Summary

This document contains staggered regressions (OLS) on arrest-release. We include the following variables in this regression:

  1. Xg-Boost risk predictor
  2. Arrest and demographic covariates p-hat
  3. Mugshot CNN p-hat

Note that the new Xg-Boost risk predictor is based on the time-varying historical arrest data. I see an increase in AUC from 0.601 to 0.63 by including these.

My main takeaway from these regressions is:

Multihead 0.1 - ResNet18 - Overfit

Multihead(ResNet18)
Dependent variable:
Release Outcome
(1) (2) (3)
risk_pred_prob -1.112*** -0.800*** -0.776***
(-1.216, -1.008) (-0.902, -0.699) (-0.877, -0.675)
p_hat_covariates 0.935*** 0.931***
(0.883, 0.988) (0.879, 0.984)
p_hat_cnn 0.078***
(0.055, 0.100)
Constant 1.105*** 0.296*** 0.226***
(1.072, 1.138) (0.241, 0.351) (0.167, 0.285)
Observations 8,835 8,835 8,835
Adjusted R2 0.034 0.119 0.122
F Statistic 307.362*** (df = 1; 8833) 597.506*** (df = 2; 8832) 410.110*** (df = 3; 8831)
Note: p<0.1; p<0.05; p<0.01

Multihead 0.3 - ResNet50 - (not) Overfit

Multihead(ResNet50)
Dependent variable:
Release Outcome
(1) (2) (3)
risk_pred_prob -1.112*** -0.800*** -0.725***
(-1.216, -1.008) (-0.902, -0.699) (-0.827, -0.624)
p_hat_covariates 0.935*** 0.873***
(0.883, 0.988) (0.819, 0.926)
p_hat_cnn 0.377***
(0.312, 0.441)
Constant 1.105*** 0.296*** 0.039
(1.072, 1.138) (0.241, 0.351) (-0.031, 0.110)
Observations 8,835 8,835 8,835
Adjusted R2 0.034 0.119 0.128
F Statistic 307.362*** (df = 1; 8833) 597.506*** (df = 2; 8832) 432.928*** (df = 3; 8831)
Note: p<0.1; p<0.05; p<0.01

Multihead 0.7 - Inception ResNet - Overfit

Multihead(Inception ResNet)
Dependent variable:
Release Outcome
(1) (2) (3)
risk_pred_prob -1.112*** -0.800*** -0.787***
(-1.216, -1.008) (-0.902, -0.699) (-0.888, -0.686)
p_hat_covariates 0.935*** 0.928***
(0.883, 0.988) (0.875, 0.981)
p_hat_cnn 0.061***
(0.034, 0.088)
Constant 1.105*** 0.296*** 0.246***
(1.072, 1.138) (0.241, 0.351) (0.186, 0.305)
Observations 8,835 8,835 8,835
Adjusted R2 0.034 0.119 0.120
F Statistic 307.362*** (df = 1; 8833) 597.506*** (df = 2; 8832) 403.596*** (df = 3; 8831)
Note: p<0.1; p<0.05; p<0.01

Multihead 0.12 - ResNet50 - Overfit

Multihead(ResNet50)
Dependent variable:
Release Outcome
(1) (2) (3)
risk_pred_prob -1.112*** -0.800*** -0.788***
(-1.216, -1.008) (-0.902, -0.699) (-0.889, -0.687)
p_hat_covariates 0.935*** 0.931***
(0.883, 0.988) (0.878, 0.983)
p_hat_cnn 0.086***
(0.059, 0.112)
Constant 1.105*** 0.296*** 0.220***
(1.072, 1.138) (0.241, 0.351) (0.161, 0.280)
Observations 8,835 8,835 8,835
Adjusted R2 0.034 0.119 0.122
F Statistic 307.362*** (df = 1; 8833) 597.506*** (df = 2; 8832) 409.142*** (df = 3; 8831)
Note: p<0.1; p<0.05; p<0.01

Plotting Model Coeficients:

This gives a good visual for differences among models for judging performance based on regression coefficients. All perform essentially the same, with he exception of model_03 which has a significantly larger p_hat_cnn coefficient (outside the C.I.’s). This may potentially be informative for choosing models going forward.

