--------------------------------------------------------------------------------------------------------------- name: log: D:\Stephens-Dougan 2022 NSE2.log log type: text opened on: 6 Dec 2022, 20:42:22 . do "C:\Users\lzigere\AppData\Local\Temp\STD2b40_000000.tmp" . * I already ran the code from Stephens-Dougan2022 to get the variables. . . * This next line creates an "nse2" variable that is the . * negative stereotype endorsement variable without the 0.5 observations: . . recode negstereotype_endorsement (0.5=.), gen(nse2) (94 differences between negstereotype_endorsement and nse2) . tab negstereotype_endorsement nse2, mi | RECODE of negative | negstereotype_endorsement stereotype | (negative stereotype endorsemen | endorsement) t | 0 1 . | Total -----------+---------------------------------+---------- 0 | 435 0 0 | 435 .5 | 0 0 94 | 94 1 | 0 62 0 | 62 -----------+---------------------------------+---------- Total | 435 62 94 | 591 . . * The next lines calculate the weighed treatment effect for the 0s and the 1s . * and then calculates the difference between the treatment effects: . . reg Follow_Blacks_dichotomous Treatment if negstereotype_endorsement==0 [pw=WEIGHT] (sum of wgt is 438.035785092) Linear regression Number of obs = 431 F(1, 429) = 1.89 Prob > F = 0.1701 R-squared = 0.0083 Root MSE = .48451 ------------------------------------------------------------------------------ | Robust Follow_Bl~us | Coefficient std. err. t P>|t| [95% conf. interval] -------------+---------------------------------------------------------------- Treatment | -.0888811 .0646736 -1.37 0.170 -.2159977 .0382355 _cons | .667467 .0463193 14.41 0.000 .5764259 .7585081 ------------------------------------------------------------------------------ . reg Follow_Blacks_dichotomous Treatment if negstereotype_endorsement==1 [pw=WEIGHT] (sum of wgt is 58.264915095) Linear regression Number of obs = 62 F(1, 60) = 2.49 Prob > F = 0.1201 R-squared = 0.0570 Root MSE = .34224 ------------------------------------------------------------------------------ | Robust Follow_Bl~us | Coefficient std. err. t P>|t| [95% conf. interval] -------------+---------------------------------------------------------------- Treatment | .1678772 .1064693 1.58 0.120 -.0450932 .3808475 _cons | .7625933 .0914077 8.34 0.000 .5797506 .945436 ------------------------------------------------------------------------------ . di -.0888811 - .1678772 -.2567583 . . * That -.2567583 difference appears in this OLS regression: . . reg Follow_Blacks_dichotomous i.Treatment##c.nse2 [pw=WEIGHT] (sum of wgt is 496.300700187) Linear regression Number of obs = 493 F(3, 489) = 8.78 Prob > F = 0.0000 R-squared = 0.0374 Root MSE = .47046 ---------------------------------------------------------------------------------- | Robust Follow_Blacks_~s | Coefficient std. err. t P>|t| [95% conf. interval] -----------------+---------------------------------------------------------------- 1.Treatment | -.0888811 .0647868 -1.37 0.171 -.2161759 .0384137 nse2 | .0951263 .1015135 0.94 0.349 -.1043301 .2945827 | Treatment#c.nse2 | 1 | .2567583 .1235197 2.08 0.038 .0140635 .4994531 | _cons | .667467 .0464004 14.38 0.000 .5762983 .7586357 ---------------------------------------------------------------------------------- . . * But the -.2567583 difference does not appear if . * using the 3-level negative stereotype endorsement variable: . . reg Follow_Blacks_dichotomous i.Treatment##c.negstereotype_endorsement [pw=WEIGHT] (sum of wgt is 588.0094088770001) Linear regression Number of obs = 586 F(3, 582) = 6.92 Prob > F = 0.0001 R-squared = 0.0354 Root MSE = .46681 ------------------------------------------------------------------------------------------------------- | Robust Follow_Blacks_dichotomous | Coefficient std. err. t P>|t| [95% conf. interval] --------------------------------------+---------------------------------------------------------------- 1.Treatment | -.0915267 .0632161 -1.45 0.148 -.2156861 .0326327 negstereotype_endorsement | .1014544 .1022733 0.99 0.322 -.0994153 .3023241 | Treatment#c.negstereotype_endorsement | 1 | .2398664 .1279354 1.87 0.061 -.011405 .4911378 | _cons | .6683277 .0451983 14.79 0.000 .5795559 .7570994 ------------------------------------------------------------------------------------------------------- . . * Below are comparisons for the four APSR letter outcomes, using logit: . . logit facemasks_dichotomous i.Treatment##c.negstereotype_endorsement [pw=WEIGHT], or Iteration 0: log pseudolikelihood = -403.19987 Iteration 1: log pseudolikelihood = -392.77238 Iteration 2: log pseudolikelihood = -392.75884 Iteration 3: log pseudolikelihood = -392.75883 Logistic regression Number of obs = 590 Wald chi2(3) = 13.38 Prob > chi2 = 0.0039 Log pseudolikelihood = -392.75883 Pseudo R2 = 0.0259 ------------------------------------------------------------------------------------------------------- | Robust facemasks_dichotomous | Odds ratio std. err. z P>|z| [95% conf. interval] --------------------------------------+---------------------------------------------------------------- 1.Treatment | .8681719 .2378766 -0.52 0.606 .5074333 1.485363 negstereotype_endorsement | 1.265442 .6293818 0.47 0.636 .4774057 3.354261 | Treatment#c.negstereotype_endorsement | 1 | 3.953402 2.650938 2.05 0.040 1.062187 14.71434 | _cons | .6854616 .1397093 -1.85 0.064 .4597196 1.022053 ------------------------------------------------------------------------------------------------------- Note: _cons estimates baseline odds. . logit facemasks_dichotomous i.Treatment##c.nse2 [pw=WEIGHT], or Iteration 0: log pseudolikelihood = -337.3936 Iteration 1: log pseudolikelihood = -328.40005 Iteration 2: log pseudolikelihood = -328.38568 Iteration 3: log pseudolikelihood = -328.38567 Logistic regression Number of obs = 496 Wald chi2(3) = 11.79 Prob > chi2 = 0.0081 Log pseudolikelihood = -328.38567 Pseudo R2 = 0.0267 --------------------------------------------------------------------------------------- | Robust facemasks_dichotomous | Odds ratio std. err. z P>|z| [95% conf. interval] ----------------------+---------------------------------------------------------------- 1.Treatment | .8929154 .250653 -0.40 0.687 .5150708 1.547938 nse2 | 1.065749 .5881876 0.12 0.908 .3613069 3.143646 | Treatment#c.nse2 | 1 | 4.84813 3.55596 2.15 0.031 1.151445 20.41293 | _cons | .6692596 .1404751 -1.91 0.056 .4435378 1.009854 --------------------------------------------------------------------------------------- Note: _cons estimates baseline odds. . . logit individualrights_dichotomous i.Treatment##c.negstereotype_endorsement [pw=WEIGHT], or Iteration 0: log pseudolikelihood = -351.63813 Iteration 1: log pseudolikelihood = -342.62242 Iteration 2: log pseudolikelihood = -342.48513 Iteration 3: log pseudolikelihood = -342.48513 Logistic regression Number of obs = 587 Wald chi2(3) = 9.83 Prob > chi2 = 0.0200 Log pseudolikelihood = -342.48513 Pseudo R2 = 0.0260 ------------------------------------------------------------------------------------------------------- | Robust individualrights_dichotomous | Odds ratio std. err. z P>|z| [95% conf. interval] --------------------------------------+---------------------------------------------------------------- 1.Treatment | .5993562 .1895615 -1.62 0.106 .3224565 1.114035 negstereotype_endorsement | .8222517 .4340863 -0.37 0.711 .2921697 2.314059 | Treatment#c.negstereotype_endorsement | 1 | 5.506455 3.955326 2.37 0.018 1.347249 22.50589 | _cons | .4505659 .1013486 -3.54 0.000 .2899294 .7002035 ------------------------------------------------------------------------------------------------------- Note: _cons estimates baseline odds. . logit individualrights_dichotomous i.Treatment##c.nse2 [pw=WEIGHT], or Iteration 0: log pseudolikelihood = -293.86869 Iteration 1: log pseudolikelihood = -285.00219 Iteration 2: log pseudolikelihood = -284.80498 Iteration 3: log pseudolikelihood = -284.80488 Iteration 4: log pseudolikelihood = -284.80488 Logistic regression Number of obs = 494 Wald chi2(3) = 9.74 Prob > chi2 = 0.0209 Log pseudolikelihood = -284.80488 