LISREL Syntax

LISREL syntax to accompany models and analyses in:

Preacher, K. J. (2006). Testing complex correlational hypotheses using structural equation modeling. Structural Equation Modeling, 13, 520-543.

LISREL syntax for computing the bivariate correlation between X and Y.

 TI bivariate correlation
 DA NI=2 NO=40
 CM
 0.958365
 0.231046 1.163310
 MO NX=2 NK=2 LX=DI,FR PH=ST TD=ZE
 LK
 X Y
 ST .5 LX 1 1 LX 2 2 PH 2 1
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF

LISREL syntax for computing the bivariate correlation between X and Y simultaneously in two groups, with equality constraint in place.

 GROUP 1 bivariate correlation
 DA NG=2 NI=2 NO=40
 CM
 0.958365
 0.231046 1.163310
 MO NX=2 NK=2 LX=DI,FR PH=ST TD=ZE
 LK
 X Y
 ST .5 LX 1 1 LX 2 2 PH 2 1
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF
 GROUP 2 bivariate correlation
 DA NI=2 NO=40
 CM
 0.923433
 0.021623 1.263412
 MO NX=2 NK=2 LX=DI,FR PH=ST TD=ZE
 LK
 X Y
 ST .5 LX 1 1 LX 2 2 PH 2 1
 EQ PH(1,2,1) PH(2,1)
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF

LISREL syntax for computing the partial correlation between X and Y, controlling both for W.

 TI partial correlation
 DA NI=3 NO=40
 CM
 1.405466
 0.633555 0.958365
 0.359973 0.231046 1.163310
 MO NX=3 NK=3 LX=FU,FI PH=SY,FI TD=ZE
 LK
 W X Y
 FR PH 1 1 PH 3 2 LX 2 1 LX 2 2 LX 3 1 LX 3 3
 VA 1 PH 2 2 PH 3 3 LX 1 1
 ST .5 PH 1 1 PH 3 2 LX 2 1 LX 2 2 LX 3 1 LX 3 3
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF

LISREL syntax for computing the partial correlation between X and Y simultaneously in two groups, controlling both X and Y for W, with equality constraint in place. This syntax includes the stereotype data discussed in Example 1.

 GROUP 1 partial correlation
 DA NG=2 NI=3 NO=65
 CM
 1.371222
 0.308131 0.875000
 0.063412 0.024306 0.085601
 MO NX=3 NK=3 LX=FU,FI PH=SY,FI TD=ZE
 LK
 W X Y
 FR PH 1 1 PH 3 2 LX 2 1 LX 2 2 LX 3 1 LX 3 3
 VA 1 PH 2 2 PH 3 3 LX 1 1
 ST .5 PH 1 1 PH 3 2 LX 2 1 LX 2 2 LX 3 1 LX 3 3
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF
 GROUP 2 partial correlation
 DA NI=3 NO=49
 CM
 1.593040
 0.104162 1.034864
 0.043919 0.123677 0.086420
 MO NX=3 NK=3 LX=FU,FI PH=SY,FI TD=ZE
 LK
 W X Y
 FR PH 1 1 PH 3 2 LX 2 1 LX 2 2 LX 3 1 LX 3 3
 VA 1 PH 2 2 PH 3 3 LX 1 1
 ST .5 PH 1 1 PH 3 2 LX 2 1 LX 2 2 LX 3 1 LX 3 3
 EQ PH(1,3,2) PH(3,2)
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF

LISREL syntax for computing the semipartial correlation between X and Y, controlling Y for W.

 TI semipartial correlation
 DA NI=3 NO=40
 CM
 1.405466
 0.633555 0.958365
 0.359973 0.231046 1.163310
 MO NX=3 NK=3 LX=FU,FI PH=SY,FI TD=ZE
 LK
 W X Y
 FR PH 3 1 PH 3 2 LX 1 1 LX 2 2 LX 3 3 LX 2 1
 VA 1 PH 1 1 PH 2 2 PH 3 3
 ST .5 PH 3 1 PH 3 2 LX 1 1 LX 2 2 LX 3 3 LX 2 1
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF

LISREL syntax for computing the semipartial correlation between X and Y simultaneously in two groups, controlling Y for W, with equality constraint in place.

