load(file='/Users/meganwilliams/Desktop/Research/HANDLS Data/MeasurementData.rdata')
load(file='/Users/meganwilliams/Desktop/Research/HANDLS Data/MeasurementData1.rdata')
Model (1) with three-ways
Model1.1 = lm(MCLcountry~Age0 + Race + Sex + Educ + Employment01 + acasiincomx01 + Neighborhood02 + CES + Race*Sex*Employment01 + Race*Employment01 + Sex*Employment01 + Race*Sex*Educ + Sex*Educ + Race*Educ + Race*Sex*acasiincomx01 + Race*acasiincomx01 + Sex*acasiincomx01 + Race*Sex*Neighborhood02 + Race*Neighborhood02 + Sex*Neighborhood02 + Race*Sex*CES + Race*CES + Sex*CES + Race*Sex, MeasurementData,x=T)
summary(Model1.1)
##
## Call:
## lm(formula = MCLcountry ~ Age0 + Race + Sex + Educ + Employment01 +
## acasiincomx01 + Neighborhood02 + CES + Race * Sex * Employment01 +
## Race * Employment01 + Sex * Employment01 + Race * Sex * Educ +
## Sex * Educ + Race * Educ + Race * Sex * acasiincomx01 + Race *
## acasiincomx01 + Sex * acasiincomx01 + Race * Sex * Neighborhood02 +
## Race * Neighborhood02 + Sex * Neighborhood02 + Race * Sex *
## CES + Race * CES + Sex * CES + Race * Sex, data = MeasurementData,
## x = T)
##
## Residuals:
## MCL: SES Standing in country
## Min 1Q Median 3Q Max
## -5.342 -1.500 -0.078 1.715 6.861
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.09675 0.63370 8.04 1.4e-15
## Age0 0.00660 0.00446 1.48 0.139
## Race1 -1.05136 0.84141 -1.25 0.212
## Sex1 -0.64027 0.86037 -0.74 0.457
## Educ 0.02075 0.03023 0.69 0.493
## Employment011 -0.14729 0.15734 -0.94 0.349
## acasiincomx01 -0.15154 0.14474 -1.05 0.295
## Neighborhood02 -0.05882 0.07127 -0.83 0.409
## CES -0.04213 0.00640 -6.58 6.0e-11
## Race1:Sex1 0.42472 1.24949 0.34 0.734
## Race1:Employment011 0.28747 0.24331 1.18 0.238
## Sex1:Employment011 0.40830 0.23554 1.73 0.083
## Race1:Educ 0.07095 0.04105 1.73 0.084
## Sex1:Educ 0.04197 0.04584 0.92 0.360
## Race1:acasiincomx01 0.05286 0.22879 0.23 0.817
## Sex1:acasiincomx01 -0.12011 0.21677 -0.55 0.580
## Race1:Neighborhood02 -0.14375 0.10364 -1.39 0.166
## Sex1:Neighborhood02 -0.09044 0.10788 -0.84 0.402
## Race1:CES -0.00320 0.00962 -0.33 0.740
## Sex1:CES 0.01339 0.01062 1.26 0.208
## Race1:Sex1:Employment011 0.30743 0.38091 0.81 0.420
## Race1:Sex1:Educ -0.01619 0.06103 -0.27 0.791
## Race1:Sex1:acasiincomx01 -0.32773 0.35934 -0.91 0.362
## Race1:Sex1:Neighborhood02 0.10932 0.15943 0.69 0.493
## Race1:Sex1:CES -0.00340 0.01610 -0.21 0.833
##
## Residual standard error: 1.86 on 2116 degrees of freedom
## Multiple R-squared: 0.147, Adjusted R-squared: 0.138
## F-statistic: 15.2 on 24 and 2116 DF, p-value: <2e-16
Model (1) with two-ways
