Appendix A:

Supplementary Analyses Including State Contextual Controls

Table A1. Regression of Support for Expanding Police Powers on Perceived Latino

Political and Economic Threat and Controls (State Contextual Controls)

Model 1 / Model 2
Variables / b / RSE / Beta / b / RSE / Beta
Latino Economic Threat / ― / ― / ― / .177** / .049 / .105
Latino Political Threat / .216** / .066 / .109 / ― / ― / ―
Black / –.937** / .264 / –.116 / –.868** / .265 / –.107
Latino / –.405 / .278 / –.043 / –.311 / .306 / –.033
Female / .175 / .128 / .033 / .166 / .133 / .032
Age / –.407* / .198 / –.245 / –.426* / .205 / –.256
Age2 / .063** / .023 / .313 / .067** / .024 / .329
Education / –.085 / .072 / –.037 / –.100 / .072 / –.043
Income / .128* / .054 / .077 / .121* / .055 / .073
Moderate / –.012 / .188 / –.002 / –.024 / .189 / –.004
Conservative / .652** / .177 / .123 / .681*** / .179 / .129
National News / .166 / .156 / .026 / .198 / .162 / .031
Local News / .173 / .218 / .023 / .159 / .211 / .021
Crime TV / –.073 / .147 / –.013 / –.074 / .147 / –.014
Punitiveness / .412*** / .043 / .325 / .412*** / .043 / .325
Symbolic Racism / .137† / .071 / .053 / .116 / .071 / .045
Perceived Risk / .183** / .051 / .152 / .181** / .050 / .151
Household Victim / –.344* / .166 / –.060 / –.327* / .161 / –.057
State Percent Latino / –.019 / .018 / –.084 / –.017 / .017 / –.076
State Latino Growth / –.036 / .095 / –.021 / –.049 / .093 / –.028
State Percent Republican / –.024* / .010 / –.074 / –.025* / .010 / –.077
State Percent Unemployed / –.045 / .041 / –.029 / –.051 / .041 / –.032
State Homicide Rate / .028 / .040 / .020 / .032 / .037 / .023
Border State / .323 / .339 / .046 / .336 / .326 / .047
Adjusted R–squared / .253 / .252
N / 932 / 932
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; DV = dependent variable; RSE = robust standard error, clustered by state.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table A2. Analysis of the Relationship BetweenPerceived Latino Political Threat and Support for Expanding Specific

Police Powers (State Contextual Controls)

Model 1: Police Profiling / Model 2: Police Searches / Model 3: Police Use of Force
Variables / b / RSE / Beta / b / RSE / Beta / b / RSE / Beta
Latino Political Threat / .284** / .077 / .122 / .211* / .088 / .087 / .147† / .085 / .067
Black / –.799** / .272 / –.084 / –.872* / .325 / –.089 / –1.143** / .320 / –.128
Latino / –.598† / .350 / –.053 / –.187 / .310 / –.016 / –.509 / .373 / –.048
Female / –.061 / .152 / –.01 / .510** / .164 / .080 / .086 / .157 / .015
Age / –.400 / .276 / –.206 / –.200 / .280 / –.099 / –.628** / .203 / –.342
Age2 / .059† / .032 / .247 / .039 / .032 / .158 / .093*** / .023 / .417
Education / –.000 / .090 / –.000 / –.132 / .101 / –.047 / –.129 / .090 / –.050
Income / .180** / .053 / .093 / .116 / .071 / .057 / .084 / .069 / .046
Moderate / .026 / .216 / .004 / .013 / .230 / .002 / –.085 / .233 / –.014
Conservative / .872** / .249 / .141 / .449* / .216 / .070 / .623** / .190 / .107
National News / .196 / .201 / .026 / .292 / .183 / .038 / .013 / .253 / .002
Local News / –.082 / .244 / –.009 / .379 / .292 / .042 / .181 / .281 / .022
Crime TV / –.144 / .204 / –.023 / –.262 / .207 / –.040 / .192 / .176 / .032
Punitiveness / .392*** / .048 / .264 / .372*** / .049 / .242 / .473*** / .054 / .337
Symbolic Racism / .167* / .078 / .055 / .087 / .101 / .028 / .155† / .083 / .054
Perceived Risk / .145** / .050 / .104 / .239*** / .060 / .165 / .164** / .058 / .124
Household Victim / –.161 / .224 / –.024 / –.589** / .178 / –.084 / –.326 / .201 / –.051
State Percent Latino / –.021 / .022 / –.082 / –.020 / .021 / –.076 / –.013 / .021 / –.055
State Latino Growth / .044 / .107 / .022 / –.100 / .112 / –.047 / –.038 / .112 / –.020
State Percent Republican / –.024 / .014 / –.063 / –.028* / .011 / –.072 / –.021* / .010 / –.059
State Percent Unemployed / .018 / .047 / .010 / –.114* / .048 / –.060 / –.049 / .050 / –.028
State Homicide Rate / –.001 / .055 / –.001 / .021 / .046 / .012 / .063† / .037 / .041
Border State / .315 / .427 / .038 / .498 / .427 / .058 / .139 / .360 / .018
Adjusted R–squared / .171 / .169 / .240
N / 930 / 929 / 929
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; RSE = robust standard error, clustered by state.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table A3. Analysis of the Relationship BetweenPerceived Latino Economic Threat and Support for Expanding Specific

Police Powers (State Contextual Controls)

Model 1: Police Profiling / Model 2: Police Searches / Model 3: Police Use of Force
Variables / b / RSE / Beta / b / RSE / Beta / b / RSE / Beta
Latino Economic Threat / .271*** / .060 / .138 / .138* / .057 / .068 / .117† / .063 / .063
Black / –.710** / .261 / –.075 / –.802* / .327 / –.082 / –1.096** / .319 / –.122
Latino / –.462 / .383 / –.041 / –.107 / .318 / –.009 / –.447 / .392 / –.042
Female / –.066 / .156 / –.011 / .493** / .171 / .077 / .080 / .160 / .014
Age / –.432 / .280 / –.222 / –.211 / .285 / –.105 / –.641** / .209 / –.349
Age2 / .064† / .032 / .269 / .041 / .032 / .168 / .096*** / .024 / .427
Education / –.019 / .097 / –.007 / –.148 / .100 / –.053 / –.140 / .088 / –.054
Income / .174** / .054 / .090 / .104 / .072 / .052 / .078 / .070 / .043
Moderate / .006 / .216 / .001 / .005 / .233 / .001 / –.093 / .233 / –.015
Conservative / .901** / .245 / .146 / .486* / .223 / .076 / .643** / .188 / .110
National News / .247 / .198 / .033 / .315 / .188 / .041 / .034 / .254 / .005
Local News / –.105 / .234 / –.012 / .372 / .284 / .041 / .172 / .279 / .021
Crime TV / –.158 / .207 / –.025 / –.253 / .202 / –.038 / .192 / .174 / .032
Punitiveness / .388*** / .047 / .261 / .376*** / .049 / .244 / .473*** / .053 / .337
Symbolic Racism / .134† / .074 / .044 / .072 / .101 / .023 / .141† / .083 / .049
Perceived Risk / .142** / .049 / .101 / .239*** / .059 / .165 / .163** / .057 / .123
Household Victim / –.131 / .215 / –.019 / –.579** / .178 / –.083 / –.315 / .199 / –.049
State Percent Latino / –.019 / .022 / –.072 / –.019 / .020 / –.070 / –.012 / .021 / –.050
State Latino Growth / .024 / .105 / .012 / –.111 / .110 / –.052 / –.047 / .112 / –.024
State Percent Republican / –.025† / .013 / –.067 / –.029* / .011 / –.073 / –.022* / .010 / –.060
State Percent Unemployed / .009 / .046 / .005 / –.119* / .048 / –.062 / –.053 / .051 / –.031
State Homicide Rate / .005 / .050 / .003 / .024 / .043 / .014 / .066† / .036 / .042
Border State / .326 / .410 / .039 / .513 / .413 / .060 / .149 / .359 / .019
Adjusted R–squared / .175 / .167 / .240
N / 930 / 929 / 929
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; RSE = robust standard error, clustered by state.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table A4. Analyses Disaggregated by Race (State Contextual Controls)