Decile Plots

Here I provide two types of plots for each of p_hat_cnn , p_hat_covariate, and risk_pred_prob:

  1. Decile Plot A - The max value in a decile vs. the mean arrest outcome in that decile
  2. Decile Plot B - The mean arrest outcome at each decile index

Decile Plot 1 - p_hat_cnn

Decile Plot 2 - p_hat_covariate

Decile Plot 3 - risk_pred_prob

MTurk Features

We now include MTurk results in our covariates. These are collected for some (not all) of the validation set, reaching 7318 arrest_ids. The included features are:

  1. Attractiveness
  2. Competence
  3. Dominance
  4. Trustworthiness
  5. Age
  6. Race (Black, White, Hispanic, Asian, Indian, Unsure/Other)
  7. Skin-color (18 variants)

Multihead 0.3 - ResNet50 - (not) Overfit

Table _01 - Model 03

  • The p_hat_features model includes 18 skin-tone variants (not super-categorized as in regression table No.2)
  • p_hat_cnn is significant throughout !
  • These effects are robust to the inclusion/exclusion of race in the covariate model on top of the skin_color levels
  • Quite confident that in this sense we are picking up signal on-top of the information gained through knowing race/skin_color !
Multihead(ResNet50)
Dependent variable:
Release Outcome
(1) (2) (3) (4)
risk_pred_prob -1.074*** -1.053*** -0.720*** -0.658***
(-1.188, -0.960) (-1.167, -0.938) (-0.831, -0.608) (-0.770, -0.547)
p_hat_features 0.773*** 0.579*** 0.395**
(0.447, 1.099) (0.266, 0.891) (0.083, 0.708)
p_hat_covariates 0.924*** 0.855***
(0.864, 0.983) (0.795, 0.916)
p_hat_cnn 0.384***
(0.313, 0.455)
Constant 1.095*** 0.498*** -0.162 -0.276*
(1.059, 1.131) (0.243, 0.752) (-0.410, 0.085) (-0.523, -0.029)
Observations 7,318 7,318 7,318 7,318
Adjusted R2 0.032 0.033 0.113 0.122
F Statistic 239.132*** (df = 1; 7316) 127.396*** (df = 2; 7315) 311.346*** (df = 3; 7314) 255.754*** (df = 4; 7313)
Note: p<0.1; p<0.05; p<0.01

Table _02 - Model 03

  • covariates_lm excludes race (so as to allow the skin_tone to account for all race signal in this test)
  • The results (p_hat_cnn being significant) are robust to the inclusion of race, though the skin_tone_cat_light become insignificant
  • skin_tone_(category) is a factor variable which encodes the 18 raw hexidecimal color variants (included in Table _01) into three categories comprised of 6 such variants into one of light, medium, and dark skin categories respectively.
Multihead(ResNet50)
Dependent variable:
Release Outcome
(1) (2) (3) (4)
risk_pred_prob -1.074*** -1.068*** -0.736*** -0.676***
(-1.188, -0.960) (-1.183, -0.953) (-0.849, -0.624) (-0.788, -0.564)
skin_tone_cat_light_skin -0.004 -0.015 -0.019*
(-0.022, 0.015) (-0.033, 0.002) (-0.037, -0.002)
skin_tone_cat_medium_skin 0.004 -0.004 -0.007
(-0.018, 0.026) (-0.025, 0.018) (-0.028, 0.014)
age -0.0005 -0.0001 0.001
(-0.002, 0.001) (-0.001, 0.001) (-0.0004, 0.002)
attractiveness -0.003 0.001 0.001
(-0.013, 0.008) (-0.010, 0.011) (-0.009, 0.011)
competence 0.003 -0.002 -0.003
(-0.010, 0.015) (-0.014, 0.010) (-0.015, 0.009)
dominance -0.002 0.003 0.005
(-0.011, 0.007) (-0.005, 0.012) (-0.003, 0.014)
trustworthiness 0.004 0.004 0.002
(-0.007, 0.015) (-0.007, 0.015) (-0.009, 0.012)
p_hat_covariates 0.931*** 0.860***
(0.872, 0.991) (0.799, 0.920)
p_hat_cnn 0.404***
(0.333, 0.476)
Constant 1.095*** 1.100*** 0.261*** -0.026
(1.059, 1.131) (1.030, 1.170) (0.175, 0.347) (-0.126, 0.073)
Observations 7,318 7,318 7,318 7,318
Adjusted R2 0.032 0.031 0.111 0.122
F Statistic 239.132*** (df = 1; 7316) 30.083*** (df = 8; 7309) 102.968*** (df = 9; 7308) 102.390*** (df = 10; 7307)
Note: p<0.1; p<0.05; p<0.01

