Descriptives

Mean Mental Health Days by Race (Unweighted)
Race_Label Mean_Poor_Menthlth_Days Mean_Any_Bad_Menthlth_Days Mean_Frequent_Bad_Menthlth_Days
Latino 3.829876 0.3219597 0.1226691
White 3.381298 0.3147087 0.1042163
Mean Mental Health Days by Race (Weighted)
Race_Label Mean_Poor_Menthlth_Days Mean_Any_Bad_Menthlth_Days Mean_Frequent_Bad_Menthlth_Days
Latino 3.694928 0.3301296 0.1166011
White 3.931572 0.3586166 0.1221424

Lagged Time Series Analysis

In this section, I test three sets models with four measures.

Measures:

  1. Rolling average hate speech

  2. Did hate speech get better or worse the month before the response? This is a measure of the average hate speech of three months prior minus the average hate speech of two months prior.

  3. Rolling average standard deviation. How far from the mean is this month?

  4. Percent negative hate speech. What percent is above a score of 1?

Models:

  1. White

  2. Latino

  3. Race interaction

Controls:

  1. age

  2. education

  3. gender

  4. state

  5. month

White Models

Rolling Average

Regression results examining 90 day rolling average and mental health - White
term estimate std.error statistic p.value
(Intercept) 11.914 0.099 120.162 0.000
X_age_g -0.639 0.003 -191.095 0.000
X_educag -0.940 0.006 -164.970 0.000
female 1.478 0.011 130.786 0.000
_STATE2 -1.255 0.127 -9.882 0.000
_STATE4 -0.476 0.060 -7.983 0.000
_STATE5 0.135 0.070 1.932 0.053
_STATE6 -0.425 0.049 -8.701 0.000
_STATE8 -0.803 0.060 -13.316 0.000
_STATE9 -0.841 0.067 -12.503 0.000
_STATE10 -0.466 0.110 -4.216 0.000
_STATE11 -1.349 0.160 -8.455 0.000
_STATE12 -0.369 0.050 -7.361 0.000
_STATE13 -0.492 0.056 -8.794 0.000
_STATE15 -1.139 0.140 -8.105 0.000
_STATE16 -1.009 0.082 -12.363 0.000
_STATE17 -0.877 0.052 -16.712 0.000
_STATE18 -0.425 0.056 -7.537 0.000
_STATE19 -1.342 0.066 -20.452 0.000
_STATE20 -1.133 0.070 -16.163 0.000
_STATE21 -0.173 0.061 -2.850 0.004
_STATE22 -0.339 0.066 -5.154 0.000
_STATE23 -0.576 0.083 -6.944 0.000
_STATE24 -0.723 0.063 -11.465 0.000
_STATE25 -0.424 0.057 -7.471 0.000
_STATE26 -0.484 0.053 -9.098 0.000
_STATE27 -1.378 0.058 -23.709 0.000
_STATE28 -0.145 0.075 -1.924 0.054
_STATE29 -0.658 0.057 -11.502 0.000
_STATE30 -1.087 0.094 -11.553 0.000
_STATE31 -1.337 0.079 -16.949 0.000
_STATE32 -0.373 0.078 -4.761 0.000
_STATE33 -0.680 0.084 -8.123 0.000
_STATE34 -0.540 0.059 -9.130 0.000
_STATE35 -0.288 0.099 -2.908 0.004
_STATE36 -0.763 0.051 -15.112 0.000
_STATE37 -0.637 0.054 -11.765 0.000
_STATE38 -1.512 0.109 -13.878 0.000
_STATE39 -0.501 0.052 -9.729 0.000
_STATE40 -0.269 0.066 -4.058 0.000
_STATE41 -0.178 0.063 -2.840 0.005
_STATE42 -0.717 0.051 -14.059 0.000
_STATE44 -0.533 0.099 -5.393 0.000
_STATE45 -0.231 0.063 -3.675 0.000
_STATE46 -1.838 0.104 -17.664 0.000
_STATE47 -0.172 0.057 -3.016 0.003
_STATE48 -0.901 0.050 -18.004 0.000
_STATE49 -0.836 0.070 -11.947 0.000
_STATE50 -0.721 0.112 -6.449 0.000
_STATE51 -0.820 0.056 -14.587 0.000
_STATE53 -0.522 0.056 -9.241 0.000
_STATE54 0.249 0.075 3.311 0.001
_STATE55 -0.955 0.057 -16.668 0.000
_STATE56 -1.114 0.123 -9.047 0.000
_STATE66 -1.999 0.776 -2.575 0.010
IMONTH2 -0.044 0.030 -1.495 0.135
IMONTH3 0.045 0.029 1.522 0.128
IMONTH4 0.165 0.030 5.517 0.000
IMONTH5 0.494 0.030 16.693 0.000
IMONTH6 0.388 0.029 13.324 0.000
IMONTH7 0.221 0.029 7.552 0.000
IMONTH8 0.300 0.029 10.235 0.000
IMONTH9 0.405 0.030 13.552 0.000
IMONTH10 0.484 0.030 16.321 0.000
IMONTH11 0.479 0.030 16.085 0.000
IMONTH12 0.547 0.030 18.157 0.000
rolling_avg_HS90 1.259 0.033 38.149 0.000

In this model I find that the rolling average of three months is significantly associated to number of bad days mental health for the White population. A one unit increase in average hate speech classifier is associated with a 1.26 day increase in bad mental health. This suggests that worsening hate speech towards the Latinx population is associated with more bad mental health among whites.

Difference

Regression results examining 2 and 3 month difference and mental health - White
term estimate std.error statistic p.value
(Intercept) 8.738 0.053 163.483 0.000
X_age_g -0.638 0.003 -190.476 0.000
X_educag -0.937 0.006 -164.320 0.000
female 1.477 0.011 130.669 0.000
_STATE2 -1.246 0.127 -9.805 0.000
_STATE4 -0.469 0.060 -7.868 0.000
_STATE5 0.122 0.070 1.747 0.081
_STATE6 -0.419 0.049 -8.581 0.000
_STATE8 -0.792 0.060 -13.142 0.000
_STATE9 -0.827 0.067 -12.301 0.000
_STATE10 -0.474 0.110 -4.292 0.000
_STATE11 -1.340 0.160 -8.396 0.000
_STATE12 -0.357 0.050 -7.129 0.000
_STATE13 -0.545 0.056 -9.731 0.000
_STATE15 -1.147 0.141 -8.164 0.000
_STATE16 -1.021 0.082 -12.508 0.000
_STATE17 -0.882 0.052 -16.805 0.000
_STATE18 -0.409 0.056 -7.244 0.000
_STATE19 -1.339 0.066 -20.404 0.000
_STATE20 -1.130 0.070 -16.118 0.000
_STATE21 -0.175 0.061 -2.878 0.004
_STATE22 -0.330 0.066 -5.015 0.000
_STATE23 -0.584 0.083 -7.028 0.000
_STATE24 -0.721 0.063 -11.426 0.000
_STATE25 -0.424 0.057 -7.463 0.000
_STATE26 -0.477 0.053 -8.955 0.000
_STATE27 -1.395 0.058 -23.990 0.000
_STATE28 -0.159 0.075 -2.101 0.036
_STATE29 -0.648 0.057 -11.328 0.000
_STATE30 -1.075 0.094 -11.410 0.000
_STATE31 -1.342 0.079 -16.997 0.000
_STATE32 -0.380 0.078 -4.840 0.000
_STATE33 -0.686 0.084 -8.192 0.000
_STATE34 -0.581 0.059 -9.820 0.000
_STATE35 -0.293 0.099 -2.961 0.003
_STATE36 -0.763 0.051 -15.101 0.000
_STATE37 -0.633 0.054 -11.684 0.000
_STATE38 -1.507 0.109 -13.821 0.000
_STATE39 -0.512 0.052 -9.940 0.000
_STATE40 -0.269 0.066 -4.053 0.000
_STATE41 -0.179 0.063 -2.841 0.005
_STATE42 -0.718 0.051 -14.066 0.000
_STATE44 -0.525 0.099 -5.305 0.000
_STATE45 -0.236 0.063 -3.747 0.000
_STATE46 -1.834 0.104 -17.625 0.000
_STATE47 -0.161 0.057 -2.815 0.005
_STATE48 -0.889 0.050 -17.759 0.000
_STATE49 -0.827 0.070 -11.822 0.000
_STATE50 -0.719 0.112 -6.426 0.000
_STATE51 -0.829 0.056 -14.729 0.000
_STATE53 -0.515 0.056 -9.112 0.000
_STATE54 0.248 0.075 3.290 0.001
_STATE55 -0.957 0.057 -16.684 0.000
_STATE56 -1.108 0.123 -8.999 0.000
_STATE66 -1.985 0.777 -2.556 0.011
IMONTH2 0.013 0.029 0.441 0.659
IMONTH3 0.088 0.029 3.010 0.003
IMONTH4 0.156 0.031 4.967 0.000
IMONTH5 0.386 0.031 12.592 0.000
IMONTH6 0.275 0.030 9.241 0.000
IMONTH7 0.162 0.029 5.496 0.000
IMONTH8 0.281 0.029 9.568 0.000
IMONTH9 0.366 0.030 12.100 0.000
IMONTH10 0.324 0.031 10.395 0.000
IMONTH11 0.295 0.031 9.503 0.000
IMONTH12 0.401 0.030 13.383 0.000
difference_HS30 0.516 0.148 3.479 0.001

If the difference between month 3 and 2 is positive then things are getting worse.This model suggests that worsening hate speech is associated with an increase of .5 days bad mental health for the White population.

