Digit Span Forward - Psychological Aggression
Model 1
DSFlog1 <- glm(PsychAggress ~ DigitSpanFwd, data=Allvars,family = "binomial")
summary(DSFlog1)
##
## Call:
## glm(formula = PsychAggress ~ DigitSpanFwd, family = "binomial",
## data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.2372 0.4749 0.5205 0.5697 0.6799
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.15245 0.39763 2.898 0.00375 **
## DigitSpanFwd 0.09729 0.05325 1.827 0.06768 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 505.40 on 640 degrees of freedom
## Residual deviance: 501.93 on 639 degrees of freedom
## AIC: 505.93
##
## Number of Fisher Scoring iterations: 4
confint(DSFlog1)
## 2.5 % 97.5 %
## (Intercept) 0.374377408 1.9358469
## DigitSpanFwd -0.004952948 0.2041858
exp(cbind(OR = coef(DSFlog1), confint(DSFlog1)))
## OR 2.5 % 97.5 %
## (Intercept) 3.165935 1.4540858 6.929910
## DigitSpanFwd 1.102177 0.9950593 1.226526
########Compare to null model
#Difference in Deviance
with(DSFlog1,null.deviance - deviance)
## [1] 3.471681
#Degrees of freedom for the difference between two models
with(DSFlog1,df.null - df.residual)
## [1] 1
#p-value
with(DSFlog1,pchisq(null.deviance-deviance,df.null-df.residual,lower.tail=FALSE))
## [1] 0.06242784
Model 2
DSFlog2 <- glm(PsychAggress ~ DigitSpanFwd + Age + Sex + PovStat + WRATtotal, data = Allvars, family = "binomial")
summary(DSFlog2)
##
## Call:
## glm(formula = PsychAggress ~ DigitSpanFwd + Age + Sex + PovStat +
## WRATtotal, family = "binomial", data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.3744 0.4082 0.4997 0.5858 0.8451
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.78187 0.95032 1.875 0.0608 .
## DigitSpanFwd 0.07980 0.06021 1.325 0.1850
## Age -0.02014 0.01315 -1.531 0.1257
## SexMen -0.44464 0.23712 -1.875 0.0608 .
## PovStatBelow 0.31763 0.27196 1.168 0.2428
## WRATtotal 0.01382 0.01643 0.841 0.4003
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 505.40 on 640 degrees of freedom
## Residual deviance: 492.71 on 635 degrees of freedom
## AIC: 504.71
##
## Number of Fisher Scoring iterations: 5
confint(DSFlog2)
## 2.5 % 97.5 %
## (Intercept) -0.05975028 3.672887543
## DigitSpanFwd -0.03597106 0.200534414
## Age -0.04613507 0.005539276
## SexMen -0.91505007 0.017156078
## PovStatBelow -0.20190212 0.868656506
## WRATtotal -0.01900186 0.045600639
exp(cbind(OR = coef(DSFlog2), confint(DSFlog2)))
## OR 2.5 % 97.5 %
## (Intercept) 5.9409367 0.9419997 39.365411
## DigitSpanFwd 1.0830707 0.9646682 1.222056
## Age 0.9800581 0.9549130 1.005555
## SexMen 0.6410525 0.4004966 1.017304
## PovStatBelow 1.3738701 0.8171749 2.383706
## WRATtotal 1.0139186 0.9811775 1.046656
########Compare to null model
#Difference in Deviance
with(DSFlog2,null.deviance - deviance)
## [1] 12.68671
#Degrees of freedom for the difference between two models
with(DSFlog2,df.null - df.residual)
## [1] 5
#p-value
with(DSFlog2,pchisq(null.deviance-deviance,df.null-df.residual,lower.tail=FALSE))
## [1] 0.02649835
Model 3
DSFlog3 <- glm(PsychAggress ~ (DigitSpanFwd + Sex + PovStat)^3 + Age + WRATtotal, data = Allvars, family = "binomial")
summary(DSFlog3)
##
## Call:
## glm(formula = PsychAggress ~ (DigitSpanFwd + Sex + PovStat)^3 +
## Age + WRATtotal, family = "binomial", data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.2630 0.4174 0.4864 0.5671 0.8936
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 2.05209 1.11257 1.844 0.0651 .