Table _03 - Model 03

Here we include the 18 raw skin-tone levels.

Multihead(ResNet50)
Dependent variable:
Release Outcome
(1) (2) (3) (4) (5)
risk_pred_prob -1.074*** -1.077*** -1.072*** -0.745*** -0.685***
(-1.188, -0.960) (-1.192, -0.962) (-1.187, -0.956) (-0.858, -0.632) (-0.798, -0.573)
skin_tonenumber_623a17 -0.002 -0.002 -0.005 -0.006
(-0.039, 0.034) (-0.039, 0.035) (-0.040, 0.030) (-0.041, 0.029)
skin_tonenumber_76441f -0.005 -0.005 -0.006 -0.009
(-0.050, 0.040) (-0.051, 0.040) (-0.049, 0.038) (-0.052, 0.034)
skin_tonenumber_80492a 0.005 0.005 -0.001 -0.008
(-0.035, 0.045) (-0.035, 0.045) (-0.040, 0.037) (-0.046, 0.031)
skin_tonenumber_885633 0.050* 0.050* 0.037 0.033
(0.002, 0.099) (0.002, 0.098) (-0.010, 0.083) (-0.013, 0.078)
skin_tonenumber_94623d 0.006 0.006 -0.008 -0.020
(-0.040, 0.053) (-0.041, 0.053) (-0.053, 0.037) (-0.065, 0.025)
skin_tonenumber_ab8b64 0.001 0.0002 -0.013 -0.022
(-0.041, 0.043) (-0.042, 0.043) (-0.053, 0.028) (-0.063, 0.018)
skin_tonenumber_b26949 0.013 0.013 0.012 0.004
(-0.044, 0.070) (-0.044, 0.070) (-0.043, 0.067) (-0.051, 0.058)
skin_tonenumber_cb9662 0.041 0.040 0.035 0.025
(-0.005, 0.088) (-0.006, 0.087) (-0.009, 0.080) (-0.019, 0.070)
skin_tonenumber_d09e7d 0.025 0.024 0.014 0.001
(-0.030, 0.079) (-0.031, 0.078) (-0.038, 0.067) (-0.051, 0.053)
skin_tonenumber_e7bc91 0.002 0.001 -0.031 -0.043
(-0.064, 0.068) (-0.065, 0.067) (-0.094, 0.032) (-0.106, 0.020)
skin_tonenumber_e9cba7 -0.026 -0.027 -0.066* -0.077**
(-0.090, 0.038) (-0.091, 0.037) (-0.127, -0.005) (-0.138, -0.016)
skin_tonenumber_ecc083 -0.022 -0.022 -0.042 -0.044
(-0.071, 0.028) (-0.072, 0.027) (-0.089, 0.006) (-0.092, 0.003)
skin_tonenumber_eed0b8 0.022 0.022 -0.009 -0.024
(-0.022, 0.066) (-0.022, 0.065) (-0.051, 0.033) (-0.065, 0.018)
skin_tonenumber_efc088 -0.084* -0.084* -0.089** -0.097**
(-0.159, -0.008) (-0.160, -0.009) (-0.162, -0.017) (-0.168, -0.025)
skin_tonenumber_efc794 -0.034 -0.034 -0.049 -0.056*
(-0.087, 0.019) (-0.087, 0.019) (-0.099, 0.002) (-0.107, -0.006)
skin_tonenumber_f6e1aa 0.014 0.013 -0.010 -0.017
(-0.029, 0.057) (-0.030, 0.056) (-0.051, 0.031) (-0.058, 0.024)
skin_tonenumber_f7ddc4 0.006 0.006 -0.020 -0.036
(-0.033, 0.045) (-0.034, 0.045) (-0.058, 0.018) (-0.074, 0.001)
age -0.0004 -0.00002 0.001
(-0.002, 0.001) (-0.001, 0.001) (-0.0003, 0.002)
attractiveness -0.002 0.002 0.003
(-0.013, 0.009) (-0.008, 0.012) (-0.008, 0.013)
competence 0.002 -0.002 -0.003
(-0.010, 0.015) (-0.014, 0.010) (-0.015, 0.009)
dominance -0.002 0.003 0.005
(-0.011, 0.007) (-0.006, 0.011) (-0.004, 0.013)
trustworthiness 0.004 0.004 0.001
(-0.007, 0.015) (-0.007, 0.014) (-0.009, 0.012)
p_hat_covariates 0.936*** 0.865***
(0.876, 0.995) (0.804, 0.925)
p_hat_cnn 0.410***
(0.338, 0.482)
Constant 1.095*** 1.091*** 1.095*** 0.258*** -0.029
(1.059, 1.131) (1.043, 1.139) (1.018, 1.171) (0.168, 0.349) (-0.132, 0.074)
Observations 7,318 7,318 7,318 7,318 7,318
Adjusted R2 0.032 0.032 0.031 0.112 0.123
F Statistic 239.132*** (df = 1; 7316) 14.271*** (df = 18; 7299) 11.215*** (df = 23; 7294) 39.518*** (df = 24; 7293) 41.899*** (df = 25; 7292)
Note: p<0.1; p<0.05; p<0.01