IV = Latino Political
Threat / IV = Latino Economic Threat
Respondent Race and Dependent Variable / b / RSE / Beta / b / RSE / Beta
Whites
Models 1–2
DV = Expand Police Powers (N = 709) / .213** / .073 / .107 / .186** / .065 / .111
Models 3–4
DV = Police Profiling (N = 708) / .280** / .092 / .118 / .288** / .083 / .145
Models 5–6
DV = Police Searches (N = 706) / .243* / .094 / .099 / .178* / .084 / .087
Models 7–8
DV = Police Use of Force (N = 707) / .120 / .085 / .054 / .094 / .069 / .050
Blacks1
Models 9–10
DV = Expand Police Powers (N = 109) / .140 / .186 / .076 / .171 / .148 / .113
Models 11–12
DV = Police Profiling (N = 109) / .342 / .219 / .169 / .216 / .158 / .131
Models 13–14
DV = Police Searches (N = 109) / –.105 / .231 / –.046 / .038 / .193 / .020
Models 15–16
DV = Police Use of Force (N = 109) / .185 / .179 / .093 / .260 / .164 / .159
Latinos1
Models 17–18
DV = Expand Police Powers (N = 76) / .241 / .261 / .122 / .131 / .242 / .076
Models 19–20
DV = Police Profiling (N = 75) / –.002 / .325 / –.001 / .121 / .271 / .062
Models 21–22
DV = Police Searches (N = 76) / .297 / .344 / .124 / .021 / .320 / .010
Models 23–24
DV = Police Use of Force (N = 75) / .279 / .304 / .131 / .139 / .270 / .075
NOTES: Estimates shown are from equations that include all control variables. 1The results for Blacks and Latinos should be interpreted with caution, because of the low sample sizes in the models. In the models for Latinos, there are less than four respondents per variable.
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; DV = dependent variable; IV = Independent variable; RSE = robust standard error, clustered by state.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Appendix B:

Supplementary Analyses for Separate Items Measuring Perceived Latino Political Threat

Table B1.Regression of Support for Expanding Police Powers on Individual Items

Measuring Perceived Latino Political Threat

Model 1 / Model 2
Variables / b / RSE / Beta / b / RSE / Beta
Latino Voting / ― / ― / ― / .140* / .057 / .080
Latino Political Influence / .187** / .056 / .104 / ― / ― / ―
Black / –.976*** / .257 / –.121 / –.975*** / .256 / –.120
Latino / –.332 / .341 / –.035 / –.357 / .346 / –.038
Female / .167 / .153 / .032 / .121 / .153 / .023
Age / –.396* / .197 / –.238 / –.386† / .199 / –.231
Age2 / .061* / .024 / .301 / .061* / .025 / .301
Education / –.083 / .080 / –.036 / –.081 / .082 / –.035
Income / .114† / .060 / .069 / .108† / .060 / .065
Moderate / .032 / .197 / .006 / .018 / .199 / .003
Conservative / .680** / .205 / .129 / .669** / .209 / .126
National News / .203 / .187 / .032 / .192 / .186 / .030
Local News / .133 / .226 / .018 / .136 / .224 / .018
Crime TV / –.056 / .165 / –.010 / –.062 / .165 / –.011
Punitiveness / .416*** / .044 / .328 / .418*** / .045 / .330
Symbolic Racism / .132 / .083 / .051 / .154† / .084 / .059
Perceived Risk / .172*** / .042 / .144 / .178*** / .042 / .148
Household Victim / –.348* / .171 / –.060 / –.372* / .172 / –.064
Ln Percent Latino / –.149 / .122 / –.064 / –.145 / .121 / –.063
Latino Growth / .031 / .047 / .029 / .032 / .046 / .030
Percent Republican / –.008 / .006 / –.041 / –.008 / .006 / –.043
Ln Percent Unemployed / –.729* / .294 / –.075 / –.727* / .298 / –.075
Ln Homicide Rate / .093 / .120 / .026 / .107 / .122 / .030
Border State / –.014 / .263 / –.002 / –.023 / .267 / –.003
Adjusted R–squared / .251 / .249
N / 930 / 928
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; DV = dependent variable; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table B2. Analysis of the Relationship Between Perceived Latino Voting and Support for Expanding Specific Police

Powers

Model 1: Police Profiling / Model 2: Police Searches / Model 3: Police Use of Force
Variables / b / RSE / Beta / b / RSE / Beta / b / RSE / Beta
Latino Voting / .164* / .069 / .080 / .133† / .071 / .062 / .115† / .065 / .059
Black / –.978** / .293 / –.103 / –.919** / .331 / –.094 / –1.033*** / .284 / –.116
Latino / –.645† / .378 / –.058 / –.087 / .410 / –.008 / –.431 / .408 / –.041
Female / –.141 / .193 / –.023 / .477* / .196 / .075 / .037 / .169 / .006
Age / –.347 / .272 / –.178 / –.195 / .258 / –.097 / –.621** / .225 / –.337
Age2 / .054 / .034 / .228 / .039 / .032 / .159 / .091** / .028 / .406
Education / –.007 / .102 / –.003 / –.139 / .104 / –.050 / –.104 / .094 / –.040
Income / .137† / .071 / .070 / .108 / .077 / .054 / .073 / .069 / .040
Moderate / .042 / .228 / .007 / .049 / .252 / .007 / –.046 / .221 / –.008
Conservative / .861** / .250 / .139 / .464† / .267 / .072 / .676** / .234 / .116
National News / .207 / .227 / .028 / .304 / .229 / .039 / .071 / .230 / .010
Local News / –.145 / .287 / –.017 / .373 / .293 / .041 / .142 / .245 / .017
Crime TV / –.107 / .211 / –.017 / –.269 / .207 / –.041 / .194 / .179 / .032
Punitiveness / .398*** / .055 / .269 / .380*** / .056 / .247 / .480*** / .049 / .342
Symbolic Racism / .199† / .102 / .066 / .095 / .106 / .030 / .163† / .094 / .057
Perceived Risk / .134** / .050 / .096 / .229*** / .051 / .158 / .170** / .049 / .128
Household Victim / –.193 / .219 / –.029 / –.620** / .207 / –.089 / –.345† / .194 / –.054
Ln Percent Latino / .015 / .145 / .005 / –.329* / .163 / –.117 / –.111 / .143 / –.043
Latino Growth / .038 / .053 / .030 / .048 / .062 / .037 / .010 / .060 / .008
Percent Republican / –.005 / .007 / –.023 / –.013† / .008 / –.058 / –.006 / .007 / –.027
Ln Percent Unemployed / –.635† / .370 / –.056 / –1.102** / .369 / –.094 / –.422 / .336 / –.040
Ln Homicide Rate / .253† / .150 / .060 / .121 / .154 / .028 / –.060 / .135 / –.015
Border State / –.255 / .304 / –.031 / .214 / .348 / .025 / –.020 / .302 / –.003
Adjusted R–squared / .171 / .168 / .239
N / 926 / 925 / 925
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table B3. Analysis of the Relationship BetweenPerceived Latino Political Influence and Support for Expanding Specific

