Trails A Regression Model 1 for Women Only
## Loading required package: Matrix
## KernSmooth 2.23 loaded
## Copyright M. P. Wand 1997-2009
##
## Attaching package: 'lmerTest'
##
## The following object is masked from 'package:lme4':
##
## lmer
##
## The following object is masked from 'package:stats':
##
## step
## Warning: number of observations <= rank(Z); variance-covariance matrix
## will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: log(TrailsAtestSec) ~ (Age + IPVstatus + PovStat)^3 + (Age | HNDid) + (1 | subclass)
## Data: WomenlogTrailsA
## REML criterion at convergence: 67.56
## Random effects:
## Groups Name Std.Dev. Corr
## HNDid (Intercept) 0.3355
## Age 0.0187 1.00
## subclass (Intercept) 0.2156
## Residual 0.2262
## Number of obs: 72, groups: HNDid, 36; subclass, 19
## Fixed Effects:
## (Intercept) Age
## 3.524032 0.017918
## IPVstatus1 PovStatBelow
## -0.001744 -0.069160
## Age:IPVstatus1 Age:PovStatBelow
## -0.000511 -0.010751
## IPVstatus1:PovStatBelow Age:IPVstatus1:PovStatBelow
## 0.683933 0.079896
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Random term (Age | HNDid) was eliminated because of having correlation +-1 or NaN
## Warning: number of observations <= rank(Z); variance-covariance matrix
## will be unidentifiable
## Random term (Age + 0 | HNDid) was eliminated because of standard deviation being equal to 0
##
## Random effects:
## Chi.sq Chi.DF elim.num p.value
## (1 | HNDid) 1.34 1 1 0.2474
## (1 | subclass) 11.97 1 kept 0.0005
##
## Fixed effects:
## Sum Sq Mean Sq NumDF DenDF F.value elim.num Pr(>F)
## Age:IPVstatus:PovStat 0.2401 0.2401 1 63.43 3.3611 1 0.0714
## Age:IPVstatus 0.0050 0.0050 1 62.78 0.0483 2 0.8268
## IPVstatus:PovStat 0.0069 0.0069 1 65.84 0.0907 3 0.7642
## Age:PovStat 0.0040 0.0040 1 64.31 0.1056 4 0.7463
## PovStat 0.0048 0.0048 1 67.22 0.0284 5 0.8667
## IPVstatus 0.0123 0.0123 1 58.14 0.1740 6 0.6781
## Age 0.8661 0.8661 1 60.07 12.6077 kept 0.0008
##
## Least squares means:
## Estimate Standard Error DF t-value Lower CI Upper CI p-value
##
## Differences of LSMEANS:
## Estimate Standard Error DF t-value Lower CI Upper CI p-value
##
## Final model:
## lme4::lmer(formula = log(TrailsAtestSec) ~ Age + (1 | subclass),
## data = WomenlogTrailsA, REML = reml, contrasts = l)
Re-run final Model 1
(mm1 = lmer(log(TrailsAtestSec) ~ Age + (Age | HNDid) + (1 | subclass), data = WomenlogTrailsA))
## Warning: number of observations <= rank(Z); variance-covariance matrix
## will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: log(TrailsAtestSec) ~ Age + (Age | HNDid) + (1 | subclass)
## Data: WomenlogTrailsA
## REML criterion at convergence: 47.28
## Random effects:
## Groups Name Std.Dev. Corr
## HNDid (Intercept) 0.2718
## Age 0.0153 1.00
## subclass (Intercept) 0.2158
## Residual 0.2334
## Number of obs: 72, groups: HNDid, 36; subclass, 19
## Fixed Effects:
## (Intercept) Age
## 3.5469 0.0192
summary(mm1)
