Recode
##INCOME
#average the income data
income_data = d[var_table$var[which(str_detect(var_table$label, "^TTL"))]]
#add years
income_years = str_match(var_table$label, "PAST YR (\\d{4})") %>% .[, 2] %>% na.omit()
colnames(income_data) = "income_" + income_years
#z score
income_data_z = income_data %>% df_standardize()
#log
income_log_data = df_colFunc(income_data, function(x) log(x + 1))
#cors
wtd.cors(income_data) %>%
MAT_half() %>%
describe() %>% unclass()
## $vars
## [1] 1
##
## $n
## [1] 120
##
## $mean
## [1] 0.37
##
## $sd
## [1] 0.21
##
## $median
## [1] 0.33
##
## $trimmed
## [1] 0.36
##
## $mad
## [1] 0.25
##
## $min
## [1] 0.064
##
## $max
## [1] 0.84
##
## $range
## [1] 0.78
##
## $skew
## [1] 0.46
##
## $kurtosis
## [1] -1
##
## $se
## [1] 0.019
##
## attr(,"row.names")
## [1] "X1"
wtd.cors(income_log_data) %>%
MAT_half() %>%
describe() %>% unclass()
## $vars
## [1] 1
##
## $n
## [1] 120
##
## $mean
## [1] 0.27
##
## $sd
## [1] 0.13
##
## $median
## [1] 0.24
##
## $trimmed
## [1] 0.26
##
## $mad
## [1] 0.15
##
## $min
## [1] 0.077
##
## $max
## [1] 0.55
##
## $range
## [1] 0.48
##
## $skew
## [1] 0.45
##
## $kurtosis
## [1] -1
##
## $se
## [1] 0.012
##
## attr(,"row.names")
## [1] "X1"
#missing data
miss_plot(income_data)
#impute
income_data_imputed = miss_impute(income_data, max_na = 10)
#average of imputed
d$income_imputed = rowMeans(income_data_imputed)
#average all data without imputing
#results in higher N, but some bias due to selective missing data
d$income_simple = rowMeans(income_data, na.rm = T)
#log version
d$income_log = rowMeans(income_log_data, na.rm = T)
#post-education income only; age >= 25
post_edu_income = income_data["income_" + c(2007:2011, 2013)]
d$income_post_edu = post_edu_income %>% df_standardize() %>% rowMeans(na.rm=T)
d$income_post_edu_log = post_edu_income %>% df_colFunc(function(x) log(x + 1) %>% standardize()) %>% rowMeans(na.rm=T)
##AGE
#born
d$born = d$R0536402
GG_denhist(d, "born")
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
#calculate age from month and year
#subtract from 2013, last year of data
d$age = (parse_date("06 2013", "%m %Y") - sprintf("%d %d", d$R0536401, d$R0536402) %>% parse_date(format = "%m %Y")) %>% as.numeric() %>% divide_by(365.24)
GG_denhist(d, "age")
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
##SEX
d$sex = if_else(d$R0536300 == 1, true = "Male", false = "Female") %>% factor()
##IQ
#AFQT
d$AFQT = d$R9829600 %>%
#recode 0 and 100 centiles
winsorise(99999, 1) %>%
#convert to fraction
divide_by(100000) %>%
#convert to Z score
qnorm()
#plot
GG_denhist(d, "AFQT")
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## Warning: Removed 1891 rows containing non-finite values (stat_bin).
## Warning: Removed 1891 rows containing non-finite values (stat_density).
#PPAT
#multiple scores
PIAT_data = d[var_table$var[which(str_detect(var_table$label, "^CV_PIAT"))]]
#z score
PIAT_data %<>% df_standardize()
#cors
wtd.cors(PIAT_data) %T>%
print() %>%
MAT_half() %>%
describe() %>% unclass()
## R5473700 R7237400 S1552700
## R5473700 1.00 0.76 0.78
## R7237400 0.76 1.00 0.77
## S1552700 0.78 0.77 1.00
## $vars
## [1] 1
##
## $n
## [1] 3
##
## $mean
## [1] 0.77
##
## $sd
## [1] 0.0092
##
## $median
## [1] 0.77
##
## $trimmed
## [1] 0.77
##
## $mad
## [1] 0.0036
##
## $min
## [1] 0.76
##
## $max
## [1] 0.78
##
## $range
## [1] 0.017
##
## $skew
## [1] 0.36
##
## $kurtosis
## [1] -2.3
##
## $se
## [1] 0.0053
##
## attr(,"row.names")
## [1] "X1"
#simple mean will do
d$PIAT = rowMeans(PIAT_data, na.rm=T)
#plot
GG_denhist(d, "PIAT")
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## Warning: Removed 7335 rows containing non-finite values (stat_bin).
## Warning: Removed 7335 rows containing non-finite values (stat_density).
#SAT and ACT for comparison
d$SAT_math = d$Z9033700
d$SAT_verbal = d$Z9033900
d$SAT = rowMeans(d[c("SAT_verbal", "SAT_math")], na.rm=T)
d$ACT = d$Z9034100
#g factor
d$GCA = rowMeans(d[c("AFQT", "PIAT")], na.rm = T)
##RACE
#self-perceived dummy version, 2002
SPRE_data = d[c("S122490" + 0:5, "R0538600")] %>%
set_colnames(c("White", "Black", "Asian", "Pacific Islander", "American Indian", "Other", "Hispanic")) %>%
df_colFunc(as.logical)
#standard coding
d$SPRE = plyr::aaply(SPRE_data, .margins = 1, .expand = F, function(x) {
#all NA? code missing
if (all(is.na(x))) return(NA)
#otherwise, impute false
x[is.na(x)] = F
#is hispanic?
if (x$Hispanic) return("Hispanic")
#is only 1 non-hipanic?