Rolling average sd

Regression results examining rolling average standard deviation and mental health - White
term estimate std.error statistic p.value
(Intercept) 4.564 0.107 42.647 0.000
X_age_g -0.640 0.003 -191.157 0.000
X_educag -0.941 0.006 -165.141 0.000
female 1.478 0.011 130.822 0.000
_STATE2 -1.246 0.127 -9.810 0.000
_STATE4 -0.481 0.060 -8.073 0.000
_STATE5 0.110 0.070 1.579 0.114
_STATE6 -0.422 0.049 -8.629 0.000
_STATE8 -0.807 0.060 -13.388 0.000
_STATE9 -0.830 0.067 -12.346 0.000
_STATE10 -0.481 0.110 -4.352 0.000
_STATE11 -1.331 0.160 -8.346 0.000
_STATE12 -0.360 0.050 -7.181 0.000
_STATE13 -0.550 0.056 -9.828 0.000
_STATE15 -1.136 0.140 -8.087 0.000
_STATE16 -1.037 0.082 -12.702 0.000
_STATE17 -0.886 0.052 -16.882 0.000
_STATE18 -0.408 0.056 -7.245 0.000
_STATE19 -1.338 0.066 -20.397 0.000
_STATE20 -1.136 0.070 -16.204 0.000
_STATE21 -0.171 0.061 -2.808 0.005
_STATE22 -0.332 0.066 -5.056 0.000
_STATE23 -0.589 0.083 -7.102 0.000
_STATE24 -0.718 0.063 -11.387 0.000
_STATE25 -0.423 0.057 -7.441 0.000
_STATE26 -0.478 0.053 -8.977 0.000
_STATE27 -1.403 0.058 -24.143 0.000
_STATE28 -0.153 0.075 -2.030 0.042
_STATE29 -0.650 0.057 -11.374 0.000
_STATE30 -1.086 0.094 -11.542 0.000
_STATE31 -1.346 0.079 -17.055 0.000
_STATE32 -0.385 0.078 -4.904 0.000
_STATE33 -0.696 0.084 -8.310 0.000
_STATE34 -0.529 0.059 -8.956 0.000
_STATE35 -0.287 0.099 -2.900 0.004
_STATE36 -0.761 0.051 -15.064 0.000
_STATE37 -0.638 0.054 -11.779 0.000
_STATE38 -1.501 0.109 -13.775 0.000
_STATE39 -0.512 0.052 -9.940 0.000
_STATE40 -0.276 0.066 -4.164 0.000
_STATE41 -0.183 0.063 -2.917 0.004
_STATE42 -0.708 0.051 -13.882 0.000
_STATE44 -0.524 0.099 -5.300 0.000
_STATE45 -0.243 0.063 -3.868 0.000
_STATE46 -1.837 0.104 -17.657 0.000
_STATE47 -0.159 0.057 -2.780 0.005
_STATE48 -0.893 0.050 -17.840 0.000
_STATE49 -0.839 0.070 -11.992 0.000
_STATE50 -0.722 0.112 -6.463 0.000
_STATE51 -0.823 0.056 -14.634 0.000
_STATE53 -0.518 0.056 -9.176 0.000
_STATE54 0.250 0.075 3.323 0.001
_STATE55 -0.960 0.057 -16.745 0.000
_STATE56 -1.103 0.123 -8.963 0.000
_STATE66 -2.007 0.776 -2.585 0.010
IMONTH2 0.040 0.029 1.347 0.178
IMONTH3 0.122 0.029 4.172 0.000
IMONTH4 0.179 0.030 5.995 0.000
IMONTH5 0.377 0.030 12.765 0.000
IMONTH6 0.271 0.029 9.317 0.000
IMONTH7 0.156 0.029 5.336 0.000
IMONTH8 0.274 0.029 9.355 0.000
IMONTH9 0.332 0.030 11.123 0.000
IMONTH10 0.200 0.030 6.748 0.000
IMONTH11 0.144 0.030 4.852 0.000
IMONTH12 0.211 0.030 6.993 0.000
rolling_sd_HS90 3.107 0.069 44.805 0.000

This model suggests that an increase in one standard deviation of hate speech is associated with 3 days increased bad mental health for whites.

Percent negative

Regression results examining rolling percent negative and mental health - White
term estimate std.error statistic p.value
(Intercept) 8.700 0.060 145.956 0.000
X_age_g -0.638 0.003 -190.497 0.000
X_educag -0.937 0.006 -164.324 0.000
female 1.477 0.011 130.673 0.000
_STATE2 -1.246 0.127 -9.804 0.000
_STATE4 -0.471 0.060 -7.891 0.000
_STATE5 0.123 0.070 1.763 0.078
_STATE6 -0.420 0.049 -8.590 0.000
_STATE8 -0.793 0.060 -13.148 0.000
_STATE9 -0.829 0.067 -12.323 0.000
_STATE10 -0.474 0.110 -4.290 0.000
_STATE11 -1.342 0.160 -8.409 0.000
_STATE12 -0.357 0.050 -7.127 0.000
_STATE13 -0.540 0.056 -9.642 0.000
_STATE15 -1.147 0.141 -8.159 0.000
_STATE16 -1.021 0.082 -12.500 0.000
_STATE17 -0.882 0.052 -16.804 0.000
_STATE18 -0.410 0.056 -7.267 0.000
_STATE19 -1.338 0.066 -20.394 0.000
_STATE20 -1.131 0.070 -16.133 0.000
_STATE21 -0.175 0.061 -2.880 0.004
_STATE22 -0.330 0.066 -5.020 0.000
_STATE23 -0.582 0.083 -7.007 0.000
_STATE24 -0.719 0.063 -11.405 0.000
_STATE25 -0.424 0.057 -7.467 0.000
_STATE26 -0.477 0.053 -8.959 0.000
_STATE27 -1.393 0.058 -23.964 0.000
_STATE28 -0.157 0.075 -2.079 0.038
_STATE29 -0.650 0.057 -11.352 0.000
_STATE30 -1.076 0.094 -11.430 0.000
_STATE31 -1.341 0.079 -16.994 0.000
_STATE32 -0.378 0.078 -4.823 0.000
_STATE33 -0.685 0.084 -8.176 0.000
_STATE34 -0.586 0.059 -9.917 0.000
_STATE35 -0.293 0.099 -2.958 0.003
_STATE36 -0.762 0.051 -15.082 0.000
_STATE37 -0.633 0.054 -11.684 0.000
_STATE38 -1.506 0.109 -13.814 0.000
_STATE39 -0.511 0.052 -9.915 0.000
_STATE40 -0.269 0.066 -4.058 0.000
_STATE41 -0.179 0.063 -2.853 0.004
_STATE42 -0.717 0.051 -14.043 0.000
_STATE44 -0.524 0.099 -5.295 0.000
_STATE45 -0.236 0.063 -3.751 0.000
_STATE46 -1.835 0.104 -17.632 0.000
_STATE47 -0.160 0.057 -2.795 0.005
_STATE48 -0.890 0.050 -17.764 0.000
_STATE49 -0.828 0.070 -11.836 0.000
_STATE50 -0.718 0.112 -6.421 0.000
_STATE51 -0.827 0.056 -14.694 0.000
_STATE53 -0.514 0.056 -9.108 0.000
_STATE54 0.248 0.075 3.294 0.001
_STATE55 -0.957 0.057 -16.698 0.000
_STATE56 -1.108 0.123 -8.995 0.000
_STATE66 -1.985 0.777 -2.556 0.011
IMONTH2 0.014 0.030 0.490 0.624
IMONTH3 0.093 0.030 3.153 0.002
IMONTH4 0.193 0.030 6.344 0.000
IMONTH5 0.420 0.030 13.923 0.000
IMONTH6 0.302 0.030 10.187 0.000
IMONTH7 0.179 0.030 5.975 0.000
IMONTH8 0.284 0.030 9.427 0.000
IMONTH9 0.390 0.031 12.702 0.000
IMONTH10 0.366 0.030 12.115 0.000
IMONTH11 0.332 0.030 11.104 0.000
IMONTH12 0.407 0.030 13.599 0.000
percent_negative_HS90 0.028 0.031 0.898 0.369

No significant association for whites.

