Objective
A longitudinal study with the assessment of the influence of MMSAS (Multidimensional Motivations to Study Abroad) on BSAS (Brief sociocultural adaptation) and drunk behaviour.
Data from: Aresi, Giovanni; Moore, Simon C.; Marta, Elena (2021), “The longitudinal health behaviours of European study abroad students sampled from forty-two countries and across three-waves”, Mendeley Data, V3, doi: 10.17632/585d2wdmtd.3
Method: PLS-SEM R’s SEMinR package.
Although somewhat concrete and observable, behaviours (health, sustainability or market related) are not necessarily directly measurable constructs. They can manifest in different ways, depending on the person, for example. And, to quantify it, we use latent or composite variables, depending on the scale, evaluating, for example, the frequency of manifestation of such modes.
This case explores a model using the behavioural construct ‘drunk behaviour’ as a composite variable.
Because the variable is composite, we use PLS-SEM (Partial Least
Squares-SEM, a specific type of SEM), using the package semin.
Data acquisition
data<-read_sav('Aresi2021.sav')
Model definition
measurement model
Note that, in this case, - Being multifaceted, the ‘variables’ (like drunk behaviour) are assessed by a combination of indicators associated with the variety of manifestations of how such behaviour. It makes up a ‘formative construct’ (also called ‘composite variable’). - The MMSAS (Multidimensional Motivations to Study Abroad) is prompted as influencing both drunk behaviours at times 1 and 2 directly. - By comprising a longitudinal study, the interrelatedness of the three assessments of the same latent variable is unavoidable.
corres<-Hmisc::rcorr(as.matrix(data[,c(paste0("MMSAS_",c(11,15,19:22,24,25,27)),
names(data[,c(34,36,42:43,50,51,78)]),
names(data[,c(85,87,90,91,93,98,126)]),
names(data[,c(229,230,233,234,236)]),
names(data[,c(133:135,139:142,149,151,159,175:177,179)])) ]))
flattenCorrMatrix <- function(cormat, pmat) {
ut <- upper.tri(cormat)
data.frame(
row = rownames(cormat)[row(cormat)[ut]],
column = rownames(cormat)[col(cormat)[ut]],
cor =(cormat)[ut],
p = pmat[ut]
)
}
flatCorr<-flattenCorrMatrix(corres$r,sapply(corres$P,function(x){round(x,3)}))
flatCorr[flatCorr$p<0.05,]
## row column cor p
## 2 MMSAS_11 MMSAS_19 0.20807827 0.000
## 3 MMSAS_15 MMSAS_19 0.12179652 0.000
## 4 MMSAS_11 MMSAS_20 0.16671987 0.000
## 5 MMSAS_15 MMSAS_20 0.10766180 0.001
## 6 MMSAS_19 MMSAS_20 0.68704401 0.000
## 7 MMSAS_11 MMSAS_21 0.18681643 0.000
## 8 MMSAS_15 MMSAS_21 0.17379178 0.000
## 9 MMSAS_19 MMSAS_21 0.71512321 0.000
## 10 MMSAS_20 MMSAS_21 0.62349157 0.000
## 12 MMSAS_15 MMSAS_22 0.33202372 0.000
## 13 MMSAS_19 MMSAS_22 0.11088208 0.001
## 14 MMSAS_20 MMSAS_22 0.09876363 0.003
## 15 MMSAS_21 MMSAS_22 0.14449405 0.000
## 17 MMSAS_15 MMSAS_24 0.26052203 0.000
## 18 MMSAS_19 MMSAS_24 0.10852007 0.001
## 19 MMSAS_20 MMSAS_24 0.14355454 0.000
## 20 MMSAS_21 MMSAS_24 0.13564145 0.000
## 21 MMSAS_22 MMSAS_24 0.59802467 0.000
## 23 MMSAS_15 MMSAS_25 0.47960643 0.000
## 24 MMSAS_19 MMSAS_25 0.07570008 0.023
## 26 MMSAS_21 MMSAS_25 0.10595370 0.001
## 27 MMSAS_22 MMSAS_25 0.29242141 0.000
## 28 MMSAS_24 MMSAS_25 0.26928593 0.000
## 29 MMSAS_11 MMSAS_27 -0.07257259 0.029
## 30 MMSAS_15 MMSAS_27 0.46736475 0.000
## 34 MMSAS_22 MMSAS_27 0.26225007 0.000
## 35 MMSAS_24 MMSAS_27 0.22846216 0.000
## 36 MMSAS_25 MMSAS_27 0.55475341 0.000
## 47 MMSAS_15 T1_LY_alcohol -0.09127138 0.006
## 48 MMSAS_19 T1_LY_alcohol 0.07626718 0.022
## 53 MMSAS_25 T1_LY_alcohol -0.11047177 0.001
## 54 