Multilevel Data/Complex Sample  
Mplus Discussion >
There are two approaches to the analysis of complex sample data in Mplus. The first approach is to compute standard errors and chi-square tests of model fit taking into account complex sampling features. These include stratification, sampling weights, and clustering. Standard error computations use a sandwich estimator. This is referred to as TYPE=COMPLEX in Mplus. For this approach, observed dependent variables can be continuous, censored, binary, ordered categorical (ordinal), unordered categorical (nominal), counts, or combinations of these variable types. When observed dependent variables are all continuous, Mplus has one estimator choice: maximum likelihood with robust standard errors and chi-square (MLR). When at least one observed dependent variable is binary or ordered categorical, Mplus has five estimator choices: weighted least squares (WLS), robust weighted least squares (WLSM, WLSMV), maximum likelihood with robust standard errors and chi-square (MLR), and unweighted least squares (ULS). When at least one observed dependent variable is censored, unordered categorical, or a count, Mplus has four estimator choices: weighted least squares (WLS), robust weighted least squares (WLSM, WLSMV), and maximum likelihood with robust standard errors and chi-square (MLR).

The second approach is to specify a model for each level of the multilevel data, commonly referred to as multilevel modeling. This is referred to as TYPE=TWOLEVEL in Mplus. The multilevel extension of the full Mplus modeling framework allows random intercepts and random slopes that vary across clusters in hierarchical data. These random effects can be specified for any of the relationships of the full model. Random effects representing across-cluster variation in intercepts and slopes or individual differences in growth can be combined with factors measured by multiple indicators on both the individual and cluster levels. In line with SEM, regressions among random effects, among factors, and between random effects and factors are allowed. For this approach, observed dependent variables can be continuous, binary, ordered categorical (ordinal), or combinations of these variable types. When observed dependent variables are all continuous, Mplus has four estimator choices: Muthen’s limited information estimator (MUML), maximum likelihood (ML), and maximum likelihood with robust standard errors and chi-square (MLR, MLF). When at least one observed dependent variable is binary or ordered categorical, Mplus has three estimator choices: maximum likelihood (ML) and maximum likelihood with robust standard errors and chi-square (MLR, MLF).

Besides controlling for clustering, TYPE=TWOLEVEL allows the use of sampling weights and stratification in the estimation of parameters, standard errors, and chi-square tests of model fit. With sampling weights, parameters are estimated by maximizing a weighted loglikelihood function.
Thread (Start New Thread) Last Post Last Poster Posts
Convergence1-05-09  11:23 amAngela Bryan17
Intraclass correlations10-28-08  7:57 amLinda K. Muthen77
Evaluation of model fit9-02-08  7:57 amBengt O. Muthen78
Categorical Outcomes10-21-07  5:34 pmLinda K. Muthen29
Factor loading MCA models3-06-06  10:20 ampoker casino3416
Total and Indirect Effects10-15-08  10:13 amLinda K. Muthen26
Building a Multilevel Structural Model 10-09-08  9:44 amLinda K. Muthen89
Hierarchical regression2-01-08  6:31 pmBengt O. Muthen35
Multilevel factor analysis8-04-08  9:05 amLinda K. Muthen69
Example data: Published?1-24-01  6:21 pmBengt O. Muthen2
Multigroup Model2-15-08  3:08 pmLinda K. Muthen67
Fit indices7-03-08  4:17 pmLinda K. Muthen20
Saving Parameter estimates in separate files9-12-08  1:56 pmHao Duong7
Multilevel SEM Examples11-04-08  10:05 amLinda K. Muthen74
Missing data7-30-07  9:13 amLinda K. Muthen29
Cluster Size10-03-07  12:38 pmSusan E. Collins23
Multilevel Time Series Question4-13-06  10:47 amBengt O. Muthen10
3-level analysis by HLM and Mplus10-20-06  8:02 amLinda K. Muthen17
Random intercepts5-29-08  4:24 pmJeremy Flaherty32