if (sum(x %>% unlist()) == 1) {
return(names(which(x %>% unlist())))
}
#multiracial
"Multi-racial"
}) %>% factor() %>% fct_relevel("White")
#table
table2(d$SPRE)
#other-perceived
d$OPRE = d$R2395200 %>% plyr::mapvalues(1:3, c("White", "Black", "Other")) %>% factor() %>% fct_relevel("White")
table2(d$OPRE)
#skin tone
skin_data = d[c("T3173000", "T4584700", "T6217800")]
#regress out interviewer effect on ratings
#code interviewer data
interviewer_race_data = d[c("T2023300", "T3614100", "T5214000")] %>% df_colFunc(function(x) plyr::mapvalues(x, 1:6, c("White", "Black", "American Indian", "Asian", "Other", "Multiracial")))
## The following `from` values were not present in `x`: 3
## The following `from` values were not present in `x`: 3, 4, 5
## The following `from` values were not present in `x`: 3
#regression
skin_data_fixed = map2_df(skin_data, interviewer_race_data, function(x, y) {
#make temp df
tmp = data.frame(x, y) %>% set_colnames(c("skin_tone", "interviewer"))
#model, resids, standardize
lm(skin_tone ~ interviewer, data = tmp, na.action = na.exclude) %>% resid() %>% standardize()
})
#use mean; no repeated measures
d$skin_tone = rowMeans(skin_data, na.rm=T)
d$skin_tone_fixed = rowMeans(skin_data_fixed, na.rm=T)
#short versions
d$income = d$income_post_edu
#standardized versions
d2 = d[c("GCA", "income", "age", "sex", "SPRE", "OPRE", "skin_tone", "skin_tone_fixed")] %>% df_standardize()
## Skipped sex because it is a factor.
## Skipped SPRE because it is a factor.
## Skipped OPRE because it is a factor.
Results
Correlations
#cors
hetcor(d[c("GCA", "AFQT", "PIAT", "SAT", "ACT", "income_imputed", "income_simple", "income_log", "income_post_edu", "income_post_edu_log", "skin_tone", "skin_tone_fixed", "age", "sex")], use = "pairwise")
## Warning in hetcor.data.frame(d[c("GCA", "AFQT", "PIAT", "SAT", "ACT",
## "income_imputed", : the correlation matrix has been adjusted to make it
## positive-definite
##
## Two-Step Estimates
##
## Correlations/Type of Correlation:
## GCA AFQT PIAT SAT ACT
## GCA 1 Pearson Pearson Pearson Pearson
## AFQT 0.958 1 Pearson Pearson Pearson
## PIAT 0.921 0.778 1 Pearson Pearson
## SAT 0.608 0.601 0.609 1 Pearson
## ACT 0.66 0.669 0.576 0.552 1
## income_imputed 0.196 0.199 0.185 0.137 0.0754
## income_simple 0.228 0.224 0.211 0.134 0.0894
## income_log 0.151 0.147 0.133 0.0161 -0.00199
## income_post_edu 0.318 0.312 0.298 0.197 0.163
## income_post_edu_log 0.285 0.283 0.242 0.131 0.123
## skin_tone -0.367 -0.364 -0.35 -0.263 -0.326
## skin_tone_fixed -0.357 -0.357 -0.341 -0.279 -0.313
## age -0.0193 -0.00663 0.0173 -0.0191 0.00159
## sex -0.0314 -0.0463 0.0762 0.0925 0.0526
## income_imputed income_simple income_log
## GCA Pearson Pearson Pearson
## AFQT Pearson Pearson Pearson
## PIAT Pearson Pearson Pearson
## SAT Pearson Pearson Pearson
## ACT Pearson Pearson Pearson
## income_imputed 1 Pearson Pearson
## income_simple 0.914 1 Pearson
## income_log 0.641 0.72 1
## income_post_edu 0.942 0.894 0.547
## income_post_edu_log 0.716 0.653 0.749
## skin_tone -0.142 -0.158 -0.112
## skin_tone_fixed -0.138 -0.153 -0.102
## age 0.29 0.149 0.176
## sex 0.277 0.238 0.159
## income_post_edu income_post_edu_log skin_tone
## GCA Pearson Pearson Pearson
## AFQT Pearson Pearson Pearson
## PIAT Pearson Pearson Pearson
## SAT Pearson Pearson Pearson
## ACT Pearson Pearson Pearson
## income_imputed Pearson Pearson Pearson
## income_simple Pearson Pearson Pearson
## income_log Pearson Pearson Pearson
## income_post_edu 1 Pearson Pearson
## income_post_edu_log 0.693 1 Pearson
## skin_tone -0.187 -0.169 1
## skin_tone_fixed -0.18 -0.161 0.993
## age 0.164 0.0943 0.018
## sex 0.224 0.143 0.0262
## skin_tone_fixed age sex
## GCA Pearson Pearson Polyserial
## AFQT Pearson Pearson Polyserial
## PIAT Pearson Pearson Polyserial
## SAT Pearson Pearson Polyserial
## ACT Pearson Pearson Polyserial
## income_imputed Pearson Pearson Polyserial
## income_simple Pearson Pearson Polyserial
## income_log Pearson Pearson Polyserial
## income_post_edu Pearson Pearson Polyserial
## income_post_edu_log Pearson Pearson Polyserial
## skin_tone Pearson Pearson Polyserial
## skin_tone_fixed 1 Pearson Polyserial
## age 0.00604 1 Polyserial
## sex 0.0197 -0.00893 1
##
## Standard Errors/Numbers of Observations:
## GCA AFQT PIAT SAT ACT income_imputed
## GCA 7396 7093 1649 2132 1771 5420
## AFQT 0.000256 7093 1346 2057 1722 5262
## PIAT 0.00263 0.0114 1649 487 409 1138
## SAT 0.0137 0.0141 0.0285 2493 575 2053
## ACT 0.0134 0.0133 0.0331 0.0291 2005 1677
## income_imputed 0.013 0.0132 0.0287 0.0217 0.0243 6340
## income_simple 0.0112 0.0114 0.024 0.0198 0.0223 0.00207
## income_log 0.0115 0.0118 0.0246 0.0201 0.0224 0.0074
## income_post_edu 0.0112 0.0115 0.024 0.0204 0.0229 0.00142
## income_post_edu_log 0.0115 0.0117 0.0248 0.0208 0.0232 0.00619
## skin_tone 