Table _04 - Model 03 - Male vs. Female

We now split the regression model into male and female.

  • The p_hat_cnn coefficient is significant and larger than the combined model for both
  • The Female p_hat_cnn is surprisingly large
  • The dominance feature for the female population becomes signficant (which is fascinating !!!)
Table _04 - Male
Multihead(ResNet50)
Dependent variable:
Release Outcome
(1) (2) (3) (4) (5)
risk_pred_prob -0.997*** -1.002*** -1.003*** -0.760*** -0.724***
(-1.125, -0.869) (-1.132, -0.873) (-1.133, -0.874) (-0.885, -0.634) (-0.849, -0.600)
skin_tonenumber_623a17 -0.011 -0.011 -0.011 -0.013
(-0.052, 0.030) (-0.052, 0.029) (-0.050, 0.028) (-0.051, 0.026)
skin_tonenumber_76441f -0.009 -0.009 -0.004 -0.007
(-0.060, 0.041) (-0.060, 0.042) (-0.053, 0.044) (-0.055, 0.041)
skin_tonenumber_80492a -0.017 -0.016 -0.016 -0.022
(-0.062, 0.028) (-0.061, 0.029) (-0.059, 0.027) (-0.064, 0.021)
skin_tonenumber_885633 0.036 0.037 0.029 0.026
(-0.018, 0.091) (-0.018, 0.091) (-0.023, 0.081) (-0.026, 0.078)
skin_tonenumber_94623d -0.015 -0.014 -0.021 -0.030
(-0.069, 0.039) (-0.068, 0.040) (-0.073, 0.030) (-0.082, 0.021)
skin_tonenumber_ab8b64 -0.019 -0.018 -0.025 -0.034
(-0.067, 0.029) (-0.066, 0.030) (-0.071, 0.021) (-0.079, 0.012)
skin_tonenumber_b26949 -0.019 -0.019 -0.010 -0.020
(-0.084, 0.045) (-0.084, 0.046) (-0.072, 0.052) (-0.081, 0.042)
skin_tonenumber_cb9662 0.029 0.030 0.036 0.028
(-0.025, 0.083) (-0.024, 0.084) (-0.016, 0.087) (-0.024, 0.079)
skin_tonenumber_d09e7d 0.032 0.033 0.027 0.009
(-0.031, 0.094) (-0.029, 0.096) (-0.033, 0.087) (-0.050, 0.069)
skin_tonenumber_e7bc91 -0.051 -0.049 -0.066 -0.078*
(-0.130, 0.029) (-0.128, 0.031) (-0.142, 0.010) (-0.154, -0.003)
skin_tonenumber_e9cba7 -0.036 -0.036 -0.057 -0.074*
(-0.112, 0.040) (-0.112, 0.041) (-0.130, 0.016) (-0.147, -0.001)
skin_tonenumber_ecc083 -0.054 -0.053 -0.061* -0.064*
(-0.112, 0.003) (-0.110, 0.005) (-0.117, -0.006) (-0.119, -0.009)
skin_tonenumber_eed0b8 0.019 0.019 -0.003 -0.023
(-0.032, 0.070) (-0.032, 0.070) (-0.052, 0.046) (-0.072, 0.025)
skin_tonenumber_efc088 -0.122** -0.121** -0.099* -0.105*
(-0.214, -0.029) (-0.213, -0.028) (-0.188, -0.011) (-0.193, -0.017)
skin_tonenumber_efc794 -0.039 -0.039 -0.046 -0.055