Police Powers

Model 1: Police Profiling / Model 2: Police Searches / Model 3: Police Use of Force
Variables / b / RSE / Beta / b / RSE / Beta / b / RSE / Beta
Latino Political Influence / .248*** / .071 / .118 / .176* / .071 / .081 / .131* / .065 / .066
Black / –.984** / .292 / –.104 / –.908** / .332 / –.092 / –1.044*** / .284 / –.117
Latino / –.614† / .371 / –.055 / –.056 / .408 / –.005 / –.427 / .404 / –.040
Female / –.067 / .193 / –.011 / .505* / .196 / .079 / .074 / .168 / .013
Age / –.351 / .265 / –.181 / –.201 / .258 / –.100 / –.640** / .224 / –.348
Age2 / .051 / .033 / .215 / .038 / .032 / .156 / .094** / .027 / .420
Education / –.003 / .101 / –.001 / –.130 / .104 / –.047 / –.122 / .093 / –.048
Income / .148* / .071 / .076 / .106 / .077 / .053 / .082 / .069 / .044
Moderate / .063 / .226 / .010 / .076 / .251 / .012 / –.053 / .223 / –.009
Conservative / .874*** / .245 / .142 / .497† / .263 / .077 / .659** / .234 / .113
National News / .254 / .227 / .034 / .337 / .229 / .044 / .022 / .234 / .003
Local News / –.172 / .287 / –.020 / .365 / .294 / .040 / .169 / .250 / .021
Crime TV / –.113 / .211 / –.018 / –.256 / .208 / –.039 / .206 / .179 / .034
Punitiveness / .394*** / .054 / .266 / .379*** / .055 / .246 / .479*** / .049 / .341
Symbolic Racism / .159 / .103 / .053 / .077 / .107 / .025 / .154 / .094 / .054
Perceived Risk / .128* / .050 / .091 / .226*** / .051 / .155 / .164** / .049 / .124
Household Victim / –.171 / .219 / –.025 / –.597** / .204 / –.085 / –.321 / .196 / –.050
Ln Percent Latino / .016 / .145 / .006 / –.341* / .166 / –.121 / –.110 / .143 / –.043
Latino Growth / .033 / .053 / .027 / .044 / .063 / .034 / .016 / .061 / .013
Percent Republican / –.004 / .007 / –.019 / –.013† / .008 / –.058 / –.006 / .007 / –.027
Ln Percent Unemployed / –.623† / .364 / –.055 / –1.107** / .367 / –.094 / –.433 / .334 / –.041
Ln Homicide Rate / .235 / .149 / .056 / .096 / .154 / .022 / –.059 / .133 / –.015
Border State / –.239 / .297 / –.029 / .249 / .345 / .029 / –.042 / .303 / –.005
Adjusted R–squared / .174 / .170 / .238
N / 928 / 927 / 927
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table B4. Analyses Disaggregated by Race for Individual Items Measuring Perceived

Latino Political Threat

IV = Latino Voting / IV = Latino Political Influence
Respondent Race and Dependent Variable / b / RSE / Beta / b / RSE / Beta
Whites
Models 1–2
DV = Expand Police Powers (N = 705, 707) / .117† / .066 / .065 / .200** / .066 / .110
Models 3–4
DV = Police Profiling (N = 704, 706) / .149† / .080 / .071 / .248** / .083 / .115
Models 5–6
DV = Police Searches (N = 702, 704) / .124 / .083 / .057 / .220* / .085 / .099
Models 7–8
DV = Police Use of Force (N = 703, 705) / .079 / .074 / .040 / .131† / .074 / .065
Blacks1
Models 9–10
DV = Expand Police Powers (N = 109) / .152 / .174 / .092 / .042 / .172 / .026
Models 11–12
DV = Police Profiling (N = 109) / .231 / .204 / .129 / .220 / .190 / .123
Models 13–14
DV = Police Searches (N = 109) / .014 / .206 / .007 / –.161 / .213 / –.080
Models 15–16
DV = Police Use of Force (N = 109) / .212 / .175 / .120 / .067 / .166 / .038
Latinos1
Models 17–18
DV = Expand Police Powers (N = 76) / .208 / .227 / .121 / .224 / .202 / .130
Models 19–20
DV = Police Profiling (N = 75) / .070 / .269 / .036 / .100 / .283 / .051
Models 21–22
DV = Police Searches (N = 76) / .295 / .287 / .142 / .220 / .282 / .106
Models 23–24
DV = Police Use of Force (N = 75) / .148 / .264 / .079 / .285 / .222 / .154
NOTES: Estimates shown are from equations that include all control variables. 1The results for Blacks and Latinos should be interpreted with caution, because of the low sample sizes in the models. In the models for Latinos, there are less than four respondents per variable.
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; DV = dependent variable; IV = Independent variable; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Appendix C:

Supplementary Models Using Dummy Coding for Age, Education, and Income

Table C1. Regression of Support for Expanding Police Powers on Perceived Latino

Political and Economic Threat and Controls (With Dummy Coding)

Model 1 / Model 2
Variables / b / RSE / Beta / b / RSE / Beta
Latino Economic Threat / ― / ― / ― / .173** / .054 / .103
Latino Political Threat / .214** / .065 / .108 / ― / ― / ―
Black / –.976*** / .263 / –.120 / –.922*** / .261 / –.114
Latino / –.391 / .353 / –.041 / –.308 / .352 / –.032
Female / .166 / .155 / .032 / .154 / .156 / .029
Age
25 to 34 / .086 / .324 / .011 / .104 / .320 / .013
35 to 44 / .090 / .323 / .013 / .121 / .316 / .017
45 to 54 / –.188 / .299 / –.034 / –.169 / .293 / –.030
55 to 64 / –.026 / .314 / –.004 / –.017 / .304 / –.002
65 to 74 / .521 / .417 / .055 / .558 / .409 / .059
75 or older / .759† / .404 / .076 / .841* / .402 / .084
Education
High School Degree / –.432 / .401 / –.071 / –.497 / .406 / –.081
Some College / –.594 / .400 / –.108 / –.655 / .406 / –.119
Bachelor’s Degree / –.547 / .410 / –.086 / –.660 / .418 / –.104
Graduate Degree / –.665 / .437 / –.092 / –.752† / .443 / –.104
Income
$15–34,999 / .294 / .329 / .042 / .263 / .331 / .038
$35–49,999 / .133 / .306 / .020 / .151 / .307 / .023
$50–74,999 / .548† / .315 / .087 / .504 / .312 / .080
$75–99,999 / .726* / .345 / .099 / .701* / .343 / .096
$100,000 or more / .565 / .354 / .082 / .526 / .354 / .076
Moderate / .028 / .197 / .005 / .014 / .198 / .003
Conservative / .661** / .209 / .125 / .689** / .207 / .130
National News / .199 / .186 / .031 / .227 / .187 / .036
Local News / .120 / .226 / .016 / .106 / .227 / .014
Crime TV / –.079 / .165 / –.015 / –.073 / .164 / –.013
Punitiveness / .406*** / .045 / .320 / .406*** / .046 / .320
Symbolic Racism / .144† / .083 / .056 / .125 / .084 / .048
Perceived Risk / .172*** / .043 / .144 / .170*** / .042 / .142
Household Victim / –.355* / .174 / –.062 / –.339† / .174 / –.059
Ln Percent Latino / –.135 / .121 / –.058 / –.126 / .123 / –.054
Latino Growth / .027 / .046 / .025 / .016 / .047 / .015
Percent Republican / –.007 / .006 / –.038 / –.007 / .006 / –.038
Ln Percent Unemployed / –.717* / .294 / –.074 / –.745* / .296 / –.077
Ln Homicide Rate / .121 / .124 / .033 / .152 / .123 / .042
Border State / –.063 / .270 / –.009 / –.022 / .270 / –.003
Adjusted R–squared / .250 / .250
N / 932 / 932
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; DV = dependent variable; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table C2. Analysis of the Relationship Between Perceived Latino Political Threat and Support for Expanding Specific