Is it possible to run "type=complex"and "type=mgroup" simultaneousl...8-27-08  8:53 amLinda K. Muthen4
FIML versus MUML2-29-08  3:08 pmLinda K. Muthen21
Version differences9-03-08  7:25 amLinda K. Muthen6
Random regression in Mplus11-19-08  9:12 amLinda K. Muthen13
Path analysis with 3-way (multilevel?) data3-06-06  11:05 ampoker casino6256
Simple multilevel problem (?)12-19-06  6:40 amLinda K. Muthen43
Testing for clustering2-06-03  10:50 ambmuthen2
Random intercept model with latent independent variable6-10-05  6:18 ambmuthen4
Using only the within or between matrix4-23-07  8:01 amLinda K. Muthen8
Autoregressive errors in multi-level model3-06-06  10:36 ampoker casino68929
MUML and MUMLM8-22-03  9:41 ambmuthen2
Adjusting standard errors for cluster sampling3-09-08  5:08 pmBengt O. Muthen15
Effective sample size8-28-03  10:20 amLinda K. Muthen2
Path Analysis using categorical latent outcome variable1-03-05  12:10 pmLinda K. Muthen13
Group level variable and low intra class correlation10-17-03  4:32 amdaan stam5
Linear dependency in ML models10-20-08  6:19 pmKwam Nuak2
Unconditional model6-19-07  5:49 pmLinda K. Muthen19
Power/monte carlo studies with multilevel data8-09-08  3:33 pmBengt O. Muthen38
Nonrecursive models1-25-04  3:30 pmAnonymous1
Rmsea with unbalanced group sizes2-10-04  2:41 pmLinda K. Muthen2
How does Mplus handle non-normal model?1-26-06  8:23 amLinda K. Muthen9
SEM and sampling weights1-07-07  8:53 amLinda K. Muthen24
Multilevel latent class analysis9-24-08  1:35 amPaul Widdop6
Two-level model without first-level covariates5-19-07  9:08 amLinda K. Muthen11
Analysis=integration, multiple groups4-20-04  6:27 pmLinda K. Muthen2
Unstable estimate of correlation5-05-04  1:17 pmLinda K. Muthen2
Multilevel data in discrete-time survival analysis7-28-04  1:37 pmbmuthen4
Possible LISREL syntax for multilevel modeling?7-13-04  5:41 pmbmuthen2
Multilevel Means Modeling7-25-04  12:06 pmbmuthen2
Numerical integration4-13-06  5:29 pmLinda K. Muthen12
Non-linear constraints and complex sample data9-03-04  8:48 ambmuthen4
Teory path analysis with multilevel data9-29-04  3:59 pmLinda K. Muthen2
Modification indices in multilevel?9-29-04  4:00 pmLinda K. Muthen3
Multiple group comparion with complex sample using cluster and weig...10-28-08  8:44 amLinda K. Muthen41
Estimated means for multilevel mixture model1-24-07  3:49 pmNordin abd Razak8
Multilevel path analysis for binary outcomes12-05-04  8:03 amLinda K. Muthen2
Random slopes for SEM?12-05-08  6:28 amLinda K. Muthen8
Weight Option and ICCs?12-17-04  3:55 pmbmuthen2
Standardized Solutions 1-29-08  2:05 pmLinda K. Muthen9
Unknown Factors in Between Level12-22-04  8:50 amThuy Nguyen4
Data Structure3-10-06  4:22 pmLinda K. Muthen11
A general comment12-23-04  11:02 amLinda K. Muthen2
Contrasts using multinomial logistic4-13-07  9:01 amBengt O. Muthen6
Robust WLS Sample size?1-18-05  3:48 pmLinda K. Muthen2
Intercept?2-18-05  5:18 pmbmuthen2
Zero-inflated Poisson regression in multilevel setting12-12-06  2:46 pmBengt O. Muthen4
Discriminant validity in multilevel model2-28-05  6:30 pmbmuthen2
Start Values in Multilevel Path Analysis8-25-05  2:37 pmbmuthen4
Is type=complex or type=multilevel the appropriate type of analysis?3-10-05  4:27 pmbmuthen2
More thann two levels?11-24-06  9:32 amBengt O. Muthen4
Level-two variables with no complement on level-one 3-20-05  1:37 pmLinda K. Muthen2
Unbiased Estimates Using Type=Complex?3-21-05  5:46 pmbmuthen2
Unit of Analysis in Aggregated Data3-06-06  10:32 ampoker casino7532
Between Level Covariance Matrix is NPD4-02-05  8:40 pmLinda K. Muthen2
Multiple membership models5-09-05  9:39 ambmuthen9
Help with a random slope interpretation3-06-06  10:48 ampoker casino866
No ICCs when using ordinal variables?4-15-05  2:01 amBMuthen5
Unbalanced multilevel data 4-15-05  1:33 amBMuthen2
Non-Positive Definite Between Level Matrix4-23-05  5:24 pmAnonymous2
Save out factor scores with complex data5-03-05  3:10 pmbmuthen4
Complex sample (BRR) and dichtomous outcomes5-17-05  3:47 pmbmuthen2
Three-level growth modeling8-03-06  3:12 pmBengt O. Muthen8
Using KNOWNCLASS3-20-07  4:08 pmLinda K. Muthen6
