1 General Structural Equation (LISREL) Models General Structural Equation (LISREL) Models Week 3 #1 Week 3 #1 Multiple Group Models Multiple Group Models An extended multiple-group An extended multiple-group model: Religiosity & Sexual model: Religiosity & Sexual Morality in 2 countries (LISREL Morality in 2 countries (LISREL example) example) Computer programming for Computer programming for multiple-group models: a) LISREL multiple-group models: a) LISREL b) AMOS b) AMOS – See Week3Examples See Week3Examples
General Structural Equation (LISREL) Models. Week 3 #1 Multiple Group Models An extended multiple-group model: Religiosity & Sexual Morality in 2 countries (LISREL example) Computer programming for multiple-group models: a) LISREL b) AMOS See Week3Examples. - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
11
General Structural Equation (LISREL) ModelsGeneral Structural Equation (LISREL) Models
Week 3 #1 Week 3 #1 Multiple Group ModelsMultiple Group ModelsAn extended multiple-group model: An extended multiple-group model: Religiosity & Sexual Morality in 2 countries Religiosity & Sexual Morality in 2 countries (LISREL example) (LISREL example) Computer programming for multiple-group Computer programming for multiple-group models: a) LISREL b) AMOSmodels: a) LISREL b) AMOS
–See Week3ExamplesSee Week3Examples
2
MOST IMPORTANT FORM OF CONSTRAINT INVOLVES CONSTRAINING PARAMETERS ACROSS GROUPS
1
1 1 1
1
1 1 1
b1
1
1 1 1
1
1 1 1
b1
Group 1
Group 2
Constraint: b1group1 = b1group2
From last Friday 3
Multiple Group Models
Group 1 (male)Group 2 (female)
Equivalence of measurement coefficients
H0: Λ[1] = Λ[2]
lambda 1 [1] = lambda 1 [2] df=2
lambda 2 [1] = lambda 2 [2]
Eta1[1]
y111
y2ly1[1] 1
y3
ly2[1]1
Eta1[2]
y1
y2
y3
11
ly1[2] 1
ly2[2]1
From last Friday 4
Multiple Group Models
Eta1[1]
y1 e111
y2 e2ly1[1] 1
y3 e3
ly2[1]1
Eta1[2]
y1 e1
y2 e2
y3 e3
11
ly1[2] 1
ly2[2]1
Other equivalence tests possible:
1. Equivalence of variances of latent variables
H0: PSI-1[1] = PSI-1[2]
• This test will depend upon which ref. indicator used
2. Equivalence of error variances *
H0: Theta-eps[1] = Theta-eps[2] {entire matrix}
df=3 *and covariances if there are correlated errors
From last Friday 5
Multiple Group Models
Measurement model equivalence does not imply same mean levelsMeasurement model for Group 1 can be
identical to Group 2, yet the two groups can differ radically in terms of level.
• It is possible to have multiple group models with both common and unique items
• Example:
• Y1 Both countries: We should always trust our elected leaders
• Y2 Both countries: If my government told me to go to war, I’d go
• Y3 Both countries: We need more respect for government & authority
•Y4 (US): George Bush commands my respect because he is our President•Y4 (Canada) Paul Martin commands my respect because he is our Prime Minister
We might expect (if measurement equivalence holds):
lambda1[1] = lambda1[2]
lambda2[1] = lambda2[2]
BUT
lambda3[1] ≠ lambda3[2]
From last Friday 7
Multiple Group Models
• Should be careful with the use of reference indicators (and/or sensitive to the fact that apparently non-equivalent models might appear to be so simply because of a single (reference) indicator
• Example:
Group 1 Group 2
Lambda-1 1.0* 1.0*
Lambda-2 .50 1.0
Lambda-3 .75 1.5
Lambda-4 1.0 2.0
• These two groups appear to have measurement models that are very different, but….
Eta1
y1lambda-1
1
y2lambda-2 1
y3lambda-3 1
y4
lambda-4
1
From last Friday 8
Multiple Group ModelsGroup 1 Group 2
Lambda-1 1.0* 1.0*
Lambda-2 .50 1.0
Lambda-3 .75 1.5
Lambda-4 1.0 2.0
• These two groups appear to have measurement models that are very different, but….