Model1.2 = lm(MCLcountry~Age0 + Race + Sex + Educ + Employment01 + acasiincomx01 + Neighborhood02 + CES + Race * Educ + Race * Employment01 + Race * acasiincomx01 + Race * Neighborhood02 + Race * CES + Sex * Educ + Sex * Employment01 + Sex * acasiincomx01 + Sex * Neighborhood02 + Sex * CES + Race * Sex, MeasurementData, x=T)
summary(Model1.2)
##
## Call:
## lm(formula = MCLcountry ~ Age0 + Race + Sex + Educ + Employment01 +
## acasiincomx01 + Neighborhood02 + CES + Race * Educ + Race *
## Employment01 + Race * acasiincomx01 + Race * Neighborhood02 +
## Race * CES + Sex * Educ + Sex * Employment01 + Sex * acasiincomx01 +
## Sex * Neighborhood02 + Sex * CES + Race * Sex, data = MeasurementData,
## x = T)
##
## Residuals:
## MCL: SES Standing in country
## Min 1Q Median 3Q Max
## -5.314 -1.168 -0.137 1.092 6.767
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.07193 0.56618 8.96 < 2e-16
## Age0 0.00656 0.00444 1.48 0.1396
## Race1 -0.95609 0.62850 -1.52 0.1284
## Sex1 -0.55533 0.62308 -0.89 0.3729
## Educ 0.02310 0.02619 0.88 0.3779
## Employment011 -0.19770 0.14306 -1.38 0.1671
## acasiincomx01 -0.10277 0.13215 -0.78 0.4368
## Neighborhood02 -0.07828 0.06370 -1.23 0.2193
## CES -0.04158 0.00587 -7.09 1.9e-12
## Race1:Educ 0.06357 0.03030 2.10 0.0360
## Race1:Employment011 0.41521 0.18678 2.22 0.0263
## Race1:acasiincomx01 -0.08103 0.17584 -0.46 0.6450
## Race1:Neighborhood02 -0.09896 0.07860 -1.26 0.2081
## Race1:CES -0.00432 0.00769 -0.56 0.5747
## Sex1:Educ 0.03514 0.02998 1.17 0.2412
## Sex1:Employment011 0.52698 0.18496 2.85 0.0044
## Sex1:acasiincomx01 -0.23762 0.17243 -1.38 0.1683
## Sex1:Neighborhood02 -0.04431 0.07913 -0.56 0.5755
## Sex1:CES 0.01137 0.00796 1.43 0.1531
## Race1:Sex1 0.23443 0.17074 1.37 0.1699
##
## Residual standard error: 1.86 on 2121 degrees of freedom
## Multiple R-squared: 0.146, Adjusted R-squared: 0.138
## F-statistic: 19.1 on 19 and 2121 DF, p-value: <2e-16
Model (1) with no interactions
Model1.3 = lm(MCLcountry~Age0 + Race + Sex + Educ + Employment01 + acasiincomx01 + Neighborhood02 + CES + Race * Educ + Race * Employment01 + Sex * Employment01 , MeasurementData,x=T)
summary(Model1.3)
##
## Call:
## lm(formula = MCLcountry ~ Age0 + Race + Sex + Educ + Employment01 +
## acasiincomx01 + Neighborhood02 + CES + Race * Educ + Race *
## Employment01 + Sex * Employment01, data = MeasurementData,
## x = T)
##
## Residuals:
## MCL: SES Standing in country
## Min 1Q Median 3Q Max
## -5.305 -1.244 -0.865 1.714 6.761
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.34450 0.43518 12.28 < 2e-16