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: log(TrailsAtestSec) ~ Age + (Age | HNDid) + (1 | subclass)
## Data: WomenlogTrailsA
##
## REML criterion at convergence: 47.28
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## HNDid (Intercept) 0.073899 0.2718
## Age 0.000233 0.0153 1.00
## subclass (Intercept) 0.046561 0.2158
## Residual 0.054481 0.2334
## Number of obs: 72, groups: HNDid, 36; subclass, 19
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 3.54689 0.09715 17.56000 36.51 <2e-16
## Age 0.01922 0.00733 27.73000 2.62 0.014
##
## Correlation of Fixed Effects:
## (Intr)
## Age 0.760
plot(st)
plot(mm1)
Trails A Regression Model 2 for Women Only (with CES)
load("~/Desktop/Megan/Research/IPV and Cognition Paper/IPV R Output/WomenlogTrailsA.rda")
library(lme4)
library(lmerTest)
(mm2 = lmer(log(TrailsAtestSec) ~ (Age + IPVstatus + PovStat + CES1)^4 + (Age |
HNDid) + (1 | subclass), data = WomenlogTrailsA))
## Warning: number of observations <= rank(Z); variance-covariance matrix
## will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: log(TrailsAtestSec) ~ (Age + IPVstatus + PovStat + CES1)^4 + (Age | HNDid) + (1 | subclass)
## Data: WomenlogTrailsA
## REML criterion at convergence: 79.23
## Random effects:
## Groups Name Std.Dev. Corr
## HNDid (Intercept) 0.3210
## Age 0.0194 1.00
## subclass (Intercept) 0.2528
## Residual 0.2081
## Number of obs: 72, groups: HNDid, 36; subclass, 19
## Fixed Effects:
## (Intercept) Age
## 3.34361 0.00713
## IPVstatus1 PovStatBelow
## 0.48466 0.03894
## CES11 Age:IPVstatus1
## 0.57356 0.04702
## Age:PovStatBelow Age:CES11
## -0.00748 0.03446
## IPVstatus1:PovStatBelow IPVstatus1:CES11
## 1.32961 -1.09698
## PovStatBelow:CES11 Age:IPVstatus1:PovStatBelow
## -0.49998 0.11102
## Age:IPVstatus1:CES11 Age:PovStatBelow:CES11
## -0.10494 -0.02199
## IPVstatus1:PovStatBelow:CES11 Age:IPVstatus1:PovStatBelow:CES11
## -0.22577 0.01375
(st = step(mm2))
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Random term (Age | HNDid) was eliminated because of having correlation +-1 or NaN
## Warning: number of observations <= rank(Z); variance-covariance matrix
## will be unidentifiable
## Random term (Age + 0 | HNDid) was eliminated because of standard deviation being equal to 0
##
## Random effects:
## Chi.sq Chi.DF elim.num p.value
## (1 | HNDid) 0.00 1 1 1e+00
## (1 | subclass) 14.79 1 kept 1e-04
##
## Fixed effects:
## Sum Sq Mean Sq NumDF DenDF F.value elim.num
## Age:IPVstatus:PovStat:CES1 0.0018 0.0018 1 43.42 0.0313 1
## Age:PovStat:CES1 0.0349 0.0349 1 56.84 0.5779 2
## IPVstatus:PovStat:CES1 0.0458 0.0458 1 55.77 0.8762 3
## Age 0.7939 0.7939 1 55.49 15.9484 kept
## IPVstatus 0.0114 0.0114 1 52.02 6.7854 kept
## PovStat 0.0165 0.0165 1 57.18 2.0943 kept
## CES1 0.0469 0.0469 1 57.93 1.3719 kept
## Age:IPVstatus 0.0004 0.0004 1 51.21 6.3637 kept
## Age:PovStat 0.0115 0.0115 1 48.87 2.9951 