If we change the reference indicator to Y2, we find:
Eta1
y1lambda-1
1
y2lambda-2 1
y3lambda-3 1
y4
lambda-4
1
Gr 1 Gr 2
Lambda1 2.0 1.0
Lambda2 1.0* 1.0*
Lambda3 1.5 1.5
Lambda4 2.0 2.0
From last Friday 9
Multiple Group Models
Modification Indices and what they mean in multiple-group models
Assuming LY[1] = LY[2]
(entire matrix)
Eta1
y11
1
y2lambda-1 1
y3lambda-2 1
y4
lambda-3
1
Example:
MODIFICATION INDICES:
Group 1 Group 2
Eta 1 Eta 1
Y1 --- Y1 ---
Y2 .382 Y2 .382
Y3 1.24 Y3 1.24
Y4 45.23 Y4 45.23
10
Multiple Group Models
Modification Indices and what they mean in multiple-group models
Assuming LY[1] = LY[2]
(entire matrix)
Eta1
y11
1
y2lambda-1 1
y3lambda-2 1
y4
lambda-3
1
Example:
MODIFICATION INDICES:
Group 1 Group 2
Eta 1 Eta 1
Y1 --- Y1 ---
Y2 .382 Y2 .382
Y3 1.24 Y3 1.24
Y4 45.23 Y4 45.23
Improvement in chi-square if equality constraint released
MULTIPLE GROUP MODELS: parameter significance tests
When a parameter is constrained to equality across 2 (or more) groups, “pooled” significance test (more power)
Possible to have a coefficient non-signif. in each of 2 groups yet significant when equality constraint imposed
Possible to have a coefficient that is not significant in each of two groups (e.g., +ve coefficient, NS, in one group, -ve, NS, in another) yet the difference between the groups is statistically significant
14
Tests in 3+ groups
2-group model can be extended to m groups (in theory, infinite number as long as minimum sample size requirements met in each group; in practice, some software packages have limits)
Must be careful about interpretation of Modification Indices In a model with [1]=[2]=[3] ≠ 0, then a MI will provide an indication of
how much the model improve if the parameter constraint is removed in the mth group only (e.g., MI in group 2 would test against a model in which the group 1 & group 3 {same} parameter are constrained to equality but the group 2 is allowed to differ)
15
MULTIPLE GROUP MODELS: Modification Indices (again)
Eta1
y11
1
y2lambda-1 1
y3lambda-2 1
y4
lambda-3
1
Group 1 MOD INDICES
Lambda 1 3.01
Lambda 2 1.52
Lambda 3 3.22
Group 2 MOD INDICES
Lambda 1 4.22
Lambda 2 3.99
Lambda 3 5.22
Group 3 MOD INDICES
Lambda 1 89.22
Lambda 2 6.11
Lambda 3 1.22
Model: LY[1]=LY[2]=LY[3]
Free LY(2,1) in group 3 but
LY(2,1) in group 1 = LY(2,1) in group 2
16
When do we have measurement equivalence STRONG equivalence:
all matrices identical, all groups (might possibly exclude variance of LV’s from this …
i.e., the PHI or PSI matrices) WEAKER equivalence (usually accepted)
Lambda matices identical, all groups Theta matrices could be different (and probably are),
either having the same form or not WEAKER YET:
Lambda matrices have the same form, some identical coefficients
17
Measurement coefficients, construct equation coefficients in multiple group models We usually need the measurement
equation coefficients to be equivalent in order to proceed with comparison of construct equation coefficients
18
Measurement coefficients, construct equation coefficients in multiple group models
We usually need the measurement equation coefficients to be equivalent in order to proceed with comparison of construct equation coefficients
For this reason, tests for measurement equivalence are usually not as rigorous as the “substantive” tests for construct equation coefficient equivalence (though instances of poor fit should be noted in any report of results)
1
1 1 1
1
1 1 1
gamma1[1]
1
1
1 1 1
1
1 1 1