MMSAS_27 T1_LY_alcohol -0.13584552 0.000
## 57 MMSAS_15 T1_Friday 0.12071058 0.000
## 61 MMSAS_22 T1_Friday 0.07510146 0.024
## 63 MMSAS_25 T1_Friday 0.12678098 0.000
## 64 MMSAS_27 T1_Friday 0.19615585 0.000
## 66 T1_LY_alcohol T1_Friday -0.40250991 0.000
## 68 MMSAS_15 T1_Saturday 0.09497584 0.004
## 72 MMSAS_22 T1_Saturday 0.07459760 0.025
## 74 MMSAS_25 T1_Saturday 0.09360457 0.005
## 75 MMSAS_27 T1_Saturday 0.13233155 0.000
## 77 T1_LY_alcohol T1_Saturday -0.44244486 0.000
## 78 T1_Friday T1_Saturday 0.35710607 0.000
## 80 MMSAS_15 T1_Binge_drinking 0.14029368 0.000
## 84 MMSAS_22 T1_Binge_drinking 0.09857368 0.003
## 86 MMSAS_25 T1_Binge_drinking 0.18660837 0.000
## 87 MMSAS_27 T1_Binge_drinking 0.22534466 0.000
## 88 T1_Sweets T1_Binge_drinking -0.10268723 0.002
## 89 T1_LY_alcohol T1_Binge_drinking -0.29767021 0.000
## 90 T1_Friday T1_Binge_drinking 0.38862750 0.000
## 91 T1_Saturday T1_Binge_drinking 0.37780707 0.000
## 93 MMSAS_15 T1_DrunkDrunkenness 0.10075318 0.003
## 94 MMSAS_19 T1_DrunkDrunkenness -0.07387252 0.027
## 99 MMSAS_25 T1_DrunkDrunkenness 0.16064913 0.000
## 100 MMSAS_27 T1_DrunkDrunkenness 0.23367496 0.000
## 101 T1_Sweets T1_DrunkDrunkenness -0.09349145 0.005
## 102 T1_LY_alcohol T1_DrunkDrunkenness -0.18502149 0.000
## 103 T1_Friday T1_DrunkDrunkenness 0.26781004 0.000
## 104 T1_Saturday T1_DrunkDrunkenness 0.23906828 0.000
## 105 T1_Binge_drinking T1_DrunkDrunkenness 0.50814130 0.000
## 108 MMSAS_19 T1_Drugs_YN -0.07423333 0.029
## 113 MMSAS_25 T1_Drugs_YN 0.07362089 0.031
## 114 MMSAS_27 T1_Drugs_YN 0.10679206 0.002
## 115 T1_Sweets T1_Drugs_YN -0.07930846 0.020
## 117 T1_Friday T1_Drugs_YN 0.09200096 0.007
## 118 T1_Saturday T1_Drugs_YN 0.07274607 0.033
## 119 T1_Binge_drinking T1_Drugs_YN 0.21546125 0.000
## 120 T1_DrunkDrunkenness T1_Drugs_YN 0.11622213 0.001
## 130 T1_Sweets T2_Sweets 0.61075675 0.000
## 134 T1_Binge_drinking T2_Sweets -0.10486038 0.004
## 136 T1_Drugs_YN T2_Sweets -0.12835283 0.001
## 138 MMSAS_15 T2_LM_alcohol -0.20672449 0.000
## 142 MMSAS_22 T2_LM_alcohol -0.12551132 0.001
## 144 MMSAS_25 T2_LM_alcohol -0.20565654 0.000
## 145 MMSAS_27 T2_LM_alcohol -0.22373038 0.000
## 147 T1_LY_alcohol T2_LM_alcohol 0.60163805 0.000
## 148 T1_Friday T2_LM_alcohol -0.42637998 0.000
## 149 T1_Saturday T2_LM_alcohol -0.41066301 0.000
## 150 T1_Binge_drinking T2_LM_alcohol -0.30858087 0.000
## 151 T1_DrunkDrunkenness T2_LM_alcohol -0.20780140 0.000
## 161 MMSAS_25 T2_Wednesday 0.12732588 0.000
## 162 MMSAS_27 T2_Wednesday 0.14781951 0.000
## 164 T1_LY_alcohol T2_Wednesday -0.15940386 0.000
## 165 T1_Friday T2_Wednesday 0.21499875 0.000
## 166 T1_Saturday T2_Wednesday 0.14762794 0.000
## 167 T1_Binge_drinking T2_Wednesday 0.21692159 0.000
## 168 T1_DrunkDrunkenness T2_Wednesday 0.20144866 0.000
## 169 T1_Drugs_YN T2_Wednesday 0.11400766 0.002
## 171 T2_LM_alcohol T2_Wednesday -0.20301660 0.000
## 173 MMSAS_15 T2_Thursday 0.14087250 0.000
## 177 MMSAS_22 T2_Thursday 0.08913901 0.014
## 179 MMSAS_25 T2_Thursday 0.10277325 0.004
## 180 MMSAS_27 T2_Thursday 0.13821348 0.000
## 182 T1_LY_alcohol T2_Thursday -0.20858473 0.000
## 183 T1_Friday T2_Thursday 0.29108980 0.000
## 184 T1_Saturday T2_Thursday 0.21820365 0.000
## 185 T1_Binge_drinking T2_Thursday 0.24176131 0.000
## 186 T1_DrunkDrunkenness T2_Thursday 0.24848709 0.000
## 187 T1_Drugs_YN T2_Thursday 0.09299185 0.012
## 189 T2_LM_alcohol T2_Thursday -0.28587180 0.000
## 190 T2_Wednesday T2_Thursday 0.16527286 0.000
## 192 MMSAS_15 T2_Saturday 0.16098073 0.000
## 194 MMSAS_20 T2_Saturday 0.07446933 0.040
## 195 