Fluency (Word) Regression Models

Fluency (Word) Regression Model 1

## 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: FluencyWord ~ (Age + IPVstatus + Sex + PovStat)^4 + (Age | HNDid) +      (1 | subclass) 
##    Data: IPVandCognitionDataSet2 
## REML criterion at convergence: 734 
## Random effects:
##  Groups   Name        Std.Dev. Corr
##  HNDid    (Intercept) 6.674879     
##           Age         0.190552 1.00
##  subclass (Intercept) 0.000207     
##  Residual             3.306109     
## Number of obs: 126, groups: HNDid, 63; subclass, 21
## Fixed Effects:
##                        (Intercept)                                 Age  
##                            17.1247                             -0.3395  
##                         IPVstatus1                              SexMen  
##                             3.8133                              6.8968  
##                       PovStatBelow                      Age:IPVstatus1  
##                            -2.7692                              0.4190  
##                         Age:SexMen                    Age:PovStatBelow  
##                             0.3221                              0.1136  
##                  IPVstatus1:SexMen             IPVstatus1:PovStatBelow  
##                            -2.5374                              0.3843  
##                SexMen:PovStatBelow               Age:IPVstatus1:SexMen  
##                            -0.4661                             -0.4154  
##        Age:IPVstatus1:PovStatBelow             Age:SexMen:PovStatBelow  
##                             0.0147                             -0.2484  
##     IPVstatus1:SexMen:PovStatBelow  Age:IPVstatus1:SexMen:PovStatBelow  
##                            -4.1289                             -0.0112
## 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 term (1 | subclass) was eliminated because of standard deviation being equal to 0
## 
## Random effects:
##             Chi.sq Chi.DF elim.num p.value
## (1 | HNDid)  32.97      1     kept < 1e-07
## 
## Fixed effects:
##                             Sum Sq  Mean Sq NumDF  DenDF F.value elim.num
## Age:IPVstatus:Sex:PovStat   0.2058   0.2058     1 109.62  0.0171        1
## IPVstatus:Sex:PovStat       4.9810   4.9810     1  57.99  0.4170        2
## Age:IPVstatus:PovStat       2.1765   2.1765     1 111.99  0.3176        3
## IPVstatus:PovStat           4.2670   4.2670     1  56.47  0.1922        4
## Age:Sex:PovStat             3.8761   3.8761     1 110.20  0.1672        5
## Sex:PovStat                 0.3764   0.3764     1  60.42  0.0054        6
## Age:PovStat                 2.8484   2.8484     1 108.75  0.0552        7
## Age:IPVstatus:Sex           8.9212   8.9212     1 111.79  1.2422        8
## Age:Sex                     1.4435   1.4435     1 114.02  0.1261        9
## IPVstatus:Sex               1.8697   1.8697     1  56.02  0.1771       10
## Age:IPVstatus               2.1851   2.1851     1 113.68  0.1953       11
## IPVstatus                   0.0000   0.0000     1  56.98  0.0396       12
## Age                        60.5725  60.5725     1 115.41  4.2756     kept
## Sex                       105.7626 105.7626     1  58.23 10.2953     kept
## PovStat                    67.5447  67.5447     1  60.11  6.1562     kept
##                           Pr(>F)
## Age:IPVstatus:Sex:PovStat 0.8961
## IPVstatus:Sex:PovStat     0.5210
## Age:IPVstatus:PovStat     0.5742
## IPVstatus:PovStat         0.6628
## Age:Sex:PovStat           0.6834
## Sex:PovStat               0.9418
## Age:PovStat               0.8147
## Age:IPVstatus:Sex         0.2674
## Age:Sex                   0.7232
## IPVstatus:Sex             0.6755
## Age:IPVstatus             0.6594
## IPVstatus                 0.8430
## Age                       0.0409
## Sex                       0.0022
## PovStat                   0.0159
## 
## Least squares means:
##                 Sex PovStat Estimate Standard Error   DF t-value Lower CI
## Sex  Women      2.0      NA   18.089          0.937 57.9    19.3     16.2
## Sex  Men        1.0      NA   22.322          0.996 58.5    22.4     20.3
## PovStat  Above   NA     1.0   21.994          0.801 58.6    27.4     20.4
## PovStat  Below   NA     2.0   18.418          1.181 59.5    15.6     16.1
##                Upper CI p-value
## Sex  Women         20.0  <2e-16
## Sex  Men           24.3  <2e-16
## PovStat  Above     23.6  <2e-16
## PovStat  Below     20.8  <2e-16
## 
##  Differences of LSMEANS:
##                     Estimate Standard Error   DF t-value Lower CI Upper CI
## Sex Women-Men           -4.2           1.32 58.2   -3.21   -6.873    -1.59
## PovStat Above-Below      3.6           1.44 60.1    2.48    0.693     6.46
##                     p-value
## Sex Women-Men         0.002
## PovStat Above-Below   0.016
## 
## Final model:
## lme4::lmer(formula = FluencyWord ~ Age + Sex + PovStat + (1 | 
##     HNDid), data = IPVandCognitionDataSet2, REML = reml, contrasts = l)