 GROUP 1 semipartial correlation
 DA NG=2 NI=3 NO=40
 CM
 1.405466
 0.633555 0.958365
 0.359973 0.231046 1.163310
 MO NX=3 NK=3 LX=FU,FI PH=SY,FI TD=ZE
 LK
 W X Y
 FR PH 3 1 PH 3 2 LX 1 1 LX 2 2 LX 3 3 LX 2 1
 VA 1 PH 1 1 PH 2 2 PH 3 3
 ST .5 PH 3 1 PH 3 2 LX 1 1 LX 2 2 LX 3 3 LX 2 1
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF
 GROUP 2 semipartial correlation
 DA NI=3 NO=40
 CM
 1.357541
 0.614773 0.923433
 0.265435 0.021623 1.263412
 MO NX=3 NK=3 LX=FU,FI PH=SY,FI TD=ZE
 LK
 W X Y
 FR PH 3 1 PH 3 2 LX 1 1 LX 2 2 LX 3 3 LX 2 1
 VA 1 PH 1 1 PH 2 2 PH 3 3
 ST .5 PH 3 1 PH 3 2 LX 1 1 LX 2 2 LX 3 3 LX 2 1
 EQ PH(1,3,2) PH(3,2)
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF

LISREL syntax for computing the squared bivariate correlation between X and Y.

 TI corr_squared
 DA NI=2 NO=40
 CM
 0.958365
 0.231046 1.163310
 MO NX=2 NK=2 LX=FU,FI PH=SY,FI TD=ZE AP=1
 LK
 X Y
 FR LX 1 1 LX 2 2 PH 2 1
 ST .5 LX 1 1 LX 2 2 PH 2 1
 VA 1 PH 1 1 PH 2 2
 CO PA(1)=PH(2,1)**2
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF

LISREL syntax for computing the unsquared and squared semipartial correlation between X and Y simultaneously in three groups, controlling X for W, with equality constraints in place. This syntax includes the data for Democrats, Republicans, and Independents discussed in Example 2.

 GROUP 1 semi corr Dem
 DA NG=3 NI=3 NO=408
 CM
 2.564644
 0.734038 5.394397
 0.418448 1.223178 2.489353
 MO NX=3 NK=3 LX=FU,FI PH=SY,FI TD=ZE AP=1
 LK
 W X Y
 FR PH 3 1 PH 3 2 LX 1 1 LX 2 2 LX 3 3 LX 2 1
 VA 1 PH 1 1 PH 2 2 PH 3 3
 ST .5 PH 3 1 PH 3 2 LX 1 1 LX 2 2 LX 3 3 LX 2 1
 CO PA(1)=PH(3,2)**2
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF
 GROUP 2 semi corr Rep
 DA NI=3 NO=459
 CM
 2.476566
 0.162600 6.525559
 0.175854 1.112253 1.752890
 MO NX=3 NK=3 LX=FU,FI PH=SY,FI TD=ZE AP=1
 LK
 W X Y
 FR PH 3 1 PH 3 2 LX 1 1 LX 2 2 LX 3 3 LX 2 1
 VA 1 PH 1 1 PH 2 2 PH 3 3
 ST .5 PH 3 1 PH 3 2 LX 1 1 LX 2 2 LX 3 3 LX 2 1
 CO PA(1)=PH(3,2)**2
 EQ PH(1,3,2) PH(3,2)
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF
 GROUP 3 semi corr Ind
 DA NI=3 NO=411
 CM
 2.138116
 0.160980 5.305798
 0.270231 1.174910 2.095828
 MO NX=3 NK=3 LX=FU,FI PH=SY,FI TD=ZE AP=1
 LK
 W X Y
 FR PH 3 1 PH 3 2 LX 1 1 LX 2 2 LX 3 3 LX 2 1
 VA 1 PH 1 1 PH 2 2 PH 3 3
 ST .5 PH 3 1 PH 3 2 LX 1 1 LX 2 2 LX 3 3 LX 2 1
 CO PA(1)=PH(3,2)**2
 EQ PH(2,3,2) PH(3,2)
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF

LISREL syntax for testing a pattern hypothesis for bivariate correlations. This syntax includes the multi-trait multi-method data discussed in Example 3.