0.0115 0.0117 0.0241 0.0215 0.0225 0.0134
## skin_tone_fixed 0.0127 0.0129 0.0265 0.0232 0.025 0.0147
## age 0.0116 0.0119 0.0246 0.02 0.0223 0.0115
## sex 0.0146 0.0149 0.0307 0.0249 0.028 0.0154
## income_simple income_log income_post_edu
## GCA 7209 7209 6413
## AFQT 6925 6925 6167
## PIAT 1593 1593 1441
## SAT 2464 2464 2232
## ACT 1988 1988 1803
## income_imputed 6340 6340 6212
## income_simple 8685 8685 7634
## income_log 0.00517 8685 7634
## income_post_edu 0.00231 0.00802 7634
## income_post_edu_log 0.00657 0.00503 0.00596
## skin_tone 0.012 0.0121 0.0122
## skin_tone_fixed 0.0131 0.0133 0.0133
## age 0.0105 0.0104 0.0111
## sex 0.0132 0.0132 0.0142
## income_post_edu_log skin_tone skin_tone_fixed age
## GCA 6413 5678 4745 7396
## AFQT 6167 5447 4549 7093
## PIAT 1441 1325 1115 1649
## SAT 2232 1872 1580 2493
## ACT 1803 1576 1307 2005
## income_imputed 6212 5309 4474 6340
## income_simple 7634 6636 5561 8685
## income_log 7634 6636 5561 8685
## income_post_edu 7634 6308 5284 7634
## income_post_edu_log 7634 6308 5284 7634
## skin_tone 0.0122 6723 5631 6723
## skin_tone_fixed 0.0134 0.000177 5631 5631
## age 0.0113 0.0122 0.0133 8984
## sex 0.0143 0.0153 0.0167 0.0132
## sex
## GCA 7396
## AFQT 7093
## PIAT 1649
## SAT 2493
## ACT 2005
## income_imputed 6340
## income_simple 8685
## income_log 8685
## income_post_edu 7634
## income_post_edu_log 7634
## skin_tone 6723
## skin_tone_fixed 5631
## age 8984
## sex 8984
##
## P-values for Tests of Bivariate Normality:
## GCA AFQT PIAT SAT
## GCA
## AFQT 1.44e-94
## PIAT 2.52e-22 1.16e-06
## SAT 4.94065645841247e-324 4.2e-298 1.1e-61
## ACT 0 0 4.91e-88 2.16e-102
## income_imputed 3.93e-81 2.45e-71 1.36e-88 1.22e-68
## income_simple 2.13e-37 1.16e-32 6.86e-36 1.09e-83
## income_log 6.4e-116 3.26e-118 9.91e-20 1.41e-131
## income_post_edu 3.24e-62 7.87e-56 5.22e-53 1.61e-102
## income_post_edu_log 6.14e-287 2.22e-285 2.26e-78 9.1e-243
## skin_tone 0 0 2.4e-85 1.12e-194
## skin_tone_fixed 2.69e-201 3.95e-195 5.83e-48 4.82e-142
## age 9.58e-83 1.14e-52 0 6.66e-97
## sex 0.00206 0.00625 1.34e-07 1.48e-71
## ACT income_imputed income_simple income_log
## GCA
## AFQT
## PIAT
## SAT
## ACT
## income_imputed 6.07e-249
## income_simple 2.33e-286 0
## income_log 0 0 0
## income_post_edu 2.7e-271 7.39e-167 1.54e-246 0
## income_post_edu_log 0 0 0 0
## skin_tone 0 0 0 0
## skin_tone_fixed 7.79e-257 1.44e-275 9.53e-250 0
## age 4.28e-298 8.04e-112 6.89e-128 1.51e-206
## sex 1.01e-284 3.56e-47 7.23e-48 2.69e-118
## income_post_edu income_post_edu_log skin_tone
## GCA
## AFQT
## PIAT
## SAT
## ACT
## income_imputed
## income_simple
## income_log
## income_post_edu
## income_post_edu_log 0
## skin_tone 0 0
## skin_tone_fixed 1.23e-283 0 0
## age 7.11e-141 0 0
## sex 1.13e-69 0 0
## skin_tone_fixed age
## GCA
## AFQT
## PIAT
## SAT
## ACT
## income_imputed
## income_simple
## income_log
## income_post_edu
## income_post_edu_log
## skin_tone
## skin_tone_fixed
## age 1.93e-268
## sex 6.41e-223 3.09e-84
#for blacks and hispanics
hetcor(d %>% filter(SPRE == "Black") %>% .[c("GCA", "AFQT", "PIAT", "SAT", "ACT", "income_imputed", "income_simple", "income_log", "income_post_edu", "income_post_edu_log", "skin_tone", "skin_tone_fixed", "age", "sex")], use = "pairwise")
## Warning in hetcor.data.frame(d %>% filter(SPRE == "Black") %>% .[c("GCA", :
## the correlation matrix has been adjusted to make it positive-definite
##
## Two-Step Estimates
##
## Correlations/Type of Correlation:
## GCA AFQT PIAT SAT ACT income_imputed
## GCA 1 Pearson Pearson Pearson Pearson Pearson
## AFQT 0.936 1 Pearson Pearson Pearson Pearson
## PIAT 0.879 0.682 1 Pearson Pearson Pearson
## SAT 0.426 0.394 0.54 1 Pearson Pearson
## ACT 0.404 0.415 0.352 0.291 1 Pearson
## income_imputed 0.327 0.343 0.295 0.114 0.069 1
## income_simple 0.375 0.376 0.354 0.17 0.0986 0.913
## income_log 0.316 0.312 0.333 0.131 0.0608 0.662
## income_post_edu 0.406 0.408 0.362 0.166 0.168 0.944
## income_post_edu_log 0.354 0.359 0.306 0.132 0.142 0.711
## skin_tone -0.114 -0.1 -0.168 -0.0792 -0.0411 -0.013
## skin_tone_fixed -0.0921 -0.0794 -0.15 -0.075 -0.0281 0.00318
## age -0.0682 -0.0325 0.0354 -0.0389 0.0655 0.265
## sex -0.0933 -0.106 0.0278 -0.0123 -0.117 0.16
## income_simple income_log income_post_edu
## GCA Pearson Pearson Pearson
## AFQT Pearson Pearson Pearson
## PIAT Pearson Pearson Pearson
## SAT Pearson Pearson Pearson
## ACT Pearson Pearson Pearson
## income_imputed Pearson Pearson Pearson
## income_simple 1 Pearson Pearson
## income_log 0.737 1 Pearson
## income_post_edu 0.915 0.602 1
## income_post_edu_log 0.668 0.796 0.695
## skin_tone -0.0364 -0.0265 -0.042
## skin_tone_fixed -0.0215 -0.011 -0.019
## age 0.121 0.15 0.147
## sex 0.102 0.0038 0.103
## income_post_edu_log skin_tone skin_tone_fixed age
## GCA Pearson Pearson Pearson