## Age0 0.00692 0.00444 1.56 0.11937
## Race1 -1.59908 0.35463 -4.51 6.9e-06
## Sex1 -0.43472 0.12757 -3.41 0.00067
## Educ 0.03052 0.02216 1.38 0.16850
## Employment011 -0.27346 0.13483 -2.03 0.04266
## acasiincomx01 -0.22577 0.08495 -2.66 0.00793
## Neighborhood02 -0.14547 0.03921 -3.71 0.00021
## CES -0.03984 0.00383 -10.40 < 2e-16
## Race1:Educ 0.07865 0.02844 2.77 0.00574
## Race1:Employment011 0.49394 0.17320 2.85 0.00439
## Sex1:Employment011 0.66202 0.16635 3.98 7.1e-05
##
## Residual standard error: 1.86 on 2129 degrees of freedom
## Multiple R-squared: 0.141, Adjusted R-squared: 0.137
## F-statistic: 31.8 on 11 and 2129 DF, p-value: <2e-16
Model 1 Plots
library(effects)
## Loading required package: colorspace
##
## Attaching package: 'effects'
##
## The following object is masked from 'package:car':
##
## Prestige
plot(effect("Sex:Employment01", Model1.3, list(wt=c(2.2,3.2,4.2))), multiline=TRUE)
plot(effect("Race:Employment01", Model1.3, list(wt=c(2.2,3.2,4.2))), multiline=TRUE)
plot(effect("Race:Educ", Model1.3, list(wt=c(2.2,3.2,4.2))), multiline=TRUE)
load(file='/Users/meganwilliams/Desktop/Research/HANDLS Data/MeasurementData.rdata')
load(file='/Users/meganwilliams/Desktop/Research/HANDLS Data/MeasurementData1.rdata')
Model (2) with three-ways
Model2.1 = lm(MCLcountry~Age0 + Race + Sex + Educ + Employment01 + acasiincomx01 + Neighborhood02 + CES + SF01 + Race*Sex*Employment01 + Race*Employment01 + Sex*Employment01 + Race*Sex*Educ + Race*Educ + Sex*Educ + Race*Sex*acasiincomx01 + Race*acasiincomx01 + Sex*acasiincomx01 + Race*Sex*Neighborhood02 + Race*Neighborhood02 + Sex*Neighborhood02 + Race*Sex*CES + Race*CES + Sex*CES + Race*Sex + Race*Sex*SF01 + Race*SF01 + Sex*SF01, MeasurementData1,x=T)
summary(Model2.1)
##
## Call:
## lm(formula = MCLcountry ~ Age0 + Race + Sex + Educ + Employment01 +
## acasiincomx01 + Neighborhood02 + CES + SF01 + Race * Sex *
## Employment01 + Race * Employment01 + Sex * Employment01 +
## Race * Sex * Educ + Race * Educ + Sex * Educ + Race * Sex *
## acasiincomx01 + Race * acasiincomx01 + Sex * acasiincomx01 +
## Race * Sex * Neighborhood02 + Race * Neighborhood02 + Sex *
## Neighborhood02 + Race * Sex * CES + Race * CES + Sex * CES +
## Race * Sex + Race * Sex * SF01 + Race * SF01 + Sex * SF01,
## data = MeasurementData1, x = T)
##
## Residuals:
## MCL: SES Standing in country
## Min 1Q Median 3Q Max
## -5.402 -1.226 -0.638 1.123 6.729
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.73131 0.68217 8.40 < 2e-16