Latinx Model

Rolling Average

Regression Results for Examining 90 day rolling average and mental health - Latino
term estimate std.error statistic p.value
(Intercept) 5.338 0.430 12.413 0.000
X_age_g -0.163 0.012 -13.605 0.000
X_educag -0.052 0.016 -3.314 0.001
female 0.882 0.036 24.202 0.000
_STATE2 0.244 0.680 0.359 0.720
_STATE4 -0.075 0.337 -0.223 0.823
_STATE5 -0.238 0.448 -0.530 0.596
_STATE6 -0.587 0.325 -1.806 0.071
_STATE8 -0.447 0.348 -1.285 0.199
_STATE9 0.146 0.370 0.395 0.693
_STATE10 -0.617 0.571 -1.082 0.279
_STATE11 -1.238 0.592 -2.090 0.037
_STATE12 -0.115 0.329 -0.349 0.727
_STATE13 -0.436 0.353 -1.234 0.217
_STATE15 0.402 0.479 0.840 0.401
_STATE16 -0.312 0.453 -0.688 0.491
_STATE17 -0.584 0.337 -1.733 0.083
_STATE18 -0.732 0.387 -1.891 0.059
_STATE19 -0.958 0.467 -2.052 0.040
_STATE20 -0.524 0.409 -1.281 0.200
_STATE21 0.349 0.492 0.709 0.479
_STATE22 1.161 0.431 2.693 0.007
_STATE23 0.289 1.019 0.283 0.777
_STATE24 -0.829 0.368 -2.250 0.024
_STATE25 0.169 0.358 0.472 0.637
_STATE26 1.431 0.383 3.740 0.000
_STATE27 -0.641 0.420 -1.528 0.126
_STATE28 -0.043 0.639 -0.067 0.947
_STATE29 0.850 0.430 1.976 0.048
_STATE30 0.504 0.782 0.645 0.519
_STATE31 -1.053 0.459 -2.295 0.022
_STATE32 -0.846 0.357 -2.370 0.018
_STATE33 1.276 0.768 1.660 0.097
_STATE34 -0.394 0.341 -1.155 0.248
_STATE35 0.295 0.350 0.841 0.400
_STATE36 0.004 0.331 0.011 0.991
_STATE37 -1.069 0.355 -3.014 0.003
_STATE38 1.733 0.957 1.810 0.070
_STATE39 2.231 0.393 5.676 0.000
_STATE40 -0.318 0.395 -0.805 0.421
_STATE41 0.398 0.379 1.049 0.294
_STATE42 1.564 0.355 4.407 0.000
_STATE44 0.284 0.475 0.598 0.550
_STATE45 -0.361 0.421 -0.856 0.392
_STATE46 -0.168 0.904 -0.186 0.852
_STATE47 0.282 0.415 0.680 0.496
_STATE48 -0.624 0.326 -1.915 0.055
_STATE49 -0.708 0.392 -1.806 0.071
_STATE50 2.391 1.338 1.786 0.074
_STATE51 -1.031 0.358 -2.878 0.004
_STATE53 -0.254 0.356 -0.714 0.475
_STATE54 3.089 0.906 3.410 0.001
_STATE55 0.581 0.398 1.461 0.144
_STATE56 0.312 0.656 0.475 0.634
_STATE66 0.348 1.326 0.263 0.793
IMONTH2 0.309 0.100 3.078 0.002
IMONTH3 0.199 0.097 2.050 0.040
IMONTH4 0.225 0.099 2.260 0.024
IMONTH5 0.163 0.098 1.657 0.097
IMONTH6 0.177 0.094 1.892 0.058
IMONTH7 0.118 0.097 1.220 0.222
IMONTH8 0.238 0.097 2.456 0.014
IMONTH9 0.484 0.095 5.099 0.000
IMONTH10 0.355 0.094 3.755 0.000
IMONTH11 0.311 0.096 3.225 0.001
IMONTH12 0.399 0.097 4.095 0.000
rolling_avg_HS90 0.547 0.106 5.171 0.000

A one unit increase is significantly associated with a .55 day increase in poor mental health days for the Latino population.

Difference

Regression results examining 2 and 3 month difference and mental health - Latino
term estimate std.error statistic p.value
(Intercept) 3.959 0.336 11.798 0.000
X_age_g -0.162 0.012 -13.512 0.000
X_educag -0.050 0.016 -3.191 0.001
female 0.882 0.036 24.213 0.000
_STATE2 0.258 0.680 0.379 0.704
_STATE4 -0.061 0.337 -0.182 0.855
_STATE5 -0.230 0.448 -0.513 0.608
_STATE6 -0.572 0.325 -1.758 0.079
_STATE8 -0.432 0.348 -1.244 0.214
_STATE9 0.163 0.370 0.440 0.660
_STATE10 -0.605 0.571 -1.060 0.289
_STATE11 -1.223 0.592 -2.065 0.039
_STATE12 -0.101 0.329 -0.308 0.758
_STATE13 -0.450 0.353 -1.274 0.203
_STATE15 0.403 0.479 0.842 0.400
_STATE16 -0.309 0.453 -0.682 0.495
_STATE17 -0.575 0.337 -1.708 0.088
_STATE18 -0.716 0.387 -1.850 0.064
_STATE19 -0.951 0.467 -2.037 0.042
_STATE20 -0.516 0.409 -1.262 0.207
_STATE21 0.352 0.492 0.715 0.475
_STATE22 1.177 0.431 2.730 0.006
_STATE23 0.294 1.019 0.288 0.773
_STATE24 -0.819 0.368 -2.223 0.026
_STATE25 0.176 0.358 0.493 0.622
_STATE26 1.442 0.383 3.767 0.000
_STATE27 -0.641 0.420 -1.527 0.127
_STATE28 -0.046 0.639 -0.072 0.943
_STATE29 0.858 0.430 1.995 0.046
_STATE30 0.520 0.782 0.665 0.506
_STATE31 -1.046 0.459 -2.278 0.023
_STATE32 -0.840 0.357 -2.354 0.019
_STATE33 1.291 0.768 1.680 0.093
_STATE34 -0.400 0.341 -1.170 0.242
_STATE35 0.300 0.350 0.855 0.392
_STATE36 0.008 0.331 0.024 0.981
_STATE37 -1.060 0.355 -2.990 0.003
_STATE38 1.758 0.957 1.836 0.066
_STATE39 2.235 0.393 5.685 0.000
_STATE40 -0.311 0.395 -0.789 0.430
_STATE41 0.408 0.379 1.075 0.282
_STATE42 1.573 0.355 4.432 0.000
_STATE44 0.300 0.475 0.630 0.528
_STATE45 -0.359 0.421 -0.852 0.394
_STATE46 -0.144 0.904 -0.159 0.873
_STATE47 0.297 0.415 0.717 0.474
_STATE48 -0.611 0.326 -1.875 0.061
_STATE49 -0.699 0.392 -1.782 0.075
_STATE50 2.411 1.339 1.801 0.072
_STATE51 -1.027 0.358 -2.866 0.004
_STATE53 -0.243 0.356 -0.684 0.494
_STATE54 3.097 0.906 3.419 0.001
_STATE55 0.587 0.398 1.475 0.140
_STATE56 0.325 0.656 0.496 0.620
_STATE66 0.347 1.326 0.261 0.794
IMONTH2 0.343 0.100 3.432 0.001
IMONTH3 0.224 0.097 2.307 0.021
IMONTH4 0.212 0.106 2.008 0.045
IMONTH5 0.105 0.103 1.024 0.306
IMONTH6 0.116 0.096 1.204 0.229
IMONTH7 0.084 0.097 0.860 0.390
IMONTH8 0.233 0.097 2.403 0.016
IMONTH9 0.455 0.097 4.693 0.000
IMONTH10 0.261 0.101 2.598 0.009
IMONTH11 0.207 0.101 2.048 0.041
IMONTH12 0.319 0.097 3.293 0.001
difference_HS30 0.459 0.488 0.941 0.346