Pearson
## AFQT Pearson Pearson Pearson Pearson
## PIAT Pearson Pearson Pearson Pearson
## SAT Pearson Pearson Pearson Pearson
## ACT Pearson Pearson Pearson Pearson
## income_imputed Pearson Pearson Pearson Pearson
## income_simple Pearson Pearson Pearson Pearson
## income_log Pearson Pearson Pearson Pearson
## income_post_edu Pearson Pearson Pearson Pearson
## income_post_edu_log 1 Pearson Pearson Pearson
## skin_tone -0.0378 1 Pearson Pearson
## skin_tone_fixed -0.0141 0.987 1 Pearson
## age 0.0786 -0.0221 -0.0263 1
## sex -0.00901 0.126 0.134 -0.0312
## sex
## GCA Polyserial
## AFQT Polyserial
## PIAT Polyserial
## SAT Polyserial
## ACT Polyserial
## income_imputed Polyserial
## income_simple Polyserial
## income_log Polyserial
## income_post_edu Polyserial
## income_post_edu_log Polyserial
## skin_tone Polyserial
## skin_tone_fixed Polyserial
## age Polyserial
## sex 1
##
## Standard Errors/Numbers of Observations:
## GCA AFQT PIAT SAT ACT income_imputed
## GCA 1697 1621 384 412 373 1124
## AFQT 0.000707 1621 308 399 359 1093
## PIAT 0.00759 0.0326 384 98 88 211
## SAT 0.0407 0.0422 0.0716 495 110 403
## ACT 0.0435 0.0437 0.0938 0.0876 435 352
## income_imputed 0.0266 0.0268 0.0631 0.0492 0.0531 1326
## income_simple 0.0212 0.0216 0.0457 0.044 0.0476 0.00456
## income_log 0.0223 0.0227 0.0464 0.0445 0.0479 0.0154
## income_post_edu 0.0218 0.0222 0.0477 0.0456 0.0485 0.00303
## income_post_edu_log 0.0227 0.0232 0.0498 0.0461 0.0489 0.0137
## skin_tone 0.0259 0.0266 0.0531 0.0487 0.0513 0.0294
## skin_tone_fixed 0.0281 0.0288 0.057 0.0522 0.0549 0.0315
## age 0.0241 0.0248 0.051 0.0449 0.0478 0.0255
## sex 0.0303 0.0309 0.064 0.0566 0.0609 0.0338
## income_simple income_log income_post_edu
## GCA 1646 1646 1475
## AFQT 1577 1577 1415
## PIAT 367 367 332
## SAT 488 488 456
## ACT 433 433 402
## income_imputed 1326 1326 1308
## income_simple 1979 1979 1769
## income_log 0.0103 1979 1769
## income_post_edu 0.00387 0.0152 1769
## income_post_edu_log 0.0132 0.0087 0.0123
## skin_tone 0.0244 0.0244 0.0254
## skin_tone_fixed 0.0263 0.0263 0.0273
## age 0.0222 0.022 0.0233
## sex 0.0282 0.0282 0.0297
## income_post_edu_log skin_tone skin_tone_fixed age
## GCA 1475 1451 1244 1697
## AFQT 1415 1386 1188 1621
## PIAT 332 335 295 384
## SAT 456 417 364 495
## ACT 402 379 332 435
## income_imputed 1308 1160 1009 1326
## income_simple 1769 1672 1443 1979
## income_log 1769 1672 1443 1979
## income_post_edu 1769 1547 1337 1769
## income_post_edu_log 1769 1547 1337 1769
## skin_tone 0.0254 1718 1481 1718
## skin_tone_fixed 0.0274 0.000666 1481 1481
## age 0.0236 0.0241 0.026 2044
## sex 0.0298 0.0298 0.032 0.0277
## sex
## GCA 1697
## AFQT 1621
## PIAT 384
## SAT 495
## ACT 435
## income_imputed 1326
## income_simple 1979
## income_log 1979
## income_post_edu 1769
## income_post_edu_log 1769
## skin_tone 1718
## skin_tone_fixed 1481
## age 2044
## sex 2044
##
## P-values for Tests of Bivariate Normality:
## GCA AFQT PIAT SAT ACT
## GCA
## AFQT 1.05e-22
## PIAT 4.5e-18 6.09e-08
## SAT 1.66e-38 1.97e-36 0.000274
## ACT 1.36e-113 4.35e-110 8.89e-26 2.06e-19
## income_imputed 7.56e-23 6.62e-22 4.97e-16 2.97e-07 1.94e-62
## income_simple 7.74e-11 1.26e-08 1.13e-11 1.69e-14 6.27e-85
## income_log 4.03e-20 1.83e-19 5.46e-07 6.47e-32 1.39e-105
## income_post_edu 1.63e-09 6.81e-08 2.73e-11 8.09e-17 9.86e-81
## income_post_edu_log 2.96e-53 2.2e-53 3.88e-14 6.96e-44 2.36e-110
## skin_tone 6.09e-10 1.3e-09 3.8e-06 1.48e-05 8.2e-70
## skin_tone_fixed 0.496 0.453 9.04e-05 0.0579 6.67e-65
## age 1.3e-19 3.15e-10 2.27e-224 5.42e-14 1.2e-86
## sex 0.144 0.36 5.39e-06 2.47e-08 7.77e-86
## income_imputed income_simple income_log
## GCA
## AFQT
## PIAT
## SAT
## ACT
## income_imputed
## income_simple 3.74e-113
## income_log 3.86e-158 4.07e-126
## income_post_edu 6.71e-79 2.19e-29 9.87e-106
## income_post_edu_log 3.49e-244 6.9e-188 4.24e-87
## skin_tone 5.35e-11 1.67e-12 4.83e-29
## skin_tone_fixed 0.00412 0.0103 3.29e-15
## age 5.95e-19 4.43e-24 9.65e-38
## sex 3.29e-05 1.61e-08 2.36e-20
## income_post_edu income_post_edu_log skin_tone
## GCA
## AFQT
## PIAT
## SAT
## ACT
## income_imputed
## income_simple
## income_log
## income_post_edu
## income_post_edu_log 0
## skin_tone 5.32e-15 2.03e-62
## skin_tone_fixed 0.000341 2.17e-44 4.65e-67
## age 1.07e-22 1.86e-78 5.32e-25
## sex 8.39e-08 7.69e-64 1.39e-12
## skin_tone_fixed age
## GCA
## AFQT
## PIAT
## SAT
## ACT
## income_imputed
## income_simple
## income_log
## income_post_edu
## income_post_edu_log
## skin_tone
## skin_tone_fixed
## age 5.35e-12
## sex 0.236 7.61e-23
#hispanic
hetcor(d %>% filter(SPRE == "Hispanic") %>% .[c("GCA", "AFQT", "PIAT", "SAT", "ACT", "income_imputed", "income_simple", "income_log", "income_post_edu", "income_post_edu_log", "skin_tone", "skin_tone_fixed", "age", "sex")], use = "pairwise")