Re-run suggested final Model 1

(mm1 = lmer(FluencyWord ~ Age + Sex + PovStat + (Age | HNDid) + (1 | subclass), 
    data = IPVandCognitionDataSet2))
## Warning: number of observations <= rank(Z); variance-covariance matrix
## will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: FluencyWord ~ Age + Sex + PovStat + (Age | HNDid) + (1 | subclass) 
##    Data: IPVandCognitionDataSet2 
## REML criterion at convergence: 754.7 
## Random effects:
##  Groups   Name        Std.Dev. Corr
##  HNDid    (Intercept) 5.360750     
##           Age         0.079632 1.00
##  subclass (Intercept) 0.000524     
##  Residual             3.290914     
## Number of obs: 126, groups: HNDid, 63; subclass, 21
## Fixed Effects:
##  (Intercept)           Age        SexMen  PovStatBelow  
##       18.938        -0.147         4.110        -3.570

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: FluencyWord ~ Age + Sex + PovStat + (Age | HNDid) + (1 | subclass) 
##    Data: IPVandCognitionDataSet2 
## 
## REML criterion at convergence: 754.7 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev. Corr
##  HNDid    (Intercept) 2.87e+01 5.360750     
##           Age         6.34e-03 0.079632 1.00
##  subclass (Intercept) 2.74e-07 0.000524     
##  Residual             1.08e+01 3.290914     
## Number of obs: 126, groups: HNDid, 63; subclass, 21
## 
## Fixed effects:
##              Estimate Std. Error      df t value Pr(>|t|)
## (Intercept)   18.9385     1.3097 34.4000   14.46  4.4e-16
## Age           -0.1470     0.0828 66.9000   -1.78   0.0803
## SexMen         4.1098     1.3357 58.6000    3.08   0.0032
## PovStatBelow  -3.5696     1.4719 57.4000   -2.43   0.0185
## 
## Correlation of Fixed Effects:
##             (Intr) Age    SexMen
## Age          0.658              
## SexMen      -0.513 -0.095       
## PovStatBelw -0.438 -0.183  0.005

plot(st)

plot of chunk unnamed-chunk-1 plot of chunk unnamed-chunk-1 plot of chunk unnamed-chunk-1 plot of chunk unnamed-chunk-1


plot(mm1)

plot of chunk unnamed-chunk-1

Fluency (Word) Regression Model 2 (with CES)

load("~/Desktop/Megan/Research/IPV and Cognition Paper/IPV R Output/IPVandCognitionDataSet2.rda")

library(lme4)
library(lmerTest)

(mm2 = lmer(FluencyWord ~ (Age + IPVstatus + Sex + PovStat + CES1)^5 + (Age | 
    HNDid) + (1 | subclass), data = IPVandCognitionDataSet2))
## Warning: number of observations <= rank(Z); variance-covariance matrix
## will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: FluencyWord ~ (Age + IPVstatus + Sex + PovStat + CES1)^5 + (Age |      HNDid) + (1 | subclass) 
##    Data: IPVandCognitionDataSet2 
## REML criterion at convergence: 672.9 
## Random effects:
##  Groups   Name        Std.Dev. Corr
##  HNDid    (Intercept) 6.29e+00     
##           Age         1.71e-01 1.00
##  subclass (Intercept) 2.28e-05     
##  Residual             3.48e+00     
## Number of obs: 126, groups: HNDid, 63; subclass, 21
## Fixed Effects:
##                              (Intercept)  
##                                  18.8238  
##                                      Age  
##                                  -0.3028  
##                               IPVstatus1  
##                                   2.5449  
##                                   SexMen  
##                                   8.1998  
##                             PovStatBelow  
##                                  -2.7206  
##                                    CES11  
##                                  -3.8723  
##                           Age:IPVstatus1  
##                                   0.6012  
##                               Age:SexMen  
##                                   0.3839  
##                         Age:PovStatBelow  
##                                   0.2195  
##                                Age:CES11  
##                                  -0.0975  
##                        IPVstatus1:SexMen  
##                                  -4.5686  
##                  IPVstatus1:PovStatBelow  
##                                  -3.0481  
##                         IPVstatus1:CES11  
##                                   3.3879  
##                      SexMen:PovStatBelow  
##                                  -5.6642  
##                             SexMen:CES11  
##                                 -10.1661  
##                       PovStatBelow:CES11  
##                                   1.1385  
##                    Age:IPVstatus1:SexMen  
##                                  -0.6824  
##              Age:IPVstatus1:PovStatBelow  
##                                  -0.3179  
##                     Age:IPVstatus1:CES11  
##                                  -0.3208  
##                  Age:SexMen:PovStatBelow  
##                                  -0.5475  
##                         Age:SexMen:CES11  
##                                  -0.3703  
##                   Age:PovStatBelow:CES11  
##                                  -0.1134  
##           IPVstatus1:SexMen:PovStatBelow  
##                                   6.1938  
##                  IPVstatus1:SexMen:CES11  
##                                  10.0368  
##            IPVstatus1:PovStatBelow:CES11  
##                                   2.4319  
##                SexMen:PovStatBelow:CES11  
##                                  19.7528  
##       Age:IPVstatus1:SexMen:PovStatBelow  
##                                   0.5514  
##              Age:IPVstatus1:SexMen:CES11  
##                                   0.6314  
##        Age:IPVstatus1:PovStatBelow:CES11  
##                                   0.3409  
##            Age:SexMen:PovStatBelow:CES11  
##                                   0.9804  
##     IPVstatus1:SexMen:PovStatBelow:CES11  
##                                 -25.5100  
## Age:IPVstatus1:SexMen:PovStatBelow:CES11  
##                                  -1.0208