gamma1[2]
1
19
LISREL PROGRAMMING FOR MULTIPLE GROUP MODELS
Basics:“Stack” the groups (one program after the next)In DA statement of first group, specify total number of groups: e.g.,
DA NG=2 NI=23 NO=1246NI specification (# of input var’s) applies onlyto group 1NO specification (# of observations) applies only to group 1
20
LISREL PROGRAMMING FOR MULTIPLE GROUP MODELS
DA NG=2 …Specify group 1 model as usualTitle for Group 2DA statement group 2 NI= NO=CM FI= [location of group 2 cov mtx]SE1 3 4 6 8 / LABELS [optional]MO NY= NX= NK= NE= + special options for matrix
specificationOU (as usual)
21
LISREL PROGRAMMING: MULTIPLE GROUPS
MO specification:LY=PS same pattern as prev. group
- for example, if group 1 specifies 2 LV’s with first three indicators on LV1, next 6 on LV2, this same specification will be copied to group 2
LY= IN invariant- same pattern and all free coefficients in this
matrix constrained to equality with corresponding coefficients in previous group
22
LISREL PROGRAMMING: MULTIPLE GROUPS
Adding equality constraints to a matrix that is otherwise allowed to differ from the same matrix in a previous group:Group 1 (e.g.) LY=FU,FI
Releasing equality constraints on a single parameter when matrix is otherwise specified as Invariant:
Group 2:LY=INFR LY 4 1 LY remains invariant
except for parameter LY(4,1)
24
Examples: Example (separate handouts) (religion & sexual morality in 2 countries)
Multiple group example #1: USA DA NO=1150 NI=19 MA=CM NG=2 CM FI=G:\CLASSES\ICPSR2005\WEEK3EXAMPLES\USA.COV LA A006 F028 F066 F063 F118 F119 F120 F121 GENDER AGE EDUC TOWNSIZE MARR1 MARR2 MARR3 OCC1 OCC2 OCC3 OCC4 SE 1 2 3 4 5 6 7 8/ MO NY=8 NE=2 LY=FU,FI PS=SY,FR TE=SY VA 1.0 LY 2 1 LY 5 2 FR LY 1 1 LY 3 1 LY 4 1 FR LY 6 2 LY 7 2 LY 8 2 FR TE 4 3 OU ND=4 SC MI
Basic program
See handout for variable list
25
Multiple group example #1: USA Number of Input Variables 19 Number of Y - Variables 8 Number of X - Variables 0 Number of ETA - Variables 2 Number of KSI - Variables 0 Number of Observations 1150 Number of Groups 2 Group #2: Canada DA NO=1763 NI=19 MA=CM CM FI=G:\CLASSES\ICPSR2005\WEEK3EXAMPLES\CDN.COV LA A006 F028 F066 F063 F118 F119 F120 F121 GENDER AGE EDUC TOWNSIZE MARR1 MARR2 MARR3 OCC1 OCC2 OCC3 OCC4 SE 1 2 3 4 5 6 7 8/ MO LY=IN PS=PS TE=PS OU ND=4 SC MI Group #2: Canada Number of Input Variables 19 Number of Y - Variables 8 Number of X - Variables 0 Number of ETA - Variables 2 Number of KSI - Variables 0 Number of Observations 1763 Number of Groups 2
ETA 1 ETA 2 -------- -------- ETA 1 2.4328 ETA 2 1.9415 4.5847
PSI
ETA 1 ETA 2 -------- -------- ETA 1 2.4328 (0.1554) 15.6531 ETA 2 1.9415 4.5847 (0.1464) (0.3118) 13.2596 14.7036
Group 1
Covariance Matrix of ETA
ETA 1 ETA 2 -------- -------- ETA 1 3.5419 ETA 2 2.3997 5.2741
PSI
ETA 1 ETA 2 -------- -------- ETA 1 3.5419 (0.1924) 18.4093 ETA 2 2.3997 5.2741 (0.1529) (0.3175) 15.6932 16.6128
Group 2
28
Group Goodness of Fit Statistics (USA)
Contribution to Chi-Square = 120.0772 Percentage Contribution to Chi-Square = 51.3912 Root Mean Square Residual (RMR) = 0.2698 Standardized RMR = 0.04414 Goodness of Fit Index (GFI) = 0.9756
Global Goodness of Fit Statistics
Degrees of Freedom = 42 Minimum Fit Function Chi-Square = 233.6530 (P = 0.0) Normal Theory Weighted Least Squares Chi-Square = 229.0169 (P = 0.0) Estimated Non-centrality Parameter (NCP) = 187.0169 90 Percent Confidence Interval for NCP = (143.2744 ; 238.2786)
