001
Park / Residential 3.158 0.025400 Inf 1 143.221 <.0001
Park / Vacant Lot 1.992 0.014100 Inf 1 97.427 <.0001
(Public Right-Of-Way) / Residential 1.896 0.013400 Inf 1 90.533 <.0001
(Public Right-Of-Way) / Vacant Lot 1.196 0.008320 Inf 1 25.722 <.0001
Residential / Vacant Lot 0.631 0.004460 Inf 1 -65.203 <.0001
P value adjustment: tukey method for comparing a family of 6 estimates
Tests are performed on the log scale
#BASAL AREA CALCULATED BY SUBSITE AREA
#MODEL OPTIONS ##OPTION 1: Lmer, with log-transformed basal area, since we are lots of small values, reduces skew. Same as basal area by subsite
#Linear model + log transformmodelgst_bapa <-lmer(log(basal_area_plant_ha) ~ GreenSpace + (1| Plot),data = subsite_table)#Residualsresid_gamma <-residuals(modelgst_bapa, type ="pearson")qqnorm(resid_gamma); qqline(resid_gamma)
hist(resid_gamma)
plot(resid_gamma)
#check effect of green space type on species richness using model outputsAnova(modelgst_bapa)
#Green space statistically significant portion of the variation#push out model summary as csv file for plottingtbl.bapa <-tidy(modelgst_bapa)tbl.bapa <- tbl.basub %>%mutate(p.value =2* (1-pt(abs(statistic), df =df.residual(modelgst_bapa))))tbl.bapa
# A tibble: 8 × 7
effect group term estimate std.error statistic p.value
<chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl>
1 fixed <NA> (Intercept) -0.0891 0.231 -0.385 7.01e- 1
2 fixed <NA> GreenSpaceInstitutio… 1.15 0.296 3.88 1.96e- 4
3 fixed <NA> GreenSpacePark 2.16 0.317 6.80 9.89e-10
4 fixed <NA> GreenSpacePublic Rig… 1.81 0.284 6.36 7.40e- 9
5 fixed <NA> GreenSpaceResidential 1.17 0.284 4.12 8.18e- 5
6 fixed <NA> GreenSpaceVacant Lot 1.19 0.347 3.43 9.00e- 4
7 ran_pars Plot sd__(Intercept) 0.328 NA NA NA
8 ran_pars Residual sd__Observation 0.840 NA NA NA
#tukey test to check pairwise comparisons of species richness of each green space typePairwise.bapa <-emmeans(modelgst_bapa, pairwise ~ GreenSpace, type ="response")# Summary of pairwise comparisonssummary(Pairwise.bapa$contrasts)
contrast ratio SE df null t.ratio p.value
Commercial / Institutional 0.612 0.1810 78.3 1 -1.663 0.5598
Commercial / Park 0.330 0.1040 82.3 1 -3.508 0.0093
Commercial / (Public Right-Of-Way) 0.189 0.0536 79.4 1 -5.875 <.0001
Commercial / Residential 0.712 0.2020 79.4 1 -1.198 0.8363
Commercial / Vacant Lot 0.793 0.2750 80.8 1 -0.667 0.9850
Institutional / Park 0.539 0.1640 83.8 1 -2.034 0.3325
Institutional / (Public Right-Of-Way) 0.309 0.0828 77.2 1 -4.384 0.0005
Institutional / Residential 1.163 0.3110 77.2 1 0.566 0.9929
Institutional / Vacant Lot 1.296 0.4360 82.5 1 0.771 0.9716
Park / (Public Right-Of-Way) 0.575 0.1660 79.9 1 -1.913 0.4018
Park / Residential 2.160 0.6260 79.9 1 2.660 0.0951
Park / Vacant Lot 2.407 0.8560 85.7 1 2.469 0.1452
(Public Right-Of-Way) / Residential 3.760 0.9490 74.6 1 5.246 <.0001
(Public Right-Of-Way) / Vacant Lot 4.189 1.3600 83.0 1 4.399 0.0004
Residential / Vacant Lot 1.114 0.3630 83.0 1 0.332 0.9994
Degrees-of-freedom method: kenward-roger
P value adjustment: tukey method for comparing a family of 6 estimates
Tests are performed on the log scale
#summarizing pairwise differences across green space typessummary(Pairwise.bapa)
$emmeans
GreenSpace response SE df lower.CL upper.CL