No significant association

Rolling average sd

Regression results examining rolling average standard deviation and mental health - Latino
term estimate std.error statistic p.value
(Intercept) 1.932 0.451 4.284 0.000
X_age_g -0.163 0.012 -13.611 0.000
X_educag -0.052 0.016 -3.273 0.001
female 0.883 0.036 24.244 0.000
_STATE2 0.262 0.680 0.385 0.700
_STATE4 -0.074 0.337 -0.220 0.826
_STATE5 -0.248 0.448 -0.554 0.580
_STATE6 -0.582 0.325 -1.790 0.073
_STATE8 -0.444 0.348 -1.277 0.202
_STATE9 0.153 0.370 0.414 0.679
_STATE10 -0.615 0.571 -1.078 0.281
_STATE11 -1.225 0.592 -2.069 0.039
_STATE12 -0.106 0.329 -0.322 0.748
_STATE13 -0.460 0.353 -1.303 0.193
_STATE15 0.404 0.479 0.843 0.399
_STATE16 -0.326 0.453 -0.720 0.471
_STATE17 -0.585 0.337 -1.736 0.083
_STATE18 -0.723 0.387 -1.870 0.061
_STATE19 -0.956 0.467 -2.050 0.040
_STATE20 -0.523 0.409 -1.281 0.200
_STATE21 0.348 0.492 0.708 0.479
_STATE22 1.163 0.431 2.697 0.007
_STATE23 0.285 1.019 0.279 0.780
_STATE24 -0.827 0.368 -2.245 0.025
_STATE25 0.169 0.358 0.473 0.637
_STATE26 1.434 0.383 3.747 0.000
_STATE27 -0.650 0.420 -1.549 0.121
_STATE28 -0.027 0.639 -0.042 0.966
_STATE29 0.846 0.430 1.968 0.049
_STATE30 0.511 0.782 0.654 0.513
_STATE31 -1.053 0.459 -2.295 0.022
_STATE32 -0.853 0.357 -2.389 0.017
_STATE33 1.281 0.768 1.667 0.096
_STATE34 -0.379 0.341 -1.110 0.267
_STATE35 0.296 0.350 0.845 0.398
_STATE36 0.005 0.331 0.015 0.988
_STATE37 -1.070 0.355 -3.016 0.003
_STATE38 1.740 0.957 1.818 0.069
_STATE39 2.227 0.393 5.665 0.000
_STATE40 -0.323 0.395 -0.818 0.413
_STATE41 0.396 0.379 1.044 0.297
_STATE42 1.572 0.355 4.430 0.000
_STATE44 0.291 0.475 0.612 0.540
_STATE45 -0.368 0.421 -0.873 0.383
_STATE46 -0.160 0.904 -0.177 0.860
_STATE47 0.289 0.415 0.698 0.485
_STATE48 -0.617 0.326 -1.895 0.058
_STATE49 -0.706 0.392 -1.801 0.072
_STATE50 2.381 1.338 1.779 0.075
_STATE51 -1.034 0.358 -2.885 0.004
_STATE53 -0.250 0.356 -0.703 0.482
_STATE54 3.094 0.906 3.416 0.001
_STATE55 0.579 0.398 1.455 0.146
_STATE56 0.322 0.656 0.491 0.624
_STATE66 0.359 1.326 0.271 0.787
IMONTH2 0.363 0.100 3.626 0.000
IMONTH3 0.242 0.097 2.496 0.013
IMONTH4 0.228 0.099 2.300 0.021
IMONTH5 0.105 0.098 1.068 0.286
IMONTH6 0.126 0.093 1.350 0.177
IMONTH7 0.090 0.096 0.936 0.349
IMONTH8 0.232 0.097 2.397 0.017
IMONTH9 0.446 0.095 4.701 0.000
IMONTH10 0.226 0.094 2.400 0.016
IMONTH11 0.177 0.096 1.844 0.065
IMONTH12 0.271 0.097 2.795 0.005
rolling_sd_HS90 1.499 0.224 6.676 0.000

A one standard deviation increase in hate speech is associated with a 1.5 day increase in bad mental health days for the Latinx population.

Percent negative

Regression results examining rolling percent negative and mental health - Latino
term estimate std.error statistic p.value
(Intercept) 3.828 0.347 11.047 0.000
X_age_g -0.162 0.012 -13.509 0.000
X_educag -0.051 0.016 -3.200 0.001
female 0.882 0.036 24.220 0.000
_STATE2 0.259 0.680 0.381 0.703
_STATE4 -0.065 0.337 -0.192 0.848
_STATE5 -0.233 0.448 -0.521 0.602
_STATE6 -0.575 0.325 -1.769 0.077
_STATE8 -0.434 0.348 -1.248 0.212
_STATE9 0.160 0.370 0.433 0.665
_STATE10 -0.607 0.571 -1.064 0.287
_STATE11 -1.221 0.592 -2.061 0.039
_STATE12 -0.102 0.329 -0.311 0.756
_STATE13 -0.445 0.353 -1.261 0.207
_STATE15 0.404 0.479 0.843 0.399
_STATE16 -0.312 0.453 -0.688 0.491
_STATE17 -0.575 0.337 -1.708 0.088
_STATE18 -0.717 0.387 -1.854 0.064
_STATE19 -0.951 0.467 -2.037 0.042
_STATE20 -0.518 0.409 -1.267 0.205
_STATE21 0.353 0.492 0.718 0.473
_STATE22 1.177 0.431 2.731 0.006
_STATE23 0.295 1.019 0.289 0.772
_STATE24 -0.819 0.368 -2.222 0.026
_STATE25 0.178 0.358 0.498 0.619
_STATE26 1.441 0.383 3.766 0.000
_STATE27 -0.639 0.420 -1.523 0.128
_STATE28 -0.041 0.639 -0.065 0.949
_STATE29 0.855 0.430 1.988 0.047
_STATE30 0.517 0.782 0.661 0.509
_STATE31 -1.045 0.459 -2.277 0.023
_STATE32 -0.839 0.357 -2.351 0.019
_STATE33 1.288 0.768 1.676 0.094
_STATE34 -0.401 0.341 -1.174 0.240
_STATE35 0.299 0.350 0.854 0.393
_STATE36 0.009 0.331 0.028 0.977
_STATE37 -1.061 0.355 -2.991 0.003
_STATE38 1.757 0.957 1.835 0.066
_STATE39 2.237 0.393 5.691 0.000
_STATE40 -0.311 0.395 -0.787 0.431
_STATE41 0.405 0.379 1.069 0.285
_STATE42 1.575 0.355 4.439 0.000
_STATE44 0.300 0.475 0.631 0.528
_STATE45 -0.359 0.421 -0.852 0.394
_STATE46 -0.149 0.904 -0.165 0.869
_STATE47 0.301 0.415 0.726 0.468
_STATE48 -0.612 0.326 -1.879 0.060
_STATE49 -0.701 0.392 -1.789 0.074
_STATE50 2.411 1.339 1.801 0.072
_STATE51 -1.025 0.358 -2.861 0.004
_STATE53 -0.243 0.356 -0.684 0.494
_STATE54 3.099 0.906 3.421 0.001
_STATE55 0.586 0.398 1.474 0.141
_STATE56 0.325 0.656 0.495 0.621
_STATE66 0.347 1.326 0.261 0.794
IMONTH2 0.358 0.101 3.558 0.000
IMONTH3 0.251 0.099 2.547 0.011
IMONTH4 0.272 0.101 2.692 0.007
IMONTH5 0.165 0.101 1.637 0.102
IMONTH6 0.171 0.097 1.768 0.077
IMONTH7 0.130 0.100 1.304 0.192
IMONTH8 0.266 0.100 2.653 0.008
IMONTH9 0.510 0.099 5.172 0.000
IMONTH10 0.331 0.097 3.398 0.001
IMONTH11 0.264 0.098 2.707 0.007
IMONTH12 0.341 0.097 3.514 0.000
percent_negative_HS90 0.128 0.096 1.336 0.181