## Age0 0.00610 0.00452 1.35 0.178
## Race1 -0.61210 0.94698 -0.65 0.518
## Sex1 -0.08560 0.95670 -0.09 0.929
## Educ 0.01835 0.03062 0.60 0.549
## Employment011 -0.23924 0.16178 -1.48 0.139
## acasiincomx01 -0.31666 0.16503 -1.92 0.055
## Neighborhood02 -0.01404 0.07278 -0.19 0.847
## CES -0.04076 0.00657 -6.20 6.7e-10
## SF01Fair -0.53060 0.31406 -1.69 0.091
## SF01Good -0.64267 0.29099 -2.21 0.027
## SF01Poor -0.40809 0.46353 -0.88 0.379
## SF01vGood -0.20674 0.30304 -0.68 0.495
## Race1:Sex1 -0.91795 1.40513 -0.65 0.514
## Race1:Employment011 0.21425 0.25133 0.85 0.394
## Sex1:Employment011 0.40239 0.24336 1.65 0.098
## Race1:Educ 0.05174 0.04181 1.24 0.216
## Sex1:Educ 0.03702 0.04625 0.80 0.424
## Race1:acasiincomx01 -0.00623 0.25782 -0.02 0.981
## Sex1:acasiincomx01 -0.03934 0.24749 -0.16 0.874
## Race1:Neighborhood02 -0.16073 0.10553 -1.52 0.128
## Sex1:Neighborhood02 -0.12472 0.10950 -1.14 0.255
## Race1:CES 0.00315 0.01000 0.32 0.752
## Sex1:CES 0.01737 0.01086 1.60 0.110
## Race1:SF01Fair -0.29352 0.49339 -0.59 0.552
## Race1:SF01Good 0.03207 0.43684 0.07 0.941
## Race1:SF01Poor -1.13728 0.66071 -1.72 0.085
## Race1:SF01vGood -0.23212 0.44948 -0.52 0.606
## Sex1:SF01Fair -0.68542 0.46660 -1.47 0.142
## Sex1:SF01Good -0.52118 0.42037 -1.24 0.215
## Sex1:SF01Poor -1.18668 0.73709 -1.61 0.108
## Sex1:SF01vGood -0.64824 0.44151 -1.47 0.142
## Race1:Sex1:Employment011 0.47201 0.39457 1.20 0.232
## Race1:Sex1:Educ 0.00614 0.06173 0.10 0.921
## Race1:Sex1:acasiincomx01 -0.13476 0.40116 -0.34 0.737
## Race1:Sex1:Neighborhood02 0.12326 0.16245 0.76 0.448
## Race1:Sex1:CES -0.01437 0.01667 -0.86 0.389
## Race1:Sex1:SF01Fair 1.25868 0.72282 1.74 0.082
## Race1:Sex1:SF01Good 0.74826 0.64043 1.17 0.243
## Race1:Sex1:SF01Poor 2.23094 1.06659 2.09 0.037
## Race1:Sex1:SF01vGood 0.81245 0.65895 1.23 0.218
##
## Residual standard error: 1.84 on 2036 degrees of freedom
## Multiple R-squared: 0.168, Adjusted R-squared: 0.152
## F-statistic: 10.3 on 40 and 2036 DF, p-value: <2e-16
Model (2) with two-ways and significant three-ways
Model2.2 = lm(MCLcountry~Age0 + Race + Sex + Educ + Employment01 + acasiincomx01 + Neighborhood02 + CES + Race * Educ + Race * Employment01 + Race * acasiincomx01 + Race * Neighborhood02 + Race * CES + Race*SF01 + Sex * Educ + Sex * Employment01 + Sex * acasiincomx01 + Sex * Neighborhood02 + Sex * CES + Sex*SF01 + Race * Sex + Race*Sex*SF01, MeasurementData1, x=T)
summary(Model2.2)
##
## Call:
## lm(formula = MCLcountry ~ Age0 + Race + Sex + Educ + Employment01 +
## acasiincomx01 + Neighborhood02 + CES + Race * Educ + Race *
## Employment01 + Race * acasiincomx01 + Race * Neighborhood02 +
## Race * CES + Race * SF01 + Sex * Educ + Sex * Employment01 +
## Sex * acasiincomx01 + Sex * Neighborhood02 + Sex * CES +
## Sex * SF01 + Race * Sex + Race * Sex * SF01, data = MeasurementData1,
## x = T)
##
## Residuals:
## MCL: SES Standing in country
## Min 1Q Median 3Q Max
## -5.367 -1.429 -0.076 1.204 6.620
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.85840 0.62020 9.45 < 2e-16