Commercial 3.93 0.891 94.4 2.50 6.16
Institutional 6.42 1.330 93.7 4.26 9.68
Park 11.92 2.800 94.7 7.48 19.00
Public Right-Of-Way 20.74 3.880 92.3 14.31 30.07
Residential 5.52 1.030 92.3 3.81 8.00
Vacant Lot 4.95 1.380 95.0 2.85 8.60
Degrees-of-freedom method: kenward-roger
Confidence level used: 0.95
Intervals are back-transformed from the log scale
$contrasts
contrast ratio SE df null t.ratio p.value
Commercial / Institutional 0.612 0.1810 78.3 1 -1.663 0.5598
Commercial / Park 0.330 0.1040 82.3 1 -3.508 0.0093
Commercial / (Public Right-Of-Way) 0.189 0.0536 79.4 1 -5.875 <.0001
Commercial / Residential 0.712 0.2020 79.4 1 -1.198 0.8363
Commercial / Vacant Lot 0.793 0.2750 80.8 1 -0.667 0.9850
Institutional / Park 0.539 0.1640 83.8 1 -2.034 0.3325
Institutional / (Public Right-Of-Way) 0.309 0.0828 77.2 1 -4.384 0.0005
Institutional / Residential 1.163 0.3110 77.2 1 0.566 0.9929
Institutional / Vacant Lot 1.296 0.4360 82.5 1 0.771 0.9716
Park / (Public Right-Of-Way) 0.575 0.1660 79.9 1 -1.913 0.4018
Park / Residential 2.160 0.6260 79.9 1 2.660 0.0951
Park / Vacant Lot 2.407 0.8560 85.7 1 2.469 0.1452
(Public Right-Of-Way) / Residential 3.760 0.9490 74.6 1 5.246 <.0001
(Public Right-Of-Way) / Vacant Lot 4.189 1.3600 83.0 1 4.399 0.0004
Residential / Vacant Lot 1.114 0.3630 83.0 1 0.332 0.9994
Degrees-of-freedom method: kenward-roger
P value adjustment: tukey method for comparing a family of 6 estimates
Tests are performed on the log scale
##OPTION 2: Gamma with log link, but I believe since we have so many small values, this is not the best option? I also get a warning when running the model (warning means the Gamma GLMM didn’t converge cleanly)
#Gamma with log linkmodelgst_bapa1 <-glmer(basal_area_plant_ha ~ GreenSpace + (1| Plot), data = subsite_table, family =Gamma(link ="log"))summary(modelgst_bapa1)
Generalized linear mixed model fit by maximum likelihood (Laplace
Approximation) [glmerMod]
Family: Gamma ( log )
Formula: basal_area_plant_ha ~ GreenSpace + (1 | Plot)
Data: subsite_table
AIC BIC logLik -2*log(L) df.resid
679.4 700.3 -331.7 663.4 93
Scaled residuals:
Min 1Q Median 3Q Max
-1.2565 -0.7048 -0.2022 0.5101 3.2797
Random effects:
Groups Name Variance Std.Dev.
Plot (Intercept) 0.1175 0.3428
Residual 0.5064 0.7116
Number of obs: 101, groups: Plot, 22
Fixed effects:
Estimate Std. Error t value Pr(>|z|)
(Intercept) 1.6253 0.2093 7.767 8.04e-15 ***
GreenSpaceInstitutional 0.5265 0.2589 2.033 0.0420 *
GreenSpacePark 1.0989 0.2677 4.106 4.03e-05 ***
GreenSpacePublic Right-Of-Way 1.6894 0.2544 6.640 3.14e-11 ***
GreenSpaceResidential 0.1757 0.2380 0.738 0.4605
GreenSpaceVacant Lot 0.5077 0.2929 1.733 0.0831 .
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Correlation of Fixed Effects:
(Intr) GrnSpI GrnSpP GSPR-O GrnSpR
GrnSpcInstt -0.673
GreenSpcPrk -0.622 0.502
GrnSpPR-O-W -0.709 0.615 0.516
GrnSpcRsdnt -0.707 0.581 0.549 0.608
GrnSpcVcntL -0.543 0.436 0.431 0.443 0.479
print(modelgst_bapa1)
Generalized linear mixed model fit by maximum likelihood (Laplace
Approximation) [glmerMod]
Family: Gamma ( log )
Formula: basal_area_plant_ha ~ GreenSpace + (1 | Plot)
Data: subsite_table
AIC BIC logLik -2*log(L) df.resid
679.3522 700.2732 -331.6761 663.3522 93
Random effects:
Groups Name Std.Dev.