kept
## Age:CES1 0.0048 0.0048 1 58.95 1.5720 kept
## IPVstatus:PovStat 0.0113 0.0113 1 58.39 11.5130 kept
## IPVstatus:CES1 0.0430 0.0430 1 52.32 10.1216 kept
## PovStat:CES1 0.2029 0.2029 1 51.79 6.0791 kept
## Age:IPVstatus:PovStat 0.3743 0.3743 1 56.42 9.5752 kept
## Age:IPVstatus:CES1 0.4767 0.4767 1 48.77 8.4274 kept
## Pr(>F)
## Age:IPVstatus:PovStat:CES1 0.8604
## Age:PovStat:CES1 0.4503
## IPVstatus:PovStat:CES1 0.3533
## Age 0.0002
## IPVstatus 0.0120
## PovStat 0.1533
## CES1 0.2463
## Age:IPVstatus 0.0148
## Age:PovStat 0.0898
## Age:CES1 0.2149
## IPVstatus:PovStat 0.0012
## IPVstatus:CES1 0.0025
## PovStat:CES1 0.0170
## Age:IPVstatus:PovStat 0.0031
## Age:IPVstatus:CES1 0.0055
##
## Least squares means:
## IPVstatus PovStat CES1 Estimate Standard Error
## IPVstatus 0 1.0 NA NA 3.3944 0.0821
## IPVstatus 1 2.0 NA NA 3.4981 0.1000
## PovStat Above NA 1.0 NA 3.4344 0.0827
## PovStat Below NA 2.0 NA 3.4582 0.1043
## CES1 0 NA NA 1.0 3.4612 0.0957
## CES1 1 NA NA 2.0 3.4313 0.0973
## IPVstatus:PovStat 0 Above 1.0 1.0 NA 3.4778 0.0888
## IPVstatus:PovStat 1 Above 2.0 1.0 NA 3.3911 0.0970
## IPVstatus:PovStat 0 Below 1.0 2.0 NA 3.3111 0.1147
## IPVstatus:PovStat 1 Below 2.0 2.0 NA 3.6052 0.1512
## IPVstatus:CES1 0 0 1.0 NA 1.0 3.3363 0.0909
## IPVstatus:CES1 1 0 2.0 NA 1.0 3.5862 0.1361
## IPVstatus:CES1 0 1 1.0 NA 2.0 3.4526 0.1085
## IPVstatus:CES1 1 1 2.0 NA 2.0 3.4101 0.1119
## PovStat:CES1 Above 0 NA 1.0 1.0 3.3511 0.0972
## PovStat:CES1 Below 0 NA 2.0 1.0 3.5714 0.1360
## PovStat:CES1 Above 1 NA 1.0 2.0 3.5178 0.1028
## PovStat:CES1 Below 1 NA 2.0 2.0 3.3449 0.1219
## DF t-value Lower CI Upper CI p-value
## IPVstatus 0 16.1 41.3 3.22 3.57 <2e-16
## IPVstatus 1 30.8 35.0 3.29 3.70 <2e-16
## PovStat Above 16.3 41.5 3.26 3.61 <2e-16
## PovStat Below 30.3 33.2 3.25 3.67 <2e-16
## CES1 0 26.7 36.2 3.26 3.66 <2e-16
## CES1 1 24.3 35.3 3.23 3.63 <2e-16
## IPVstatus:PovStat 0 Above 20.2 39.2 3.29 3.66 <2e-16
## IPVstatus:PovStat 1 Above 29.1 35.0 3.19 3.59 <2e-16
## IPVstatus:PovStat 0 Below 30.4 28.9 3.08 3.55 <2e-16
## IPVstatus:PovStat 1 Below 54.8 23.8 3.30 3.91 <2e-16
## IPVstatus:CES1 0 0 20.6 36.7 3.15 3.53 <2e-16
## IPVstatus:CES1 1 0 51.2 26.3 3.31 3.86 <2e-16
## IPVstatus:CES1 0 1 32.3 31.8 3.23 3.67 <2e-16
## IPVstatus:CES1 1 1 39.0 30.5 3.18 3.64 <2e-16
## PovStat:CES1 Above 0 24.8 34.5 3.15 3.55 <2e-16
## PovStat:CES1 Below 0 47.4 26.3 3.30 3.85 <2e-16
## PovStat:CES1 Above 1 27.3 34.2 3.31 3.73 <2e-16
## PovStat:CES1 Below 1 44.2 27.4 3.10 3.59 <2e-16
##
## Differences of LSMEANS:
## Estimate Standard Error DF t-value
## IPVstatus 0-1 -0.1 0.0884 54.2 -1.17
## PovStat Above-Below 0.0 0.0988 59.0 -0.24
## CES1 0-1 0.0 0.1077 58.8 0.28
## IPVstatus:PovStat 0 Above- 1 Above 0.1 0.0848 48.4 1.02
## IPVstatus:PovStat 0 Above- 0 Below 0.2 0.1230 56.2 1.35
## IPVstatus:PovStat 0 Above- 1 Below -0.1 0.1418 53.4 -0.90
## IPVstatus:PovStat 1 Above- 