 TI bivariate correlation pattern hypothesis
 DA NI=6 NO=113
 CM
 1.00
 0.53 1.00
 0.56 0.44 1.00
 0.65 0.38 0.40 1.00
 0.42 0.52 0.30 0.56 1.00
 0.40 0.31 0.53 0.56 0.40 1.00
 MO NX=6 NK=6 LX=DI,FR PH=ST TD=ZE
 LK
 QS AS ES QP AP EP
 ST .5 LX 1 1 LX 2 2 LX 3 3 LX 4 4 LX 5 5 LX 6 6
 ST .5 PH 2 1 PH 3 1 PH 3 2 PH 4 1 PH 4 2 PH 4 3
 ST .5 PH 5 1 PH 5 2 PH 5 3 PH 5 4 PH 6 1 PH 6 2 PH 6 3 PH 6 4 PH 6 5
 EQ PH 2 1 PH 5 1 PH 4 2 PH 5 4
 EQ PH 3 1 PH 6 1 PH 4 3 PH 6 4
 EQ PH 3 2 PH 6 2 PH 5 3 PH 6 5
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF

LISREL syntax for testing a pattern hypothesis for partial correlations. This syntax includes the NLSY PIAT data discussed in Example 4.

 TI partial correlation pattern hypothesis
 DA NI=10 NO=1071
 CM
 0.249
 0.003 0.229
 0.025 0.157 0.283
 0.032 0.141 0.213 0.258
 -0.006 0.146 0.143 0.134 0.223
 0.024 0.171 0.269 0.217 0.178 0.377
 0.017 0.137 0.201 0.181 0.143 0.248 0.283
 -0.009 0.146 0.144 0.128 0.166 0.172 0.149 0.246
 0.028 0.178 0.278 0.231 0.180 0.342 0.257 0.195 0.459
 0.012 0.143 0.194 0.174 0.146 0.231 0.203 0.166 0.275 0.316
 MO NX=10 NK=10 LX=FU,FI PH=SY,FI TD=ZE
 LK
 SEX M08 R08 C08 M10 R10 C10 M12 R12 C12
 FR PH 1 1 PH 3 2 PH 4 2 PH 4 3 PH 5 2 PH 5 3 PH 5 4 PH 6 2 PH 6 3
 FR PH 6 4 PH 6 5 PH 7 2 PH 7 3 PH 7 4 PH 7 5 PH 7 6 PH 8 2 PH 8 3
 FR PH 8 4 PH 8 5 PH 8 6 PH 8 7 PH 9 2 PH 9 3 PH 9 4 PH 9 5 PH 9 6
 FR PH 9 7 PH 9 8 PH 10 2 PH 10 3 PH 10 4 PH 10 5 PH 10 6 PH 10 7
 FR PH 10 8 PH 10 9 LX 2 1 LX 3 1 LX 4 1 LX 5 1 LX 6 1 LX 7 1 LX 8 1
 FR LX 9 1 LX 10 1 LX 2 2 LX 3 3 LX 4 4 LX 5 5 LX 6 6 LX 7 7 LX 8 8
 FR LX 9 9 LX 10 10
 VA 1 PH 2 2 PH 3 3 PH 4 4 PH 5 5 PH 6 6 PH 7 7 PH 8 8 PH 9 9 PH 10 10
 VA 1 LX 1 1
 ST .5 PH 1 1 PH 3 2 PH 4 2 PH 4 3 PH 5 2 PH 5 3 PH 5 4 PH 6 2 PH 6 3
 ST .5 PH 6 4 PH 6 5 PH 7 2 PH 7 3 PH 7 4 PH 7 5 PH 7 6 PH 8 2 PH 8 3
 ST .5 PH 8 4 PH 8 5 PH 8 6 PH 8 7 PH 9 2 PH 9 3 PH 9 4 PH 9 5 PH 9 6
 ST .5 PH 9 7 PH 9 8 PH 10 2 PH 10 3 PH 10 4 PH 10 5 PH 10 6 PH 10 7
 ST .5 PH 10 8 PH 10 9 LX 2 2 LX 3 3 LX 4 4 LX 5 5 LX 6 6 LX 7 7 LX 8 8
 ST .5 LX 9 9 LX 10 10
 EQ PH 3 2 PH 6 5 PH 9 8
 EQ PH 4 2 PH 7 5 PH 10 8
 EQ PH 4 3 PH 7 6 PH 10 9
 PD
 OU ME=ML ND=4 XM EP=0.00001 IT=1000 NS AD=OFF