Police Powers (With Dummy Coding)

Model 1: Police Profiling / Model 2: Police Searches / Model 3: Police Use of Force
Variables / b / RSE / Beta / b / RSE / Beta / b / RSE / Beta
Latino Political Threat / .281** / .080 / .121 / .220** / .080 / .091 / .134† / .077 / .061
Black / –1.011** / .293 / –.107 / –.856* / .343 / –.087 / –1.064*** / .289 / –.119
Latino / –.730† / .384 / –.065 / –.124 / .419 / –.011 / –.416 / .418 / –.039
Female / –.058 / .194 / –.009 / .521* / .202 / .082 / .043 / .173 / .007
Age
25 to 34 / .410 / .384 / .044 / .057 / .438 / .006 / –.212 / .422 / –.024
35 to 44 / –.037 / .423 / –.005 / .314 / .440 / .037 / –.001 / .379 / –.000
45 to 54 / –.168 / .387 / –.026 / –.051 / .413 / –.007 / –.361 / .369 / –.058
55 to 64 / .029 / .393 / .004 / .150 / .424 / .018 / –.263 / .392 / –.034
65 to 74 / .561 / .513 / .051 / .569 / .536 / .050 / .431 / .501 / .041
75 or older / .544 / .519 / .046 / .755 / .533 / .062 / .967* / .485 / .087
Education
High School Degree / –.870† / .522 / –.122 / –.427 / .490 / –.057 / –.022 / .488 / –.003
Some College / –.872† / .515 / –.135 / –.780 / .487 / –.116 / –.130 / .476 / –.021
Bachelor’s Degree / –.476 / .529 / –.064 / –.804 / .511 / –.104 / –.381 / .497 / –.054
Graduate Degree / –.858 / .553 / –.101 / –.751 / .545 / –.086 / –.442 / .517 / –.055
Income
$15–34,999 / .093 / .376 / .011 / .848* / .426 / .100 / –.064 / .389 / –.008
$35–49,999 / .030 / .351 / .004 / .409 / .397 / .051 / –.065 / .358 / –.009
$50–74,999 / .561 / .370 / .076 / 1.165** / .408 / .152 / –.071 / .348 / –.010
$75–99,999 / .848* / .402 / .099 / .883* / .444 / .100 / .441 / .384 / .054
$100,000 or more / .614 / .411 / .076 / .819† / .462 / .098 / .223 / .405 / .029
Moderate / .054 / .229 / .008 / .106 / .249 / .016 / –.087 / .221 / –.014
Conservative / .811** / .249 / .131 / .507† / .270 / .079 / .654** / .235 / .112
National News / .246 / .225 / .033 / .326 / .229 / .042 / .030 / .237 / .004
Local News / –.175 / .287 / –.020 / .338 / .293 / .037 / .162 / .252 / .020
Crime TV / –.144 / .208 / –.023 / –.309 / .208 / –.047 / .216 / .180 / .036
Punitiveness / .384*** / .055 / .259 / .362*** / .056 / .235 / .475*** / .049 / .338
Symbolic Racism / .181† / .102 / .060 / .099 / .105 / .032 / .146 / .093 / .051
Perceived Risk / .126* / .052 / .090 / .228*** / .051 / .157 / .162** / .050 / .123
Household Victim / –.195 / .223 / –.029 / –.589** / .208 / –.084 / –.332† / .195 / –.052
Ln Percent Latino / .019 / .145 / .007 / –.299† / .164 / –.106 / –.114 / .144 / –.045
Latino Growth / .027 / .052 / .022 / .035 / .063 / .027 / .018 / .061 / .016
Percent Republican / –.004 / .007 / –.017 / –.012 / .008 / –.051 / –.005 / .007 / –.027
Ln Percent Unemployed / –.596 / .363 / –.053 / –1.121** / .364 / –.096 / –.410 / .335 / –.038
Ln Homicide Rate / .258† / .152 / .061 / .149 / .158 / .034 / –.052 / .137 / –.013
Border State / –.244 / .306 / –.029 / .145 / .353 / .017 / –.078 / .307 / –.010
Adjusted R–squared / .177 / .173 / .235
N / 930 / 929 / 929
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table C3. Analysis of the Relationship Between Perceived Latino Economic Threat and Support for Expanding Specific

Police Powers (With Dummy Coding)