0 Below 0.1 0.1228 58.3 0.65
## IPVstatus:PovStat 1 Above- 1 Below -0.2 0.1567 55.1 -1.37
## IPVstatus:PovStat 0 Below- 1 Below -0.3 0.1691 57.8 -1.74
## IPVstatus:CES1 0 0- 1 0 -0.2 0.1304 56.9 -1.92
## IPVstatus:CES1 0 0- 0 1 -0.1 0.1144 59.0 -1.02
## IPVstatus:CES1 0 0- 1 1 -0.1 0.1168 58.3 -0.63
## IPVstatus:CES1 1 0- 0 1 0.1 0.1586 57.1 0.84
## IPVstatus:CES1 1 0- 1 1 0.2 0.1487 55.2 1.18
## IPVstatus:CES1 0 1- 1 1 0.0 0.1033 44.9 0.41
## PovStat:CES1 Above 0- Below 0 -0.2 0.1389 59.0 -1.59
## PovStat:CES1 Above 0- Above 1 -0.2 0.1124 58.8 -1.48
## PovStat:CES1 Above 0- Below 1 0.0 0.1372 58.8 0.04
## PovStat:CES1 Below 0- Above 1 0.1 0.1545 58.9 0.35
## PovStat:CES1 Below 0- Below 1 0.2 0.1525 54.3 1.49
## PovStat:CES1 Above 1- Below 1 0.2 0.1139 50.4 1.52
## Lower CI Upper CI p-value
## IPVstatus 0-1 -0.2810 0.0735 0.25
## PovStat Above-Below -0.2215 0.1740 0.81
## CES1 0-1 -0.1855 0.2453 0.78
## IPVstatus:PovStat 0 Above- 1 Above -0.0839 0.2573 0.31
## IPVstatus:PovStat 0 Above- 0 Below -0.0798 0.4131 0.18
## IPVstatus:PovStat 0 Above- 1 Below -0.4118 0.1569 0.37
## IPVstatus:PovStat 1 Above- 0 Below -0.1658 0.3257 0.52
## IPVstatus:PovStat 1 Above- 1 Below -0.5281 0.0998 0.18
## IPVstatus:PovStat 0 Below- 1 Below -0.6326 0.0444 0.09
## IPVstatus:CES1 0 0- 1 0 -0.5109 0.0111 0.06
## IPVstatus:CES1 0 0- 0 1 -0.3453 0.1127 0.31
## IPVstatus:CES1 0 0- 1 1 -0.3076 0.1600 0.53
## IPVstatus:CES1 1 0- 0 1 -0.1841 0.4513 0.40
## IPVstatus:CES1 1 0- 1 1 -0.1219 0.4741 0.24
## IPVstatus:CES1 0 1- 1 1 -0.1656 0.2506 0.68
## PovStat:CES1 Above 0- Below 0 -0.4982 0.0575 0.12
## PovStat:CES1 Above 0- Above 1 -0.3917 0.0582 0.14
## PovStat:CES1 Above 0- Below 1 -0.2685 0.2808 0.96
## PovStat:CES1 Below 0- Above 1 -0.2556 0.3629 0.73
## PovStat:CES1 Below 0- Below 1 -0.0792 0.5322 0.14
## PovStat:CES1 Above 1- Below 1 -0.0558 0.4016 0.14
##
## Final model:
## lme4::lmer(formula = log(TrailsAtestSec) ~ Age + IPVstatus +
## PovStat + CES1 + (1 | subclass) + Age:IPVstatus + Age:PovStat +
## Age:CES1 + IPVstatus:PovStat + IPVstatus:CES1 + PovStat:CES1 +
## Age:IPVstatus:PovStat + Age:IPVstatus:CES1, data = WomenlogTrailsA,
## REML = reml, contrasts = l)
Re-run suggested final Model 2
(mm2 = lmer(log(TrailsAtestSec) ~ Age + IPVstatus + PovStat + CES1 + (Age |
HNDid) + (1 | subclass) + Age:IPVstatus + Age:PovStat + Age:CES1 + IPVstatus:PovStat +
IPVstatus:CES1 + PovStat:CES1 + Age:IPVstatus:PovStat + Age:IPVstatus:CES1,
data = WomenlogTrailsA, contrasts = 1))