MMSAS_21 T2_Saturday 0.08415513 0.020
## 198 MMSAS_25 T2_Saturday 0.12871652 0.000
## 199 MMSAS_27 T2_Saturday 0.18347935 0.000
## 201 T1_LY_alcohol T2_Saturday -0.36032942 0.000
## 202 T1_Friday T2_Saturday 0.32128385 0.000
## 203 T1_Saturday T2_Saturday 0.42923785 0.000
## 204 T1_Binge_drinking T2_Saturday 0.32082704 0.000
## 205 T1_DrunkDrunkenness T2_Saturday 0.25070681 0.000
## 208 T2_LM_alcohol T2_Saturday -0.51764861 0.000
## 209 T2_Wednesday T2_Saturday 0.17189503 0.000
## 210 T2_Thursday T2_Saturday 0.24193039 0.000
## 212 MMSAS_15 T2_N_shot_spirit 0.10973024 0.003
## 219 MMSAS_27 T2_N_shot_spirit 0.08271904 0.023
## 221 T1_LY_alcohol T2_N_shot_spirit -0.11588033 0.001
## 222 T1_Friday T2_N_shot_spirit 0.11833972 0.001
## 223 T1_Saturday T2_N_shot_spirit 0.09494923 0.009
## 224 T1_Binge_drinking T2_N_shot_spirit 0.24818324 0.000
## 225 T1_DrunkDrunkenness T2_N_shot_spirit 0.17095008 0.000
## 228 T2_LM_alcohol T2_N_shot_spirit -0.18837478 0.000
## 229 T2_Wednesday T2_N_shot_spirit 0.08643241 0.018
## 230 T2_Thursday T2_N_shot_spirit 0.16647633 0.000
## 231 T2_Saturday T2_N_shot_spirit 0.16692939 0.000
## 233 MMSAS_15 T2_Cannabis_YN 0.10222366 0.005
## 237 MMSAS_22 T2_Cannabis_YN 0.07747565 0.034
## 239 MMSAS_25 T2_Cannabis_YN 0.12805502 0.000
## 240 MMSAS_27 T2_Cannabis_YN 0.15809339 0.000
## 241 T1_Sweets T2_Cannabis_YN -0.10092397 0.006
## 242 T1_LY_alcohol T2_Cannabis_YN -0.12646246 0.001
## 243 T1_Friday T2_Cannabis_YN 0.16945902 0.000
## 244 T1_Saturday T2_Cannabis_YN 0.13474585 0.000
## 245 T1_Binge_drinking T2_Cannabis_YN 0.26072382 0.000
## 246 T1_DrunkDrunkenness T2_Cannabis_YN 0.22909250 0.000
## 247 T1_Drugs_YN T2_Cannabis_YN 0.17790716 0.000
## 248 T2_Sweets T2_Cannabis_YN -0.08149113 0.025
## 249 T2_LM_alcohol T2_Cannabis_YN -0.13999271 0.000
## 250 T2_Wednesday T2_Cannabis_YN 0.13220007 0.000
## 251 T2_Thursday T2_Cannabis_YN 0.21672497 0.000
## 252 T2_Saturday T2_Cannabis_YN 0.16909344 0.000
## 253 T2_N_shot_spirit T2_Cannabis_YN 0.19390976 0.000
## 255 MMSAS_15 BPAS_1 0.24142104 0.000
## 256 MMSAS_19 BPAS_1 0.13388005 0.000
## 257 MMSAS_20 BPAS_1 0.13945555 0.000
## 258 MMSAS_21 BPAS_1 0.18434702 0.000
## 259 MMSAS_22 BPAS_1 0.15947533 0.000
## 260 MMSAS_24 BPAS_1 0.15512518 0.000
## 261 MMSAS_25 BPAS_1 0.24835652 0.000
## 262 MMSAS_27 BPAS_1 0.22149147 0.000
## 267 T1_Binge_drinking BPAS_1 0.12661522 0.000
## 269 T1_Drugs_YN BPAS_1 0.08747776 0.018
## 271 T2_LM_alcohol BPAS_1 -0.11424850 0.002
## 273 T2_Thursday BPAS_1 0.12085183 0.001
## 274 T2_Saturday BPAS_1 0.14642864 0.000
## 275 T2_N_shot_spirit BPAS_1 0.07978843 0.029
## 276 T2_Cannabis_YN BPAS_1 0.10983795 0.003
## 278 MMSAS_15 BPAS_2 -0.14029807 0.000
## 285 MMSAS_27 BPAS_2 -0.13343226 0.000
## 294 T2_LM_alcohol BPAS_2 0.10155361 0.005
## 295 T2_Wednesday BPAS_2 -0.09296094 0.010
## 299 T2_Cannabis_YN BPAS_2 -0.08541837 0.019
## 300 BPAS_1 BPAS_2 -0.17990378 0.000
## 302 MMSAS_15 BPAS_5 -0.21695442 0.000
## 308 MMSAS_25 BPAS_5 -0.11860811 0.001
## 309 MMSAS_27 BPAS_5 -0.12623475 0.000
## 311 T1_LY_alcohol BPAS_5 0.08753555 0.016
## 318 T2_LM_alcohol BPAS_5 0.14013853 0.000
## 319 T2_Wednesday BPAS_5 -0.10658197 0.003
## 320 T2_Thursday BPAS_5 -0.10275051 0.004
## 321 T2_Saturday BPAS_5 -0.07339299 0.042
## 324 BPAS_1 BPAS_5 -0.26892102 0.000
## 325 BPAS_2 BPAS_5 0.44604738 0.000
## 327 MMSAS_15 BPAS_6 -0.18565100 0.000
## 333 MMSAS_25 BPAS_6 -0.09058984 0.012
## 334 MMSAS_27 BPAS_6 -0.11385211 0.002
## 339 T1_Binge_drinking BPAS_6 0.07274365 0.045
## 349 BPAS_1 BPAS_6 -0.23884618 0.000
## 350 BPAS_2 BPAS_6 0.42333523 