Pseudo R2 = 0.0308 ---------------------------------------------------------------------------------------------- | Robust individualrights_dichotomous | Odds ratio std. err. z P>|z| [95% conf. interval] -----------------------------+---------------------------------------------------------------- 1.Treatment | .6136675 .198757 -1.51 0.132 .325269 1.157773 nse2 | .7757365 .4510568 -0.44 0.662 .2481863 2.424659 | Treatment#c.nse2 | 1 | 6.219409 4.835023 2.35 0.019 1.355244 28.54175 | _cons | .4473843 .1036839 -3.47 0.001 .2840599 .7046145 ---------------------------------------------------------------------------------------------- Note: _cons estimates baseline odds. . . logit visitparks_dichotomous i.Treatment##c.negstereotype_endorsement [pw=WEIGHT], or Iteration 0: log pseudolikelihood = -308.12978 Iteration 1: log pseudolikelihood = -305.90783 Iteration 2: log pseudolikelihood = -305.86799 Iteration 3: log pseudolikelihood = -305.86798 Logistic regression Number of obs = 589 Wald chi2(3) = 2.27 Prob > chi2 = 0.5175 Log pseudolikelihood = -305.86798 Pseudo R2 = 0.0073 ------------------------------------------------------------------------------------------------------- | Robust visitparks_dichotomous | Odds ratio std. err. z P>|z| [95% conf. interval] --------------------------------------+---------------------------------------------------------------- 1.Treatment | .8534378 .2742785 -0.49 0.622 .4545819 1.602255 negstereotype_endorsement | .3886906 .2925473 -1.26 0.209 .0889096 1.699259 | Treatment#c.negstereotype_endorsement | 1 | 3.870488 3.548643 1.48 0.140 .641719 23.34461 | _cons | .3161425 .0779066 -4.67 0.000 .1950394 .5124404 ------------------------------------------------------------------------------------------------------- Note: _cons estimates baseline odds. . logit visitparks_dichotomous i.Treatment##c.nse2 [pw=WEIGHT], or Iteration 0: log pseudolikelihood = -262.50014 Iteration 1: log pseudolikelihood = -261.62814 Iteration 2: log pseudolikelihood = -261.6231 Iteration 3: log pseudolikelihood = -261.6231 Logistic regression Number of obs = 495 Wald chi2(3) = 0.91 Prob > chi2 = 0.8226 Log pseudolikelihood = -261.6231 Pseudo R2 = 0.0033 ---------------------------------------------------------------------------------------- | Robust visitparks_dichotomous | Odds ratio std. err. z P>|z| [95% conf. interval] -----------------------+---------------------------------------------------------------- 1.Treatment | .8118372 .2643086 -0.64 0.522 .4288887 1.536715 nse2 | .5768814 .4517466 -0.70 0.482 .1243152 2.677003 | Treatment#c.nse2 | 1 | 2.368148 2.317873 0.88 0.378 .3477584 16.12649 | _cons | .3256969 .0804168 -4.54 0.000 .2007457 .5284222 ---------------------------------------------------------------------------------------- Note: _cons estimates baseline odds. . . logit Follow_Blacks_dichotomous i.Treatment##c.negstereotype_endorsement [pw=WEIGHT], or Iteration 0: log pseudolikelihood = -376.88364 Iteration 1: log pseudolikelihood = -365.65576 Iteration 2: log pseudolikelihood = -365.3901 Iteration 3: log pseudolikelihood = -365.3892 Iteration 4: log pseudolikelihood = -365.3892 Logistic regression Number of obs = 586 Wald chi2(3) = 12.42 Prob > chi2 = 0.0061 Log pseudolikelihood = -365.3892 Pseudo R2 = 0.0305 ------------------------------------------------------------------------------------------------------- | Robust Follow_Blacks_dichotomous | Odds ratio std. err. z P>|z| [95% conf. interval] --------------------------------------+---------------------------------------------------------------- 1.Treatment | .6703608 .1835596 -1.46 0.144 .3919501 1.146533 negstereotype_endorsement | 1.643643 .8650749 0.94 0.345 .5858778 4.611138 | Treatment#c.negstereotype_endorsement | 1 | 4.115432 3.268415 1.78 0.075 .8677686 19.51763 | _cons | 2.012985 .4112941 3.42 0.001 1.348723 3.004405 ------------------------------------------------------------------------------------------------------- Note: _cons estimates baseline odds. . logit Follow_Blacks_dichotomous i.Treatment##c.nse2 [pw=WEIGHT], or Iteration 0: log pseudolikelihood = -321.90815 Iteration 1: log pseudolikelihood = -311.51246 Iteration 2: log pseudolikelihood = -310.99221 Iteration 3: log pseudolikelihood = -310.98705 Iteration 4: log pseudolikelihood = -310.98705 Logistic regression Number of obs = 493 Wald chi2(3) = 9.60 Prob > chi2 = 0.0222 Log pseudolikelihood = -310.98705 Pseudo R2 = 0.0339 ------------------------------------------------------------------------------------------- | Robust Follow_Blacks_dichotomous | Odds ratio std. err. z P>|z| [95% conf. interval] --------------------------+---------------------------------------------------------------- 1.Treatment | .684012 .1905613 -1.36 0.173 .3962092 1.180872 nse2 | 1.600313 .8627296 0.87 0.383 .5563164 4.6035 | Treatment#c.nse2 | 1 | 6.090728 6.137165 1.79 0.073 .8452358 43.88949 | _cons | 2.00722 .4183346 3.34 0.001 1.334111 3.019938 ------------------------------------------------------------------------------------------- Note: _cons estimates baseline odds. . . * Below are comparisons for the four APSR letter outcomes, using OLS: . . reg facemasks_dichotomous i.Treatment##c.negstereotype_endorsement [pw=WEIGHT] (sum of wgt is 590.475482536) Linear regression Number of obs = 590 F(3, 586) = 5.94 Prob > F = 0.0005 R-squared = 0.0351 Root MSE = .48773 ------------------------------------------------------------------------------------------------------- | Robust facemasks_dichotomous | Coefficient std. err. t P>|t| [95% conf. interval] --------------------------------------+---------------------------------------------------------------- 1.Treatment | -.0329816 .0654106 -0.50 0.614 -.1614494 .0954862 negstereotype_endorsement | .057674 .1227409 0.47 0.639 -.1833916 .2987396 | Treatment#c.negstereotype_endorsement | 1 | .3214472 .153688 2.09 0.037 .0196007 .6232936 | _cons | .4066724 .0493513 8.24 0.000 .3097454 .5035994 ------------------------------------------------------------------------------------------------------- . reg facemasks_dichotomous i.Treatment##c.nse2 [pw=WEIGHT] (sum of wgt is 497.65235185) Linear regression Number of obs = 496 F(3, 492) = 5.82 Prob > F = 0.0007 R-squared = 0.0361 Root MSE = .48537 ---------------------------------------------------------------------------------- | Robust facemasks_dich~s | Coefficient std. err. t P>|t| [95% conf. interval] -----------------+---------------------------------------------------------------- 1.Treatment | -.026874 .0668831 -0.40 0.688 -.1582857 .1045378 nse2 | .0153864 .1342946 0.11 0.909 -.2484754 .2792481 | Treatment#c.nse2 | 1 | .3659189 .1636329 2.24 0.026 .0444134 .6874244 | _cons | .400932 .0505675 7.93 0.000 .3015771 .5002869 ---------------------------------------------------------------------------------- . . reg individualrights_dichotomous i.Treatment##c.negstereotype_endorsement [pw=WEIGHT] (sum of wgt is 588.553299849) Linear regression Number of obs = 587 F(3, 583) = 3.27 Prob > F = 0.0209 R-squared = 0.0327 Root MSE = .44542 ------------------------------------------------------------------------------------------------------- | Robust individualrights_dichotomous | Coefficient std. err. t P>|t| [95% conf. interval] --------------------------------------+---------------------------------------------------------------- 1.Treatment | -.0989357 .0612329 -1.62 0.107 -.2191997 .0213283 negstereotype_endorsement | -.0405394 .1079904 -0.38 0.708 -.2526369 .1715582 | Treatment#c.negstereotype_endorsement | 1 | .3719977 .1539394 2.42 0.016 .0696544 .674341 | _cons | .3105795 .0481871 6.45 0.000 .215938 .405221 ------------------------------------------------------------------------------------------------------- . reg individualrights_dichotomous i.Treatment##c.nse2 [pw=WEIGHT] (sum of wgt is 495.891357221) Linear regression Number of obs = 494 F(3, 490) = 3.19 Prob > F = 0.0234 R-squared = 0.0394 Root MSE = .44168 ---------------------------------------------------------------------------------- | Robust individualrig~us | Coefficient std. err. t