## Warning in hetcor.data.frame(d %>% filter(SPRE == "Hispanic") %>% .
## [c("GCA", : the correlation matrix has been adjusted to make it positive-
## definite
##
## Two-Step Estimates
##
## Correlations/Type of Correlation:
## GCA AFQT PIAT SAT ACT income_imputed
## GCA 1 Pearson Pearson Pearson Pearson Pearson
## AFQT 0.946 1 Pearson Pearson Pearson Pearson
## PIAT 0.913 0.739 1 Pearson Pearson Pearson
## SAT 0.448 0.447 0.459 1 Pearson Pearson
## ACT 0.472 0.459 0.454 0.587 1 Pearson
## income_imputed 0.161 0.161 0.196 0.176 0.0751 1
## income_simple 0.181 0.177 0.204 0.158 0.0626 0.926
## income_log 0.131 0.126 0.133 0.0603 -0.108 0.669
## income_post_edu 0.243 0.241 0.259 0.2 0.126 0.937
## income_post_edu_log 0.222 0.22 0.207 0.103 0.0823 0.723
## skin_tone -0.138 -0.124 -0.124 -0.0523 -0.0882 0.00948
## skin_tone_fixed -0.132 -0.128 -0.104 -0.0485 -0.0538 -0.00768
## age -0.0112 0.00874 0.0175 0.028 0.0332 0.282
## sex -0.02 -0.0462 0.101 0.0918 -0.0872 0.353
## income_simple income_log income_post_edu
## GCA Pearson Pearson Pearson
## AFQT Pearson Pearson Pearson
## PIAT Pearson Pearson Pearson
## SAT Pearson Pearson Pearson
## ACT Pearson Pearson Pearson
## income_imputed Pearson Pearson Pearson
## income_simple 1 Pearson Pearson
## income_log 0.72 1 Pearson
## income_post_edu 0.901 0.567 1
## income_post_edu_log 0.642 0.768 0.693
## skin_tone -0.0116 -0.0144 -0.0153
## skin_tone_fixed -0.0251 -0.0113 -0.031
## age 0.155 0.164 0.161
## sex 0.303 0.236 0.292
## income_post_edu_log skin_tone skin_tone_fixed age
## GCA Pearson Pearson Pearson Pearson
## AFQT Pearson Pearson Pearson Pearson
## PIAT Pearson Pearson Pearson Pearson
## SAT Pearson Pearson Pearson Pearson
## ACT Pearson Pearson Pearson Pearson
## income_imputed Pearson Pearson Pearson Pearson
## income_simple Pearson Pearson Pearson Pearson
## income_log Pearson Pearson Pearson Pearson
## income_post_edu Pearson Pearson Pearson Pearson
## income_post_edu_log 1 Pearson Pearson Pearson
## skin_tone -0.0333 1 Pearson Pearson
## skin_tone_fixed -0.0156 0.982 1 Pearson
## age 0.0972 0.0473 0.046 1
## sex 0.181 0.106 0.0976 0.0247
## sex
## GCA Polyserial
## AFQT Polyserial
## PIAT Polyserial
## SAT Polyserial
## ACT Polyserial
## income_imputed Polyserial
## income_simple Polyserial
## income_log Polyserial
## income_post_edu Polyserial
## income_post_edu_log Polyserial
## skin_tone Polyserial
## skin_tone_fixed Polyserial
## age Polyserial
## sex 1
##
## Standard Errors/Numbers of Observations:
## GCA AFQT PIAT SAT ACT income_imputed
## GCA 1457 1358 358 333 164 1007
## AFQT 0.00076 1358 259 310 154 955
## PIAT 0.00547 0.0302 358 91 45 222
## SAT 0.044 0.0455 0.0832 411 76 317
## ACT 0.061 0.0639 0.119 0.0758 208 165
## income_imputed 0.0307 0.0315 0.0647 0.0545 0.0776 1274
## income_simple 0.0257 0.0267 0.052 0.0487 0.07 0.00398
## income_log 0.0261 0.0271 0.0533 0.0498 0.0695 0.0155
## income_post_edu 0.0263 0.0271 0.0539 0.0505 0.0717 0.00346
## income_post_edu_log 0.0266 0.0274 0.0553 0.052 0.0724 0.0134
## skin_tone 0.0292 0.0302 0.0578 0.0555 0.0774 0.0303
## skin_tone_fixed 0.0317 0.033 0.0623 0.0601 0.0848 0.0329
## age 0.0262 0.0271 0.0529 0.0493 0.0694 0.0258
## sex 0.0329 0.034 0.0656 0.0614 0.0874 0.0336
## income_simple income_log income_post_edu
## GCA 1414 1414 1281
## AFQT 1320 1320 1205
## PIAT 341 341 301
## SAT 401 401 363
## ACT 203 203 189
## income_imputed 1274 1274 1262
## income_simple 1828 1828 1642
## income_log 0.0113 1828 1642
## income_post_edu 0.00465 0.0168 1642
## income_post_edu_log 0.0145 0.0101 0.0128
## skin_tone 0.0265 0.0265 0.0271
## skin_tone_fixed 0.0288 0.0289 0.0296
## age 0.0228 0.0228 0.024
## sex 0.0283 0.0295 0.0306
## income_post_edu_log skin_tone skin_tone_fixed age
## GCA 1281 1135 959 1457
## AFQT 1205 1060 892 1358
## PIAT 301 291 253 358
## SAT 363 324 276 411
## ACT 189 165 139 208
## income_imputed 1262 1089 923 1274
## income_simple 1642 1429 1201 1828
## income_log 1642 1429 1201 1828
## income_post_edu 1642 1357 1137 1642
## income_post_edu_log 1642 1357 1137 1642
## skin_tone 0.0271 1445 1213 1445
## skin_tone_fixed 0.0297 