0.000
## 351 BPAS_5 BPAS_6 0.71603620 0.000
## 353 MMSAS_15 BPAS_8 0.25597481 0.000
## 354 MMSAS_19 BPAS_8 0.07862029 0.030
## 356 MMSAS_21 BPAS_8 0.09125592 0.012
## 357 MMSAS_22 BPAS_8 0.08987709 0.013
## 358 MMSAS_24 BPAS_8 0.08523208 0.018
## 359 MMSAS_25 BPAS_8 0.19432750 0.000
## 360 MMSAS_27 BPAS_8 0.24647492 0.000
## 361 T1_Sweets BPAS_8 0.09076326 0.012
## 365 T1_Binge_drinking BPAS_8 0.07540936 0.038
## 368 T2_Sweets BPAS_8 0.07767400 0.032
## 369 T2_LM_alcohol BPAS_8 -0.09574987 0.008
## 370 T2_Wednesday BPAS_8 0.12374107 0.001
## 371 T2_Thursday BPAS_8 0.11238116 0.002
## 372 T2_Saturday BPAS_8 0.09713734 0.007
## 373 T2_N_shot_spirit BPAS_8 0.09128053 0.012
## 374 T2_Cannabis_YN BPAS_8 0.14873891 0.000
## 375 BPAS_1 BPAS_8 0.51878748 0.000
## 376 BPAS_2 BPAS_8 -0.38527453 0.000
## 377 BPAS_5 BPAS_8 -0.42462878 0.000
## 378 BPAS_6 BPAS_8 -0.39440599 0.000
## 397 T2_Wednesday T3_Fruit 0.13391658 0.019
## 406 BPAS_8 T3_Fruit 0.11657991 0.042
## 415 MMSAS_27 T3_Vegetables 0.09732272 0.043
## 435 T3_Fruit T3_Vegetables 0.67275868 0.000
## 443 MMSAS_25 T3_Sweets 0.09583213 0.046
## 464 T3_Fruit T3_Sweets 0.14862890 0.002
## 489 BPAS_1 T3_Tuesday -0.11365170 0.047
## 493 BPAS_8 T3_Tuesday -0.12620651 0.027
## 510 T1_Binge_drinking T3_Wednesday 0.09829035 0.042
## 525 T3_Fruit T3_Wednesday -0.11753118 0.014
## 527 T3_Sweets T3_Wednesday -0.11389855 0.017
## 528 T3_Tuesday T3_Wednesday 0.30186620 0.000
## 560 T3_Tuesday T3_Thursday 0.31880626 0.000
## 561 T3_Wednesday T3_Thursday 0.26032390 0.000
## 564 MMSAS_19 T3_Friday -0.09887548 0.039
## 592 T3_Sweets T3_Friday -0.10745242 0.025
## 594 T3_Wednesday T3_Friday 0.14225738 0.003
## 595 T3_Thursday T3_Friday 0.14583719 0.002
## 604 MMSAS_27 T3_N_mixed_drinks -0.11531612 0.017
## 607 T1_Friday T3_N_mixed_drinks -0.11571984 0.017
## 630 T3_Friday T3_N_mixed_drinks 0.14482862 0.003
## 662 T3_Tuesday T3_Drunkenness 0.14564159 0.002
## 663 T3_Wednesday T3_Drunkenness 0.25455954 0.000
## 664 T3_Thursday T3_Drunkenness 0.22764361 0.000
## 665 T3_Friday T3_Drunkenness 0.24018512 0.000
## 666 T3_N_mixed_drinks T3_Drunkenness 0.15600840 0.001
## 699 T3_Wednesday T3_BYAACQ_8 0.14340957 0.003
## 701 T3_Friday T3_BYAACQ_8 0.16511706 0.001
## 703 T3_Drunkenness T3_BYAACQ_8 0.29529706 0.000
## 706 MMSAS_19 T3_Q125_24 -0.12507628 0.011
## 731 BPAS_8 T3_Q125_24 -0.16827973 0.004
## 736 T3_Wednesday T3_Q125_24 0.12630138 0.010
## 737 T3_Thursday T3_Q125_24 0.19905472 0.000
## 738 T3_Friday T3_Q125_24 0.17479003 0.000
## 739 T3_N_mixed_drinks T3_Q125_24 0.12377228 0.012
## 740 T3_Drunkenness T3_Q125_24 0.22077268 0.000
## 741 T3_BYAACQ_8 T3_Q125_24 0.23781340 0.000
## 761 T2_Thursday T3_Cannabis_YN 0.12813406 0.031
## 774 T3_Wednesday T3_Cannabis_YN 0.17104568 0.000
## 775 T3_Thursday T3_Cannabis_YN 0.14201563 0.004
## 776 T3_Friday T3_Cannabis_YN 0.12494998 0.011
## 777 T3_N_mixed_drinks T3_Cannabis_YN 0.20780191 0.000
## 778 T3_Drunkenness T3_Cannabis_YN 0.23469386 0.000
## 779 T3_BYAACQ_8 T3_Cannabis_YN 0.14304015 0.004
## 780 T3_Q125_24 T3_Cannabis_YN 0.16824050 0.001
## 783 MMSAS_19 T3_Cannabis_frequency 0.10030216 0.042
## 800 T2_Thursday T3_Cannabis_frequency 0.14859856 0.012
## 813 T3_Wednesday T3_Cannabis_frequency 0.17012898 0.001
## 814 T3_Thursday T3_Cannabis_frequency 0.13726799 0.005
## 815 T3_Friday T3_Cannabis_frequency 0.11484298 0.019
## 816 T3_N_mixed_drinks T3_Cannabis_frequency 0.25808567 0.000
## 817 T3_Drunkenness T3_Cannabis_frequency 0.17572605 0.000
## 819 T3_Q125_24 T3_Cannabis_frequency 0.17188846 0.000
## 820 T3_Cannabis_YN