EFA/CFA with complex sample data7-18-06  5:51 amBengt O. Muthen16
Multilevel SEM --everything random?6-09-05  6:31 pmbmuthen4
How many degrees of freedom?8-02-07  1:00 pmLinda K. Muthen4
Weight variable in SEM5-02-07  10:14 amLinda K. Muthen12
Formative vs. Reflective constructs in MSEM7-07-05  1:19 pmbmuthen10
Minimum number of level-1 units within level-2 units6-28-05  8:19 amLinda K. Muthen2
Memory shortage7-07-05  1:12 pmMichael J. Zyphur3
No SIGB and SAMPLE matrices2-06-06  8:46 amLinda K. Muthen13
ICC with Latent variable SEM - 2 Level Model2-05-08  12:05 pmBengt O. Muthen21
"Negative Nonindependence"/"Frog Pond Effect"7-18-05  5:14 pmbmuthen2
Correlations7-18-05  5:38 pmLinda K. Muthen2
How many clusters?12-16-05  4:39 pmLinda K. Muthen13
Censored Two Level Model?8-08-05  1:24 pmbmuthen2
How to consider this dataset / which model is good?8-11-05  10:58 ambmuthen4
Question more / How to consider this dataset8-11-05  10:50 ambmuthen2
Ex 9.13 Question. 8-12-05  2:31 pmbmuthen2
Multilevel measurement/structural invariance9-04-08  10:00 amLinda K. Muthen20
How to fix this problem? ex 9.138-15-05  10:05 amAnonymous4
Multilevel Model Example on Web Site6-16-06  12:30 pmxue li3
The Big help for your comment about model/ output/ concept(1)  0
Example for type=complex2-02-07  2:22 pmLinda K. Muthen5
Two-level model with cross-classified random effects8-20-05  3:17 pmMichael J. Zyphur3
To Integrate or not in two-level modeling8-22-05  6:05 pmbmuthen2
Information matrix: expected vs. observed vs. combination8-24-05  2:37 pmLinda K. Muthen2
A few question in Mplus example.8-24-05  6:02 pmbmuthen2
Using "Design Effect" weight variables in NCES datasets8-24-05  8:04 pmAnonymous5
Cross-level and between-group interactions7-08-08  10:12 amLinda K. Muthen21
Using Mplus: the Multilevel model Question.8-26-05  3:22 pmLinda K. Muthen4
Multilevel model -> Assume Logistic Regression9-01-05  6:08 pmbmuthen2
Multilevel model <- Cluster level one more than9-12-05  2:15 pmLinda K. Muthen2
Random slope with fixed intercept9-13-05  4:29 pmbmuthen2
Error message Output 11-11-08  1:08 pmLinda K. Muthen6
Multilevel model and eroor message / Lots of missing data9-21-05  7:32 amLinda K. Muthen2
Weights 11-11-08  1:33 pmTihomir Asparouhov7
Cluster Issues3-06-06  10:40 ampoker casino5868
Binary Data at Level-27-09-08  5:19 pmLinda K. Muthen13
Non-positive Definite Matrix--Between Level10-13-05  10:04 amLinda K. Muthen2
MONTECARLO and weights10-17-05  8:02 amLinda K. Muthen2
Using Sigb 12-07-05  6:03 amLinda K. Muthen8
Multilevel Regression8-16-07  8:53 amLinda K. Muthen10
Latent random slope models10-21-05  6:02 amLinda K. Muthen2
Level-2 variables on level-1 independent variables10-22-05  8:38 amLinda K. Muthen2
Monte Carlo Question10-30-05  7:39 amLinda K. Muthen2
Scaling in CFA/SEM with complex data10-30-05  6:37 pmbmuthen2
Multiple sampling weights for growth model?11-03-05  9:09 ambmuthen6
Multilevel SEM for repeated surveys11-03-05  10:21 amTom5
Simulation of Multilevel SEM1-05-09  8:33 amHolmes Finch10
Constraints11-03-05  5:38 pmBMuthen2
Convergence Number in Simulations11-12-05  6:05 pmLinda K. Muthen2
R-Square3-03-08  3:12 pmStephan17
Zero-Loadings on Level 2 Factor12-06-05  10:41 amLinda K. Muthen8
Invariance across reporters (multilevel v. not)11-28-05  4:37 pmLinda K. Muthen2
Within and Between Specification4-17-08  11:16 amTihomir Asparouhov14
Combination of MFA and multilevel path analysis12-01-05  8:34 amLinda K. Muthen5
Recommended Estimator12-04-08  9:22 amBengt O. Muthen28
Specifities12-19-05  6:39 pmbmuthen13
Relationships between latent constructs12-07-05  6:26 pmbmuthen2
Can MPLUS analyze a general three-level model?12-12-05  2:49 pmLinda K. Muthen2
Std vs. StdXY12-18-05  6:01 pmbmuthen3
Mplus 4 Multilevel features?12-31-05  7:54 pmMichael J. Zyphur4
Categorical dependent variable using complex?1-06-06  11:30 amLinda K. Muthen2
Shouldn't within-groups variables be latent too?1-21-06  6:24 pmmike zyphur3
StdYX in Multilevel Regression2-05-06  5:09 pmLinda K. Muthen7
Saving Multilevel Residuals 1-27-06  8:48 amLinda K. Muthen2
Between-Level Question1-30-06  2:06 pmanonymous3
Estimating interactions with sig-w1-3