No significant assocation

Interaction Model

Rolling Average

Regression Results for Examining 90 day rolling average and mental health - interaction
term estimate std.error statistic p.value
(Intercept) 10.865 0.104 104.519 0.000
X_age_g -0.564 0.003 -173.350 0.000
X_educag -0.713 0.005 -136.288 0.000
female 1.349 0.011 124.979 0.000
_STATE2 -1.149 0.130 -8.858 0.000
_STATE4 -0.494 0.058 -8.466 0.000
_STATE5 0.094 0.072 1.306 0.192
_STATE6 -0.746 0.049 -15.119 0.000
_STATE8 -0.865 0.060 -14.312 0.000
_STATE9 -0.794 0.068 -11.752 0.000
_STATE10 -0.579 0.112 -5.169 0.000
_STATE11 -1.478 0.154 -9.616 0.000
_STATE12 -0.330 0.051 -6.471 0.000
_STATE13 -0.573 0.057 -9.978 0.000
_STATE15 -0.806 0.130 -6.191 0.000
_STATE16 -0.998 0.083 -11.990 0.000
_STATE17 -0.945 0.054 -17.613 0.000
_STATE18 -0.461 0.059 -7.869 0.000
_STATE19 -1.362 0.068 -19.917 0.000
_STATE20 -1.138 0.072 -15.878 0.000
_STATE21 -0.123 0.064 -1.936 0.053
_STATE22 -0.211 0.068 -3.094 0.002
_STATE23 -0.580 0.088 -6.621 0.000
_STATE24 -0.837 0.064 -13.050 0.000
_STATE25 -0.449 0.058 -7.699 0.000
_STATE26 -0.409 0.056 -7.368 0.000
_STATE27 -1.410 0.061 -23.260 0.000
_STATE28 -0.134 0.079 -1.697 0.090
_STATE29 -0.596 0.060 -9.970 0.000
_STATE30 -1.057 0.099 -10.704 0.000
_STATE31 -1.399 0.081 -17.294 0.000
_STATE32 -0.707 0.073 -9.622 0.000
_STATE33 -0.666 0.088 -7.558 0.000
_STATE34 -0.618 0.059 -10.496 0.000
_STATE35 -0.088 0.080 -1.097 0.273
_STATE36 -0.691 0.052 -13.392 0.000
_STATE37 -0.782 0.056 -13.989 0.000
_STATE38 -1.440 0.115 -12.577 0.000
_STATE39 -0.400 0.054 -7.435 0.000
_STATE40 -0.337 0.068 -4.957 0.000
_STATE41 -0.194 0.064 -3.016 0.003
_STATE42 -0.574 0.053 -10.814 0.000
_STATE44 -0.507 0.099 -5.122 0.000
_STATE45 -0.284 0.065 -4.350 0.000
_STATE46 -1.804 0.109 -16.507 0.000
_STATE47 -0.149 0.060 -2.496 0.013
_STATE48 -0.969 0.050 -19.302 0.000
_STATE49 -0.891 0.071 -12.554 0.000
_STATE50 -0.699 0.118 -5.912 0.000
_STATE51 -0.922 0.058 -15.926 0.000
_STATE53 -0.595 0.058 -10.259 0.000
_STATE54 0.344 0.079 4.325 0.000
_STATE55 -0.907 0.060 -15.210 0.000
_STATE56 -1.020 0.126 -8.124 0.000
_STATE66 -0.839 0.595 -1.410 0.159
IMONTH2 0.019 0.028 0.661 0.509
IMONTH3 0.069 0.028 2.451 0.014
IMONTH4 0.178 0.029 6.196 0.000
IMONTH5 0.427 0.028 15.016 0.000
IMONTH6 0.337 0.028 12.101 0.000
IMONTH7 0.199 0.028 7.094 0.000
IMONTH8 0.278 0.028 9.885 0.000
IMONTH9 0.418 0.028 14.725 0.000
IMONTH10 0.457 0.028 16.174 0.000
IMONTH11 0.439 0.028 15.435 0.000
IMONTH12 0.512 0.029 17.788 0.000
X_race8 -1.857 0.188 -9.871 0.000
rolling_avg_HS90 1.178 0.035 33.706 0.000
X_race8:rolling_avg_HS90 -0.271 0.073 -3.693 0.000

A one unit increase in hate speech is associated with 1.18 increased days bad mental health for Whites and a .91 increase for latinos.

Difference

Regression results examining 2 and 3 month difference and mental health - interaction
term estimate std.error statistic p.value
(Intercept) 7.891 0.054 145.036 0.000
X_age_g -0.563 0.003 -172.775 0.000
X_educag -0.710 0.005 -135.658 0.000
female 1.348 0.011 124.894 0.000
_STATE2 -1.140 0.130 -8.787 0.000
_STATE4 -0.487 0.058 -8.358 0.000
_STATE5 0.083 0.072 1.143 0.253
_STATE6 -0.738 0.049 -14.955 0.000
_STATE8 -0.855 0.060 -14.149 0.000
_STATE9 -0.781 0.068 -11.555 0.000
_STATE10 -0.586 0.112 -5.226 0.000
_STATE11 -1.469 0.154 -9.558 0.000
_STATE12 -0.319 0.051 -6.262 0.000
_STATE13 -0.622 0.057 -10.816 0.000
_STATE15 -0.815 0.130 -6.260 0.000
_STATE16 -1.009 0.083 -12.125 0.000
_STATE17 -0.948 0.054 -17.669 0.000
_STATE18 -0.445 0.059 -7.600 0.000
_STATE19 -1.359 0.068 -19.876 0.000
_STATE20 -1.135 0.072 -15.835 0.000
_STATE21 -0.124 0.064 -1.954 0.051
_STATE22 -0.203 0.068 -2.973 0.003
_STATE23 -0.588 0.088 -6.699 0.000
_STATE24 -0.834 0.064 -13.013 0.000
_STATE25 -0.449 0.058 -7.685 0.000
_STATE26 -0.402 0.056 -7.234 0.000
_STATE27 -1.425 0.061 -23.509 0.000
_STATE28 -0.148 0.079 -1.865 0.062
_STATE29 -0.587 0.060 -9.812 0.000
_STATE30 -1.044 0.099 -10.570 0.000
_STATE31 -1.403 0.081 -17.333 0.000
_STATE32 -0.712 0.073 -9.694 0.000
_STATE33 -0.671 0.088 -7.616 0.000
_STATE34 -0.653 0.059 -11.098 0.000
_STATE35 -0.094 0.080 -1.168 0.243
_STATE36 -0.692 0.052 -13.414 0.000
_STATE37 -0.779 0.056 -13.917 0.000
_STATE38 -1.435 0.115 -12.524 0.000
_STATE39 -0.410 0.054 -7.616 0.000
_STATE40 -0.337 0.068 -4.950 0.000
_STATE41 -0.194 0.064 -3.011 0.003
_STATE42 -0.575 0.053 -10.823 0.000
_STATE44 -0.498 0.099 -5.031 0.000
_STATE45 -0.289 0.065 -4.417 0.000
_STATE46 -1.799 0.109 -16.461 0.000
_STATE47 -0.137 0.060 -2.307 0.021
_STATE48 -0.959 0.050 -19.104 0.000
_STATE49 -0.883 0.071 -12.445 0.000
_STATE50 -0.696 0.118 -5.885 0.000
_STATE51 -0.930 0.058 -16.056 0.000
_STATE53 -0.589 0.058 -10.148 0.000
_STATE54 0.343 0.079 4.313 0.000
_STATE55 -0.908 0.060 -15.220 0.000
_STATE56 -1.015 0.126 -8.080 0.000
_STATE66 -0.843 0.595 -1.416 0.157
IMONTH2 0.074 0.028 2.603 0.009
IMONTH3 0.110 0.028 3.932 0.000
IMONTH4 0.171 0.030 5.651 0.000
IMONTH5 0.329 0.029 11.169 0.000
IMONTH6 0.234 0.028 8.220 0.000
IMONTH7 0.145 0.028 5.128 0.000
IMONTH8 0.262 0.028 9.320 0.000
IMONTH9 0.382 0.029 13.259 0.000
IMONTH10 0.310 0.030 10.417 0.000
IMONTH11 0.269 0.030 9.082 0.000
IMONTH12 0.377 0.029 13.190 0.000
X_race8 -1.150 0.016 -74.087 0.000
difference_HS30 0.701 0.153 4.576 0.000
X_race8:difference_HS30 -0.967 0.286 -3.378 0.001

Tweets becoming less hateful is associated with a decrease of bad mental health days for Latinos relative to whites.