Model 1: Police Profiling / Model 2: Police Searches / Model 3: Police Use of Force
Variables / b / RSE / Beta / b / RSE / Beta / b / RSE / Beta
Latino Economic Threat / .270*** / .066 / .137 / .135† / .071 / .066 / .108† / .060 / .058
Black / –.941** / .295 / –.099 / –.801* / .340 / –.081 / –1.030*** / .287 / –.115
Latino / –.608 / .380 / –.054 / –.052 / .420 / –.004 / –.365 / .420 / –.034
Female / –.064 / .195 / –.010 / .500* / .203 / .078 / .036 / .174 / .006
Age
25 to 34 / .413 / .373 / .044 / .095 / .434 / .010 / –.200 / .419 / –.022
35 to 44 / –.019 / .412 / –.002 / .368 / .433 / .044 / .019 / .375 / .003
45 to 54 / –.163 / .376 / –.025 / –.012 / .410 / –.002 / –.349 / .365 / –.056
55 to 64 / .019 / .378 / .002 / .181 / .414 / .021 / –.256 / .387 / –.033
65 to 74 / .589 / .499 / .054 / .624 / .532 / .054 / .454 / .494 / .044
75 or older / .630 / .515 / .054 / .860 / .532 / .071 / 1.019* / .482 / .092
Education
High School Degree / –.969† / .528 / –.135 / –.479 / .497 / –.065 / –.063 / .487 / –.009
Some College / –.969† / .524 / –.150 / –.827† / .494 / –.123 / –.168 / .476 / –.028
Bachelor’s Degree / –.627 / .542 / –.084 / –.915† / .519 / –.119 / –.451 / .496 / –.064
Graduate Degree / –.979† / .565 / –.116 / –.831 / .550 / –.095 / –.497 / .516 / –.062
Income
$15–34,999 / .060 / .373 / .007 / .808† / .429 / .096 / –.083 / .392 / –.011
$35–49,999 / .068 / .348 / .009 / .411 / .400 / .051 / –.054 / .359 / –.007
$50–74,999 / .522 / .366 / .071 / 1.101** / .407 / .144 / –.099 / .345 / –.014
$75–99,999 / .835* / .398 / .098 / .836† / .444 / .094 / .425 / .384 / .052
$100,000 or more / .588 / .407 / .073 / .751 / .463 / .090 / .198 / .406 / .026
Moderate / .032 / .231 / .005 / .096 / .251 / .015 / –.096 / .222 / –.016
Conservative / .839** / .247 / .136 / .547* / .268 / .085 / .672** / .234 / .115
National News / .292 / .223 / .039 / .346 / .231 / .045 / .047 / .237 / .007
Local News / –.200 / .289 / –.023 / .330 / .294 / .036 / .153 / .252 / .019
Crime TV / –.149 / .207 / –.023 / –.291 / .207 / –.044 / .221 / .178 / .037
Punitiveness / .380*** / .056 / .256 / .365*** / .056 / .237 / .475*** / .050 / .338
Symbolic Racism / .149 / .102 / .049 / .086 / .107 / .027 / .134 / .094 / .047
Perceived Risk / .122* / .051 / .087 / .228*** / .051 / .157 / .161** / .050 / .122
Household Victim / –.163 / .224 / –.024 / –.581** / .209 / –.083 / –.322† / .194 / –.050
Ln Percent Latino / .032 / .144 / .012 / –.290† / .166 / –.103 / –.108 / .145 / –.042
Latino Growth / .012 / .052 / .010 / .026 / .063 / .020 / .012 / .063 / .010
Percent Republican / –.004 / .007 / –.018 / –.012 / .008 / –.051 / –.006 / .007 / –.027
Ln Percent Unemployed / –.633† / .361 / –.056 / –1.148** / .366 / –.098 / –.427 / .337 / –.040
Ln Homicide Rate / .297* / .149 / .070 / .183 / .159 / .042 / –.033 / .137 / –.008
Border State / –.191 / .302 / –.023 / .186 / .352 / .022 / –.053 / .307 / –.007
Adjusted R–squared / .182 / .170 / .235
N / 930 / 929 / 929
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table C4. Analyses Disaggregated by Race (With Dummy Coding)

IV = Latino Political
Threat / IV = Latino Economic Threat
Respondent Race and Dependent Variable / b / RSE / Beta / b / RSE / Beta
Whites
Models 1–2
DV = Expand Police Powers (N = 709) / .205** / .075 / .103 / .171** / .065 / .102
Models 3–4
DV = Police Profiling (N = 708) / .259** / .095 / .110 / .269** / .082 / .135
Models 5–6
DV = Police Searches (N = 706) / .239* / .095 / .098 / .166* / .083 / .081
Models 7–8
DV = Police Use of Force (N = 707) / .120 / .086 / .054 / .078 / .068 / .042
Blacks1
Models 9–10
DV = Expand Police Powers (N = 109) / –.016 / .197 / –.009 / .177 / .146 / .117
Models 11–12
DV = Police Profiling (N = 109) / .142 / .203 / .070 / .267† / .139 / .161
Models 13–14
DV = Police Searches (N = 109) / –.196 / .259 / –.086 / .062 / .192 / .033
Models 15–16
DV = Police Use of Force (N = 109) / .006 / .205 / .003 / .203 / .169 / .124
Latinos1
Models 17–18
DV = Expand Police Powers (N = 76) / .265 / .260 / .134 / .219 / .302 / .127
Models 19–20
DV = Police Profiling (N = 75) / –.003 / .360 / –.001 / .291 / .330 / .149
Models 21–22
DV = Police Searches (N = 76) / .395 / .331 / .165 / .064 / .398 / .031
Models 23–24
DV = Police Use of Force (N = 75) / .328 / .328 / .154 / .229 / .318 / .123
NOTES: Estimates shown are from equations that include all control variables. 1The results for Blacks and Latinos should be interpreted with caution, because of the low sample sizes in the models. In the models for Latinos, there are less than four respondents per variable.
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; DV = dependent variable; IV = Independent variable; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Appendix D:

Supplementary Analyses with Both Threat Measures Included Simultaneously

Table D1. Regression of Support for Expanding Police Powers on Perceived Latino

Threat and Controls (Both Threat Measures Included)

Variables / b / RSE / Beta
Latino Political Threat / .146* / .071 / .073
Latino Economic Threat / .115† / .060 / .068
Black / –.977*** / .255 / –.120
Latino / –.306 / .340 / –.032
Female / .164 / .154 / .031
Age / –.405* / .196 / –.244
Age2 / .063* / .024 / .310
Education / –.082 / .080 / –.035
Income / .121* / .060 / .073
Moderate / .014 / .198 / .002
Conservative / .657** / .208 / .124
National News / .213 / .186 / .034
Local News / .123 / .224 / .017
Crime TV / –.087 / .163 / –.016
Punitiveness / .405*** / .045 / .319
Symbolic Racism / .126 / .083 / .049
Perceived Risk / .170*** / .041 / .142
Household Victim / –.343* / .172 / –.059
Ln Percent Latino / –.144 / .121 / –.062
Latino Growth / .027 / .047 / .026
Percent Republican / –.008 / .006 / –.044
Ln Percent Unemployed / –.723* / .294 / –.075
Ln Homicide Rate / .100 / .121 / .028
Border State / –.024 / .263 / –.003
Adjusted R–squared / .255
N / 932
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; DV = dependent variable; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table D2. Analysis of the Relationship Between Perceived Latino Threat and Support for Expanding Specific Police

Powers (Both Threat Measures Included)