## DigitSpanFwd 0.04840 0.09789 0.494 0.6210
## SexMen -0.95692 0.94536 -1.012 0.3114
## PovStatBelow 0.19631 1.31460 0.149 0.8813
## Age -0.02016 0.01320 -1.527 0.1267
## WRATtotal 0.01412 0.01657 0.852 0.3939
## DigitSpanFwd:SexMen 0.05684 0.12434 0.457 0.6476
## DigitSpanFwd:PovStatBelow -0.01234 0.17920 -0.069 0.9451
## SexMen:PovStatBelow 0.20818 1.84272 0.113 0.9100
## DigitSpanFwd:SexMen:PovStatBelow 0.02781 0.25559 0.109 0.9134
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 505.40 on 640 degrees of freedom
## Residual deviance: 491.88 on 631 degrees of freedom
## AIC: 511.88
##
## Number of Fisher Scoring iterations: 5
confint(DSFlog3)
## 2.5 % 97.5 %
## (Intercept) -0.09264104 4.279188872
## DigitSpanFwd -0.13709894 0.248989556
## SexMen -2.82517597 0.891305947
## PovStatBelow -2.37156876 2.815163307
## Age -0.04624726 0.005614341
## WRATtotal -0.01894722 0.046169635
## DigitSpanFwd:SexMen -0.18968591 0.299953467
## DigitSpanFwd:PovStatBelow -0.35703285 0.351928834
## SexMen:PovStatBelow -3.43757513 3.815705044
## DigitSpanFwd:SexMen:PovStatBelow -0.47184254 0.537277577
exp(cbind(OR = coef(DSFlog3), confint(DSFlog3)))
## OR 2.5 % 97.5 %
## (Intercept) 7.7841672 0.91152064 72.181868
## DigitSpanFwd 1.0495926 0.87188396 1.282729
## SexMen 0.3840725 0.05929822 2.438312
## PovStatBelow 1.2169073 0.09333419 16.695902
## Age 0.9800401 0.95480585 1.005630
## WRATtotal 1.0142231 0.98123115 1.047252
## DigitSpanFwd:SexMen 1.0584906 0.82721892 1.349796
## DigitSpanFwd:PovStatBelow 0.9877360 0.69974951 1.421807
## SexMen:PovStatBelow 1.2314393 0.03214253 45.408760
## DigitSpanFwd:SexMen:PovStatBelow 1.0281966 0.62385173 1.711342
########Compare to null model
#Difference in Deviance
with(DSFlog3,null.deviance - deviance)
## [1] 13.52276
#Degrees of freedom for the difference between two models
with(DSFlog3,df.null - df.residual)
## [1] 9
#p-value
with(DSFlog3,pchisq(null.deviance-deviance,df.null-df.residual,lower.tail=FALSE))
## [1] 0.1403431
Compare Models 1,2, & 3
anova(DSFlog1,DSFlog2,test = "LR")
## Analysis of Deviance Table
##
## Model 1: PsychAggress ~ DigitSpanFwd
## Model 2: PsychAggress ~ DigitSpanFwd + Age + Sex + PovStat + WRATtotal
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 639 501.93
## 2 635 492.71 4 9.215 0.05594 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
anova(DSFlog2,DSFlog3,test = "LR")
## Analysis of Deviance Table
##
## Model 1: PsychAggress ~ DigitSpanFwd + Age + Sex + PovStat + WRATtotal
## Model 2: PsychAggress ~ (DigitSpanFwd + Sex + PovStat)^3 + Age + WRATtotal
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 635 492.71
## 2 631 491.88 4 0.83605 0.9336
anova(DSFlog1,DSFlog3,test = "LR")
## Analysis of Deviance Table
##
## Model 1: PsychAggress ~ DigitSpanFwd
## Model 2: PsychAggress ~ (DigitSpanFwd + Sex + PovStat)^3 + Age + WRATtotal
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 639 501.93
## 2 631 491.88 8 10.051 0.2615
Suggested Model by Predictors
anova(DSFlog3, test="Chisq")
## Analysis of Deviance Table
##
## Model: binomial, link: logit
##
## Response: PsychAggress
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev Pr(>Chi)
## NULL 640 505.40
## DigitSpanFwd 1 3.4717 639 501.93 0.06243 .
## Sex 1 4.2385 638 497.69 0.03952 *
## PovStat 1 1.9145 637 495.78 0.16647
## Age 1 2.3652 636 493.41 0.12407
## WRATtotal 1 0.6969 635 492.71 0.40381
## DigitSpanFwd:Sex 1 0.2701 634 492.44 0.60330
## DigitSpanFwd:PovStat 1 0.0003 633 492.44 0.98737
## Sex:PovStat 1 0.5539 632 491.89 0.45673
## DigitSpanFwd:Sex:PovStat 1 0.0118 631 491.88 0.91332
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
DSFlog4 <- glm(PsychAggress ~ DigitSpanFwd + Sex, data = Allvars, family = "binomial")
summary(DSFlog4)
##
## Call:
## glm(formula = PsychAggress ~ DigitSpanFwd + Sex, family = "binomial",
## data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.3541 0.4389 0.5084 0.5776 0.7632
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.35916 0.41403 3.283 0.00103 **