(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 term (1 | subclass) was eliminated because of standard deviation being equal to 0
## 
## Random effects:
##             Chi.sq Chi.DF elim.num p.value
## (1 | HNDid)  23.08      1     kept       0
## 
## Fixed effects:
##                                  Sum Sq  Mean Sq NumDF  DenDF F.value
## Age:IPVstatus:Sex:PovStat:CES1   1.4233   1.4233     1  67.96  0.1094
## Age:IPVstatus:Sex:PovStat        0.6362   0.6362     1  72.27  0.0003
## Age:IPVstatus:PovStat:CES1       0.8953   0.8953     1  76.06  0.0011
## Age:IPVstatus:PovStat            0.0019   0.0019     1  83.45  0.0018
## Age:IPVstatus:Sex:CES1           0.7355   0.7355     1  95.86  0.0098
## Age:IPVstatus:Sex                8.4292   8.4292     1  97.60  0.2338
## Age:IPVstatus:CES1              14.5610  14.5610     1 100.00  0.2092
## Age:Sex:PovStat:CES1            13.8055  13.8055     1  93.12  0.8377
## Age:Sex:CES1                     3.6410   3.6410     1  99.82  0.0442
## Age:Sex:PovStat                  4.9768   4.9768     1  97.81  0.1678
## Age:Sex                          1.2138   1.2138     1 103.87  0.0775
## Age:PovStat:CES1                 6.2860   6.2860     1 104.45  0.7376
## Age:PovStat                      2.4022   2.4022     1  98.62  0.0215
## Age:IPVstatus                    3.4952   3.4952     1 103.06  0.4520
## Age:CES1                         3.1371   3.1371     1  99.20  1.0587
## IPVstatus:Sex:PovStat:CES1       7.6855   7.6855     1  44.65  2.4357
## IPVstatus:Sex:CES1               4.8691   4.8691     1  46.01  0.1713
## IPVstatus:Sex:PovStat            8.1055   8.1055     1  46.66  1.1244
## IPVstatus:Sex                    1.0108   1.0108     1  47.47  0.0301
## Sex:PovStat:CES1                18.9302  18.9302     1  48.60  0.9252
## Sex:PovStat                      2.2484   2.2484     1  51.05  0.5597
## Sex:CES1                         8.5038   8.5038     1  50.56  0.7283
## IPVstatus:PovStat:CES1          13.6571  13.6571     1  51.58  2.0764
## IPVstatus:PovStat                6.1151   6.1151     1  52.74  0.6488
## PovStat:CES1                    37.4172  37.4172     1  53.72  1.4555
## IPVstatus:CES1                  35.0329  35.0329     1  55.12  3.4278
## IPVstatus                        0.0004   0.0004     1  56.04  0.1486
## CES1                             7.8188   7.8188     1  57.57  0.5323
## Age                             72.3776  72.3776     1 115.41  4.2756
## Sex                            120.5149 120.5149     1  58.23 10.2953
## PovStat                         74.7668  74.7668     1  60.11  6.1562
##                                elim.num Pr(>F)
## Age:IPVstatus:Sex:PovStat:CES1        1 0.7418
## Age:IPVstatus:Sex:PovStat             2 0.9853
## Age:IPVstatus:PovStat:CES1            3 0.9742
## Age:IPVstatus:PovStat                 4 0.9659
## Age:IPVstatus:Sex:CES1                5 0.9213
## Age:IPVstatus:Sex                     6 0.6298
## Age:IPVstatus:CES1                    7 0.6484
## Age:Sex:PovStat:CES1                  8 0.3624
## Age:Sex:CES1                          9 0.8339
## Age:Sex:PovStat                      10 0.6830
## Age:Sex                              11 0.7813
## Age:PovStat:CES1                     12 0.3924
## Age:PovStat                          13 0.8837
## Age:IPVstatus                        14 0.5029
## Age:CES1                             15 0.3060
## IPVstatus:Sex:PovStat:CES1           16 0.1257
## IPVstatus:Sex:CES1                   17 0.6809
## IPVstatus:Sex:PovStat                18 0.2944
## IPVstatus:Sex                        19 0.8630
## Sex:PovStat:CES1                     20 0.3409
## Sex:PovStat                          21 0.4578
## Sex:CES1                             22 0.3975
## IPVstatus:PovStat:CES1               23 0.1556
## IPVstatus:PovStat                    24 0.4242
## PovStat:CES1                         25 0.2329
## IPVstatus:CES1                       26 0.0695
## IPVstatus                            27 0.7013
## CES1                                 28 0.4686
## Age                                kept 0.0409
## Sex                                kept 0.0022
## PovStat                            kept 0.0159
## 
## Least squares means:
##                 Sex PovStat Estimate Standard Error   DF t-value Lower CI
## Sex  Women      2.0      NA   18.089          0.937 57.9    19.3     16.2
## Sex  Men        1.0      NA   22.322          0.996 58.5    22.4     20.3
## PovStat  Above   NA     1.0   21.994          0.801 58.6    27.4     20.4
## PovStat  Below   NA     2.0   18.418          1.181 59.5    15.6     16.1
##                Upper CI p-value
## Sex  Women         20.0  <2e-16
## Sex  Men           24.3  <2e-16
## PovStat  Above     23.6  <2e-16
## PovStat  Below     20.8  <2e-16
## 
##  Differences of LSMEANS:
##                     Estimate Standard Error   DF t-value Lower CI Upper CI
## Sex Women-Men           -4.2           1.32 58.2   -3.21   -6.873    -1.59
## PovStat Above-Below      3.6           1.44 60.1    2.48    0.693     6.46
##                     p-value
## Sex Women-Men         0.002
## PovStat Above-Below   0.016
## 
## Final model:
## lme4::lmer(formula = FluencyWord ~ Age + Sex + PovStat + (1 | 
##     HNDid), data = IPVandCognitionDataSet2, REML = reml, contrasts = l)

