Nome: SmartPLS
Produtor: SmartPLS GmbH
Versão: 4 (Agosto 2022)
Plataformas: Windows e Mac
Versão de teste: Sim
Áreas: Gestão | Gestão de Projectos | Gestão de Marketing | Análise de Dados | Estatística

Informação: Visão Geral | What’s New | Features | Resources | Informações Adicionais

O SmartPLS é um software com interface gráfica para modelação de equações estruturais (SEM) baseada em variância utilizando o método de modelação de caminhos de Partial Least Square (PLS). Além de estimar modelos de caminho com variáveis latentes usando o algoritmo PLS-SEM, o SmartPLS calcula critérios de avaliação de resultados padrão (por exemplo, para modelos de medição reflexivos e formativos, o modelo estrutural e a adequação do ajuste) e suporta análises estatísticas adicionais (por exemplo, análise confirmatória de tetrad, análise de mapas de importância-desempenho, segmentação, multigrupos.

 O SmartPLS 3 é um marco na modelação de variáveis latentes. O SmartPLS combina métodos de última geração (por exemplo, PLS-POS, IPMA, rotinas complexas de bootstrapping) com uma interface gráfica fácil de usar e intuitiva.

 Executar as suas análises com o SmartPLS é divertido e sem complicações. Obtenha informações sobre os seus dados de uma forma fácil e simples.

SmartPLS 4 introduces the following notable new features. As part of an early access phase, many videos on SmartPLS4 have already been created to help you with your transition. 

– Redesigned GUI

You can expect a fundamentally renewed and optimized GUI. We have developed this over the last few years based on the latest technologies, so SmartPLS will have a long-term future.

– Improved performance

Performance has improved significantly in many areas. Algorithms like bootstrapping or FIMIX are now much faster.

– Improved data import

Data sets are often available in Excel or SPSS formats. SmartPLS can now read these directly, without prior conversion to a CSV file.

– Generate new data files from your results

New datafiles can now be created from calculation results directly which is useful especiall for higher-order models. You can select which data should be included in your new file – e.g. the scores from a PLS-SEM calculation. Immediately after that you can use the file for new calculations.

– Side by side comparison

You can now compare reports more easily by simply opening them in our Compare View page.

– Save your reports

Finally, SmartPLS reports can now be saved and opened again later. Exporting to Excel or to a website is of course still possible as well.

– Customize charts

Diagrams can now be customized. For example, new colors or sizes for dots can be defined – but also many other properties can be changed and can also be saved as new defaults.

– Many new methods

In addition to the algorithms and models known from SmartPLS 3, SmartPLS 4 now also supports new model types, techniques and algorithms. These include:

• Multiple moderation (e.g., three-way interactions)
• PROCESS type analysis, including bootstrapping of conditional direct and indirect effects
• Accounting for scale type of variables in most algorithms
• Standardized, unstandardized and mean-centered PLS-SEM analysis
• Testing endogeneity with Gaussian copulas is now possible.
• Regression models
• Necessary condition analysis (NCA) including significance testing

– Many included example projects

There are now many different example model directly integrated in the software. So you can directly try out many of the functionalities, even without your own data.

– 32-bit support is back

32-bit support for Windows is back. With SmartPLS 3, we unfortunately had to abandon support for 32-bit systems for security and technology reasons

• The powerful modeling environment lets you create a path model in minutes.
• The project manager helps you to keep track of all your analyzes and files.
• Customize your model with colors, borders and fonts to underline your ideas individually!
• In-built explanations of the algorithms and meaningful defaults give you an easy start into the PLS-SEM world.
• Well-organized reports provide full insights into your results.
• Save your results permanently as HTML report or Excel file.
• Create data groups to run multi group analyses effortlessly.
• Create interaction terms and run moderator analyses without any problems.

SmartPLS is the workhorse for all PLS-SEM analyses – for beginners as well as experts

Here is our (constantly growing) list of all available calculation methods. Relevant innovative algorithms will also be made available in SmartPLS within a short time. We promise.

• Partial least squares (PLS) path modeling
• Ordinary least squares (OLS) regression based on sumscores
• Consistent PLS (PLSc)
• Weighted PLS (WPLS), weighted OLS (WOLS) and weighted consistent PLS (WPLSc)
• Bootstrapping and the use of advanced bootstrapping options
• Blindfolding
• Importance-performance map analysis (IPMA)
• PLS multi-group analysis (MGA): Analyses the difference and significance of group-specific PLS path model estimations
• Higher-order Models
• Mediation: Estimation of indirect effects and their bootstrap-based significance testing
• Moderation: Estimation of interaction effects and their bootstrap-based significance testing
• Nonlinear relationships: Estimation of quadratic effects and their bootstrap-based significance testing
• Confirmatory tetrad analysis (CTA): A statistical technique which allows for empirical testing the measurement model setup
• Finite mixture (FIMIX) segmentation: A latent class approach which allows identifying and treating unobserved heterogeneity in path models
• Prediction-oriented segmentation (POS): An approach to identify groups of data
• PLS Predict: A technique to determine the predictive quality of the PLS path model
• Prediction-oriented model selection