0.00103 1213 1213
## age 0.0245 0.0263 0.0287 1899
## sex 0.0314 0.0326 0.0357 0.0288
## sex
## GCA 1457
## AFQT 1358
## PIAT 358
## SAT 411
## ACT 208
## income_imputed 1274
## income_simple 1828
## income_log 1828
## income_post_edu 1642
## income_post_edu_log 1642
## skin_tone 1445
## skin_tone_fixed 1213
## age 1899
## sex 1899
##
## P-values for Tests of Bivariate Normality:
## GCA AFQT PIAT SAT ACT
## GCA
## AFQT 1.4e-19
## PIAT 2.93e-09 0.000193
## SAT 1.47e-31 1.78e-27 1.7e-06
## ACT 5.5e-34 3e-32 3.57e-07 2.48e-09
## income_imputed 3.51e-09 8.79e-06 1.03e-14 4.96e-07 1.97e-23
## income_simple 2.13e-06 0.00106 9.97e-08 1.05e-05 3.31e-25
## income_log 3.24e-23 1.17e-22 5.92e-07 1.57e-16 1.43e-29
## income_post_edu 8.79e-09 6.15e-05 2.12e-12 1.9e-09 1.19e-24
## income_post_edu_log 2.04e-68 6.93e-64 6.85e-25 9.76e-36 7.83e-40
## skin_tone 1.38e-23 3.89e-21 1.84e-10 1.09e-06 5.04e-22
## skin_tone_fixed 6.43e-10 1.79e-08 2.7e-07 0.000126 1.16e-17
## age 8.76e-19 1.41e-07 1.1e-209 9.69e-07 4.75e-27
## sex 0.00121 0.0118 4.26e-06 0.00024 1.8e-28
## income_imputed income_simple income_log
## GCA
## AFQT
## PIAT
## SAT
## ACT
## income_imputed
## income_simple 3.82e-54
## income_log 5.22e-115 1.88e-128
## income_post_edu 2.38e-27 5.48e-26 1.38e-96
## income_post_edu_log 1.88e-215 3.99e-204 5.78e-120
## skin_tone 3.72e-25 4.39e-28 9.9e-58
## skin_tone_fixed 4.05e-12 2.18e-10 3.04e-34
## age 7.18e-13 1.26e-16 1.71e-36
## sex 6.52e-06 9.09e-06 4.5e-25
## income_post_edu income_post_edu_log skin_tone
## GCA
## AFQT
## PIAT
## SAT
## ACT
## income_imputed
## income_simple
## income_log
## income_post_edu
## income_post_edu_log 0
## skin_tone 5.84e-31 7.8e-96
## skin_tone_fixed 6.5e-14 3.23e-67 1.8e-74
## age 3.3e-16 2.3e-92 7.73e-39
## sex 1.22e-08 2.59e-89 4.76e-31
## skin_tone_fixed age
## GCA
## AFQT
## PIAT
## SAT
## ACT
## income_imputed
## income_simple
## income_log
## income_post_edu
## income_post_edu_log
## skin_tone
## skin_tone_fixed
## age 7.59e-19
## sex 5.98e-13 1.45e-15
Modeling
#baseline
lm(income ~ SPRE + age + sex, data = d2) %T>% {summary(.) %>% print()} %>% MOD_summary()
##
## Call:
## lm(formula = income ~ SPRE + age + sex, data = d2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.288 -0.621 -0.135 0.438 5.708
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.00894 0.01975 0.45 0.6506
## SPREAmerican Indian -0.54663 0.17394 -3.14 0.0017 **
## SPREAsian 0.44186 0.09577 4.61 4.0e-06 ***
## SPREBlack -0.49071 0.02776 -17.68 < 2e-16 ***
## SPREHispanic -0.25769 0.02845 -9.06 < 2e-16 ***
## SPREMulti-racial -0.19180 0.04215 -4.55 5.4e-06 ***
## SPREOther -0.13319 0.18677 -0.71 0.4758
## SPREPacific Islander 0.18968 0.19843 0.96 0.3392
## age 0.16968 0.01090 15.57 < 2e-16 ***
## sexMale 0.33732 0.02175 15.51 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.95 on 7622 degrees of freedom
## (1352 observations deleted due to missingness)
## Multiple R-squared: 0.102, Adjusted R-squared: 0.101
## F-statistic: 95.8 on 9 and 7622 DF, p-value: <2e-16
##
## ---- Model summary ----
## Model coefficients
## Beta SE CI_lower CI_upper
## SPRE: White 0.00 NA NA NA
## SPRE: American Indian -0.55 0.174 -0.89 -0.21
## SPRE: Asian 0.44 0.096 0.25 0.63
## SPRE: Black -0.49 0.028 -0.55 -0.44
## SPRE: Hispanic -0.26 0.028 -0.31 -0.20
## SPRE: Multi-racial -0.19 0.042 -0.27 -0.11
## SPRE: Other -0.13 0.187 -0.50 0.23
## SPRE: Pacific Islander 0.19 0.198 -0.20 0.58
## age 0.17 0.011 0.15 0.19
## sex: Female 0.00 NA NA NA
## sex: Male 0.34 0.022 0.29 0.38
##
##
## Model meta-data
## outcome N df R2 R2-adj. R2-cv
## 1 income 7632 7622 0.1 0.1 NA
##
##
## Etas from analysis of variance
## Eta Eta_partial
## SPRE 0.21 0.22
## age 0.17 0.18
## sex 0.17 0.17
lm(income ~ OPRE + age + sex, data = d2) %T>% {summary(.) %>% print()} %>% MOD_summary()
##
## Call:
## lm(formula = income ~ OPRE + age + sex, data = d2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.945 -0.626 -0.141 0.423 5.692
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.0617 0.0182 -3.39 0.0007 ***