## Age0 0.00591 0.00451 1.31 0.1898
## Race1 -0.84637 0.74739 -1.13 0.2576
## Sex1 -0.29614 0.73599 -0.40 0.6875
## Educ 0.01464 0.02655 0.55 0.5814
## Employment011 -0.31226 0.14730 -2.12 0.0341
## acasiincomx01 -0.29971 0.15019 -2.00 0.0461
## Neighborhood02 -0.03805 0.06504 -0.58 0.5586
## CES -0.03876 0.00603 -6.43 1.6e-10
## SF01Fair -0.55846 0.31235 -1.79 0.0739
## SF01Good -0.64491 0.29033 -2.22 0.0264
## SF01Poor -0.46229 0.46116 -1.00 0.3162
## SF01vGood -0.21078 0.30266 -0.70 0.4862
## Race1:Educ 0.05668 0.03070 1.85 0.0650
## Race1:Employment011 0.40123 0.19335 2.08 0.0381
## Race1:acasiincomx01 -0.06521 0.19686 -0.33 0.7405
## Race1:Neighborhood02 -0.10842 0.08004 -1.35 0.1757
## Race1:CES -0.00204 0.00798 -0.26 0.7982
## Race1:SF01Fair -0.15910 0.48151 -0.33 0.7411
## Race1:SF01Good 0.07497 0.43413 0.17 0.8629
## Race1:SF01Poor -0.95109 0.64714 -1.47 0.1418
## Race1:SF01vGood -0.19595 0.44845 -0.44 0.6622
## Sex1:Educ 0.04082 0.03045 1.34 0.1802
## Sex1:Employment011 0.58051 0.19143 3.03 0.0025
## Sex1:acasiincomx01 -0.08421 0.19439 -0.43 0.6649
## Sex1:Neighborhood02 -0.07311 0.08068 -0.91 0.3649
## Sex1:CES 0.01132 0.00822 1.38 0.1690
## Sex1:SF01Fair -0.57413 0.45985 -1.25 0.2120
## Sex1:SF01Good -0.48413 0.41854 -1.16 0.2475
## Sex1:SF01Poor -1.03863 0.73132 -1.42 0.1557
## Sex1:SF01vGood -0.61812 0.44069 -1.40 0.1609
## Race1:Sex1 -0.41579 0.57450 -0.72 0.4693
## Race1:Sex1:SF01Fair 0.94190 0.68518 1.37 0.1694
## Race1:Sex1:SF01Good 0.64296 0.63091 1.02 0.3083
## Race1:Sex1:SF01Poor 1.78549 1.02828 1.74 0.0826
## Race1:Sex1:SF01vGood 0.77158 0.65670 1.17 0.2402
##
## Residual standard error: 1.84 on 2041 degrees of freedom
## Multiple R-squared: 0.167, Adjusted R-squared: 0.152
## F-statistic: 11.7 on 35 and 2041 DF, p-value: <2e-16
Model(2) with only significant interactions
Model2.3 = lm(MCLcountry~Age0 + Race + Sex + Educ + Employment01 + acasiincomx01 + Neighborhood02 + CES + Race * Educ + Race * Employment01 + Sex * Employment01 , MeasurementData1,x=T)
summary(Model2.3)
##
## Call:
## lm(formula = MCLcountry ~ Age0 + Race + Sex + Educ + Employment01 +
## acasiincomx01 + Neighborhood02 + CES + Race * Educ + Race *
## Employment01 + Sex * Employment01, data = MeasurementData1,
## x = T)
##
## Residuals:
## MCL: SES Standing in country
## Min 1Q Median 3Q Max
## -5.331 -1.529 -0.438 1.408 6.786
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.67638 0.44386 12.79 < 2e-16