Model 1: Police Profiling / Model 2: Police Searches / Model 3: Police Use of Force
Variables / b / RSE / Beta / b / RSE / Beta / b / RSE / Beta
Latino Political Threat / .152† / .091 / .065 / .169† / .088 / .070 / .111 / .081 / .050
Latino Economic Threat / .204** / .074 / .104 / .067 / .078 / .033 / .071 / .064 / .038
Black / –.972** / .290 / –.103 / –.919** / .333 / –.093 / –1.044*** / .282 / –.117
Latino / –.546 / .368 / –.049 / –.052 / .410 / –.004 / –.409 / .403 / –.039
Female / –.070 / .194 / –.011 / .499* / .197 / .078 / .074 / .170 / .013
Age / –.373 / .268 / –.192 / –.212 / .258 / –.105 / –.636** / .223 / –.346
Age2 / .055† / .033 / .231 / .040 / .032 / .164 / .094** / .027 / .418
Education / –.004 / .101 / –.001 / –.128 / .103 / –.046 / –.121 / .094 / –.047
Income / .160* / .071 / .082 / .115 / .077 / .057 / .083 / .069 / .045
Moderate / .036 / .228 / .006 / .064 / .252 / .010 / –.070 / .223 / –.012
Conservative / .845** / .248 / .136 / .475† / .266 / .074 / .643** / .235 / .110
National News / .267 / .224 / .036 / .329 / .229 / .043 / .045 / .232 / .006
Local News / –.186 / .287 / –.021 / .364 / .294 / .040 / .153 / .248 / .019
Crime TV / –.161 / .209 / –.025 / –.282 / .207 / –.043 / .185 / .177 / .031
Punitiveness / .376*** / .056 / .254 / .369*** / .056 / .240 / .471*** / .050 / .336
Symbolic Racism / .148 / .101 / .049 / .080 / .106 / .026 / .146 / .094 / .051
Perceived Risk / .124* / .050 / .088 / .226*** / .050 / .155 / .161** / .049 / .122
Household Victim / –.152 / .222 / –.022 / –.597** / .206 / –.085 / –.325† / .194 / –.051
Ln Percent Latino / .026 / .143 / .009 / –.335* / .164 / –.119 / –.111 / .143 / –.043
Latino Growth / .025 / .053 / .020 / .042 / .063 / .032 / .016 / .061 / .013
Percent Republican / –.005 / .007 / –.021 / –.013† / .008 / –.059 / –.006 / .007 / –.029
Ln Percent Unemployed / –.626† / .364 / –.055 / –1.101** / .366 / –.094 / –.418 / .334 / –.039
Ln Homicide Rate / .253† / .148 / .060 / .104 / .154 / .024 / –.065 / .134 / –.016
Border State / –.247 / .293 / –.030 / .228 / .346 / .026 / –.043 / .303 / –.006
Adjusted R–squared / .181 / .171 / .239
N / 930 / 929 / 929
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table D3. Analyses Disaggregated by Race (Both Threat Measures Included)

Both IVs Included in the Same Model
IV1 = Latino Political
Threat / IV2 = Latino Economic Threat
Respondent Race and Dependent Variable / b / RSE / Beta / b / RSE / Beta
Whites
Models 1
DV = Expand Police Powers (N = 709) / .132† / .079 / .066 / .124† / .070 / .074
Models 2
DV = Police Profiling (N = 708) / .132 / .101 / .056 / .219* / .089 / .110
Models 3
DV = Police Searches (N = 706) / .170† / .101 / .070 / .104 / .090 / .051
Models 4
DV = Police Use of Force (N = 707) / .095 / .090 / .043 / .050 / .072 / .027
Blacks1
Models 5
DV = Expand Police Powers (N = 109) / .022 / .285 / .012 / .150 / .211 / .099
Models 6
DV = Police Profiling (N = 109) / .218 / .331 / .108 / .100 / .236 / .060
Models 7
DV = Police Searches (N = 109) / –.181 / .315 / –.079 / .131 / .247 / .070
Models 8
DV = Police Use of Force (N = 109) / .030 / .279 / .015 / .220 / .224 / .135
Latinos1
Models 9
DV = Expand Police Powers (N = 76) / .246 / .292 / .124 / .111 / .271 / .064
Models 10
DV = Police Profiling (N = 75) / .009 / .337 / .004 / .240 / .276 / .123
Models 11
DV = Police Searches (N = 76) / .390 / .425 / .163 / –.070 / .358 / –.034
Models 12
DV = Police Use of Force (N = 75) / .254 / .312 / .119 / .105 / .309 / .056
NOTES: Estimates shown are from equations that include all control variables. 1The results for Blacks and Latinos should be interpreted with caution, because of the low sample sizes in the models. In the models for Latinos, there are less than four respondents per variable.
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; DV = dependent variable; IV = Independent variable; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Appendix E:

Supplementary Analyses with Indices (Punitiveness, Symbolic Racism and Perceived Risk) Constructed Using Principle Components Analysis

Table E1. Regression of Support for Expanding Police Powers on Perceived Latino

Political and Economic Threat and Controls (With PCA Indices)

Model 1 / Model 2
Variables / b / RSE / Beta / b / RSE / Beta
Latino Economic Threat / ― / ― / ― / .176** / .056 / .105
Latino Political Threat / .221** / .066 / .111 / ― / ― / ―
Black / –1.025*** / .259 / –.128 / –.973*** / .257 / –.121
Latino / –.348 / .341 / –.037 / –.257 / .342 / –.027
Female / .212 / .153 / .041 / .203 / .155 / .039
Age / –.454* / .199 / –.275 / –.467* / .198 / –.283
Age2 / .068** / .025 / .337 / .071** / .025 / .351
Education / –.067 / .081 / –.029 / –.086 / .082 / –.037
Income / .131* / .061 / .079 / .124* / .061 / .075
Moderate / .066 / .201 / .012 / .042 / .203 / .008
Conservative / .664** / .210 / .126 / .683** / .207 / .130
National News / .195 / .188 / .031 / .215 / .188 / .034
Local News / .173 / .227 / .023 / .157 / .228 / .021
Crime TV / –.069 / .166 / –.013 / –.063 / .164 / –.012
Punitiveness / .452*** / .051 / .322 / .453*** / .051 / .323
Symbolic Racism / .085 / .053 / .051 / .071 / .054 / .043
Perceived Risk / .181*** / .046 / .140 / .178*** / .045 / .138
Household Victim / –.337† / .175 / –.059 / –.319† / .174 / –.055
Ln Percent Latino / –.145 / .121 / –.063 / –.132 / .123 / –.057
Latino Growth / .036 / .047 / .034 / .025 / .049 / .024
Percent Republican / –.007 / .006 / –.040 / –.007 / .006 / –.040
Ln Percent Unemployed / –.702* / .294 / –.073 / –.730* / .296 / –.076
Ln Homicide Rate / .073 / .118 / .020 / .104 / .117 / .029
Border State / –.037 / .271 / –.005 / .012 / .270 / .002
Adjusted R–squared / .250 / .249
N / 912 / 912
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; DV = dependent variable; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table E2. Analysis of the Relationship BetweenPerceived Latino Political Threat and Support for Expanding Specific

Police Powers (With PCA Indices)

Model 1: Police Profiling / Model 2: Police Searches / Model 3: Police Use of Force
Variables / b / RSE / Beta / b / RSE / Beta / b / RSE / Beta
Latino Political Threat / .277** / .082 / .119 / .203* / .080 / .084 / .176* / .077 / .081
Black / –1.002** / .299 / –.106 / –.980** / .336 / –.101 / –1.098*** / .284 / –.124
Latino / –.659† / .371 / –.059 / –.083 / .411 / –.007 / –.390 / .410 / –.037
Female / –.047 / .196 / –.008 / .572** / .197 / .090 / .123 / .170 / .021
Age / –.388 / .273 / –.200 / –.292 / .260 / –.146 / –.687** / .223 / –.379
Age2 / .056† / .034 / .236 / .050 / .032 / .205 / .098*** / .027 / .443
Education / .014 / .101 / .005 / –.131 / .104 / –.047 / –.091 / .094 / –.036
Income / .164* / .072 / .084 / .130† / .077 / .065 / .095 / .070 / .052
Moderate / .089 / .227 / .014 / .068 / .254 / .010 / .034 / .226 / .006
Conservative / .895*** / .249 / .145 / .440 / .269 / .069 / .648** / .238 / .112
National News / .229 / .227 / .031 / .277 / .229 / .036 / .082 / .235 / .012
Local News / –.089 / .290 / –.010 / .410 / .296 / .045 / .160 / .250 / .020
Crime TV / –.168 / .213 / –.026 / –.241 / .207 / –.037 / .209 / .179 / .035
Punitiveness / .413*** / .061 / .251 / .428*** / .063 / .251 / .516*** / .055 / .334
Symbolic Racism / .123† / .066 / .063 / .034 / .068 / .017 / .096 / .061 / .052
Perceived Risk / .132* / .055 / .087 / .236*** / .056 / .151 / .175** / .053 / .124
Household Victim / –.189 / .224 / –.028 / –.566** / .206 / –.081 / –.299 / .197 / –.047
Ln Percent Latino / .026 / .146 / .010 / –.333* / .164 / –.119 / –.116 / .142 / –.046
Latino Growth / .035 / .054 / .028 / .056 / .063 / .043 / .018 / .061 / .015
Percent Republican / –.004 / .007 / –.019 / –.013† / .008 / –.059 / –.004 / .007 / –.022
Ln Percent Unemployed / –.521 / .368 / –.046 / –1.136** / .365 / –.098 / –.427 / .332 / –.040
Ln Homicide Rate / .195 / .150 / .046 / .074 / .150 / .017 / –.057 / .129 / –.014
Border State / –.287 / .300 / –.034 / .230 / .355 / .027 / –.048 / .311 / –.006
Adjusted R–squared / .171 / .173 / .236
N / 910 / 909 / 909
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table E3. Analysis of the Relationship Between Perceived Latino Economic Threat and Support for Expanding Specific