## OPREBlack -0.4028 0.0260 -15.49 <2e-16 ***
## OPREOther -0.0342 0.0387 -0.88 0.3774
## age 0.1691 0.0112 15.04 <2e-16 ***
## sexMale 0.3409 0.0224 15.20 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.95 on 7240 degrees of freedom
## (1739 observations deleted due to missingness)
## Multiple R-squared: 0.0876, Adjusted R-squared: 0.0871
## F-statistic: 174 on 4 and 7240 DF, p-value: <2e-16
##
## ---- Model summary ----
## Model coefficients
## Beta SE CI_lower CI_upper
## OPRE: White 0.000 NA NA NA
## OPRE: Black -0.404 0.026 -0.45 -0.353
## OPRE: Other -0.034 0.039 -0.11 0.042
## age 0.169 0.011 0.15 0.191
## sex: Female 0.000 NA NA NA
## sex: Male 0.342 0.022 0.30 0.386
##
##
## Model meta-data
## outcome N df R2 R2-adj. R2-cv
## 1 income 7245 7240 0.088 0.087 NA
##
##
## Etas from analysis of variance
## Eta Eta_partial
## OPRE 0.18 0.18
## age 0.17 0.17
## sex 0.17 0.18
lm(income ~ skin_tone + age + sex, data = d2) %T>% {summary(.) %>% print()} %>% MOD_summary()
##
## Call:
## lm(formula = income ~ skin_tone + age + sex, data = d2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.968 -0.600 -0.133 0.421 5.189
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.2160 0.0166 -13.0 <2e-16 ***
## skin_tone -0.1894 0.0118 -16.1 <2e-16 ***
## age 0.1604 0.0116 13.8 <2e-16 ***
## sexMale 0.3712 0.0233 16.0 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.92 on 6304 degrees of freedom
## (2676 observations deleted due to missingness)
## Multiple R-squared: 0.0981, Adjusted R-squared: 0.0977
## F-statistic: 229 on 3 and 6304 DF, p-value: <2e-16
##
## ---- Model summary ----
## Model coefficients
## Beta SE CI_lower CI_upper
## skin_tone -0.19 0.012 -0.22 -0.17
## age 0.17 0.012 0.14 0.19
## sex: Female 0.00 NA NA NA
## sex: Male 0.38 0.024 0.34 0.43
##
##
## Model meta-data
## outcome N df R2 R2-adj. R2-cv
## 1 income 6308 6304 0.098 0.098 NA
##
##
## Etas from analysis of variance
## Eta Eta_partial
## skin_tone 0.19 0.20
## age 0.17 0.17
## sex 0.19 0.20
#full model
lm(income ~ GCA + OPRE + SPRE + skin_tone + age + sex, data = d2) %T>% {summary(.) %>% print()} %>% MOD_summary()
##
## Call:
## lm(formula = income ~ GCA + OPRE + SPRE + skin_tone + age + sex,
## data = d2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.608 -0.557 -0.111 0.409 4.854
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.2018 0.0268 -7.54 5.6e-14 ***
## GCA 0.2975 0.0136 21.89 < 2e-16 ***
## OPREBlack 0.0445 0.0826 0.54 0.5896
## OPREOther 0.1034 0.0517 2.00 0.0455 *
## SPREAmerican Indian -0.4073 0.1905 -2.14 0.0326 *
## SPREAsian 0.4482 0.1096 4.09 4.4e-05 ***
## SPREBlack -0.0842 0.0841 -1.00 0.3169
## SPREHispanic -0.0197 0.0380 -0.52 0.6039
## SPREMulti-racial -0.0791 0.0623 -1.27 0.2043
## SPREOther 0.2466 0.2240 1.10 0.2709
## SPREPacific Islander 0.1472 0.2240 0.66 0.5110
## skin_tone -0.0570 0.0214 -2.67 0.0076 **
## age 0.1699 0.0121 14.08 < 2e-16 ***
## sexMale 0.3745 0.0246 15.22 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.88 on 5186 degrees of freedom
## (3784 observations deleted due to missingness)
## Multiple R-squared: 0.185, Adjusted R-squared: 0.183
## F-statistic: 90.6 on 13 and 5186 DF, p-value: <2e-16
##
## ---- Model summary ----
## Model coefficients
## Beta SE CI_lower CI_upper
## GCA 0.305 0.014 0.2781 0.333
## OPRE: White 0.000 NA NA NA
## OPRE: Black 0.046 0.085 -0.1203 0.212
## OPRE: Other 0.106 0.053 0.0021 0.210
## SPRE: White 0.000 NA NA NA
## SPRE: American Indian -0.418 0.195 -0.8007 -0.035
## SPRE: Asian 0.460 0.112 0.2393 0.680
## SPRE: Black -0.086 0.086 -0.2554 0.083
## SPRE: Hispanic -0.020 0.039 -0.0965 0.056
## SPRE: Multi-racial -0.081 0.064 -0.2063 0.044
## SPRE: Other 0.253 0.230 -0.1974 0.703
## SPRE: Pacific Islander 0.151 0.230 -0.2993 0.601
## skin_tone -0.057 0.022 -0.0996 -0.015
## age 0.177 0.013 0.1522 0.201
## sex: Female 0.000 NA NA NA
## sex: Male 0.384 0.025 0.3346 0.434
##
##
## Model meta-data
## outcome N df R2 R2-adj. R2-cv
## 1 income 5200 5186 0.19 0.18 NA
##
##
## Etas from analysis of variance
## Eta Eta_partial
## GCA 0.274 0.291
## OPRE 0.025 0.028
## SPRE 0.066 0.073
## skin_tone 0.033 0.037
## age 0.177 0.192
## sex 0.191 0.207
##robustness
#fixed skin tone?