Note: the interpretation of this one is a little confusing, I am interpreting as if month three has a score of 4 and then month two has a score of 3, this would be a positive score where things are improving. If month three has a score of 3 and month 2 a score of 3 then this would be a negative score where things are getting worse. Thus i am interpreting the negative interaction to refer to when scores are getting better then bad mental health days are lower. Happy to discuss this interpretation

Rolling average standard deviations from the mean

Regression results examining rolling average standard deviation and mental health - Interaction
term estimate std.error statistic p.value
(Intercept) 3.758 0.113 33.392 0.000
X_age_g -0.565 0.003 -173.424 0.000
X_educag -0.714 0.005 -136.403 0.000
female 1.349 0.011 125.041 0.000
_STATE2 -1.139 0.130 -8.784 0.000
_STATE4 -0.499 0.058 -8.557 0.000
_STATE5 0.071 0.072 0.981 0.326
_STATE6 -0.740 0.049 -14.991 0.000
_STATE8 -0.868 0.060 -14.377 0.000
_STATE9 -0.785 0.068 -11.619 0.000
_STATE10 -0.593 0.112 -5.290 0.000
_STATE11 -1.462 0.154 -9.514 0.000
_STATE12 -0.323 0.051 -6.332 0.000
_STATE13 -0.627 0.057 -10.916 0.000
_STATE15 -0.806 0.130 -6.192 0.000
_STATE16 -1.025 0.083 -12.313 0.000
_STATE17 -0.952 0.054 -17.762 0.000
_STATE18 -0.447 0.059 -7.628 0.000
_STATE19 -1.358 0.068 -19.873 0.000
_STATE20 -1.141 0.072 -15.925 0.000
_STATE21 -0.121 0.064 -1.898 0.058
_STATE22 -0.206 0.068 -3.026 0.002
_STATE23 -0.594 0.088 -6.773 0.000
_STATE24 -0.833 0.064 -12.994 0.000
_STATE25 -0.449 0.058 -7.695 0.000
_STATE26 -0.403 0.056 -7.260 0.000
_STATE27 -1.432 0.061 -23.638 0.000
_STATE28 -0.142 0.079 -1.792 0.073
_STATE29 -0.591 0.060 -9.892 0.000
_STATE30 -1.057 0.099 -10.707 0.000
_STATE31 -1.407 0.081 -17.396 0.000
_STATE32 -0.718 0.073 -9.780 0.000
_STATE33 -0.681 0.088 -7.728 0.000
_STATE34 -0.609 0.059 -10.354 0.000
_STATE35 -0.090 0.080 -1.130 0.258
_STATE36 -0.691 0.052 -13.394 0.000
_STATE37 -0.784 0.056 -14.019 0.000
_STATE38 -1.430 0.114 -12.491 0.000
_STATE39 -0.411 0.054 -7.628 0.000
_STATE40 -0.345 0.068 -5.070 0.000
_STATE41 -0.200 0.064 -3.102 0.002
_STATE42 -0.565 0.053 -10.652 0.000
_STATE44 -0.498 0.099 -5.035 0.000
_STATE45 -0.297 0.065 -4.540 0.000
_STATE46 -1.803 0.109 -16.503 0.000
_STATE47 -0.135 0.060 -2.272 0.023
_STATE48 -0.964 0.050 -19.201 0.000
_STATE49 -0.894 0.071 -12.605 0.000
_STATE50 -0.701 0.118 -5.927 0.000
_STATE51 -0.925 0.058 -15.983 0.000
_STATE53 -0.593 0.058 -10.221 0.000
_STATE54 0.344 0.079 4.336 0.000
_STATE55 -0.912 0.060 -15.288 0.000
_STATE56 -1.011 0.126 -8.052 0.000
_STATE66 -0.844 0.595 -1.418 0.156
IMONTH2 0.099 0.028 3.497 0.000
IMONTH3 0.141 0.028 5.022 0.000
IMONTH4 0.192 0.029 6.689 0.000
IMONTH5 0.322 0.028 11.356 0.000
IMONTH6 0.232 0.028 8.358 0.000
IMONTH7 0.141 0.028 5.042 0.000
IMONTH8 0.257 0.028 9.128 0.000
IMONTH9 0.353 0.028 12.403 0.000
IMONTH10 0.205 0.028 7.252 0.000
IMONTH11 0.145 0.028 5.096 0.000
IMONTH12 0.218 0.029 7.555 0.000
X_race8 0.918 0.220 4.173 0.000
rolling_sd_HS90 3.072 0.073 41.826 0.000
X_race8:rolling_sd_HS90 -1.527 0.161 -9.494 0.000

A one unit increase in the standard deviation is associated with 3.07 increased bad mental health days for Whites and 1.54 for Latinos.

Percent negative

Regression results examining rolling percent negative and mental health - Interaction
term estimate std.error statistic p.value
(Intercept) 7.857 0.061 129.224 0.000
X_age_g -0.563 0.003 -172.783 0.000
X_educag -0.710 0.005 -135.669 0.000
female 1.349 0.011 124.911 0.000
_STATE2 -1.139 0.130 -8.781 0.000
_STATE4 -0.488 0.058 -8.371 0.000
_STATE5 0.084 0.072 1.165 0.244
_STATE6 -0.740 0.049 -14.983 0.000
_STATE8 -0.855 0.060 -14.151 0.000
_STATE9 -0.783 0.068 -11.577 0.000
_STATE10 -0.585 0.112 -5.220 0.000
_STATE11 -1.470 0.154 -9.563 0.000
_STATE12 -0.319 0.051 -6.255 0.000
_STATE13 -0.616 0.057 -10.713 0.000
_STATE15 -0.814 0.130 -6.252 0.000
_STATE16 -1.008 0.083 -12.114 0.000
_STATE17 -0.948 0.054 -17.667 0.000
_STATE18 -0.447 0.059 -7.623 0.000
_STATE19 -1.358 0.068 -19.862 0.000
_STATE20 -1.137 0.072 -15.851 0.000
_STATE21 -0.125 0.064 -1.958 0.050
_STATE22 -0.202 0.068 -2.966 0.003
_STATE23 -0.585 0.088 -6.671 0.000
_STATE24 -0.833 0.064 -12.989 0.000
_STATE25 -0.449 0.058 -7.686 0.000
_STATE26 -0.402 0.056 -7.237 0.000
_STATE27 -1.424 0.061 -23.483 0.000
_STATE28 -0.145 0.079 -1.830 0.067
_STATE29 -0.588 0.060 -9.836 0.000
_STATE30 -1.046 0.099 -10.591 0.000
_STATE31 -1.402 0.081 -17.327 0.000
_STATE32 -0.710 0.073 -9.666 0.000
_STATE33 -0.669 0.088 -7.595 0.000
_STATE34 -0.657 0.059 -11.164 0.000
_STATE35 -0.093 0.080 -1.159 0.246
_STATE36 -0.691 0.052 -13.387 0.000
_STATE37 -0.778 0.056 -13.914 0.000
_STATE38 -1.433 0.115 -12.511 0.000
_STATE39 -0.409 0.054 -7.592 0.000
_STATE40 -0.337 0.068 -4.952 0.000
_STATE41 -0.195 0.064 -3.026 0.002
_STATE42 -0.573 0.053 -10.790 0.000
_STATE44 -0.496 0.099 -5.019 0.000
_STATE45 -0.289 0.065 -4.419 0.000
_STATE46 -1.800 0.109 -16.469 0.000
_STATE47 -0.136 0.060 -2.286 0.022
_STATE48 -0.960 0.050 -19.110 0.000
_STATE49 -0.884 0.071 -12.459 0.000
_STATE50 -0.695 0.118 -5.880 0.000
_STATE51 -0.927 0.058 -16.012 0.000
_STATE53 -0.588 0.058 -10.135 0.000
_STATE54 0.343 0.079 4.320 0.000
_STATE55 -0.909 0.060 -15.237 0.000
_STATE56 -1.014 0.126 -8.072 0.000
_STATE66 -0.841 0.595 -1.413 0.158
IMONTH2 0.078 0.029 2.719 0.007
IMONTH3 0.119 0.028 4.191 0.000
IMONTH4 0.211 0.029 7.218 0.000
IMONTH5 0.367 0.029 12.664 0.000
IMONTH6 0.266 0.028 9.354 0.000
IMONTH7 0.167 0.029 5.795 0.000
IMONTH8 0.270 0.029 9.339 0.000
IMONTH9 0.411 0.029 14.041 0.000
IMONTH10 0.356 0.029 12.343 0.000
IMONTH11 0.310 0.029 10.833 0.000
IMONTH12 0.386 0.029 13.491 0.000
X_race8 -1.250 0.049 -25.275 0.000
percent_negative_HS90 0.017 0.033 0.532 0.595
X_race8:percent_negative_HS90 0.136 0.066 2.052 0.040

For every one unit increase in percent negative Latinos have .136 increase in bad days mental health. No significance for whites