Re-run suggested final Model 2

(mm2 = lmer(FluencyWord ~ Age + Sex + PovStat + (Age | HNDid) + (1 | subclass), 
    data = IPVandCognitionDataSet2))
## Warning: number of observations <= rank(Z); variance-covariance matrix
## will be unidentifiable
## Linear mixed model fit by REML ['merModLmerTest']
## Formula: FluencyWord ~ Age + Sex + PovStat + (Age | HNDid) + (1 | subclass) 
##    Data: IPVandCognitionDataSet2 
## REML criterion at convergence: 754.7 
## Random effects:
##  Groups   Name        Std.Dev. Corr
##  HNDid    (Intercept) 5.360750     
##           Age         0.079632 1.00
##  subclass (Intercept) 0.000524     
##  Residual             3.290914     
## Number of obs: 126, groups: HNDid, 63; subclass, 21
## Fixed Effects:
##  (Intercept)           Age        SexMen  PovStatBelow  
##       18.938        -0.147         4.110        -3.570

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: FluencyWord ~ Age + Sex + PovStat + (Age | HNDid) + (1 | subclass) 
##    Data: IPVandCognitionDataSet2 
## 
## REML criterion at convergence: 754.7 
## 
## Random effects:
##  Groups   Name        Variance Std.Dev. Corr
##  HNDid    (Intercept) 2.87e+01 5.360750     
##           Age         6.34e-03 0.079632 1.00
##  subclass (Intercept) 2.74e-07 0.000524     
##  Residual             1.08e+01 3.290914     
## Number of obs: 126, groups: HNDid, 63; subclass, 21
## 
## Fixed effects:
##              Estimate Std. Error      df t value Pr(>|t|)
## (Intercept)   18.9385     1.3097 34.4000   14.46  4.4e-16
## Age           -0.1470     0.0828 66.9000   -1.78   0.0803
## SexMen         4.1098     1.3357 58.6000    3.08   0.0032
## PovStatBelow  -3.5696     1.4719 57.4000   -2.43   0.0185
## 
## Correlation of Fixed Effects:
##             (Intr) Age    SexMen
## Age          0.658              
## SexMen      -0.513 -0.095       
## PovStatBelw -0.438 -0.183  0.005

plot(st)

plot of chunk unnamed-chunk-3 plot of chunk unnamed-chunk-3 plot of chunk unnamed-chunk-3 plot of chunk unnamed-chunk-3


plot(mm2)

plot of chunk unnamed-chunk-3