## DigitSpanFwd 0.10361 0.05357 1.934 0.05310 .
## SexMen -0.48183 0.23569 -2.044 0.04092 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 505.40 on 640 degrees of freedom
## Residual deviance: 497.69 on 638 degrees of freedom
## AIC: 503.69
##
## Number of Fisher Scoring iterations: 4
Digit Span Forward - Physical Assault
Model 1
DSFlog1 <- glm(PhysAssault ~ DigitSpanFwd, data=Allvars,family = "binomial")
summary(DSFlog1)
##
## Call:
## glm(formula = PhysAssault ~ DigitSpanFwd, family = "binomial",
## data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.5494 -0.5236 -0.5136 -0.5038 2.0889
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.10336 0.40909 -5.141 2.73e-07 ***
## DigitSpanFwd 0.02063 0.05157 0.400 0.689
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 482.53 on 640 degrees of freedom
## Residual deviance: 482.37 on 639 degrees of freedom
## AIC: 486.37
##
## Number of Fisher Scoring iterations: 4
confint(DSFlog1)
## 2.5 % 97.5 %
## (Intercept) -2.91625120 -1.3096827
## DigitSpanFwd -0.08208486 0.1205407
exp(cbind(OR = coef(DSFlog1), confint(DSFlog1)))
## OR 2.5 % 97.5 %
## (Intercept) 0.1220461 0.05413625 0.2699057
## DigitSpanFwd 1.0208459 0.92119379 1.1281067
########Compare to null model
#Difference in Deviance
with(DSFlog1,null.deviance - deviance)
## [1] 0.1590922
#Degrees of freedom for the difference between two models
with(DSFlog1,df.null - df.residual)
## [1] 1
#p-value
with(DSFlog1,pchisq(null.deviance-deviance,df.null-df.residual,lower.tail=FALSE))
## [1] 0.6899937
Model 2
DSFlog2 <- glm(PhysAssault ~ DigitSpanFwd + Age + Sex + PovStat + WRATtotal, data = Allvars, family = "binomial")
summary(DSFlog2)
##
## Call:
## glm(formula = PhysAssault ~ DigitSpanFwd + Age + Sex + PovStat +
## WRATtotal, family = "binomial", data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.8265 -0.5482 -0.4738 -0.3940 2.3908
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.298525 1.038992 -1.250 0.2114
## DigitSpanFwd -0.003763 0.059056 -0.064 0.9492
## Age -0.033411 0.013806 -2.420 0.0155 *
## SexMen -0.154818 0.244320 -0.634 0.5263
## PovStatBelow 0.486848 0.253372 1.921 0.0547 .
## WRATtotal 0.018102 0.018832 0.961 0.3364
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 482.53 on 640 degrees of freedom
## Residual deviance: 469.42 on 635 degrees of freedom
## AIC: 481.42
##
## Number of Fisher Scoring iterations: 5
confint(DSFlog2)
## 2.5 % 97.5 %
## (Intercept) -3.37067738 0.71074063
## DigitSpanFwd -0.12148521 0.11053782
## Age -0.06090493 -0.00666178
## SexMen -0.63831369 0.32239763
## PovStatBelow -0.01492452 0.98116664
## WRATtotal -0.01785196 0.05611976
exp(cbind(OR = coef(DSFlog2), confint(DSFlog2)))
## OR 2.5 % 97.5 %
## (Intercept) 0.2729341 0.03436635 2.0354983
## DigitSpanFwd 0.9962436 0.88560415 1.1168786
## Age 0.9671408 0.94091269 0.9933604
## SexMen 0.8565711 0.52818235 1.3804336
## PovStatBelow 1.6271794 0.98518630 2.6675665
## WRATtotal 1.0182669 0.98230645 1.0577243
########Compare to null model
#Difference in Deviance
with(DSFlog2,null.deviance - deviance)
## [1] 13.11099
#Degrees of freedom for the difference between two models
with(DSFlog2,df.null - df.residual)
## [1] 5
#p-value
with(DSFlog2,pchisq(null.deviance-deviance,df.null-df.residual,lower.tail=FALSE))
## [1] 0.02236057
Model 3
DSFlog3 <- glm(PhysAssault ~ (DigitSpanFwd + Sex + PovStat)^3 + Age + WRATtotal, data = Allvars, family = "binomial")
summary(DSFlog3)
##
## Call:
## glm(formula = PhysAssault ~ (DigitSpanFwd + Sex + PovStat)^3 +
## Age + WRATtotal, family = "binomial", data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.8189 -0.5530 -0.4756 -0.3929 2.4095
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.19141 1.18123 -1.009 0.3132
## DigitSpanFwd -0.01788 0.09490 -0.188 0.8505
## SexMen -0.47145 1.09257 -0.432 0.6661
## PovStatBelow 0.30840 1.14433 0.270 0.7875
## Age -0.03343 0.01381 -2.421 0.0155 *
## WRATtotal 0.01844 0.01896 0.973 0.3306
## DigitSpanFwd:SexMen 0.03636 0.13301 0.273 0.7846
## DigitSpanFwd:PovStatBelow 0.01956 0.14872 0.131 0.8954
## SexMen:PovStatBelow 0.55778 1.74829 0.319 0.7497
## DigitSpanFwd:SexMen:PovStatBelow -0.06336 0.22405 -0.283 0.7773
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 482.53 on 640 degrees of freedom
## Residual deviance: 469.30 on 631 degrees of freedom
## AIC: 489.3
##
## Number of Fisher Scoring iterations: 5
confint(DSFlog3)