## Age0 0.00398 0.00448 0.89 0.37444
## Race1 -1.61200 0.36084 -4.47 8.3e-06
## Sex1 -0.41459 0.12973 -3.20 0.00142
## Educ 0.02800 0.02246 1.25 0.21268
## Employment011 -0.31011 0.13651 -2.27 0.02320
## acasiincomx01 -0.33941 0.09571 -3.55 0.00040
## Neighborhood02 -0.14314 0.03964 -3.61 0.00031
## CES -0.03965 0.00387 -10.25 < 2e-16
## Race1:Educ 0.07804 0.02876 2.71 0.00671
## Race1:Employment011 0.49787 0.17430 2.86 0.00433
## Sex1:Employment011 0.67174 0.16805 4.00 6.6e-05
##
## Residual standard error: 1.85 on 2065 degrees of freedom
## Multiple R-squared: 0.147, Adjusted R-squared: 0.142
## F-statistic: 32.4 on 11 and 2065 DF, p-value: <2e-16
Model 2 Plots
library(effects)
plot(effect("Sex:Employment01", Model2.3, list(wt=c(2.2,3.2,4.2))), multiline=TRUE)
plot(effect("Race:Employment01", Model2.3, list(wt=c(2.2,3.2,4.2))), multiline=TRUE)
plot(effect("Race:Educ", Model2.3, list(wt=c(2.2,3.2,4.2))), multiline=TRUE)
load(file='/Users/meganwilliams/Desktop/Research/HANDLS Data/MeasurementData.rdata')
load(file='/Users/meganwilliams/Desktop/Research/HANDLS Data/MeasurementData1.rdata')
Model (3) with three-ways
Model3.1 = lm(MCLcountry~Age0 + Race + Sex + Educ + Employment01 + acasiincomx01 + Neighborhood02 + CES + sHealth + Race*Sex*Employment01 + Race*Employment01 + Sex*Employment01 + Race*Sex*Educ + Race*Educ + Sex*Educ + Race*Sex*acasiincomx01 + Race*acasiincomx01 + Sex*acasiincomx01 + Race*Sex*Neighborhood02 + Race*Neighborhood02 + Sex*Neighborhood02 + Race*Sex*CES + Race*CES + Sex*CES + Race*Sex + Race*Sex*sHealth + Race*sHealth + Sex*sHealth, MeasurementData1,x=T)
summary(Model3.1)
##
## Call:
## lm(formula = MCLcountry ~ Age0 + Race + Sex + Educ + Employment01 +
## acasiincomx01 + Neighborhood02 + CES + sHealth + Race * Sex *
## Employment01 + Race * Employment01 + Sex * Employment01 +
## Race * Sex * Educ + Race * Educ + Sex * Educ + Race * Sex *
## acasiincomx01 + Race * acasiincomx01 + Sex * acasiincomx01 +
## Race * Sex * Neighborhood02 + Race * Neighborhood02 + Sex *
## Neighborhood02 + Race * Sex * CES + Race * CES + Sex * CES +
## Race * Sex + Race * Sex * sHealth + Race * sHealth + Sex *
## sHealth, data = MeasurementData1, x = T)
##
## Residuals:
## MCL: SES Standing in country
## Min 1Q Median 3Q Max
## -5.413 -1.424 -0.649 1.316 6.773
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.565137 0.649412 8.57 < 2e-16
## Age0 0.006380 0.004523 1.41 0.1586
## Race1 -0.858885 0.873598 -0.98 0.3256
## Sex1 -0.670226 0.893407 -0.75 0.4532
## Educ 0.017642 0.030588 0.58 0.5642
## Employment011 -0.233139 0.161373 -1.44 0.1487
## acasiincomx01 -0.314342 0.165193 -1.90 0.0572
## Neighborhood02 -0.016810 0.072750 -0.23 0.8173
## CES -0.040562 0.006568 -6.18 7.9e-10
## sHealthLow -0.352570 0.197977 -1.78 0.0751
## sHealthMiddle -0.481895 0.169186 -2.85 0.0044
## Race1:Sex1 -0.168538 1.293572 -0.13 0.8964