## Warning: number of observations <= rank(Z); variance-covariance matrix
## will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: log(TrailsAtestSec) ~ Age + IPVstatus + PovStat + CES1 + (Age | HNDid) + (1 | subclass) + Age:IPVstatus + Age:PovStat + Age:CES1 + IPVstatus:PovStat + IPVstatus:CES1 + PovStat:CES1 + Age:IPVstatus:PovStat + Age:IPVstatus:CES1
## Data: WomenlogTrailsA
## REML criterion at convergence: 72.71
## Random effects:
## Groups Name Std.Dev. Corr
## HNDid (Intercept) 0.307
## Age 0.019 1.00
## subclass (Intercept) 0.252
## Residual 0.206
## Number of obs: 72, groups: HNDid, 36; subclass, 19
## Fixed Effects:
## (Intercept) Age
## 3.3549 0.0108
## IPVstatus1 PovStatBelow
## 0.5179 -0.0470
## CES11 Age:IPVstatus1
## 0.5437 0.0457
## Age:PovStatBelow Age:CES11
## -0.0195 0.0263
## IPVstatus1:PovStatBelow IPVstatus1:CES11
## 1.1428 -1.1419
## PovStatBelow:CES11 Age:IPVstatus1:PovStatBelow
## -0.3517 0.1122
## Age:IPVstatus1:CES11
## -0.1001
summary(mm2)
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: log(TrailsAtestSec) ~ Age + IPVstatus + PovStat + CES1 + (Age | HNDid) + (1 | subclass) + Age:IPVstatus + Age:PovStat + Age:CES1 + IPVstatus:PovStat + IPVstatus:CES1 + PovStat:CES1 + Age:IPVstatus:PovStat + Age:IPVstatus:CES1
## Data: WomenlogTrailsA
##
## REML criterion at convergence: 72.71
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## HNDid (Intercept) 0.094021 0.307
## Age 0.000363 0.019 1.00
## subclass (Intercept) 0.063744 0.252
## Residual 0.042422 0.206
## Number of obs: 72, groups: HNDid, 36; subclass, 19
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 3.3549 0.1673 14.2000 20.05 8.1e-12
## Age 0.0108 0.0124 16.3000 0.87 0.3949
## IPVstatus1 0.5179 0.2980 19.3000 1.74 0.0981
## PovStatBelow -0.0470 0.2542 22.4000 -0.19 0.8549
## CES11 0.5437 0.2359 15.9000 2.30 0.0350
## Age:IPVstatus1 0.0457 0.0253 32.4000 1.81 0.0796
## Age:PovStatBelow -0.0195 0.0180 17.2000 -1.09 0.2924
## Age:CES11 0.0263 0.0174 22.6000 1.51 0.1452
## IPVstatus1:PovStatBelow 1.1428 0.4827 26.9000 2.37 0.0254
## IPVstatus1:CES11 -1.1419 0.3943 18.5000 -2.90 0.0094
## PovStatBelow:CES11 -0.3517 0.1655 20.4000 -2.12 0.0460
## Age:IPVstatus1:PovStatBelow 0.1122 0.0427 41.3000 2.63 0.0119
## Age:IPVstatus1:CES11 -0.1001 0.0331 28.0000 -3.02 0.0053
##
## Correlation of Fixed Effects:
## (Intr) Age IPVst1 PvSttB CES11 Ag:IPV1 Ag:PSB A:CES1
## Age 0.800
## IPVstatus1 -0.424 -0.384
## PovStatBelw -0.413 -0.254 0.203
## CES11 -0.497 -0.422 0.218 -0.086
## Ag:IPVstts1 -0.316 -0.340 0.904 0.157 0.165
## Ag:PvSttBlw -0.357 -0.407 0.176 0.833 -0.128 0.156
## Age:CES11 -0.391 -0.535 0.201 -0.228 0.807 0.172 -0.145
## IPVstt1:PSB 0.151 0.133 -0.226 -0.474 0.148 -0.200 -0.531 0.127
## IPVs1:CES11 0.249 0.219 -0.679 0.074 -0.545 -0.618 0.097 -0.457
## PvStB:CES11 0.087 -0.147 0.013 -0.247 -0.214 -0.011 0.054 0.111
## Ag:IPV1:PSB 0.082 0.125 -0.171 -0.330 0.111 -0.222 -0.483 0.095
## A:IPV1:CES1 0.184 0.186 -0.630 0.078 -0.446 -0.692 0.109 -0.423
## IPV1:P IPV1:C PSB:CE A:IPV1:P
## Age
## IPVstatus1
## PovStatBelw
## CES11
## Ag:IPVstts1
## Ag:PvSttBlw
## Age:CES11
## IPVstt1:PSB
## IPVs1:CES11 -0.177
## PvStB:CES11 -0.216 0.062
## Ag:IPV1:PSB 0.896 -0.146 -0.148
## A:IPV1:CES1 -0.147 0.909 0.051 -0.123
plot(st)
plot(mm2)
Trails A Regression Model 1 for Men Only