P>|t| [95% conf. interval] -----------------+---------------------------------------------------------------- 1.Treatment | -.0936921 .0627348 -1.49 0.136 -.2169545 .0295703 nse2 | -.0514601 .1137149 -0.45 0.651 -.2748892 .1719689 | Treatment#c.nse2 | 1 | .4058679 .1655444 2.45 0.015 .0806035 .7311324 | _cons | .3090985 .0496442 6.23 0.000 .2115566 .4066403 ---------------------------------------------------------------------------------- . . reg visitparks_dichotomous i.Treatment##c.negstereotype_endorsement [pw=WEIGHT] (sum of wgt is 584.5646108780001) Linear regression Number of obs = 589 F(3, 585) = 0.96 Prob > F = 0.4135 R-squared = 0.0072 Root MSE = .41426 ------------------------------------------------------------------------------------------------------- | Robust visitparks_dichotomous | Coefficient std. err. t P>|t| [95% conf. interval] --------------------------------------+---------------------------------------------------------------- 1.Treatment | -.0270027 .0563863 -0.48 0.632 -.1377468 .0837415 negstereotype_endorsement | -.1373966 .096823 -1.42 0.156 -.3275596 .0527664 | Treatment#c.negstereotype_endorsement | 1 | .2129357 .1410229 1.51 0.132 -.0640372 .4899086 | _cons | .239365 .0444028 5.39 0.000 .1521567 .3265733 ------------------------------------------------------------------------------------------------------- . reg visitparks_dichotomous i.Treatment##c.nse2 [pw=WEIGHT] (sum of wgt is 491.741480192) Linear regression Number of obs = 495 F(3, 491) = 0.33 Prob > F = 0.8072 R-squared = 0.0035 Root MSE = .4189 ---------------------------------------------------------------------------------- | Robust visitparks_dic~s | Coefficient std. err. t P>|t| [95% conf. interval] -----------------+---------------------------------------------------------------- 1.Treatment | -.0365607 .0578429 -0.63 0.528 -.1502108 .0770894 nse2 | -.0875097 .1093508 -0.80 0.424 -.302363 .1273436 | Treatment#c.nse2 | 1 | .1437582 .1570286 0.92 0.360 -.1647727 .4522892 | _cons | .2456798 .0458966 5.35 0.000 .1555018 .3358578 ---------------------------------------------------------------------------------- . . reg Follow_Blacks_dichotomous i.Treatment##c.negstereotype_endorsement [pw=WEIGHT] (sum of wgt is 588.0094088770001) Linear regression Number of obs = 586 F(3, 582) = 6.92 Prob > F = 0.0001 R-squared = 0.0354 Root MSE = .46681 ------------------------------------------------------------------------------------------------------- | Robust Follow_Blacks_dichotomous | Coefficient std. err. t P>|t| [95% conf. interval] --------------------------------------+---------------------------------------------------------------- 1.Treatment | -.0915267 .0632161 -1.45 0.148 -.2156861 .0326327 negstereotype_endorsement | .1014544 .1022733 0.99 0.322 -.0994153 .3023241 | Treatment#c.negstereotype_endorsement | 1 | .2398664 .1279354 1.87 0.061 -.011405 .4911378 | _cons | .6683277 .0451983 14.79 0.000 .5795559 .7570994 ------------------------------------------------------------------------------------------------------- . reg Follow_Blacks_dichotomous i.Treatment##c.nse2 [pw=WEIGHT] (sum of wgt is 496.300700187) Linear regression Number of obs = 493 F(3, 489) = 8.78 Prob > F = 0.0000 R-squared = 0.0374 Root MSE = .47046 ---------------------------------------------------------------------------------- | Robust Follow_Blacks_~s | Coefficient std. err. t P>|t| [95% conf. interval] -----------------+---------------------------------------------------------------- 1.Treatment | -.0888811 .0647868 -1.37 0.171 -.2161759 .0384137 nse2 | .0951263 .1015135 0.94 0.349 -.1043301 .2945827 | Treatment#c.nse2 | 1 | .2567583 .1235197 2.08 0.038 .0140635 .4994531 | _cons | .667467 .0464004 14.38 0.000 .5762983 .7586357 ---------------------------------------------------------------------------------- . . * . end of do-file . log close name: log: D:\Stephens-Dougan 2022 NSE2.log log type: text closed on: 6 Dec 2022, 20:42:31 ---------------------------------------------------------------------------------------------------------------