## Race1:Employment011 0.244699 0.250796 0.98 0.3293
## Sex1:Employment011 0.453590 0.242360 1.87 0.0614
## Race1:Educ 0.055437 0.041707 1.33 0.1839
## Sex1:Educ 0.038522 0.046257 0.83 0.4051
## Race1:acasiincomx01 0.020166 0.257829 0.08 0.9377
## Sex1:acasiincomx01 -0.014352 0.247607 -0.06 0.9538
## Race1:Neighborhood02 -0.160644 0.105584 -1.52 0.1283
## Sex1:Neighborhood02 -0.115465 0.109504 -1.05 0.2918
## Race1:CES 0.000757 0.009947 0.08 0.9393
## Sex1:CES 0.016829 0.010856 1.55 0.1213
## Race1:sHealthLow -0.283742 0.322573 -0.88 0.3792
## Race1:sHealthMiddle 0.209928 0.262201 0.80 0.4234
## Sex1:sHealthLow -0.264385 0.308139 -0.86 0.3910
## Sex1:sHealthMiddle -0.047221 0.252276 -0.19 0.8515
## Race1:Sex1:Employment011 0.396372 0.393441 1.01 0.3138
## Race1:Sex1:Educ 0.001965 0.061664 0.03 0.9746
## Race1:Sex1:acasiincomx01 -0.170436 0.400744 -0.43 0.6707
## Race1:Sex1:Neighborhood02 0.114077 0.162548 0.70 0.4829
## Race1:Sex1:CES -0.011883 0.016638 -0.71 0.4752
## Race1:Sex1:sHealthLow 0.823827 0.490850 1.68 0.0934
## Race1:Sex1:sHealthMiddle 0.146064 0.395698 0.37 0.7121
##
## Residual standard error: 1.84 on 2044 degrees of freedom
## Multiple R-squared: 0.162, Adjusted R-squared: 0.149
## F-statistic: 12.4 on 32 and 2044 DF, p-value: <2e-16
Model (3) with two-ways
Model3.2 = lm(MCLcountry~Age0 + Race + Sex + Educ + Employment01 + acasiincomx01 + Neighborhood02 + CES + sHealth + Race * Educ + Race * Employment01 + Race * acasiincomx01 + Race * Neighborhood02 + Race * CES + Race*sHealth + Sex * Educ + Sex * Employment01 + Sex * acasiincomx01 + Sex * Neighborhood02 + Sex * CES + Sex*sHealth + Race * Sex, MeasurementData1, x=T)
summary(Model3.2)
##
## Call:
## lm(formula = MCLcountry ~ Age0 + Race + Sex + Educ + Employment01 +
## acasiincomx01 + Neighborhood02 + CES + sHealth + Race * Educ +
## Race * Employment01 + Race * acasiincomx01 + Race * Neighborhood02 +
## Race * CES + Race * sHealth + Sex * Educ + Sex * Employment01 +
## Sex * acasiincomx01 + Sex * Neighborhood02 + Sex * CES +
## Sex * sHealth + Race * Sex, data = MeasurementData1, x = T)
##
## Residuals:
## MCL: SES Standing in country
## Min 1Q Median 3Q Max
## -5.310 -1.138 -0.636 1.170 6.808
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.667816 0.580837 9.76 < 2e-16