load("~/Desktop/Megan/Research/IPV and Cognition Paper/IPV R Output/MenlogTrailsA.rda")
library(lme4)
library(lmerTest)
(mm1 = lmer(log(TrailsAtestSec) ~ (Age + IPVstatus + PovStat)^3 + (Age | HNDid) +
(1 | subclass), data = MenlogTrailsA))
## Warning: number of observations <= rank(Z); variance-covariance matrix
## will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: log(TrailsAtestSec) ~ (Age + IPVstatus + PovStat)^3 + (Age | HNDid) + (1 | subclass)
## Data: MenlogTrailsA
## REML criterion at convergence: 27.03
## Random effects:
## Groups Name Std.Dev. Corr
## HNDid (Intercept) 2.95e-01
## Age 1.65e-02 0.60
## subclass (Intercept) 8.80e-07
## Residual 1.33e-01
## Number of obs: 54, groups: HNDid, 27; subclass, 17
## Fixed Effects:
## (Intercept) Age
## 3.409584 0.012204
## IPVstatus1 PovStatBelow
## 0.037338 0.217925
## Age:IPVstatus1 Age:PovStatBelow
## -0.000739 0.003319
## IPVstatus1:PovStatBelow Age:IPVstatus1:PovStatBelow
## 0.230800 0.011404
(st = step(mm1))
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
##
## Random effects:
## Chi.sq Chi.DF elim.num p.value
## (1 | subclass) 0.06 1 1 0.8054
## (Age | HNDid) 0.11 1 2 0.7398
## (Age + 0 | HNDid) 0.00 1 3 1.0000
## (1 | HNDid) 16.97 1 kept 0
##
## Fixed effects:
## Sum Sq Mean Sq NumDF DenDF F.value elim.num Pr(>F)
## Age:IPVstatus:PovStat 0.0001 0.0001 1 45.85 0.0034 1 0.9535
## Age:IPVstatus 0.0106 0.0106 1 46.80 0.0707 2 0.7914
## IPVstatus:PovStat 0.0059 0.0059 1 19.70 0.4822 3 0.4955
## Age:PovStat 0.0132 0.0132 1 48.87 0.9837 4 0.3262
## IPVstatus 0.0623 0.0623 1 23.95 1.5478 5 0.2255
## Age 0.2653 0.2653 1 49.06 6.5204 kept 0.0138
## PovStat 0.1168 0.1168 1 28.12 6.5818 kept 0.0159
##
## Least squares means:
## PovStat Estimate Standard Error DF t-value Lower CI
## PovStat Above 1.0 3.3597 0.0672 25.9 50.0000 3.22
## PovStat Below 2.0 3.6693 0.0968 27.0 37.9100 3.47
## Upper CI p-value
## PovStat Above 3.50 <2e-16
## PovStat Below 3.87 <2e-16
##
## Differences of LSMEANS:
## Estimate Standard Error DF t-value Lower CI Upper CI
## PovStat Above-Below -0.3 0.121 28.1 -2.57 -0.557 -0.0624
## p-value
## PovStat Above-Below 0.02
##
## Final model:
## lme4::lmer(formula = log(TrailsAtestSec) ~ Age + PovStat + (1 |
## HNDid), data = MenlogTrailsA, REML = reml, contrasts = l)
Re-run final Model 1
(mm1 = lmer(log(TrailsAtestSec) ~ Age + PovStat + (Age | HNDid) + (1 | subclass),
data = MenlogTrailsA))
## Warning: number of observations <= rank(Z); variance-covariance matrix
## will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: log(TrailsAtestSec) ~ Age + PovStat + (Age | HNDid) + (1 | subclass)
## Data: MenlogTrailsA
## REML criterion at convergence: 6.663
## Random effects:
## Groups Name Std.Dev. Corr
## HNDid (Intercept) 3.06e-01
## Age 1.67e-02 0.65
## subclass (Intercept) 5.96e-07
## Residual 1.23e-01
## Number of obs: 54, groups: HNDid, 27; subclass, 17
## Fixed Effects:
## (Intercept) Age PovStatBelow
## 3.4471 0.0149 0.2674
summary(mm1)