Police Powers (With PCA Indices)

Model 1: Police Profiling / Model 2: Police Searches / Model 3: Police Use of Force
Variables / b / RSE / Beta / b / RSE / Beta / b / RSE / Beta
Latino Economic Threat / .265*** / .067 / .135 / .135† / .072 / .067 / .124* / .061 / .067
Black / –.938** / .299 / –.099 / –.931** / .333 / –.096 / –1.056*** / .282 / –.120
Latino / –.530 / .366 / –.047 / –.009 / .412 / –.001 / –.325 / .412 / –.031
Female / –.049 / .197 / –.008 / .558** / .198 / .088 / .113 / .170 / .020
Age / –.411 / .269 / –.212 / –.299 / .259 / –.149 / –.695** / .223 / –.383
Age2 / .060† / .034 / .255 / .052 / .032 / .213 / .100*** / .027 / .452
Education / –.008 / .104 / –.003 / –.149 / .105 / –.053 / –.107 / .094 / –.042
Income / .159* / .073 / .082 / .120 / .077 / .059 / .087 / .069 / .048
Moderate / .053 / .230 / .008 / .050 / .256 / .008 / .017 / .227 / .003
Conservative / .908*** / .246 / .147 / .466† / .267 / .073 / .668** / .236 / .115
National News / .263 / .226 / .036 / .290 / .230 / .038 / .094 / .234 / .014
Local News / –.115 / .292 / –.013 / .400 / .297 / .044 / .150 / .250 / .018
Crime TV / –.173 / .212 / –.027 / –.230 / .205 / –.035 / .218 / .177 / .036
Punitiveness / .409*** / .062 / .249 / .432*** / .063 / .254 / .519*** / .055 / .336
Symbolic Racism / .102 / .065 / .052 / .024 / .069 / .012 / .087 / .061 / .047
Perceived Risk / .125* / .055 / .083 / .234*** / .055 / .150 / .173** / .053 / .122
Household Victim / –.157 / .225 / –.023 / –.555** / .206 / –.080 / –.289 / .196 / –.046
Ln Percent Latino / .042 / .145 / .015 / –.321† / .165 / –.115 / –.106 / .144 / –.042
Latino Growth / .020 / .055 / .016 / .047 / .064 / .036 / .010 / .063 / .008
Percent Republican / –.004 / .007 / –.021 / –.013† / .008 / –.059 / –.004 / .007 / –.022
Ln Percent Unemployed / –.557 / .368 / –.049 / –1.161** / .367 / –.100 / –.449 / .336 / –.043
Ln Homicide Rate / .233 / .147 / .055 / .103 / .150 / .023 / –.031 / .130 / –.008
Border State / –.224 / .299 / –.027 / .275 / .353 / .032 / –.008 / .311 / –.001
Adjusted R–squared / .175 / .171 / .234
N / 910 / 909 / 909
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table E4. Analyses Disaggregated by Race (With PCA Indices)

IV = Latino Political
Threat / IV = Latino Economic Threat
Respondent Race and Dependent Variable / b / RSE / Beta / b / RSE / Beta
Whites
Models 1–2
DV = Expand Police Powers (N = 693) / .211** / .076 / .105 / .180** / .066 / .107
Models 3–4
DV = Police Profiling (N = 692) / .262** / .096 / .111 / .271** / .084 / .136
Models 5–6
DV = Police Searches (N = 690) / .223* / .095 / .091 / .175* / .084 / .085
Models 7–8
DV = Police Use of Force (N = 691) / .151† / .087 / .069 / .096 / .070 / .052
Blacks1
Models 9–10
DV = Expand Police Powers (N = 108) / .126 / .198 / .070 / .176 / .143 / .120
Models 11–12
DV = Police Profiling (N = 108) / .286 / .230 / .145 / .204 / .154 / .127
Models 13–14
DV = Police Searches (N = 108) / –.089 / .235 / –.040 / .064 / .185 / .035
Models 15–16
DV = Police Use of Force (N = 108) / .181 / .194 / .094 / .260 / .160 / .165
Latinos1
Models 17–18
DV = Expand Police Powers (N = 75) / .287 / .262 / .145 / .248 / .250 / .143
Models 19–20
DV = Police Profiling (N = 74) / .134 / .336 / .060 / .259 / .281 / .133
Models 21–22
DV = Police Searches (N = 75) / .343 / .355 / .144 / .116 / .305 / .055
Models 23–24
DV = Police Use of Force (N = 74) / .273 / .296 / .128 / .286 / .294 / .153
NOTES: Estimates shown are from equations that include all control variables. 1The results for Blacks and Latinos should be interpreted with caution, because of the low sample sizes in the models. In the models for Latinos, there are less than four respondents per variable.
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; DV = dependent variable; IV = Independent variable; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Appendix F:

Supplementary Analyses without Imputing Income

Table F1. Regression of Support for Expanding Police Powers on Perceived Latino

Political and Economic Threat and Controls (Without Imputing Income)