lm(income ~ GCA + OPRE + SPRE + skin_tone_fixed + age + sex, data = d2) %T>% {summary(.) %>% print()} %>% MOD_summary()
##
## Call:
## lm(formula = income ~ GCA + OPRE + SPRE + skin_tone_fixed + age +
## sex, data = d2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.651 -0.555 -0.106 0.418 4.841
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.1901 0.0291 -6.53 7.4e-11 ***
## GCA 0.2992 0.0147 20.35 < 2e-16 ***
## OPREBlack 0.0310 0.0856 0.36 0.717
## OPREOther 0.1134 0.0557 2.04 0.042 *
## SPREAmerican Indian -0.4222 0.1928 -2.19 0.029 *
## SPREAsian 0.5768 0.1170 4.93 8.6e-07 ***
## SPREBlack -0.0933 0.0881 -1.06 0.290
## SPREHispanic -0.0322 0.0410 -0.79 0.432
## SPREMulti-racial -0.1300 0.0668 -1.95 0.052 .
## SPREOther 0.2640 0.2286 1.15 0.248
## SPREPacific Islander 0.1760 0.2370 0.74 0.458
## skin_tone_fixed -0.0336 0.0225 -1.49 0.136
## age 0.1807 0.0130 13.86 < 2e-16 ***
## sexMale 0.3884 0.0266 14.60 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.87 on 4332 degrees of freedom
## (4638 observations deleted due to missingness)
## Multiple R-squared: 0.196, Adjusted R-squared: 0.193
## F-statistic: 81 on 13 and 4332 DF, p-value: <2e-16
##
## ---- Model summary ----
## Model coefficients
## Beta SE CI_lower CI_upper
## GCA 0.309 0.015 0.2791 0.339
## OPRE: White 0.000 NA NA NA
## OPRE: Black 0.032 0.088 -0.1411 0.205
## OPRE: Other 0.117 0.057 0.0043 0.230
## SPRE: White 0.000 NA NA NA
## SPRE: American Indian -0.436 0.199 -0.8257 -0.046
## SPRE: Asian 0.595 0.121 0.3584 0.832
## SPRE: Black -0.096 0.091 -0.2745 0.082
## SPRE: Hispanic -0.033 0.042 -0.1161 0.050
## SPRE: Multi-racial -0.134 0.069 -0.2692 0.001
## SPRE: Other 0.272 0.236 -0.1901 0.735
## SPRE: Pacific Islander 0.182 0.245 -0.2978 0.661
## skin_tone_fixed -0.034 0.023 -0.0784 0.011
## age 0.189 0.014 0.1625 0.216
## sex: Female 0.000 NA NA NA
## sex: Male 0.401 0.027 0.3469 0.455
##
##
## Model meta-data
## outcome N df R2 R2-adj. R2-cv
## 1 income 4346 4332 0.2 0.19 NA
##
##
## Etas from analysis of variance
## Eta Eta_partial
## GCA 0.277 0.295
## OPRE 0.028 0.031
## SPRE 0.086 0.096
## skin_tone_fixed 0.020 0.023
## age 0.189 0.206
## sex 0.199 0.217
#no skin tone?
lm(income ~ GCA + OPRE + SPRE + age + sex, data = d2) %T>% {summary(.) %>% print()} %>% MOD_summary()
##
## Call:
## lm(formula = income ~ GCA + OPRE + SPRE + age + sex, data = d2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.653 -0.572 -0.116 0.412 5.381
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.1241 0.0216 -5.75 9.1e-09 ***
## GCA 0.2962 0.0129 22.94 < 2e-16 ***
## OPREBlack -0.0792 0.0708 -1.12 0.26347
## OPREOther 0.0937 0.0488 1.92 0.05488 .
## SPREAmerican Indian -0.4390 0.1716 -2.56 0.01056 *
## SPREAsian 0.3644 0.1064 3.43 0.00062 ***
## SPREBlack -0.1035 0.0751 -1.38 0.16803
## SPREHispanic -0.0746 0.0356 -2.10 0.03620 *
## SPREMulti-racial -0.0650 0.0538 -1.21 0.22747
## SPREOther 0.0576 0.2062 0.28 0.78016
## SPREPacific Islander 0.2801 0.2071 1.35 0.17630
## age 0.1787 0.0115 15.52 < 2e-16 ***
## sexMale 0.3657 0.0233 15.73 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.91 on 6176 degrees of freedom
## (2795 observations deleted due to missingness)
## Multiple R-squared: 0.173, Adjusted R-squared: 0.172
## F-statistic: 108 on 12 and 6176 DF, p-value: <2e-16
##
## ---- Model summary ----
## Model coefficients
## Beta SE CI_lower CI_upper
## GCA 0.294 0.013 0.269 0.3189
## OPRE: White 0.000 NA NA NA
## OPRE: Black -0.079 0.071 -0.218 0.0595
## OPRE: Other 0.094 0.049 -0.002 0.1891
## SPRE: White 0.000 NA NA NA
## SPRE: American Indian -0.438 0.171 -0.774 -0.1023
## SPRE: Asian 0.364 0.106 0.156 0.5718
## SPRE: Black -0.103 0.075 -0.250 0.0436
## SPRE: Hispanic -0.074 0.036 -0.144 -0.0048
## SPRE: Multi-racial -0.065 0.054 -0.170 0.0405
## SPRE: Other 0.057 0.206 -0.346 0.4609
## SPRE: Pacific Islander 0.280 0.207 -0.126 0.6849
## age 0.180 0.012 0.157 0.2029
## sex: Female 0.000 NA NA NA
## sex: Male 0.365 0.023 0.320 0.4105
##
##
## Model meta-data
## outcome N df R2 R2-adj. R2-cv
## 1 income 6189 6176 0.17 0.17 NA
##
##
## Etas from analysis of variance
## Eta Eta_partial
## GCA 0.265 0.280
## OPRE 0.028 0.031
## SPRE 0.064 0.070
## age 0.180 0.194
## sex 0.182 0.196