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: log(TrailsAtestSec) ~ Age + PovStat + (Age | HNDid) + (1 | subclass)
## Data: MenlogTrailsA
##
## REML criterion at convergence: 6.663
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## HNDid (Intercept) 9.39e-02 3.06e-01
## Age 2.80e-04 1.67e-02 0.65
## subclass (Intercept) 3.56e-13 5.96e-07
## Residual 1.51e-02 1.23e-01
## Number of obs: 54, groups: HNDid, 27; subclass, 17
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 3.44710 0.08646 30.13000 39.87 <2e-16
## Age 0.01487 0.00645 17.11000 2.30 0.034
## PovStatBelow 0.26740 0.12071 27.12000 2.22 0.035
##
## Correlation of Fixed Effects:
## (Intr) Age
## Age 0.696
## PovStatBelw -0.538 -0.241
plot(st)
plot(mm1)
Trails A Regression Model 2 for Men Only (with CES)
load("~/Desktop/Megan/Research/IPV and Cognition Paper/IPV R Output/MenlogTrailsA.rda")
library(lme4)
library(lmerTest)
(mm2 = lmer(log(TrailsAtestSec) ~ (Age + IPVstatus + PovStat + CES1)^4 + (Age |
HNDid) + (1 | subclass), data = MenlogTrailsA))
## Warning: number of observations <= rank(Z); variance-covariance matrix
## will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: log(TrailsAtestSec) ~ (Age + IPVstatus + PovStat + CES1)^4 + (Age | HNDid) + (1 | subclass)
## Data: MenlogTrailsA
## REML criterion at convergence: 43.18
## Random effects:
## Groups Name Std.Dev. Corr
## HNDid (Intercept) 0.3157
## Age 0.0132 0.73
## subclass (Intercept) 0.0000
## Residual 0.1391
## Number of obs: 54, groups: HNDid, 27; subclass, 17
## Fixed Effects:
## (Intercept) Age
## 3.447849 0.012913
## IPVstatus1 PovStatBelow
## 0.189737 0.260888
## CES11 Age:IPVstatus1
## -0.192804 -0.012913
## Age:PovStatBelow Age:CES11
## -0.000304 -0.008476
## IPVstatus1:PovStatBelow IPVstatus1:CES11
## -0.235889 0.072469
## PovStatBelow:CES11 Age:IPVstatus1:PovStatBelow
## 0.015006 0.010065
## Age:IPVstatus1:CES11 Age:PovStatBelow:CES11
## 0.039516 0.015395
## IPVstatus1:PovStatBelow:CES11 Age:IPVstatus1:PovStatBelow:CES11
## 0.466554 -0.010697
(st = step(mm2))
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Random term (1 | subclass) was eliminated because of standard deviation being equal to 0
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
##
## Random effects:
## Chi.sq Chi.DF elim.num p.value
## (Age | HNDid) 0.71 1 1 0.3985
## (Age + 0 | HNDid) 0.00 1 2 1.0000
## (1 | HNDid) 12.53 1 kept 0.0004
##
## Fixed effects:
## Sum Sq Mean Sq NumDF DenDF F.value elim.num
## Age:IPVstatus:PovStat:CES1 0.0002 0.0002 1 22.56 0.0102 1
## Age:IPVstatus:PovStat 0.0002 0.0002 1 23.37 0.0012 2
## Age:PovStat:CES1 0.0004 0.0004 1 30.40 0.0378 3
## Age:PovStat 0.0135 0.0135 1 37.70 0.4003 4
## Age:IPVstatus:CES1 0.0671 0.0671 1 37.50 1.0879 5
## Age:IPVstatus 0.0159 0.0159 1 42.72 0.3558 6
## Age:CES1 0.0317 0.0317 1 37.14 0.9375 7
## IPVstatus:PovStat:CES1 