Model 1 / Model 2
Variables / b / RSE / Beta / b / RSE / Beta
Latino Economic Threat / ― / ― / ― / .168** / .057 / .101
Latino Political Threat / .170* / .066 / .086 / ― / ― / ―
Black / –.986*** / .256 / –.125 / –.936*** / .253 / –.119
Latino / –.294 / .336 / –.032 / –.207 / .335 / –.022
Female / .075 / .157 / .014 / .076 / .158 / .015
Age / –.353† / .213 / –.208 / –.371† / .210 / –.219
Age2 / .056* / .027 / .263 / .058* / .026 / .276
Education / –.078 / .084 / –.034 / –.084 / .085 / –.037
Income / .102† / .060 / .064 / .099 / .060 / .063
Moderate / –.032 / .209 / –.006 / –.046 / .211 / –.009
Conservative / .769** / .222 / .146 / .784*** / .220 / .149
National News / .297 / .192 / .047 / .326† / .191 / .052
Local News / .206 / .237 / .027 / .212 / .237 / .028
Crime TV / –.038 / .177 / –.007 / –.035 / .175 / –.006
Punitiveness / .431*** / .048 / .337 / .427*** / .048 / .334
Symbolic Racism / .087 / .087 / .034 / .073 / .087 / .028
Perceived Risk / .193*** / .043 / .162 / .190*** / .042 / .160
Household Victim / –.343† / .180 / –.060 / –.322† / .180 / –.056
Ln Percent Latino / –.160 / .129 / –.070 / –.150 / .130 / –.065
Latino Growth / .046 / .047 / .044 / .038 / .048 / .036
Percent Republican / –.007 / .006 / –.035 / –.007 / .006 / –.036
Ln Percent Unemployed / –.829** / .295 / –.086 / –.857** / .295 / –.089
Ln Homicide Rate / .091 / .121 / .025 / .109 / .120 / .030
Border State / –.035 / .285 / –.005 / –.010 / .285 / –.001
Adjusted R–squared / .267 / .270
N / 832 / 832
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; DV = dependent variable; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table F2. Analysis of the Relationship BetweenPerceived Latino Political Threat and Support for Expanding Specific

Police Powers (Without Imputing Income)

Model 1: Police Profiling / Model 2: Police Searches / Model 3: Police Use of Force
Variables / b / RSE / Beta / b / RSE / Beta / b / RSE / Beta
Latino Political Threat / .206* / .085 / .089 / .158† / .082 / .065 / .138† / .079 / .063
Black / –1.046*** / .289 / –.114 / –.906** / .336 / –.094 / –1.024*** / .291 / –.117
Latino / –.524 / .365 / –.049 / –.099 / .413 / –.009 / –.362 / .397 / –.035
Female / –.166 / .198 / –.027 / .401* / .202 / .063 / .001 / .174 / .000
Age / –.363 / .283 / –.184 / –.076 / .284 / –.037 / –.631* / .249 / –.336
Age2 / .055 / .035 / .224 / .021 / .035 / .083 / .091** / .031 / .390
Education / –.020 / .106 / –.007 / –.118 / .107 / –.042 / –.105 / .097 / –.041
Income / .141† / .072 / .076 / .090 / .078 / .047 / .069 / .069 / .039
Moderate / –.092 / .240 / –.015 / .055 / .267 / .008 / –.061 / .232 / –.010
Conservative / .881** / .265 / .144 / .582* / .285 / .091 / .830** / .245 / .143
National News / .319 / .232 / .043 / .364 / .242 / .047 / .207 / .231 / .030
Local News / –.036 / .303 / –.004 / .475 / .310 / .051 / .170 / .265 / .020
Crime TV / –.174 / .224 / –.027 / –.213 / .222 / –.032 / .290 / .190 / .048
Punitiveness / .424*** / .057 / .285 / .388*** / .060 / .249 / .483*** / .053 / .341
Symbolic Racism / .120 / .106 / .040 / .065 / .110 / .021 / .069 / .100 / .024
Perceived Risk / .155** / .051 / .112 / .239*** / .053 / .165 / .185*** / .051 / .141
Household Victim / –.138 / .228 / –.021 / –.585** / .218 / –.084 / –.329 / .206 / –.052
Ln Percent Latino / –.032 / .151 / –.012 / –.321† / .176 / –.115 / –.119 / .150 / –.047
Latino Growth / .066 / .054 / .053 / .049 / .065 / .038 / .024 / .062 / .020
Percent Republican / –.004 / .008 / –.019 / –.012 / .008 / –.054 / –.003 / .007 / –.015
Ln Percent Unemployed / –.775* / .357 / –.069 / –1.278** / .384 / –.110 / –.436 / .349 / –.041
Ln Homicide Rate / .304* / .151 / .072 / .109 / .158 / .025 / –.136 / .135 / –.034
Border State / –.337 / .316 / –.041 / .209 / .371 / .025 / .038 / .321 / .005
Adjusted R–squared / .197 / .172 / .246
N / 831 / 830 / 830
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table F3. Analysis of the Relationship Between Perceived Latino Economic Threat and Support for Expanding Specific

Police Powers (Without Imputing Income)

Model 1: Police Profiling / Model 2: Police Searches / Model 3: Police Use of Force
Variables / b / RSE / Beta / b / RSE / Beta / b / RSE / Beta
Latino Economic Threat / .254*** / .069 / .131 / .127† / .074 / .062 / .117† / .063 / .064
Black / –.981** / .289 / –.107 / –.863* / .333 / –.090 / –.986** / .289 / –.113
Latino / –.392 / .358 / –.036 / –.031 / .415 / –.003 / –.300 / .399 / –.029
Female / –.155 / .199 / –.026 / .395† / .203 / .062 / –.002 / .175 / .000
Age / –.397 / .279 / –.201 / –.085 / .283 / –.041 / –.641* / .248 / –.342
Age2 / .060† / .035 / .244 / .023 / .035 / .090 / .093** / .031 / .398
Education / –.023 / .108 / –.009 / –.126 / .107 / –.045 / –.112 / .098 / –.044
Income / .143* / .072 / .077 / .085 / .078 / .044 / .065 / .069 / .037
Moderate / –.113 / .243 / –.018 / .044 / .269 / .007 / –.072 / .233 / –.012
Conservative / .892** / .260 / .146 / .602* / .284 / .094 / .845** / .244 / .145
National News / .365 / .229 / .050 / .385 / .243 / .050 / .227 / .229 / .032
Local News / –.028 / .303 / –.003 / .482 / .310 / .052 / .175 / .264 / .021
Crime TV / –.183 / .223 / –.029 / –.205 / .220 / –.031 / .296 / .189 / .049
Punitiveness / .413*** / .057 / .278 / .388*** / .061 / .249 / .482*** / .054 / .340
Symbolic Racism / .097 / .104 / .032 / .055 / .111 / .018 / .059 / .100 / .021
Perceived Risk / .150** / .050 / .108 / .237*** / .053 / .163 / .184*** / .051 / .139
Household Victim / –.103 / .229 / –.015 / –.571** / .219 / –.082 / –.316 / .205 / –.050
Ln Percent Latino / –.021 / .150 / –.008 / –.311† / .177 / –.111 / –.111 / .152 / –.044
Latino Growth / .055 / .054 / .045 / .042 / .065 / .033 / .018 / .063 / .015
Percent Republican / –.005 / .007 / –.022 / –.012 / .008 / –.054 / –.003 / .007 / –.015
Ln Percent Unemployed / –.813* / .354 / –.073 / –1.302** / .385 / –.112 / –.457 / .351 / –.043
Ln Homicide Rate / .320* / .149 / .076 / .130 / .158 / .029 / –.119 / .136 / –.030
Border State / –.305 / .313 / –.037 / .232 / .370 / .027 / .058 / .321 / .007
Adjusted R–squared / .206 / .171 / .246
N / 831 / 830 / 830
ABBREVIATIONS: b = unstandardized coefficient; Beta = standardized coefficient; RSE = robust standard error, clustered by county.
†p < .10; *p < .05; **p < .01; ***p < .001 (two-tailed).

Table F4. Analyses Disaggregated by Race (Without Imputing Income)

IV = Latino Political
Threat / IV = Latino Economic Threat
Respondent Race and Dependent Variable / b / RSE / Beta / b / RSE / Beta