Changepoint Creation and Validation

## 
## INFO: To supress printing the parameers in beast(),      set print.options = 0 
## INFO: To supress printing the parameers in beast.irreg(),set print.options = 0 
## INFO: To supress printing the parameers in beast123(),   set extra$printOptions = 0  
## INFO: To supress warning messages in beast(),            set quiet = 1 
## INFO: To supress warning messages in beast.irreg(),      set quiet = 1 
## INFO: To supress warning messages in beast123(),         set extra$quiet = 1  
## 
## #--------------------------------------------------#
## #       Brief summary of Input Data                #
## #--------------------------------------------------#
## Data Dimension: One signal of length 75
## IsOrdered     : Yes, ordered in time
## IsRegular     : Yes, evenly spaced at interval of  1 (unknown unit)
## hasSeasonCmpnt: FALSE | no periodic or seasonal component. The model Y=Trend+Error is fitted.
## HasOutlierCmpt: FALSE | If true, Y=Trend+Outlier+Error (experimental) is fitted instead of Y=Trend+Error 
## Detrend       : FALSE | If true, remove a global trend component before running BEAST & add it back after BEAST
## MissingValue  : NaN  flagged as missing values 
## MaxMissingRate: if more than 75% of data is missing, BEAST will skip it.
## 
## 
## #--------------------------------------------------#
## #      OPTIONS used in the MCMC inference          #
## #--------------------------------------------------#
## 
## #......Start of displaying 'MetaData' ......
## metadata                = list()     # metadata is used to interpret the input data Y
## metadata$season         = 'none'     # trend-only data with no periodic variation
## metadata$startTime      = 1          # unknown unit
## metadata$deltaTime      = 1          # unknown unit
## metadata$maxMissingRate = 0.75       # if more than 75% of data is missing, BEAST will skip it.
## metadata$detrend        = FALSE      # if true,remove a global trend  cmpnt before running BEAST & add it back later
## #........End of displaying MetaData ........
## 
## #......Start of displaying 'prior' ......
## prior                   = list()     # prior is the true model parameters of BEAST
## prior$trendMinOrder     = 0          # torder.minmax[1]: min trend polynomial order alllowed
## prior$trendMaxOrder     = 1          # torder.minmax[2]: max trend polynomial order alllowed
## prior$trendMinKnotNum   = 0          # tcp.minmax[1]   : min num of chngpts in trend allowed
## prior$trendMaxKnotNum   = 10         # tcp.minmax[2]   : max num of chngpts in trend allowed
## prior$trendMinSepDist   = 3          # tseg.min        : min trend segment length in terms of datapoints
## prior$trendLeftMargin   = 3          # tseg.leftmargin : no trend chngpts in the first 3 datapoints
## prior$trendRightMargin  = 3          # tseg.rightmargin: no trend chngpts in the last 3 datapoints
## prior$K_MAX             = 22         # max number of terms in general linear model (relevant only at small values)
## prior$precValue         = 1.5        # useful mainly when precPriorType='constant'
## prior$modelPriorType    = 1         
## prior$precPriorType     = 'uniform'
## #......End of displaying prior ......
## 
## #......Start of displaying 'mcmc' ......
## mcmc                           = list()     # mcmc is not BEAST parameters but MCMC sampler options
## mcmc$seed                      = 0          # A nonzero seed to replicate among runs
## mcmc$samples                   = 8000       # Number of samples saved per chain: the larger, the better
## mcmc$thinningFactor            = 5          # Thinning the chain: the larger, the better 
## mcmc$burnin                    = 200        # Number of initial samples discarded: the larger, the better
## mcmc$chainNumber               = 3          # Number of chains: the larger, the better
## mcmc$maxMoveStepSize           = 4          # Max step of jumping from current changepoint: No need to change
## mcmc$trendResamplingOrderProb  = 0.1        # Proposal probability of sampling trend polynominal order 
## mcmc$credIntervalAlphaLevel    = 0.95       # The alphal level for Credible Intervals
## # Total number of models randomly visited in BEAST is (burnin+sampples*thinFactor)*chainNumber=120600
## #......End of displaying mcmc ......
## 
## #......Start of displaying 'extra' ......
## extra                      = list() # extra is used to configure output/computing options
## extra$dumpInputData        = TRUE  # if true, dump a copy of the input data as o$data 
## extra$whichOutputDimIsTime = 1     # 1,2 or 3; which dim of the result is time; used for a 2D/3D input Y
## extra$computeCredible      = FALSE # if true, compute  credibiel interval of estimated Y (e.g., o$trend$CI)
## extra$fastCIComputation    = TRUE  # if true, do not sort but approximiate CI 
## extra$computeTrendOrder    = TRUE  # if true, dump the estimated trend polynomial order 
## extra$computeTrendChngpt   = TRUE  # if true, dump the trend changepoints (tcp) in the output 
## extra$computeTrendSlope    = TRUE  # if true, dump the time-varying slope in trend
## extra$tallyPosNegTrendJump = TRUE  # differentiate postive/negative jumps at tcp
## extra$tallyIncDecTrendJump = TRUE  # differentiate increased/decreased slopes at tcp
## extra$printProgressBar     = TRUE  # if true, show an ascii progressbar
## extra$printOptions         = TRUE  # if true, print the option of the BEAST run
## extra$consoleWidth         = 80    # an integer specifying the console width for printing
## extra$numThreadsPerCPU     = 2     # each cpu core spawns 2 concurrent threads (for beast123())
## extra$numParThreads        = 0     # total number of threads (for beast123() only)
## #......End of displaying extra ......
## 
## -Progress:  0.0% done[>********************************************************]\Progress:  4.2% done[==>******************************************************]|Progress:  8.3% done[=====>***************************************************]/Progress: 12.5% done[=======>*************************************************]-Progress: 16.7% done[==========>**********************************************]\Progress: 20.8% done[============>********************************************]|Progress: 25.0% done[==============>******************************************]/Progress: 29.2% done[=================>***************************************]-Progress: 33.3% done[===================>*************************************]\Progress: 37.5% done[=====================>***********************************]|Progress: 41.7% done[========================>********************************]/Progress: 45.8% done[==========================>******************************]-Progress: 50.0% done[=============================>***************************]\Progress: 54.2% done[===============================>*************************]|Progress: 58.3% done[=================================>***********************]/Progress: 62.5% done[====================================>********************]-Progress: 66.7% done[======================================>******************]\Progress: 70.8% done[========================================>****************]|Progress: 75.0% done[===========================================>*************]/Progress: 79.2% done[=============================================>***********]-Progress: 83.3% done[================================================>********]\Progress: 87.5% done[==================================================>******]|Progress: 91.7% done[====================================================>****]/Progress: 95.8% done[=======================================================>*]-Progress:100.0% done[=========================================================]

## #####################################################################
## #                      Seasonal  Changepoints                       #
## #####################################################################
##  No seasonal/periodic component present (i.e., season='none')
## 
## 
## #####################################################################
## #                      Trend  Changepoints                          #
## #####################################################################
## .-------------------------------------------------------------------.
## | Ascii plot of probability distribution for number of chgpts (ncp) |
## .-------------------------------------------------------------------.
## |Pr(ncp = 0 )=0.000|*                                               |
## |Pr(ncp = 1 )=0.000|*                                               |
## |Pr(ncp = 2 )=0.085|*************                                   |
## |Pr(ncp = 3 )=0.228|**********************************              |
## |Pr(ncp = 4 )=0.322|*********************************************** |
## |Pr(ncp = 5 )=0.220|*********************************               |
## |Pr(ncp = 6 )=0.099|***************                                 |
## |Pr(ncp = 7 )=0.034|*****                                           |
## |Pr(ncp = 8 )=0.009|**                                              |
## |Pr(ncp = 9 )=0.002|*                                               |
## |Pr(ncp = 10)=0.000|*                                               |
## .-------------------------------------------------------------------.
## |    Summary for number of Trend ChangePoints (tcp)                 |
## .-------------------------------------------------------------------.
## |ncp_max    = 10   | MaxTrendKnotNum: A parameter you set           |
## |ncp_mode   = 4    | Pr(ncp= 4)=0.32: There is a 32.2% probability  |
## |                  | that the trend component has  4 changepoint(s).|
## |ncp_mean   = 4.17 | Sum{ncp*Pr(ncp)} for ncp = 0,...,10            |
## |ncp_pct10  = 3.00 | 10% percentile for number of changepoints      |
## |ncp_median = 4.00 | 50% percentile: Median number of changepoints  |
## |ncp_pct90  = 6.00 | 90% percentile for number of changepoints      |
## .-------------------------------------------------------------------.
## | List of probable trend changepoints ranked by probability of      |
## | occurrence: Please combine the ncp reported above to determine    |
## | which changepoints below are  practically meaningful              |
## '-------------------------------------------------------------------'
## |tcp#              |time (cp)                  |prob(cpPr)          |
## |------------------|---------------------------|--------------------|
## |1                 |50.000000                  |0.99625             |
## |2                 |71.000000                  |0.96696             |
## |3                 |60.000000                  |0.88163             |
## |4                 |65.000000                  |0.48013             |
## |5                 |39.000000                  |0.18246             |
## |6                 |56.000000                  |0.15979             |
## |7                 |7.000000                   |0.09750             |
## |8                 |19.000000                  |0.03000             |
## .-------------------------------------------------------------------.
## 
## 
## 
## NOTE: the beast output object 'o' is a LIST. Type 'str(o)' to see all 
## the elements in it. Or use 'plot(o)' or 'plot(o,interactive=TRUE)' to 
## plot the model output.

Monthly changepoints are all post Trump presidency and related to political events. Final decrease of hate speech takes place after impeachment inquiry begins for Trump.