T3_Cannabis_frequency 0.81587140 0.000
## 822 MMSAS_15 T3_Drugs_frequency -0.10556981 0.032
## 826 MMSAS_22 T3_Drugs_frequency -0.09720097 0.049
## 828 MMSAS_25 T3_Drugs_frequency -0.10274107 0.037
## 835 T1_DrunkDrunkenness T3_Drugs_frequency 0.10950050 0.028
## 844 BPAS_1 T3_Drugs_frequency -0.14472332 0.014
## 852 T3_Tuesday T3_Drugs_frequency 0.20802998 0.000
## 853 T3_Wednesday T3_Drugs_frequency 0.14066226 0.004
## 856 T3_N_mixed_drinks T3_Drugs_frequency 0.10939942 0.027
## 857 T3_Drunkenness T3_Drugs_frequency 0.21840021 0.000
## 859 T3_Q125_24 T3_Drugs_frequency 0.12824909 0.009
## 860 T3_Cannabis_YN T3_Drugs_frequency 0.21394644 0.000
## 861 T3_Cannabis_frequency T3_Drugs_frequency 0.15943255 0.001
The problem is that the package does not reverse automatically negatively correlated indicators to compute, for example, Chrombach’s Alpha. So, reversing…
#reverse
data$BPAS_1=8-data$BPAS_1
data$BPAS_8=8-data$BPAS_8
measurements<-constructs(
reflective('MMSAS',paste0("MMSAS_",c(1:3, 13:18, 23, 24,26))),
composite('DrunkB_t1',c('T1_Binge_drinking','T1_DrunkDrunkenness'), weights = mode_A),
composite('DrunkB_t2',c("T2_Binge_drinking","T2_Drunkenness" ), weights = mode_A),
reflective('BSAS',paste0("BSAS_",c(1:11))),
composite('DrunkB_t3',c("T3_Binge_drinking","T3_Drunkenness"), weights = mode_A)
)
causal model
modelSEM <- relationships(
paths(from = c("MMSAS",'BSAS','DrunkB_t1','DrunkB_t2'), to = c("DrunkB_t3")),
paths(from = c("MMSAS",'BSAS','DrunkB_t1'), to = c("DrunkB_t2")),
paths(from = "MMSAS", to = c("DrunkB_t1",'BSAS')),
paths(from = "BSAS", to = c("DrunkB_t1"))
)
SEM
fit_SEM=estimate_pls(data=data,measurement_model=measurements,structural_model=modelSEM)
## Generating the seminr model
## All 908 observations are valid.
Path coefficients
fit_summary<-summary(fit_SEM,fit.meas=TRUE)
## Warning in sqrt(rho %*% t(rho)): NaNs produced
fit_summary$paths
## DrunkB_t3 DrunkB_t2 DrunkB_t1 BSAS
## R^2 0.012 0.531 0.055 0.059
## AdjR^2 0.008 0.530 0.052 0.058
## MMSAS -0.011 0.089 0.216 0.243
## BSAS 0.022 0.100 0.050 .
## DrunkB_t1 0.156 0.684 . .
## DrunkB_t2 -0.123 . . .
Bootstraped results
boot_mobi_pls <- bootstrap_model(seminr_model = fit_SEM,
nboot = 50,
cores = 5)
## Bootstrapping model using seminr...
## Bootstrapping encountered a WARNING: 8 bootstrap iterations failed to converge (possibly due to PLSc).
## These failed iterations are excluded from the reported bootstrap statistics.
## SEMinR Model successfully bootstrapped
boot_summary<-summary(boot_mobi_pls)
boot_summary$bootstrapped_paths
## Original Est. Bootstrap Mean Bootstrap SD T Stat.
## MMSAS -> BSAS 0.243 0.248 0.038 6.457
## MMSAS -> DrunkB_t1 0.216 0.214 0.037 5.923
## MMSAS -> DrunkB_t2 0.089 0.091 0.034 2.616
## MMSAS -> DrunkB_t3 -0.011 -0.034 0.057 -0.189
## BSAS -> DrunkB_t1 0.050 0.062 0.038 1.297
## BSAS -> DrunkB_t2 0.100 0.112 0.033 2.991
## BSAS -> DrunkB_t3 0.022 0.027 0.040 0.560
## DrunkB_t1 -> DrunkB_t2 0.684 0.675 0.043 16.085
## DrunkB_t1 -> DrunkB_t3 0.156 0.197 0.106 1.463
## DrunkB_t2 -> DrunkB_t3 -0.123 -0.146 0.084 -1.465
## 2.5% CI 97.5% CI
## MMSAS -> BSAS 0.165 0.303
## MMSAS -> DrunkB_t1 0.136 0.280
## MMSAS -> DrunkB_t2 0.024 0.144
## MMSAS -> DrunkB_t3 -0.126 0.062
## BSAS -> DrunkB_t1 0.000 0.119
## BSAS -> DrunkB_t2 0.050 0.172
## BSAS -> DrunkB_t3 -0.041 0.100
## DrunkB_t1 -> DrunkB_t2 0.584 0.753
## DrunkB_t1 -> DrunkB_t3 -0.011 0.445
## DrunkB_t2 -> DrunkB_t3 -0.286 0.008
specific_effect_significance(boot_seminr_model = boot_mobi_pls,
from = "MMSAS",
through = c("DrunkB_t1"),
to = "DrunkB_t2",
alpha = 0.05)