Normed Fit Index (NFI) = 0.9840 Non-Normed Fit Index (NNFI) = 0.9824 Parsimony Normed Fit Index (PNFI) = 0.7380 Comparative Fit Index (CFI) = 0.9868 Incremental Fit Index (IFI) = 0.9868 Relative Fit Index (RFI) = 0.9787
Group Goodness of Fit Statistics (CANADA)
Contribution to Chi-Square = 113.5759 Percentage Contribution to Chi-Square = 48.6088 Root Mean Square Residual (RMR) = 0.2222 Standardized RMR = 0.03069 Goodness of Fit Index (GFI) = 0.9841
ETA 1 ETA 2 -------- -------- ETA 1 1.1410 ETA 2 0.6090 1.0544
More variance in Canada
Connection stronger in Canada
34
Testing measurement equivalence
Group #2: Canada DA NO=1763 NI=19 MA=CM CM FI=G:\CLASSES\ICPSR2005\WEEK3EXAMPLES\CDN.COV LA A006 F028 F066 F063 F118 F119 F120 F121 GENDER AGE EDUC TOWNSIZE MARR1
MARR2 MARR3 OCC1 OCC2 OCC3 OCC4 SE 1 2 3 4 5 6 7 8 / MO LY=PS PS=PS TE=PS OU ND=4 SC MI Global Goodness of Fit Statistics
Degrees of Freedom = 36 Minimum Fit Function Chi-Square = 209.7783 (P = 0.0) Normal Theory Weighted Least Squares Chi-Square = 207.3408 (P = 0.0) Estimated Non-centrality Parameter (NCP) = 171.3408 90 Percent Confidence Interval for NCP = (129.7727 ; 220.4219)
Normed Fit Index (NFI) = 0.9856 Non-Normed Fit Index (NNFI) = 0.9814 Parsimony Normed Fit Index (PNFI) = 0.6336 Comparative Fit Index (CFI) = 0.9881 Incremental Fit Index (IFI) = 0.9881 Relative Fit Index (RFI) = 0.9777
35
Some Model/Test Results:Matrices Chi-square df IFI
From TwoGroup1d, in Group #2 (Canada):Modification Indices for PSI ETA 1 ETA 2 -------- -------- ETA 1 31.1069 ETA 2 4.8672 4.0391
Expected Change for PSI ETA 1 ETA 2 -------- -------- ETA 1 0.3466 ETA 2 -0.1158 0.2188
36
Models with exogenous variables in construct equations:
Multiple group example #1: USA DA NO=1150 NI=19 MA=CM NG=2 CM FI=G:\CLASSES\ICPSR2005\WEEK3EXAMPLES\USA.COV LA A006 F028 F066 F063 F118 F119 F120 F121 GENDER AGE EDUC TOWNSIZE MARR1 MARR2
MARR3 OCC1 OCC2 OCC3 OCC4 SE 1 2 3 4 5 6 7 8 9 10 11 12/ MO NY=8 NE=2 LY=FU,FI PS=SY,FR TE=SY NX=4 NK=4 LX=ID C PH=SY,FR TD=ZE GA=FU,FR LE RELIG MORAL VA 1.0 LY 2 1 LY 5 2 FR LY 1 1 LY 3 1 LY 4 1 FR LY 6 2 LY 7 2 LY 8 2 FR TE 4 3 OU ND=4 SC MI
Group #2: Canada DA NO=1763 NI=19 MA=CM CM FI=G:\CLASSES\ICPSR2005\WEEK3EXAMPLES\CDN.COV LA A006 F028 F066 F063 F118 F119 F120 F121 GENDER AGE EDUC TOWNSIZE MARR1 MARR2 MARR3 OCC1 OCC2 OCC3 OCC4 SE 1 2 3 4 5 6 7 8 9 10 11 12/ MO LY=IN PS=PS TE=PS LX=IN PH=PS TD=IN GA=PS LE RELIG MORAL OU ND=4 SC MI
37
LAMBDA-Y EQUALS LAMBDA-Y IN THE FOLLOWING GROUP
GAMMA
GENDER AGE EDUC TOWNSIZE -------- -------- -------- -------- RELIG 7 8 9 10 MORAL 11 12 13 14
From Model Group2B (Group #2, Canada):Modification Indices for GAMMA GENDER AGE EDUC TOWNSIZE -------- -------- -------- -------- RELIG 7.8388 7.6672 -0.0001 0.0965 MORAL 3.1242 0.0477 41.6948 1.3675
Modification Indices for GAMMA (Group #1, USA) GENDER AGE EDUC TOWNSIZE -------- -------- -------- -------- RELIG 7.7220 7.4528 0.0002 0.2770 MORAL 3.0729 0.0516 5.5255 4.2116
Free occ. in group 2Group3D GA=PS 720.507 146 .972
Occ FI in group 2Group3EGA=PS 731.286 154 .972
Occ Fi both groupsGroup3F GA=PS 715.659 142 .972
Occ coefficients = For 1st Eta variable
Tests: Group3A vs. Group 3B Equality of all GA coefficientsGroup 3A vs Group 3B Equality of occupation GA coefficientsGroup 3B vs. Group 3D Is occupation stat. significant in group 2
(Canada)?Group 3B vs. Group 3E Is occupation stat. sign. in both groups?Group 3D vs. Group 3E Is occupation stat. sign. in group 1?Group 3B vs. Group 3F Equality of occupation effect for