Chptr1_sxth_update

• I tried to keep this update fairly focused for this update. things of note:
  • I added three new contextual variables (would love to hear your thoughts on the two variables with only two time points and the appropriateness of including them in the analysis):
    • dev01, dev11, and dev19 that is the light wavelength analysis for development of land in the appropriate year
    • la10 and la19 are the low food access variables
    • gq10 and gq19 are the percentage of the population in the tract that live in group quarters.
  • after runniong the PCA, there were seven dimensions. I used that for all the LPA analysis
  • for the LPA i looked at 5, 6, and 7 classes with the five dimensions. I went with five classes as I there wasn’t much separating the three and I thought it would be less confusing to write about five classes than six or seven.
    
  • Did a huge table showing the breakdown by class/variable. I’m a bit confused on how to properly describe the classes.
    
  • Also made maps of the three cities with 7 dimensions and 5 classes.

PCA

LPA

Number of recommended dimensions and classes

## Compare tidyLPA solutions:
## 
##  Model Classes AIC      BIC      Warnings
##  1     1       30498.94 30569.07         
##  1     2       29364.03 29474.23 Warning 
##  1     3       27869.20 28019.48 Warning 
##  1     4       27376.16 27566.52 Warning 
##  1     5       27525.76 27756.19 Warning 
##  1     6       27740.53 28011.04 Warning 
##  1     7       27608.85 27919.43 Warning 
##  6     1       30540.94 30716.27         
##  6     2       24950.01 25305.68         
##  6     3       23293.70 23829.71         
##  6     4       22600.31 23316.65         
##  6     5       21698.96 22595.64         
##  6     6       21573.02 22650.04         
##  6     7       21059.33 22316.69         
## 
## Best model according to AIC is Model 6 with 7 classes.
## Best model according to BIC is Model 6 with 7 classes.
## 
## An analytic hierarchy process, based on the fit indices AIC, AWE, BIC, CLC, and KIC (Akogul & Erisoglu, 2017), suggests the best solution is Model 6 with 7 classes.

y1 <- dat_tx_dim%>% 
  dplyr::select(dim1,dim2,dim3,dim4,dim5,dim6,dim7) %>% 
  single_imputation() %>% 
  poms() %>% 
  estimate_profiles(5,
                    variances = "varying",
                    covariances = "varying") %>% 
  plot_profiles()

y2 <- dat_tx_dim%>% 
  dplyr::select(dim1,dim2,dim3,dim4,dim5,dim6,dim7) %>% 
  single_imputation() %>% 
  poms() %>% 
  estimate_profiles(6,
                    variances = "varying",
                    covariances = "varying") %>% 
  plot_profiles()

y3 <- dat_tx_dim%>% 
  dplyr::select(dim1,dim2,dim3,dim4,dim5,dim6,dim7) %>% 
  single_imputation() %>% 
  poms() %>% 
  estimate_profiles(7,
                    variances = "varying",
                    covariances = "varying") %>% 
  plot_profiles()

the number of tracts in each class:

table(dat_tx_dim$class)

analyzing classes

race/ethnicity

	V1	V2	V3	V4	V5
class	1.000	2.000	3.000	4.000	5.000
y00_04	215.914	144.269	649.192	767.340	272.751
y05_09	236.402	104.686	804.090	1003.074	324.857
y10_14	227.910	84.622	800.436	1089.680	397.443
y15_19	256.365	94.167	1189.885	1206.537	477.021
inc	1.560	1.077	1.782	1.862	1.941
pnhw00	0.160	0.111	0.198	0.327	0.229
pnhw12	0.253	0.187	0.389	0.680	0.374
pnhw19	0.214	0.170	0.361	0.637	0.354
pnhb00	0.100	0.018	0.061	0.012	0.043
pnhb12	0.281	0.043	0.183	0.032	0.098
pnhb19	0.272	0.045	0.184	0.037	0.104
phisp00	0.001	0.002	0.001	0.001	0.001
phisp12	0.408	0.743	0.361	0.213	0.461
phisp19	0.446	0.752	0.377	0.227	0.465
mhhinc00	43717.719	33056.941	35458.037	74895.061	43888.396
mhhinc12	49050.052	37983.088	45134.067	94622.835	50322.588
mhhinc19	57641.305	46096.594	60454.882	113841.726	63079.668
pedu00	0.208	0.139	0.264	0.533	0.282
pedu12	0.204	0.137	0.331	0.587	0.298
pedu19	0.228	0.162	0.378	0.629	0.339
pop00	8.225	8.376	7.252	7.886	8.272
pop12	8.466	8.391	6.565	8.300	8.265
pop19	8.586	8.459	6.679	8.457	8.342
dev01	0.621	0.955	0.721	0.652	0.987
dev11	0.621	0.955	0.721	0.652	0.987
dev19	0.621	0.955	0.721	0.652	0.987
gq10	0.002	0.014	0.162	0.004	0.001
gq19	0.002	0.014	0.162	0.004	0.001
la10	0.814	0.714	0.760	0.803	0.554
la19	0.764	0.671	0.736	0.752	0.489

maps