## 2.5 % 97.5 %
## (Intercept) -3.55771445 1.082542539
## DigitSpanFwd -0.21071775 0.163745918
## SexMen -2.63659095 1.663594559
## PovStatBelow -1.95436295 2.551997385
## Age -0.06093180 -0.006678362
## WRATtotal -0.01776098 0.056701218
## DigitSpanFwd:SexMen -0.22512922 0.298520850
## DigitSpanFwd:PovStatBelow -0.27523533 0.311318253
## SexMen:PovStatBelow -2.87426110 3.999016174
## DigitSpanFwd:SexMen:PovStatBelow -0.50694634 0.374844062
########Compare to null model
#Difference in Deviance
with(DSFlog3,null.deviance - deviance)
## [1] 13.23718
#Degrees of freedom for the difference between two models
with(DSFlog3,df.null - df.residual)
## [1] 9
#p-value
with(DSFlog3,pchisq(null.deviance-deviance,df.null-df.residual,lower.tail=FALSE))
## [1] 0.1521644
Compare Models 1,2, & 3
anova(DSFlog1,DSFlog2,test = "LR")
## Analysis of Deviance Table
##
## Model 1: PhysAssault ~ DigitSpanFwd
## Model 2: PhysAssault ~ DigitSpanFwd + Age + Sex + PovStat + WRATtotal
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 639 482.37
## 2 635 469.42 4 12.952 0.01151 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
anova(DSFlog2,DSFlog3,test = "LR")
## Analysis of Deviance Table
##
## Model 1: PhysAssault ~ DigitSpanFwd + Age + Sex + PovStat + WRATtotal
## Model 2: PhysAssault ~ (DigitSpanFwd + Sex + PovStat)^3 + Age + WRATtotal
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 635 469.42
## 2 631 469.30 4 0.12619 0.9981
anova(DSFlog1,DSFlog3,test = "LR")
## Analysis of Deviance Table
##
## Model 1: PhysAssault ~ DigitSpanFwd
## Model 2: PhysAssault ~ (DigitSpanFwd + Sex + PovStat)^3 + Age + WRATtotal
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 639 482.37
## 2 631 469.30 8 13.078 0.1092
Suggested Model by predictors
anova(DSFlog3, test="Chisq")
## Analysis of Deviance Table
##
## Model: binomial, link: logit
##
## Response: PhysAssault
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev Pr(>Chi)
## NULL 640 482.53
## DigitSpanFwd 1 0.1591 639 482.37 0.68999
## Sex 1 0.8210 638 481.55 0.36490
## PovStat 1 5.1322 637 476.42 0.02349 *
## Age 1 6.0493 636 470.37 0.01391 *
## WRATtotal 1 0.9494 635 469.42 0.32988
## DigitSpanFwd:Sex 1 0.0134 634 469.41 0.90769
## DigitSpanFwd:PovStat 1 0.0047 633 469.40 0.94509
## Sex:PovStat 1 0.0279 632 469.38 0.86732
## DigitSpanFwd:Sex:PovStat 1 0.0801 631 469.30 0.77717
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
DSFlog4 <- glm(PhysAssault ~Sex + Age, data = Allvars, family = "binomial")
summary(DSFlog4)
##
## Call:
## glm(formula = PhysAssault ~ Sex + Age, family = "binomial", data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.6976 -0.5635 -0.4814 -0.4062 2.3320
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.14711 0.61516 -0.239 0.81100
## SexMen -0.16634 0.24267 -0.685 0.49306
## Age -0.03795 0.01357 -2.797 0.00516 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 482.53 on 640 degrees of freedom
## Residual deviance: 473.65 on 638 degrees of freedom
## AIC: 479.65
##
## Number of Fisher Scoring iterations: 5
Digit Span Backward - Psychological Aggression
Model 1
DSBlog1 <- glm(PsychAggress ~ DigitSpanBck, data=Allvars,family = "binomial")
summary(DSBlog1)
##
## Call:
## glm(formula = PsychAggress ~ DigitSpanBck, family = "binomial",
## data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.1240 0.5008 0.5328 0.5666 0.6395
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.48331 0.32547 4.557 5.18e-06 ***
## DigitSpanBck 0.06618 0.05387 1.228 0.219
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 505.40 on 640 degrees of freedom
## Residual deviance: 503.85 on 639 degrees of freedom
## AIC: 507.85
##
## Number of Fisher Scoring iterations: 4
confint(DSBlog1)
## 2.5 % 97.5 %
## (Intercept) 0.84949262 2.1271670
## DigitSpanBck -0.03708871 0.1744529
exp(cbind(OR = coef(DSBlog1), confint(DSBlog1)))
## OR 2.5 % 97.5 %
## (Intercept) 4.407523 2.3384601 8.391061
## DigitSpanBck 1.068418 0.9635906 1.190595
########Compare to null model
#Difference in Deviance
with(DSBlog1,null.deviance - deviance)
## [1] 1.553456
#Degrees of freedom for the difference between two models
with(DSBlog1,df.null - df.residual)
## [1] 1
#p-value
with(DSBlog1,pchisq(null.deviance-deviance,df.null-df.residual,lower.tail=FALSE))
## [1] 0.2126259
Model 2
DSBlog2 <- glm(PsychAggress ~ DigitSpanBck + Age + Sex + PovStat + WRATtotal, data = Allvars, family = "binomial")
summary(DSBlog2)
##
## Call:
## glm(formula = PsychAggress ~ DigitSpanBck + Age + Sex + PovStat +
## WRATtotal, family = "binomial", data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.3545 0.4188 0.4988 0.5806 0.8067
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 2.00097 0.93371 2.143 0.0321 *
