SSI is pleased to announce an additional feature now available in LISREL12. Users with active SSI Live™ LISREL subscription licenses can obtain this update free of charge by logging into My Account.
The Gauss-Newton iterative estimation algorithm is now available for single group LISREL models without and with intercepts (mean structures) and multiple group LISREL models without and with intercepts (mean structures).
SSI is pleased to announce additional features now available in LISREL12. Users with active SSI Live™ LISREL subscription licenses can obtain this update free of charge by logging into My Account.
The newly released version of LISREL 12 includes two new features:
- HTML tables for selected LISREL results are generated for single group LISREL models to make inclusion of tabled results in publications easier.
- For multiple group analyses, the CM option is now available. More specifically, the common metric completely standardized solutions for multiple group LISREL models with no equal parameters across groups can now be requested, and the computation of the standard error and confidence interval estimates for the within group completely standardized solutions for these models can be suppressed.
An example of the new output is given on the Examples page.
Vector Psychometric Group and Scientific Software International have created a YouTube Channel that includes educational videos on Item Response Theory Modeling, Introductory Multilevel Modeling and Advanced Multilevel Modeling.
SSI is pleased to announce the immediate availability of LISREL12. Users with active SSI Live™ LISREL subscription licenses can obtain this update free of charge by logging into My Account.
LISREL 12 introduces several new features that were not available in previous versions.
- Two-stage multiple imputation Structural Equation Modeling for continuous, ordinal, and a mixture of continuous and ordinal variables are now available.
- Confidence interval estimates for the parameters of LISREL models are now included.
- Standard error estimates and confidence interval estimates for standardized and completely standardized solutions have been implemented.
- An estimation algorithm for single group LISREL models with or without variance constraints for the endogenous latent variables has been added.
- Path diagram files can be exported as enhanced metafiles which can be imported into other documents.
- The multilevel generalized linear modeling application includes more link functions and computes estimates for the intra-class correlation coefficients.
- The technical details of the new statistical methods as well as illustrative examples are available via the New Features option on the LISREL Help menu.
The latest LISREL 11 version (188.8.131.52) includes the expansion of two-stage multiple imputation SEM for ordinal variables for single group LISREL models introduced in LISREL 11.0, released in June 2021. The expansion includes ordinal variables with different number of categories, continuous variables, and a mixture of continuous and ordinal variables for single and multiple group LISREL models. See the support page for technical details and the example page for worked examples.
In addition, the following updates were made:
1. Corrections to the reading and writing of variable labels.
2. Adjustments to the LISREL and SIMPLIS output formats.
Steve Raudenbush, Lewis-Sebring Professor of Sociology, Chairman of the Committee on Education at the University of Chicago, and main author of the well-known HLM program, discusses the fixed effect random coefficient (FIRC) model and its implications for causal inference for multilevel data. A new feature in HLM, this feature is computationally convenient, with no dummy variables and no centering. The Tennessee STAR data is used as illustration in this 45 minute webinar.
This is the first in a series of webinars by Professor Raudenbush. A second webinar, discussing the use of the newly implemented profile likelihood plots feature in HLM in inference for variance components in hierarchical linear modeling, will be available soon.
SSI is pleased to announce an update to HLM 8.2. The update contains a new feature and a bug fix. Users with active SSI Live™ HLM subscription licenses can obtain this update free of charge by logging into My Account and downloading the updated version.
List of updates:
- Previously, users could only import older Excel spreadsheets (*.xls extension). The option to import current Excel spreadsheets (*.xlsx) has been added to the program.
- A bug with some spatial models running out of memory has been reported. This has been fixed.
SSI is pleased to announce the immediate availability of HLM 8.2. The update contains new features and bug fixes. Users with active SSI Live™ HLM subscription licenses can obtain this update free of charge by logging into My Account.
List of updates:
- Profile likelihood plots for 2-level models with scalar tau has been added to the program. This new feature is available for all types of outcome variables.
- A bug with adding interaction terms in crossed-and-nested models have been addressed.
- A bug with weighting in very large data sets have been fixed.
- The Fixed-Intercept-Random-Coefficient (FIRC) model implementation has been revised and improved.
SSI is pleased to announce the immediate availability of LISREL11. Users with active SSI Live™ LISREL subscription licenses can obtain this update free of charge by logging into My Account.
LISREL 11 introduces several new features that were not available in previous versions.
- Two-stage multiple imputation SEM for ordinal variables is now available (Chung and Cai, 2019; MBR). In previous versions of LISREL, the MCMC multiple imputation method for continuous variables was used to impute missing data values for ordinal variables under the assumption of underlying normal distributions.
- Estimation, model fit testing, as well as speed and stability improvements.
- Variable names can now be up to 16 characters. The length of the path names that LISREL can accommodate has also been extended to 192 characters. The path diagrams can also accommodate the longer variable names in the display.
- The *.PTH or path diagram file is now self contained, allowing users to share these files with fellow researchers. It offers a cleaner display and users will no longer be prompted to save this file if no changes have been made to the path diagram. Path diagrams for adaptive quadrature analyses now include the display of the – 2 ln L and number of parameters estimated (nfree) on the path diagram.
- To avoid accidentally running the wrong program, only the Run LISREL button will be enabled for files with file extension *.lis (LISREL syntax), *.spl (SIMPLIS syntax), *.lpj (LISREL syntax generated through the GUI), and *.spj (SIMPLIS syntax generated through the GUI). The Run PRELIS button will become active when a *.prl file (PRELIS, Multilevel, Multilevel GLIM, Survey GLIM syntax files) is active.
SSI is pleased to announce the immediate availability of HLM 8.1. The update contains substantial improvements and bug fixes over and above HLM 8.0, which was released in 2019. Users are encouraged to update their installations. Users with active SSI Live™ HLM subscription licenses can obtain this update free of charge by logging into My Account.
List of updates:
- Improvements and corrections to model specification in GUI for multiple outcome models, HLM4 etc.
- Corrected problems with the Exploratory Statistics options in HLM2 and 3.
- Problems with mixed model formulation in output file corrected for a number of cases, for example HLM4 and multiple outcome models.
- Confidence intervals for taus refined.
- Problems regarding production of and correct writing to residual files fixed.
- Problems with degrees of freedom in HLM4 models has been fixed.
- Revision of the weights implementation has been undertaken: only marginal weighting now implemented.
- Graphing related crashes for categorical outcomes have been fixed.
- Diagonalization of matrices revised.
- Missing data issues in HLM3, HLM4, HCM2, HCM3 and HLMHCM addressed.
- Additional information added to output files for the AI routine in the case of improper imputation.