Plot (Intercept) 0.3428
Residual 0.7116
Number of obs: 101, groups: Plot, 22
Fixed Effects:
(Intercept) GreenSpaceInstitutional
1.6253 0.5265
GreenSpacePark GreenSpacePublic Right-Of-Way
1.0989 1.6894
GreenSpaceResidential GreenSpaceVacant Lot
0.1757 0.5077
#Residualsresid_gamma1 <-residuals(modelgst_bapa1, type ="pearson")qqnorm(resid_gamma1); qqline(resid_gamma1)
hist(resid_gamma1)
plot(resid_gamma1)
#check effect of green space type on species richness using model outputsAnova(modelgst_bapa1)
#Green space statistically significant portion of the variation#push out model summary as csv file for plottingtbl.bapa1 <-tidy(modelgst_bapa1)tbl.bapa1 <- tbl.bapa1 %>%mutate(p.value =ifelse(p.value <0.0001, "<0.0001", round(p.value, 4)))tbl.bapa1
#tukey test to check pairwise comparisons of species richness of each green space typePairwise.bapa1 <-emmeans(modelgst_bapa1, pairwise ~ GreenSpace, type ="response")# Summary of pairwise comparisonssummary(Pairwise.bapa1$contrasts)
contrast ratio SE df null z.ratio p.value
Commercial / Institutional 0.591 0.1530 Inf 1 -2.033 0.3232
Commercial / Park 0.333 0.0892 Inf 1 -4.106 0.0006
Commercial / (Public Right-Of-Way) 0.185 0.0470 Inf 1 -6.640 <.0001
Commercial / Residential 0.839 0.2000 Inf 1 -0.738 0.9772
Commercial / Vacant Lot 0.602 0.1760 Inf 1 -1.733 0.5098
Institutional / Park 0.564 0.1480 Inf 1 -2.178 0.2482
Institutional / (Public Right-Of-Way) 0.313 0.0704 Inf 1 -5.161 <.0001
Institutional / Residential 1.420 0.3240 Inf 1 1.538 0.6396
Institutional / Vacant Lot 1.019 0.3000 Inf 1 0.064 1.0000
Park / (Public Right-Of-Way) 0.554 0.1420 Inf 1 -2.297 0.1951
Park / Residential 2.517 0.6080 Inf 1 3.821 0.0018
Park / Vacant Lot 1.806 0.5420 Inf 1 1.972 0.3585
(Public Right-Of-Way) / Residential 4.544 0.9930 Inf 1 6.925 <.0001
(Public Right-Of-Way) / Vacant Lot 3.260 0.9480 Inf 1 4.066 0.0007
Residential / Vacant Lot 0.718 0.1970 Inf 1 -1.207 0.8337
P value adjustment: tukey method for comparing a family of 6 estimates
Tests are performed on the log scale
#summarizing pairwise differences across green space typessummary(Pairwise.bapa1)
$emmeans
GreenSpace response SE df asymp.LCL asymp.UCL
Commercial 5.08 1.06 Inf 3.37 7.66
Institutional 8.60 1.67 Inf 5.87 12.59
Park 15.24 3.26 Inf 10.02 23.19
Public Right-Of-Way 27.51 5.00 Inf 19.27 39.29
Residential 6.06 1.05 Inf 4.31 8.50
Vacant Lot 8.44 2.12 Inf 5.16 13.80
Confidence level used: 0.95
Intervals are back-transformed from the log scale
$contrasts
contrast ratio SE df null z.ratio p.value
Commercial / Institutional 0.591 0.1530 Inf 1 -2.033 0.3232
Commercial / Park 0.333 0.0892 Inf 1 -4.106 0.0006
Commercial / (Public Right-Of-Way) 0.185 0.0470 Inf 1 -6.640 <.0001
Commercial / Residential 0.839 0.2000 Inf 1 -0.738 0.9772
Commercial / Vacant Lot 0.602 0.1760 Inf 1 -1.733 0.5098
Institutional / Park 0.564 0.1480 Inf 1 -2.178 0.2482
Institutional / (Public Right-Of-Way) 0.313 0.0704 Inf 1 -5.161 <.0001
Institutional / Residential 1.420 0.3240 Inf 1 1.538 0.6396
Institutional / Vacant Lot 1.019 0.3000 Inf 1 0.064 1.0000
Park / (Public Right-Of-Way) 0.554 0.1420 Inf 1 -2.297 0.1951
Park / Residential 2.517 0.6080 Inf 1 3.821 0.0018
Park / Vacant Lot 1.806 0.5420 Inf 1 1.972 0.3585
(Public Right-Of-Way) / Residential 4.544 0.9930 Inf 1 6.925 <.0001
(Public Right-Of-Way) / Vacant Lot 3.260 0.9480 Inf 1 4.066 0.0007
Residential / Vacant Lot 0.718 0.1970 Inf 1 -1.207 0.8337
P value adjustment: tukey method for comparing a family of 6 estimates
Tests are performed on the log scale
#In conclusions, I am leaning towards lmer (although I did not use lmer for any other models). I think the residuals look best, but let me know your thoughts/suggestions?