(-0.101, 0.022) (-0.100, 0.023) (-0.105, 0.013) (-0.113, 0.004)
skin_tonenumber_f6e1aa 0.012 0.013 0.004 -0.007
(-0.039, 0.063) (-0.038, 0.064) (-0.045, 0.053) (-0.056, 0.041)
skin_tonenumber_f7ddc4 -0.004 -0.003 -0.020 -0.045*
(-0.049, 0.042) (-0.049, 0.043) (-0.064, 0.024) (-0.089, -0.001)
age 0.0005 0.0001 0.001
(-0.001, 0.002) (-0.001, 0.001) (-0.0004, 0.002)
attractiveness -0.005 0.002 0.003
(-0.018, 0.008) (-0.011, 0.014) (-0.009, 0.015)
competence 0.004 -0.0003 -0.001
(-0.011, 0.019) (-0.014, 0.014) (-0.014, 0.013)
dominance -0.001 -0.003 -0.004
(-0.012, 0.009) (-0.013, 0.007) (-0.014, 0.006)
trustworthiness 0.001 0.003 0.002
(-0.013, 0.014) (-0.010, 0.016) (-0.011, 0.014)
p_hat_covariates 0.946*** 0.904***
(0.877, 1.015) (0.834, 0.973)
p_hat_cnn 0.477***
(0.389, 0.565)
Constant 1.055*** 1.064*** 1.053*** 0.285*** -0.049
(1.014, 1.096) (1.010, 1.118) (0.964, 1.142) (0.183, 0.387) (-0.168, 0.070)
Observations 5,725 5,725 5,725 5,725 5,725
Adjusted R2 0.028 0.028 0.027 0.106 0.118
F Statistic 164.322*** (df = 1; 5723) 10.197*** (df = 18; 5706) 8.033*** (df = 23; 5701) 29.298*** (df = 24; 5700) 31.704*** (df = 25; 5699)
Note: p<0.1; p<0.05; p<0.01
Table _04 - Female
Multihead(ResNet50)
Dependent variable:
Release Outcome
(1) (2) (3) (4) (5)
risk_pred_prob -0.996*** -0.987*** -0.963*** -0.722*** -0.661***
(-1.280, -0.712) (-1.272, -0.701) (-1.250, -0.677) (-1.002, -0.441) (-0.939, -0.382)
skin_tonenumber_623a17 0.051 0.048 0.041 0.045
(-0.037, 0.139) (-0.040, 0.135) (-0.043, 0.126) (-0.039, 0.129)
skin_tonenumber_76441f 0.020 0.017 -0.0005 -0.004
(-0.084, 0.123) (-0.087, 0.120) (-0.101, 0.100) (-0.104, 0.095)
skin_tonenumber_80492a 0.101* 0.097* 0.078 0.080
(0.010, 0.191) (0.006, 0.187) (-0.009, 0.166) (-0.007, 0.167)
skin_tonenumber_885633 0.108* 0.103 0.088 0.092
(0.004, 0.213) (-0.002, 0.207) (-0.013, 0.189) (-0.008, 0.192)
skin_tonenumber_94623d 0.065 0.057 0.058 0.054
(-0.031, 0.162) (-0.039, 0.154) (-0.035, 0.151) (-0.038, 0.147)
skin_tonenumber_ab8b64 0.069 0.062 0.055 0.055
(-0.022, 0.161) (-0.030, 0.153) (-0.034, 0.143) (-0.033, 0.143)
skin_tonenumber_b26949 0.156** 0.149** 0.129* 0.137*
(0.035, 0.277) (0.028, 0.270) (0.012, 0.247) (0.021, 0.254)
skin_tonenumber_cb9662 0.077 0.067 0.058 0.061
(-0.019, 0.172) (-0.029, 0.163) (-0.035, 0.151) (-0.031, 0.153)