## Original Est. Bootstrap Mean Bootstrap SD T Stat. 2.5% CI
## 0.14794820 0.14451272 0.02680431 5.51956857 0.09560063
## 97.5% CI
## 0.19637456
boot_summary$bootstrapped_paths
## Original Est. Bootstrap Mean Bootstrap SD T Stat.
## MMSAS -> BSAS 0.243 0.248 0.038 6.457
## MMSAS -> DrunkB_t1 0.216 0.214 0.037 5.923
## MMSAS -> DrunkB_t2 0.089 0.091 0.034 2.616
## MMSAS -> DrunkB_t3 -0.011 -0.034 0.057 -0.189
## BSAS -> DrunkB_t1 0.050 0.062 0.038 1.297
## BSAS -> DrunkB_t2 0.100 0.112 0.033 2.991
## BSAS -> DrunkB_t3 0.022 0.027 0.040 0.560
## DrunkB_t1 -> DrunkB_t2 0.684 0.675 0.043 16.085
## DrunkB_t1 -> DrunkB_t3 0.156 0.197 0.106 1.463
## DrunkB_t2 -> DrunkB_t3 -0.123 -0.146 0.084 -1.465
## 2.5% CI 97.5% CI
## MMSAS -> BSAS 0.165 0.303
## MMSAS -> DrunkB_t1 0.136 0.280
## MMSAS -> DrunkB_t2 0.024 0.144
## MMSAS -> DrunkB_t3 -0.126 0.062
## BSAS -> DrunkB_t1 0.000 0.119
## BSAS -> DrunkB_t2 0.050 0.172
## BSAS -> DrunkB_t3 -0.041 0.100
## DrunkB_t1 -> DrunkB_t2 0.584 0.753
## DrunkB_t1 -> DrunkB_t3 -0.011 0.445
## DrunkB_t2 -> DrunkB_t3 -0.286 0.008
visualization
plot(boot_mobi_pls)
<div id="htmlwidget-1e77aa8128ae60a01122" style="width:672px;height:480px;" class="grViz html-widget"></div>
<script type="application/json" data-for="htmlwidget-1e77aa8128ae60a01122">{"x":{"diagram":"digraph G {\n\n// ----------------------\n// General graph settings\n// ----------------------\ngraph [\ncharset = \"UTF-8\",\nlayout = dot,\nlabel = \"\",\nfontsize = 24,\nfontcolor = black,\nfontname = helvetica,\nrankdir = LR,\nlabelloc = t,\nsplines = TRUE\nbgcolor = transparent\n]\n\n// --------------------\n// The structural model\n// --------------------\nsubgraph sm {\nrankdir = LR;\nnode [\nshape = ellipse,\ncolor = black,\nfillcolor = white,\nstyle = filled,\nfontsize = 12,\nfontcolor = black,\nheight = 1.05,\nwidth = 1.19166666666667,\nfontname = helvetica,\nfixedsize = true\n]\n\"MMSAS\" [label=<<B>MMSAS <\/B>>, shape = ellipse]\n\"BSAS\" [label=<<B>BSAS <\/B><BR /><FONT POINT-SIZE=\"10\">r² = 0.059<\/FONT>>, shape = ellipse]\n\"DrunkB_t1\" [label=<<B>DrunkB_t1 <\/B><BR /><FONT POINT-SIZE=\"10\">r² = 0.055<\/FONT>>, shape = hexagon]\n\"DrunkB_t2\" [label=<<B>DrunkB_t2 <\/B><BR /><FONT POINT-SIZE=\"10\">r² = 0.531<\/FONT>>, shape = hexagon]\n\"DrunkB_t3\" [label=<<B>DrunkB_t3 <\/B><BR /><FONT POINT-SIZE=\"10\">r² = 0.012<\/FONT>>, shape = hexagon]\nedge [\ncolor = black,\nfontsize = 9,\nfontcolor = black,\nfontname = helvetica,\ndir = both,\narrowhead = normal,\narrowtail = none\n]\n\"MMSAS\" -> {\"DrunkB_t3\"}[weight = 1, label = < 𝛽 = -0.011<BR /><FONT POINT-SIZE=\"7\">95% CI [-0.126, 0.062] <\/FONT> >, penwidth = 0.555, style = dashed, color = black]\n\"BSAS\" -> {\"DrunkB_t3\"}[weight = 1, label = < 𝛽 = 0.022<BR /><FONT POINT-SIZE=\"7\">95% CI [-0.041, 0.1] <\/FONT> >, penwidth = 0.61, style = solid, color = black]\n\"DrunkB_t1\" -> {\"DrunkB_t3\"}[weight = 1, label = < 𝛽 = 0.156<BR /><FONT POINT-SIZE=\"7\">95% CI [-0.011, 0.445] <\/FONT> >, penwidth = 1.28, style = solid, color = black]\n\"DrunkB_t2\" -> {\"DrunkB_t3\"}[weight = 1, label = < 𝛽 = -0.123<BR /><FONT POINT-SIZE=\"7\">95% CI [-0.286, 0.008] <\/FONT> >, penwidth = 1.115, style = dashed, color = black]\n\"MMSAS\" -> {\"DrunkB_t2\"}[weight = 1, label = < 𝛽 = 0.089**<BR /><FONT POINT-SIZE=\"7\">95% CI [0.024, 0.144] <\/FONT> >, penwidth = 0.945, style = solid, color = black]\n\"BSAS\" -> {\"DrunkB_t2\"}[weight = 1, label = < 𝛽 = 0.1**<BR /><FONT POINT-SIZE=\"7\">95% CI [0.05, 0.172] <\/FONT> >, penwidth = 1, style = solid, color = black]\n\"DrunkB_t1\" -> {\"DrunkB_t2\"}[weight = 1, label = < 𝛽 = 0.684***<BR /><FONT POINT-SIZE=\"7\">95% CI [0.584, 0.753] <\/FONT> >, penwidth = 3.92, style = solid, color = black]\n\"MMSAS\" -> {\"DrunkB_t1\"}[weight = 1, label = < 𝛽 = 0.216***<BR /><FONT POINT-SIZE=\"7\">95% CI [0.136, 0.28] <\/FONT> >, penwidth = 1.58, style = solid, color = black]\n\"MMSAS\" -> {\"BSAS\"}[weight = 1, label = < 𝛽 = 0.243***<BR /><FONT POINT-SIZE=\"7\">95% CI [0.165, 