0.0094 0.0094 1 18.86 2.8892 8
## IPVstatus:PovStat 0.0011 0.0011 1 19.92 0.0138 9
## IPVstatus:CES1 0.0009 0.0009 1 22.33 0.0151 10
## PovStat:CES1 0.0023 0.0023 1 21.82 0.9086 11
## CES1 0.0231 0.0231 1 22.93 1.0363 12
## IPVstatus 0.0642 0.0642 1 23.95 1.5478 13
## Age 0.2705 0.2705 1 49.06 6.5204 kept
## PovStat 0.1203 0.1203 1 28.12 6.5818 kept
## Pr(>F)
## Age:IPVstatus:PovStat:CES1 0.9204
## Age:IPVstatus:PovStat 0.9725
## Age:PovStat:CES1 0.8472
## Age:PovStat 0.5307
## Age:IPVstatus:CES1 0.3036
## Age:IPVstatus 0.5540
## Age:CES1 0.3392
## IPVstatus:PovStat:CES1 0.1056
## IPVstatus:PovStat 0.9077
## IPVstatus:CES1 0.9032
## PovStat:CES1 0.3509
## CES1 0.3193
## IPVstatus 0.2255
## Age 0.0138
## PovStat 0.0159
##
## Least squares means:
## PovStat Estimate Standard Error DF t-value Lower CI
## PovStat Above 1.0 3.3597 0.0672 25.9 50.0000 3.22
## PovStat Below 2.0 3.6693 0.0968 27.0 37.9100 3.47
## Upper CI p-value
## PovStat Above 3.50 <2e-16
## PovStat Below 3.87 <2e-16
##
## Differences of LSMEANS:
## Estimate Standard Error DF t-value Lower CI Upper CI
## PovStat Above-Below -0.3 0.121 28.1 -2.57 -0.557 -0.0624
## p-value
## PovStat Above-Below 0.02
##
## Final model:
## lme4::lmer(formula = log(TrailsAtestSec) ~ Age + PovStat + (1 |
## HNDid), data = MenlogTrailsA, REML = reml, contrasts = l)
Re-run suggested final Model 2
(mm2 = lmer(log(TrailsAtestSec) ~ Age + PovStat + (Age | HNDid) + (1 | subclass),
data = MenlogTrailsA, contrasts = 1))
## Warning: number of observations <= rank(Z); variance-covariance matrix
## will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: log(TrailsAtestSec) ~ Age + PovStat + (Age | HNDid) + (1 | subclass)
## Data: MenlogTrailsA
## REML criterion at convergence: 6.663
## Random effects:
## Groups Name Std.Dev. Corr
## HNDid (Intercept) 3.06e-01
## Age 1.67e-02 0.65
## subclass (Intercept) 5.96e-07
## Residual 1.23e-01
## Number of obs: 54, groups: HNDid, 27; subclass, 17
## Fixed Effects:
## (Intercept) Age PovStatBelow
## 3.4471 0.0149 0.2674
summary(mm2)
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Warning: number of observations <= rank(Z); variance-covariance matrix will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: log(TrailsAtestSec) ~ Age + PovStat + (Age | HNDid) + (1 | subclass)
## Data: MenlogTrailsA
##
## REML criterion at convergence: 6.663
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## HNDid (Intercept) 9.39e-02 3.06e-01
## Age 2.80e-04 1.67e-02 0.65
## subclass (Intercept) 3.56e-13 5.96e-07
## Residual 1.51e-02 1.23e-01
## Number of obs: 54, groups: HNDid, 27; subclass, 17
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 3.44710 0.08646 30.13000 39.87 <2e-16
## Age 0.01487 0.00645 17.11000 2.30 0.034
## PovStatBelow 0.26740 0.12071 27.12000 2.22 0.035
##
## Correlation of Fixed Effects:
## (Intr) Age
## Age 0.696
## PovStatBelw -0.538 -0.241
plot(st)
plot(mm2)