skin_tonenumber_d09e7d -0.003 -0.011 -0.013 -0.002
(-0.115, 0.109) (-0.122, 0.101) (-0.121, 0.096) (-0.109, 0.106)
skin_tonenumber_e7bc91 0.118* 0.109 0.083 0.097
(0.001, 0.235) (-0.008, 0.226) (-0.031, 0.196) (-0.016, 0.210)
skin_tonenumber_e9cba7 -0.006 -0.017 -0.063 -0.035
(-0.123, 0.110) (-0.134, 0.100) (-0.176, 0.051) (-0.148, 0.078)
skin_tonenumber_ecc083 0.071 0.061 0.038 0.060
(-0.028, 0.170) (-0.039, 0.160) (-0.058, 0.134) (-0.036, 0.155)
skin_tonenumber_eed0b8 0.023 0.017 0.0004 0.023
(-0.066, 0.112) (-0.073, 0.107) (-0.087, 0.087) (-0.064, 0.109)
skin_tonenumber_efc088 -0.007 -0.019 -0.041 -0.020
(-0.135, 0.121) (-0.148, 0.109) (-0.166, 0.083) (-0.144, 0.103)
skin_tonenumber_efc794 -0.021 -0.026 -0.033 -0.015
(-0.126, 0.084) (-0.132, 0.079) (-0.135, 0.070) (-0.116, 0.087)
skin_tonenumber_f6e1aa 0.012 0.002 -0.010 0.026
(-0.074, 0.098) (-0.086, 0.089) (-0.095, 0.074) (-0.059, 0.110)
skin_tonenumber_f7ddc4 0.018 0.017 0.008 0.038
(-0.065, 0.100) (-0.066, 0.100) (-0.073, 0.088) (-0.042, 0.118)
age -0.002* -0.002 -0.001
(-0.005, -0.00004) (-0.004, 0.001) (-0.003, 0.001)
attractiveness 0.007 0.009 0.007
(-0.013, 0.027) (-0.010, 0.028) (-0.012, 0.026)
competence -0.009 -0.011 -0.013
(-0.032, 0.014) (-0.033, 0.011) (-0.035, 0.009)
dominance 0.018* 0.017* 0.017*
(0.002, 0.035) (0.001, 0.033) (0.002, 0.033)
trustworthiness 0.004 0.004 0.002
(-0.016, 0.025) (-0.015, 0.024) (-0.018, 0.021)
p_hat_covariates 0.761*** 0.750***
(0.640, 0.882) (0.631, 0.870)
p_hat_cnn 0.554***
(0.382, 0.726)
Constant 1.131*** 1.084*** 1.059*** 0.406*** -0.069
(1.048, 1.213) (0.977, 1.190) (0.906, 1.211) (0.225, 0.586) (-0.301, 0.163)
Observations 1,593 1,593 1,593 1,593 1,593
Adjusted R2 0.020 0.022 0.026 0.087 0.103
F Statistic 33.263*** (df = 1; 1591) 3.024*** (df = 18; 1574) 2.819*** (df = 23; 1569) 7.351*** (df = 24; 1568) 8.303*** (df = 25; 1567)
Note: p<0.1; p<0.05; p<0.01
Table _05 - Including p_hat_cnn first
Multihead(ResNet50)
Dependent variable:
Release Outcome
(1) (2) (3) (4) (5)
risk_pred_prob -1.074*** -0.929*** -0.942*** -0.944*** -0.685***
(-1.188, -0.960) (-1.043, -0.815) (-1.056, -0.827) (-1.059, -0.829) (-0.798, -0.573)
p_hat_cnn 0.608*** 0.613*** 0.621*** 0.410***
(0.536, 0.679) (0.541, 0.685) (0.548, 0.694) (0.338, 0.482)