0.303] <\/FONT> >, penwidth = 1.715, style = solid, color = black]\n\"BSAS\" -> {\"DrunkB_t1\"}[weight = 1, label = < 𝛽 = 0.05<BR /><FONT POINT-SIZE=\"7\">95% CI [0, 0.119] <\/FONT> >, penwidth = 0.75, style = solid, color = black]\n}\n// ---------------------\n// The measurement model\n// ---------------------\n\nsubgraph construct_1 {\nnode [\nshape = box,\ncolor = dimgrey,\nfillcolor = white,\nstyle = filled,\nfontsize = 8,\nfontcolor = black,\nheight = 0.175,\nwidth = 1.72916666666667,\nfontname = helvetica,\nfixedsize = true\n]\n\"MMSAS_1\" [label = \"MMSAS_1\", shape = box]\n\"MMSAS_2\" [label = \"MMSAS_2\", shape = box]\n\"MMSAS_3\" [label = \"MMSAS_3\", shape = box]\n\"MMSAS_13\" [label = \"MMSAS_13\", shape = box]\n\"MMSAS_14\" [label = \"MMSAS_14\", shape = box]\n\"MMSAS_15\" [label = \"MMSAS_15\", shape = box]\n\"MMSAS_16\" [label = \"MMSAS_16\", shape = box]\n\"MMSAS_17\" [label = \"MMSAS_17\", shape = box]\n\"MMSAS_18\" [label = \"MMSAS_18\", shape = box]\n\"MMSAS_23\" [label = \"MMSAS_23\", shape = box]\n\"MMSAS_24\" [label = \"MMSAS_24\", shape = box]\n\"MMSAS_26\" [label = \"MMSAS_26\", shape = box]\nedge [\ncolor = dimgrey,\nfontsize = 7,\nfontcolor = black,\nfontname = helvetica,\nminlen = 1,\ndir = both\narrowhead = none\narrowtail = normal\n]\n\"MMSAS_1\" -> {\"MMSAS\"}[weight = 1000, label = < 𝜆 = 0.121 >, penwidth = 0.863, style = solid, color = dimgrey]\n\"MMSAS_2\" -> {\"MMSAS\"}[weight = 1000, label = < 𝜆 = 0.062 >, penwidth = 0.686, style = solid, color = dimgrey]\n\"MMSAS_3\" -> {\"MMSAS\"}[weight = 1000, label = < 𝜆 = -0.002 >, penwidth = 0.506, style = dashed, color = dimgrey]\n\"MMSAS_13\" -> {\"MMSAS\"}[weight = 1000, label = < 𝜆 = 0.787*** >, penwidth = 2.861, style = solid, color = dimgrey]\n\"MMSAS_14\" -> {\"MMSAS\"}[weight = 1000, label = < 𝜆 = 0.636*** >, penwidth = 2.408, style = solid, color = dimgrey]\n\"MMSAS_15\" -> {\"MMSAS\"}[weight = 1000, label = < 𝜆 = 0.885*** >, penwidth = 3.155, style = solid, color = dimgrey]\n\"MMSAS_16\" -> {\"MMSAS\"}[weight = 1000, label = < 𝜆 = 0.486*** >, penwidth = 1.958, style = solid, color = dimgrey]\n\"MMSAS_17\" -> {\"MMSAS\"}[weight = 1000, label = < 𝜆 = 0.465*** >, penwidth = 1.895, style = solid, color = dimgrey]\n\"MMSAS_18\" -> {\"MMSAS\"}[weight = 1000, label = < 𝜆 = 0.46*** >, penwidth = 1.88, style = solid, color = dimgrey]\n\"MMSAS_23\" -> {\"MMSAS\"}[weight = 1000, label = < 𝜆 = 0.425*** >, penwidth = 1.775, style = solid, color = dimgrey]\n\"MMSAS_24\" -> {\"MMSAS\"}[weight = 1000, label = < 𝜆 = 0.069 >, penwidth = 0.707, style = solid, color = dimgrey]\n\"MMSAS_26\" -> {\"MMSAS\"}[weight = 1000, label = < 𝜆 = 0.888*** >, penwidth = 3.164, style = solid, color = dimgrey]\n\n}\nsubgraph construct_2 {\nnode [\nshape = box,\ncolor = dimgrey,\nfillcolor = white,\nstyle = filled,\nfontsize = 8,\nfontcolor = black,\nheight = 0.175,\nwidth = 1.72916666666667,\nfontname = helvetica,\nfixedsize = true\n]\n\"BSAS_1\" [label = \"BSAS_1\", shape = box]\n\"BSAS_2\" [label = \"BSAS_2\", shape = box]\n\"BSAS_3\" [label = \"BSAS_3\", shape = box]\n\"BSAS_4\" [label = \"BSAS_4\", shape = box]\n\"BSAS_5\" [label = \"BSAS_5\", shape = box]\n\"BSAS_6\" [label = \"BSAS_6\", shape = box]\n\"BSAS_7\" [label = \"BSAS_7\", shape = box]\n\"BSAS_8\" [label = \"BSAS_8\", shape = box]\n\"BSAS_9\" [label = \"BSAS_9\", shape = box]\n\"BSAS_10\" [label = \"BSAS_10\", shape = box]\n\"BSAS_11\" [label = \"BSAS_11\", shape = box]\nedge [\ncolor = dimgrey,\nfontsize = 7,\nfontcolor = black,\nfontname = helvetica,\nminlen = 1,\ndir = both\narrowhead = none\narrowtail = normal\n]\n\"BSAS_1\" -> {\"BSAS\"}[weight = 1000, label = < 𝜆 = 0.305** >, penwidth = 1.415, style = solid, color = dimgrey]\n\"BSAS_2\" -> {\"BSAS\"}[weight = 1000, label = < 𝜆 = 0.601*** >, penwidth = 2.303, style = solid, color = dimgrey]\n\"BSAS_3\" -> {\"BSAS\"}[weight = 1000, label = < 𝜆 = 0.726*** >, penwidth = 2.678, style = solid, color = dimgrey]\n\"BSAS_4\" -> {\"BSAS\"}[weight = 1000, label = < 𝜆 = 