## DigitSpanBck 0.03633 0.06183 0.588 0.5569
## Age -0.02133 0.01313 -1.625 0.1042
## SexMen -0.43391 0.23663 -1.834 0.0667 .
## PovStatBelow 0.30340 0.27136 1.118 0.2635
## WRATtotal 0.01871 0.01690 1.107 0.2681
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 505.40 on 640 degrees of freedom
## Residual deviance: 494.17 on 635 degrees of freedom
## AIC: 506.17
##
## Number of Fisher Scoring iterations: 4
confint(DSBlog2)
## 2.5 % 97.5 %
## (Intercept) 0.19390855 3.861708942
## DigitSpanBck -0.08303013 0.159844631
## Age -0.04727809 0.004313792
## SexMen -0.90331488 0.027038043
## PovStatBelow -0.21507588 0.853305067
## WRATtotal -0.01488368 0.051545135
exp(cbind(OR = coef(DSBlog2), confint(DSBlog2)))
## OR 2.5 % 97.5 %
## (Intercept) 7.3961984 1.2139853 47.546536
## DigitSpanBck 1.0369930 0.9203234 1.173329
## Age 0.9788925 0.9538221 1.004323
## SexMen 0.6479681 0.4052242 1.027407
## PovStatBelow 1.3544571 0.8064802 2.347392
## WRATtotal 1.0188890 0.9852265 1.052897
########Compare to null model
#Difference in Deviance
with(DSBlog2,null.deviance - deviance)
## [1] 11.23034
#Degrees of freedom for the difference between two models
with(DSBlog2,df.null - df.residual)
## [1] 5
#p-value
with(DSBlog2,pchisq(null.deviance-deviance,df.null-df.residual,lower.tail=FALSE))
## [1] 0.04699944
Model 3
DSBlog3 <- glm(PsychAggress ~ (DigitSpanBck + Sex + PovStat)^3 + Age + WRATtotal, data = Allvars, family = "binomial")
summary(DSBlog3)
##
## Call:
## glm(formula = PsychAggress ~ (DigitSpanBck + Sex + PovStat)^3 +
## Age + WRATtotal, family = "binomial", data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.3576 0.3941 0.4862 0.5675 0.9101
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.86140 1.04549 1.780 0.0750 .