skin_tonenumber_623a17 -0.004 -0.004 -0.006
(-0.040, 0.033) (-0.040, 0.032) (-0.041, 0.029)
skin_tonenumber_76441f -0.011 -0.011 -0.009
(-0.056, 0.034) (-0.055, 0.034) (-0.052, 0.034)
skin_tonenumber_80492a -0.006 -0.006 -0.008
(-0.045, 0.034) (-0.045, 0.034) (-0.046, 0.031)
skin_tonenumber_885633 0.042 0.043 0.033
(-0.006, 0.089) (-0.005, 0.090) (-0.013, 0.078)
skin_tonenumber_94623d -0.014 -0.014 -0.020
(-0.061, 0.032) (-0.060, 0.033) (-0.065, 0.025)
skin_tonenumber_ab8b64 -0.017 -0.016 -0.022
(-0.058, 0.025) (-0.058, 0.026) (-0.063, 0.018)
skin_tonenumber_b26949 0.001 0.001 0.004
(-0.056, 0.057) (-0.056, 0.057) (-0.051, 0.058)
skin_tonenumber_cb9662 0.023 0.025 0.025
(-0.023, 0.069) (-0.022, 0.071) (-0.019, 0.070)
skin_tonenumber_d09e7d 0.002 0.002 0.001
(-0.052, 0.056) (-0.051, 0.056) (-0.051, 0.053)
skin_tonenumber_e7bc91 -0.021 -0.020 -0.043
(-0.086, 0.044) (-0.085, 0.045) (-0.106, 0.020)
skin_tonenumber_e9cba7 -0.048 -0.048 -0.077**
(-0.111, 0.015) (-0.111, 0.015) (-0.138, -0.016)
skin_tonenumber_ecc083 -0.029 -0.029 -0.044
(-0.078, 0.019) (-0.077, 0.020) (-0.092, 0.003)
skin_tonenumber_eed0b8 -0.003 -0.004 -0.024
(-0.046, 0.040) (-0.047, 0.039) (-0.065, 0.018)
skin_tonenumber_efc088 -0.096** -0.096** -0.097**
(-0.171, -0.022) (-0.170, -0.021) (-0.168, -0.025)
skin_tonenumber_efc794 -0.047 -0.047 -0.056*
(-0.100, 0.005) (-0.100, 0.005) (-0.107, -0.006)
skin_tonenumber_f6e1aa -0.001 -0.0001 -0.017
(-0.043, 0.042) (-0.043, 0.043) (-0.058, 0.024)
skin_tonenumber_f7ddc4 -0.021 -0.022 -0.036
(-0.060, 0.018) (-0.061, 0.018) (-0.074, 0.001)
age 0.001 0.001
(-0.0003, 0.002) (-0.0003, 0.002)
attractiveness -0.001 0.003
(-0.011, 0.010) (-0.008, 0.013)
competence 0.0005 -0.003
(-0.012, 0.013) (-0.015, 0.009)
dominance 0.001 0.005
(-0.007, 0.010) (-0.004, 0.013)
trustworthiness 0.0002 0.001
(-0.011, 0.011) (-0.009, 0.012)
p_hat_covariates 0.865***
(0.804, 0.925)
Constant 1.095*** 0.596*** 0.605*** 0.564*** -0.029
(1.059, 1.131) (0.528, 0.665) (0.530, 0.679) (0.467, 0.662) (-0.132, 0.074)
Observations 7,318 7,318 7,318 7,318 7,318
Adjusted R2 0.032 0.057 0.057 0.056 0.123
F Statistic 239.132*** (df = 1; 7316) 220.280*** (df = 2; 7315) 24.137*** (df = 19; 7298) 19.172*** (df = 24; 7293) 41.899*** (df = 25; 7292)
Note: p<0.1; p<0.05; p<0.01