[1] “2018-01-01” - Federal government shut down threatened if border wall is not funded. https://en.wikipedia.org/wiki/Category:January_2018_events_in_the_United_States

[2] “2018-11-01” - Midterm election. https://en.wikipedia.org/wiki/Category:November_2018_events_in_the_United_States

[3] “2019-03-01” - Trump speech at 2019 Conservative Political Action Conference. https://en.wikipedia.org/wiki/Timeline_of_the_Donald_Trump_presidency_(2019_Q1)

[4] “2019-09-01” - Impeachment inquiry into Donald Trump. https://en.wikipedia.org/wiki/Category:September_2019_events_in_the_United_States

Difference-in-Differences Tests - one changepoint

90 Days Pre- and Post-

Difference-in-differences for days bad mental health - 90 days pre and post changepoint
term estimate std.error statistic p.value
(Intercept) 7.807 0.080 97.838 0.000
X_age_g -0.585 0.012 -50.800 0.000
X_educag -0.766 0.019 -40.281 0.000
female 1.479 0.038 38.786 0.000
treatment -1.580 0.067 -23.630 0.000
post -0.189 0.043 -4.431 0.000
interaction 0.276 0.096 2.890 0.004

Treatment (-1.57953): This indicates that Latinos (the “treatment group”) report 1.58 fewer poor mental health days compared to Whites (the “control group”) before January 1, 2018, holding other variables constant.

Post (-0.18872): Individuals in the post-intervention period (after January 1, 2018) report about 0.19 fewer poor mental health days compared to those in the pre-intervention period, holding other variables constant.

Interaction (0.27611): This interaction term captures the additional effect of being both Latino and in the post-intervention period. Specifically, Latinos in the post-intervention period report 0.087 more poor mental health days than what would be expected from the separate effects of being Latino and being in the post-intervention period alone. This suggests that, while Latinos generally report fewer poor mental health days (as seen from the treatment coefficient), their mental health worsens slightly after January 1, 2018, relative to Whites, when considering the interaction. Although it seems like a small effect, the Lancet paper found that police killings of black americans led to an excess of 0.14 bad mental health days aka a relative increase of 3.3%. .087 days increase for Latinos is a 2.36% increase relative to weighted average bad days mental health.

Difference-in-differences for frequent days bad mental health - 90 days pre and post changepoint
term estimate std.error statistic p.value
(Intercept) 0.259 0.003 80.319 0.000
X_age_g -0.018 0.000 -38.422 0.000
female 0.049 0.002 32.041 0.000
X_educag -0.031 0.001 -40.098 0.000
treatment -0.053 0.003 -19.778 0.000
post -0.006 0.002 -3.416 0.001
interaction 0.011 0.004 2.973 0.003

Treatment (-0.0534540): Latinos (the treatment group) report 5.35 percentage points fewer frequent bad mental health days compared to Whites (the control group), before January 1, 2018. This suggests that Latinos tend to have better mental health outcomes than Whites in terms of frequent bad mental health days.

Post (-0.0058823): In the post-intervention period (after January 1, 2018), there is a decrease of 0.59 percentage points in frequent bad mental health days, holding other factors constant. This suggests a small overall improvement in mental health during the post-intervention period.

Interaction (0.0114819): The interaction term indicates that Latinos in the post-intervention period experience an increase of 0.5 (0.11-0.06) percentage points in frequent bad mental health days compared to what would be expected from the additive effects of being Latino and being in the post-intervention period alone. This suggests that although Latinos generally report fewer frequent bad mental health days, this advantage decreases slightly in the post-intervention period compared to Whites.

Difference-in-differences for any days bad mental health - 90 days pre and post changepoint
term estimate std.error statistic p.value
(Intercept) 0.576 0.005 123.670 0.000
X_age_g -0.062 0.001 -92.374 0.000
female 0.115 0.002 51.514 0.000
X_educag -0.008 0.001 -7.108 0.000
treatment -0.091 0.004 -23.335 0.000
post -0.016 0.002 -6.472 0.000
interaction -0.004 0.006 -0.802 0.422

Treatment (-0.0910991): Latinos (treatment group) report 9.11 percentage points fewer bad mental health days compared to Whites (control group), before January 1, 2018. This implies that Latinos tend to report better mental health (in terms of fewer bad mental health days) compared to Whites

Post (-0.0160990): Being in the post-intervention period (after January 1, 2018) is associated with a small decrease of 1.61 percentage points in any bad mental health days, suggesting an improvement in mental health after the intervention.

Interaction (-0.0044754): The interaction term between treatment (Latinos) and post (post-intervention period) is not statistically significant (p-value = 0.422). The estimated coefficient is small and negative (-0.0044754), indicating no meaningful change in the mental health of Latinos relative to Whites in the post-intervention period.

Difference-in-Differences Tests - Multiple Changepoints

Difference-in-differences for days bad mental health - multiple changepoints
term estimate std.error statistic p.value conf.low conf.high
(Intercept) 7.370 0.022 339.995 0.000 7.328 7.413
X_age_g -0.561 0.003 -172.657 0.000 -0.567 -0.554
X_educag -0.730 0.005 -140.324 0.000 -0.740 -0.720
female 1.353 0.011 125.236 0.000 1.331 1.374
treatment -1.176 0.017 -68.165 0.000 -1.210 -1.142
post_1 0.283 0.018 15.844 0.000 0.248 0.318
post_2 0.438 0.027 16.106 0.000 0.384 0.491
post_3 0.697 0.022 32.236 0.000 0.655 0.739
post_4 0.800 0.027 29.218 0.000 0.747 0.854
interaction_1 -0.175 0.039 -4.433 0.000 -0.252 -0.098
interaction_2 0.089 0.061 1.457 0.145 -0.031 0.210
interaction_3 -0.298 0.047 -6.377 0.000 -0.389 -0.206
interaction_4 -0.248 0.060 -4.121 0.000 -0.366 -0.130

In this model we see that in periods 1, 3, and 4 both Whites and Latinos have worse mental health post treatment, but that Whites have a bigger effect. In period 2 we see Latinos have a bigger effect than whites, but still a mental health benefit.

Difference-in-differences for frequent days bad mental health - multiple changepoints
term estimate std.error statistic p.value conf.low conf.high
(Intercept) 0.244 0.001 277.610 0.000 0.243 0.246
X_age_g -0.017 0.000 -128.604 0.000 -0.017 -0.017
X_educag -0.029 0.000 -138.389 0.000 -0.030 -0.029
female 0.044 0.000 100.771 0.000 0.043 0.045
treatment -0.039 0.001 -56.310 0.000 -0.041 -0.038
post_1 0.010 0.001 13.155 0.000 0.008 0.011
post_2 0.016 0.001 14.269 0.000 0.014 0.018
post_3 0.024 0.001 27.669 0.000 0.023 0.026
post_4 0.027 0.001 24.629 0.000 0.025 0.030
interaction_1 -0.005 0.002 -3.357 0.001 -0.009 -0.002
interaction_2 0.003 0.002 1.098 0.272 -0.002 0.008
interaction_3 -0.011 0.002 -5.760 0.000 -0.015 -0.007
interaction_4 -0.009 0.002 -3.878 0.000 -0.014 -0.005

In this model we see that in periods 1, 3, and 4 both Whites and Latinos have worse mental health post treatment, but that Whites have a bigger effect. In period 2 we see Latinos have a bigger effect than whites, but still a mental health benefit.

Difference-in-differences for any days bad mental health - multiple changepoints
term estimate std.error statistic p.value conf.low conf.high
(Intercept) 0.553 0.001 436.019 0.000 0.550 0.555
X_age_g -0.060 0.000 -314.712 0.000 -0.060 -0.059
X_educag -0.010 0.000 -32.245 0.000 -0.010 -0.009
female 0.111 0.001 175.830 0.000 0.110 0.112
treatment -0.070 0.001 -69.254 0.000 -0.072 -0.068
post_1 0.019 0.001 17.955 0.000 0.017 0.021
post_2 0.028 0.002 17.382 0.000 0.025 0.031
post_3 0.048 0.001 38.291 0.000 0.046 0.051
post_4 0.060 0.002 37.284 0.000 0.057 0.063
interaction_1 -0.021 0.002 -8.989 0.000 -0.025 -0.016
interaction_2 -0.007 0.004 -2.047 0.041 -0.014 0.000
interaction_3 -0.036 0.003 -13.096 0.000 -0.041 -0.030
interaction_4 -0.034 0.004 -9.683 0.000 -0.041 -0.027

In period one, White mental health worsens but Latino mental health improves. In periods 2, 3, and 4 both white and latino mental health worsen but the effect is larger for Latinos.