## DigitSpanBck 0.09323 0.11541 0.808 0.4192
## SexMen -0.62912 0.78860 -0.798 0.4250
## PovStatBelow 1.65113 1.19085 1.387 0.1656
## Age -0.02285 0.01323 -1.728 0.0841 .
## WRATtotal 0.01775 0.01714 1.036 0.3003
## DigitSpanBck:SexMen 0.01376 0.13382 0.103 0.9181
## DigitSpanBck:PovStatBelow -0.27451 0.19329 -1.420 0.1555
## SexMen:PovStatBelow 0.18041 1.54837 0.117 0.9072
## DigitSpanBck:SexMen:PovStatBelow 0.03102 0.24940 0.124 0.9010
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 505.40 on 640 degrees of freedom
## Residual deviance: 489.29 on 631 degrees of freedom
## AIC: 509.29
##
## Number of Fisher Scoring iterations: 5
confint(DSBlog3)
## 2.5 % 97.5 %
## (Intercept) -0.15637002 3.95271949
## DigitSpanBck -0.12406374 0.32908387
## SexMen -2.18681354 0.91147349
## PovStatBelow -0.63382657 4.07184741
## Age -0.04900142 0.00297188
## WRATtotal -0.01629415 0.05105948
## DigitSpanBck:SexMen -0.25325179 0.27307703
## DigitSpanBck:PovStatBelow -0.65478254 0.10995568
## SexMen:PovStatBelow -2.88490444 3.21551659
## DigitSpanBck:SexMen:PovStatBelow -0.45930083 0.52358557
exp(cbind(OR = coef(DSBlog3), confint(DSBlog3)))
## OR 2.5 % 97.5 %
## (Intercept) 6.4327376 0.85524267 52.076797
## DigitSpanBck 1.0977113 0.88332354 1.389694
## SexMen 0.5330596 0.11227394 2.487986
## PovStatBelow 5.2128734 0.53055769 58.665242
## Age 0.9774056 0.95217978 1.002976
## WRATtotal 1.0179052 0.98383788 1.052385
## DigitSpanBck:SexMen 1.0138573 0.77627240 1.314001
## DigitSpanBck:PovStatBelow 0.7599464 0.51955503 1.116229
## SexMen:PovStatBelow 1.1977140 0.05586013 24.916160
## DigitSpanBck:SexMen:PovStatBelow 1.0315050 0.63172517 1.688070
########Compare to null model
#Difference in Deviance
with(DSBlog3,null.deviance - deviance)
## [1] 16.11477
#Degrees of freedom for the difference between two models
with(DSBlog3,df.null - df.residual)
## [1] 9
#p-value
with(DSBlog3,pchisq(null.deviance-deviance,df.null-df.residual,lower.tail=FALSE))
## [1] 0.06452314
Compare Models 1,2, & 3
anova(DSBlog1,DSBlog2,test = "LR")
## Analysis of Deviance Table
##
## Model 1: PsychAggress ~ DigitSpanBck
## Model 2: PsychAggress ~ DigitSpanBck + Age + Sex + PovStat + WRATtotal
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 639 503.85
## 2 635 494.17 4 9.6769 0.04624 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
anova(DSBlog2,DSBlog3,test = "LR")
## Analysis of Deviance Table
##
## Model 1: PsychAggress ~ DigitSpanBck + Age + Sex + PovStat + WRATtotal
## Model 2: PsychAggress ~ (DigitSpanBck + Sex + PovStat)^3 + Age + WRATtotal
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 635 494.17
## 2 631 489.29 4 4.8844 0.2994
anova(DSBlog1,DSBlog3,test = "LR")
## Analysis of Deviance Table
##
## Model 1: PsychAggress ~ DigitSpanBck
## Model 2: PsychAggress ~ (DigitSpanBck + Sex + PovStat)^3 + Age + WRATtotal
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 639 503.85
## 2 631 489.29 8 14.561 0.06826 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Suggested Model by Predictors
anova(DSBlog3, test="Chisq")
## Analysis of Deviance Table
##
## Model: binomial, link: logit
##
## Response: PsychAggress
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev Pr(>Chi)
## NULL 640 505.40
## DigitSpanBck 1 1.5535 639 503.85 0.21263
## Sex 1 4.0437 638 499.80 0.04434 *
## PovStat 1 1.7275 637 498.08 0.18874
## Age 1 2.6974 636 495.38 0.10051
## WRATtotal 1 1.2083 635 494.17 0.27168
## DigitSpanBck:Sex 1 0.1052 634 494.07 0.74568
## DigitSpanBck:PovStat 1 4.3122 633 489.75 0.03784 *
## Sex:PovStat 1 0.4516 632 489.30 0.50158
## DigitSpanBck:Sex:PovStat 1 0.0155 631 489.29 0.90101
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
DSBlog4 <- glm(PsychAggress ~ (DigitSpanBck + PovStat)^2 + Sex, data = Allvars, family = "binomial")
summary(DSBlog4)
##
## Call:
## glm(formula = PsychAggress ~ (DigitSpanBck + PovStat)^2 + Sex,
## family = "binomial", data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.3307 0.4135 0.4907 0.5743 0.7848
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.21783 0.39486 3.084 0.00204 **
## DigitSpanBck 0.13974 0.06319 2.211 0.02701 *
## PovStatBelow 1.77644 0.75525 2.352 0.01867 *
## SexMen -0.47738 0.23636 -2.020 0.04342 *
## DigitSpanBck:PovStatBelow -0.25530 0.12136 -2.104 0.03540 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 505.4 on 640 degrees of freedom
## Residual deviance: 493.8 on 636 degrees of freedom
## AIC: 503.8
##
## Number of Fisher Scoring iterations: 4
sim_slopes(DSBlog4 , pred = DigitSpanBck, modx = PovStat, centered = "all",jnplot = TRUE)