0.537*** >, penwidth = 2.111, style = solid, color = dimgrey]\n\"BSAS_5\" -> {\"BSAS\"}[weight = 1000, label = < 𝜆 = 0.278* >, penwidth = 1.334, style = solid, color = dimgrey]\n\"BSAS_6\" -> {\"BSAS\"}[weight = 1000, label = < 𝜆 = 0.326** >, penwidth = 1.478, style = solid, color = dimgrey]\n\"BSAS_7\" -> {\"BSAS\"}[weight = 1000, label = < 𝜆 = 0.47*** >, penwidth = 1.91, style = solid, color = dimgrey]\n\"BSAS_8\" -> {\"BSAS\"}[weight = 1000, label = < 𝜆 = 0.213 >, penwidth = 1.139, style = solid, color = dimgrey]\n\"BSAS_9\" -> {\"BSAS\"}[weight = 1000, label = < 𝜆 = -0.034 >, penwidth = 0.602, style = dashed, color = dimgrey]\n\"BSAS_10\" -> {\"BSAS\"}[weight = 1000, label = < 𝜆 = 0.349** >, penwidth = 1.547, style = solid, color = dimgrey]\n\"BSAS_11\" -> {\"BSAS\"}[weight = 1000, label = < 𝜆 = 1.115*** >, penwidth = 3.845, style = solid, color = dimgrey]\n\n}\nsubgraph construct_3 {\nnode [\nshape = box,\ncolor = dimgrey,\nfillcolor = white,\nstyle = filled,\nfontsize = 8,\nfontcolor = black,\nheight = 0.175,\nwidth = 1.72916666666667,\nfontname = helvetica,\nfixedsize = true\n]\n\"T1_Binge_drinking\" [label = \"T1_Binge_drinking\", shape = box]\n\"T1_DrunkDrunkenness\" [label = \"T1_DrunkDrunkenness\", shape = box]\nedge [\ncolor = dimgrey,\nfontsize = 7,\nfontcolor = black,\nfontname = helvetica,\nminlen = 1,\ndir = both\narrowhead = none\narrowtail = normal\n]\n\"T1_Binge_drinking\" -> {\"DrunkB_t1\"}[weight = 1000, label = < 𝜆 = 0.879*** >, penwidth = 3.137, style = solid, color = dimgrey]\n\"T1_DrunkDrunkenness\" -> {\"DrunkB_t1\"}[weight = 1000, label = < 𝜆 = 0.856*** >, penwidth = 3.068, style = solid, color = dimgrey]\n\n}\nsubgraph construct_4 {\nnode [\nshape = box,\ncolor = dimgrey,\nfillcolor = white,\nstyle = filled,\nfontsize = 8,\nfontcolor = black,\nheight = 0.175,\nwidth = 1.72916666666667,\nfontname = helvetica,\nfixedsize = true\n]\n\"T2_Binge_drinking\" [label = \"T2_Binge_drinking\", shape = box]\n\"T2_Drunkenness\" [label = \"T2_Drunkenness\", shape = box]\nedge [\ncolor = dimgrey,\nfontsize = 7,\nfontcolor = black,\nfontname = helvetica,\nminlen = 1,\ndir = both\narrowhead = none\narrowtail = normal\n]\n\"T2_Binge_drinking\" -> {\"DrunkB_t2\"}[weight = 1000, label = < 𝜆 = 0.899*** >, penwidth = 3.197, style = solid, color = dimgrey]\n\"T2_Drunkenness\" -> {\"DrunkB_t2\"}[weight = 1000, label = < 𝜆 = 0.882*** >, penwidth = 3.146, style = solid, color = dimgrey]\n\n}\nsubgraph construct_5 {\nnode [\nshape = box,\ncolor = dimgrey,\nfillcolor = white,\nstyle = filled,\nfontsize = 8,\nfontcolor = black,\nheight = 0.175,\nwidth = 1.72916666666667,\nfontname = helvetica,\nfixedsize = true\n]\n\"T3_Binge_drinking\" [label = \"T3_Binge_drinking\", shape = box]\n\"T3_Drunkenness\" [label = \"T3_Drunkenness\", shape = box]\nedge [\ncolor = dimgrey,\nfontsize = 7,\nfontcolor = black,\nfontname = helvetica,\nminlen = 1,\ndir = both\narrowhead = normal\narrowtail = none\n]\n\"DrunkB_t3\" -> {\"T3_Binge_drinking\"}[weight = 1000, label = < 𝜆 = 0.72*** >, penwidth = 2.66, style = solid, color = dimgrey]\n\"DrunkB_t3\" -> {\"T3_Drunkenness\"}[weight = 1000, label = < 𝜆 = 0.948*** >, penwidth = 3.344, style = solid, color = dimgrey]\n\n}\n}","config":{"engine":"dot","options":null}},"evals":[],"jsHooks":[]}</script>
save_plot('./2_Aresi2021_files/Aresi2021_PLS.png')
## NULL
The results substantiate higher motivation to study abroad (MMSAS) is associated with higher sociocultural adaptation (BSAS), which, in turn, is associated also with drunken behaviour at time 2. So, the BSAS is portrayed as a mediator in the MMSAS’ influence on drunken behaviour.
Reliability
But the above results are valid just if the scales used are valid and reliable.
fit_summary$reliability
## alpha rhoC AVE rhoA
## MMSAS 0.875 0.766 0.289 0.889
## BSAS 0.863 0.751 0.280 0.909
## DrunkB_t1 0.672 0.859 0.753 0.675
## DrunkB_t2 0.740 0.885 0.793 0.743
## DrunkB_t3 0.631 0.827 0.708 0.901
##
## Alpha, rhoC, and rhoA should exceed 0.7 while AVE should exceed 0.5