## Age0 0.006125 0.004508 1.36 0.1744
## Race1 -1.060060 0.650633 -1.63 0.1034
## Sex1 -0.837575 0.647246 -1.29 0.1958
## Educ 0.016552 0.026438 0.63 0.5313
## Employment011 -0.290292 0.146822 -1.98 0.0482
## acasiincomx01 -0.290572 0.150100 -1.94 0.0530
## Neighborhood02 -0.037581 0.065004 -0.58 0.5632
## CES -0.039034 0.006025 -6.48 1.2e-10
## sHealthLow -0.470297 0.181062 -2.60 0.0095
## sHealthMiddle -0.497687 0.152459 -3.26 0.0011
## Race1:Educ 0.058689 0.030654 1.91 0.0557
## Race1:Employment011 0.401558 0.192529 2.09 0.0371
## Race1:acasiincomx01 -0.055197 0.196607 -0.28 0.7789
## Race1:Neighborhood02 -0.117662 0.079968 -1.47 0.1413
## Race1:CES -0.003719 0.007946 -0.47 0.6398
## Race1:sHealthLow 0.057775 0.242453 0.24 0.8117
## Race1:sHealthMiddle 0.267438 0.195653 1.37 0.1718
## Sex1:Educ 0.037076 0.030244 1.23 0.2204
## Sex1:Employment011 0.596623 0.190658 3.13 0.0018
## Sex1:acasiincomx01 -0.077974 0.194046 -0.40 0.6878
## Sex1:Neighborhood02 -0.062600 0.080614 -0.78 0.4375
## Sex1:CES 0.012379 0.008199 1.51 0.1313
## Sex1:sHealthLow 0.047624 0.238837 0.20 0.8420
## Sex1:sHealthMiddle 0.000622 0.193555 0.00 0.9974
## Race1:Sex1 0.314666 0.171512 1.83 0.0667
##
## Residual standard error: 1.84 on 2051 degrees of freedom
## Multiple R-squared: 0.16, Adjusted R-squared: 0.15
## F-statistic: 15.7 on 25 and 2051 DF, p-value: <2e-16
Model (3) with only significant interactions
Model3.3 = lm(MCLcountry~Age0 + Race + Sex + Educ + Employment01 + acasiincomx01 + Neighborhood02 + CES + sHealth + Race * Educ + Race * Employment01 + Sex * Employment01 , MeasurementData1,x=T)
summary(Model3.3)
##
## Call:
## lm(formula = MCLcountry ~ Age0 + Race + Sex + Educ + Employment01 +
## acasiincomx01 + Neighborhood02 + CES + sHealth + Race * Educ +
## Race * Employment01 + Sex * Employment01, data = MeasurementData1,
## x = T)
##
## Residuals:
## MCL: SES Standing in country
## Min 1Q Median 3Q Max
## -5.192 -1.071 0.358 1.816 6.800
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.72466 0.44257 12.93 < 2e-16
## Age0 0.00634 0.00450 1.41 0.15916
## Race1 -1.50678 0.36023 -4.18 3.0e-05
## Sex1 -0.39813 0.12929 -3.08 0.00210
## Educ 0.02681 0.02239 1.20 0.23126
## Employment011 -0.32309 0.13668 -2.36 0.01818
## acasiincomx01 -0.32957 0.09545 -3.45 0.00057
## Neighborhood02 -0.12130 0.03988 -3.04 0.00238
## CES -0.03688 0.00395 -9.35 < 2e-16
## sHealthLow -0.40700 0.11889 -3.42 0.00063
## sHealthMiddle -0.38487 0.09630 -4.00 6.7e-05
## Race1:Educ 0.06985 0.02871 2.43 0.01506
## Race1:Employment011 0.48126 0.17375 2.77 0.00566
## Sex1:Employment011 0.64193 0.16756 3.83 0.00013
##
## Residual standard error: 1.84 on 2063 degrees of freedom
## Multiple R-squared: 0.155, Adjusted R-squared: 0.149
## F-statistic: 29 on 13 and 2063 DF, p-value: <2e-16
Model 3 Plots
library(effects)
plot(effect("Sex:Employment01", Model3.3, list(wt=c(2.2,3.2,4.2))), multiline=TRUE)
plot(effect("Race:Employment01", Model3.3, list(wt=c(2.2,3.2,4.2))), multiline=TRUE)
plot(effect("Race:Educ", Model3.3, list(wt=c(2.2,3.2,4.2))), multiline=TRUE)