## Warning: Johnson-Neyman intervals are not available for factor moderators.
## SIMPLE SLOPES ANALYSIS
##
## Slope of DigitSpanBck when PovStat = Below:
##
## Est. S.E. z val. p
## ------- ------ -------- ------
## -0.12 0.10 -1.12 0.26
##
## Slope of DigitSpanBck when PovStat = Above:
##
## Est. S.E. z val. p
## ------ ------ -------- ------
## 0.14 0.06 2.21 0.03
interact_plot(model = DSBlog4 , pred = DigitSpanBck, modx = PovStat)
Digit Span Backward - Physical Assault
Model 1
DSBlog1 <- glm(PhysAssault ~ DigitSpanBck, data=Allvars,family = "binomial")
summary(DSBlog1)
##
## Call:
## glm(formula = PhysAssault ~ DigitSpanBck, family = "binomial",
## data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.6461 -0.5332 -0.4995 -0.4834 2.2166
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.36700 0.33815 -7.000 2.56e-12 ***
## DigitSpanBck 0.06971 0.05143 1.355 0.175
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 482.53 on 640 degrees of freedom
## Residual deviance: 480.73 on 639 degrees of freedom
## AIC: 484.73
##
## Number of Fisher Scoring iterations: 4
confint(DSBlog1)
## 2.5 % 97.5 %
## (Intercept) -3.04161374 -1.7138403
## DigitSpanBck -0.03277605 0.1693352
exp(cbind(OR = coef(DSBlog1), confint(DSBlog1)))
## OR 2.5 % 97.5 %
## (Intercept) 0.09376143 0.04775776 0.1801725
## DigitSpanBck 1.07220115 0.96775526 1.1845171
########Compare to null model
#Difference in Deviance
with(DSBlog1,null.deviance - deviance)
## [1] 1.797797
#Degrees of freedom for the difference between two models
with(DSBlog1,df.null - df.residual)
## [1] 1
#p-value
with(DSBlog1,pchisq(null.deviance-deviance,df.null-df.residual,lower.tail=FALSE))
## [1] 0.179979
Model 2
DSBlog2 <- glm(PhysAssault ~ DigitSpanBck + Age + Sex + PovStat + WRATtotal, data = Allvars, family = "binomial")
summary(DSBlog2)
##
## Call:
## glm(formula = PhysAssault ~ DigitSpanBck + Age + Sex + PovStat +
## WRATtotal, family = "binomial", data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.8505 -0.5507 -0.4663 -0.3877 2.4000
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.345694 1.024586 -1.313 0.1890
## DigitSpanBck 0.059014 0.060632 0.973 0.3304
## Age -0.031851 0.013777 -2.312 0.0208 *
## SexMen -0.170212 0.244791 -0.695 0.4868
## PovStatBelow 0.502315 0.253971 1.978 0.0479 *
## WRATtotal 0.008856 0.019220 0.461 0.6450
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 482.53 on 640 degrees of freedom
## Residual deviance: 468.49 on 635 degrees of freedom
## AIC: 480.49
##
## Number of Fisher Scoring iterations: 5
confint(DSBlog2)
## 2.5 % 97.5 %
## (Intercept) -3.391665037 0.633726337
## DigitSpanBck -0.061320692 0.176938898
## Age -0.059299715 -0.005165805
## SexMen -0.654805236 0.307756720
## PovStatBelow -0.000432283 0.998046083
## WRATtotal -0.027993557 0.047521446
exp(cbind(OR = coef(DSBlog2), confint(DSBlog2)))
## OR 2.5 % 97.5 %
## (Intercept) 0.2603591 0.0336526 1.8846202
## DigitSpanBck 1.0607902 0.9405216 1.1935582
## Age 0.9686509 0.9424243 0.9948475
## SexMen 0.8434860 0.5195432 1.3603700
## PovStatBelow 1.6525430 0.9995678 2.7129757
## WRATtotal 1.0088955 0.9723946 1.0486687
########Compare to null model
#Difference in Deviance
with(DSBlog2,null.deviance - deviance)
## [1] 14.04369
#Degrees of freedom for the difference between two models
with(DSBlog2,df.null - df.residual)
## [1] 5
#p-value
with(DSBlog2,pchisq(null.deviance-deviance,df.null-df.residual,lower.tail=FALSE))
## [1] 0.01533425
Model 3
DSBlog3 <- glm(PhysAssault ~ (DigitSpanBck + Sex + PovStat)^3 + Age + WRATtotal, data = Allvars, family = "binomial")
summary(DSBlog3)
##
## Call:
## glm(formula = PhysAssault ~ (DigitSpanBck + Sex + PovStat)^3 +
## Age + WRATtotal, family = "binomial", data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.8330 -0.5503 -0.4685 -0.3884 2.4253
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.255613 1.120611 -1.120 0.2625
## DigitSpanBck 0.044520 0.107315 0.415 0.6782
## SexMen -0.455390 0.908308 -0.501 0.6161
## PovStatBelow 0.408942 1.027509 0.398 0.6906
## Age -0.031862 0.013778 -2.313 0.0207 *
## WRATtotal 0.009189 0.019285 0.476 0.6337
## DigitSpanBck:SexMen 0.038519 0.133091 0.289 0.7723
## DigitSpanBck:PovStatBelow 0.008996 0.165178 0.054 0.9566
## SexMen:PovStatBelow 0.440727 1.437951 0.306 0.7592
## DigitSpanBck:SexMen:PovStatBelow -0.056578 0.225189 -0.251 0.8016
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 482.53 on 640 degrees of freedom
## Residual deviance: 468.32 on 631 degrees of freedom
## AIC: 488.32
##
## Number of Fisher Scoring iterations: 5
confint(DSBlog3)
## 2.5 % 97.5 %
## (Intercept) -3.50061722 0.901710583
## DigitSpanBck -0.17306222 0.249958353
## SexMen -2.25539642 1.319990640
## PovStatBelow -1.63003035 2.417495430
## Age -0.05931304 -0.005176632
## WRATtotal -0.02780734 0.047973668
## DigitSpanBck:SexMen -0.22130482 0.302664130
## DigitSpanBck:PovStatBelow -0.31741438 0.334113806
## SexMen:PovStatBelow -2.38215259 3.271157548
## DigitSpanBck:SexMen:PovStatBelow -0.50186032 0.384406408
exp(cbind(OR = coef(DSBlog3), confint(DSBlog3)))
## OR 2.5 % 97.5 %
## (Intercept) 0.2849011 0.03017875 2.4638141
## DigitSpanBck 1.0455263 0.84108528 1.2839719
## SexMen 0.6342003 0.10483198 3.7433863
## PovStatBelow 1.5052248 0.19592363 11.2177285
## Age 0.9686402 0.94241171 0.9948367
## WRATtotal 1.0092314 0.97257573 1.0491430
## DigitSpanBck:SexMen 1.0392708 0.80147234 1.3534598
## DigitSpanBck:PovStatBelow 1.0090362 0.72802901 1.3967021
## SexMen:PovStatBelow 1.5538360 0.09235157 26.3418136
## DigitSpanBck:SexMen:PovStatBelow 0.9449929 0.60540337 1.4687422
########Compare to null model
#Difference in Deviance
with(DSBlog3,null.deviance - deviance)
## [1] 14.20855
#Degrees of freedom for the difference between two models
with(DSBlog3,df.null - df.residual)
## [1] 9
#p-value
with(DSBlog3,pchisq(null.deviance-deviance,df.null-df.residual,lower.tail=FALSE))
## [1] 0.1150977
Compare Models 1,2, & 3
anova(DSBlog1,DSBlog2,test = "LR")
## Analysis of Deviance Table
##
## Model 1: PhysAssault ~ DigitSpanBck
## Model 2: PhysAssault ~ DigitSpanBck + Age + Sex + PovStat + WRATtotal
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 639 480.73
## 2 635 468.49 4 12.246 0.01561 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
anova(DSBlog2,DSBlog3,test = "LR")
## Analysis of Deviance Table
##
## Model 1: PhysAssault ~ DigitSpanBck + Age + Sex + PovStat + WRATtotal
## Model 2: PhysAssault ~ (DigitSpanBck + Sex + PovStat)^3 + Age + WRATtotal
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 635 468.49
## 2 631 468.32 4 0.16486 0.9968
anova(DSBlog1,DSBlog3,test = "LR")
## Analysis of Deviance Table
##
## Model 1: PhysAssault ~ DigitSpanBck
## Model 2: PhysAssault ~ (DigitSpanBck + Sex + PovStat)^3 + Age + WRATtotal
## Resid. Df Resid. Dev Df Deviance Pr(>Chi)
## 1 639 480.73
## 2 631 468.32 8 12.411 0.1338
Suggested Model by Predictors
anova(DSBlog3, test="Chisq")
## Analysis of Deviance Table
##
## Model: binomial, link: logit
##
## Response: PhysAssault
##
## Terms added sequentially (first to last)
##
##
## Df Deviance Resid. Df Resid. Dev Pr(>Chi)
## NULL 640 482.53
## DigitSpanBck 1 1.7978 639 480.73 0.17998
## Sex 1 0.9489 638 479.79 0.33001
## PovStat 1 5.5965 637 474.19 0.01800 *
## Age 1 5.4858 636 468.70 0.01917 *
## WRATtotal 1 0.2147 635 468.49 0.64310
## DigitSpanBck:Sex 1 0.0270 634 468.46 0.86958
## DigitSpanBck:PovStat 1 0.0336 633 468.43 0.85458
## Sex:PovStat 1 0.0411 632 468.39 0.83932
## DigitSpanBck:Sex:PovStat 1 0.0632 631 468.32 0.80152
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
DSBlog4 <- glm(PhysAssault ~ PovStat + Age, data = Allvars, family = "binomial")
summary(DSBlog4)
##
## Call:
## glm(formula = PhysAssault ~ PovStat + Age, family = "binomial",
## data = Allvars)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -0.7492 -0.5539 -0.4770 -0.3956 2.3402
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.53355 0.64202 -0.831 0.4059
## PovStatBelow 0.45113 0.24862 1.815 0.0696 .
## Age -0.03471 0.01372 -2.530 0.0114 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 482.53 on 640 degrees of freedom
## Residual deviance: 470.90 on 638 degrees of freedom
## AIC: 476.9
##
## Number of Fisher Scoring iterations: 5