2024
Quantitative Data Analysis in Psychology
Name: Quantitative Data Analysis in Psychology
Code: PSI11137M
3 ECTS
Duration: 15 weeks/78 hours
Scientific Area:
Psychology
Teaching languages: Portuguese
Languages of tutoring support: Portuguese, English
Regime de Frequência: Presencial
Sustainable Development Goals
Learning Goals
General
? Demonstrate recognizing the potential of the instrumentation of testing structural equation models (SEM) for both theoretical thinking and empirical research in psychology.
? Demonstrate competence in tailoring the test of SEM to single-sample cross-sectional correlational designs.
Specific
? Demonstrate competence in tailoring the test of SEM to single-sample cross-sectional correlational designs.
? Demonstrate competence to generate SEM in LISREL according to the observed variables? metric nature (continuous or discrete).
? Demonstrate competence in analyzing the results of the SEM test with latent variables (LV) denoted by discrete observed variables in LISREL.
? Demonstrate recognizing the potential of the instrumentation of testing structural equation models (SEM) for both theoretical thinking and empirical research in psychology.
? Demonstrate competence in tailoring the test of SEM to single-sample cross-sectional correlational designs.
Specific
? Demonstrate competence in tailoring the test of SEM to single-sample cross-sectional correlational designs.
? Demonstrate competence to generate SEM in LISREL according to the observed variables? metric nature (continuous or discrete).
? Demonstrate competence in analyzing the results of the SEM test with latent variables (LV) denoted by discrete observed variables in LISREL.
Contents
1. Epistemological framework.
1.1. Instrumentation of multivariate data analysis (IMDA) of first (e.g., SPSS) vs. second generation (e.g., LISREL).
1.2. Advantages of second-generation IMDA and SEM types.
1.2.1. Uni and multifactorial SEM (oblique, orthogonal, and hierarchical), comparison between alternative models, structural validity examination, and computation of factor scores.
1.2.2. SEM of multiple causes and multiple indicators.
1.2.3. SEM of multiple and multivariate regression with LV.
1.2.4. SEM of trajectories with LV and analysis of mediation effects.
2. Analysis of SEM with PRELIS/LISREL.
2.1. Infrastructure with PRELIS: sample size and type II error, correlation vs. covariance matrices, admissibility diagnosis, and estimation methods.
2.2. Testing with LISREL-SIMPLIS: specification, identification, estimation, admissibility diagnosis, goodness of fit, and substantive evaluation of SEM.
1.1. Instrumentation of multivariate data analysis (IMDA) of first (e.g., SPSS) vs. second generation (e.g., LISREL).
1.2. Advantages of second-generation IMDA and SEM types.
1.2.1. Uni and multifactorial SEM (oblique, orthogonal, and hierarchical), comparison between alternative models, structural validity examination, and computation of factor scores.
1.2.2. SEM of multiple causes and multiple indicators.
1.2.3. SEM of multiple and multivariate regression with LV.
1.2.4. SEM of trajectories with LV and analysis of mediation effects.
2. Analysis of SEM with PRELIS/LISREL.
2.1. Infrastructure with PRELIS: sample size and type II error, correlation vs. covariance matrices, admissibility diagnosis, and estimation methods.
2.2. Testing with LISREL-SIMPLIS: specification, identification, estimation, admissibility diagnosis, goodness of fit, and substantive evaluation of SEM.
Teaching Methods
Except in the eminently theoretical classes relating to the contents in point 1. of the syllabus, in the following classes, an eminently theoretical-practical approach is taken to teach the remaining contents, integrating those in point 2. with those in points 1.2.
It is intended that the analytical decomposition of the procedures used around the sequential organization of increasing complexity of contents allows an integrative synthesis for generating and interpreting test results of structural equation models. To support the development of the contents, students are provided with two reading sheets, one for point 1. and another for point 2., with synopses of the syllabus and respective references, and processed data files with structural modeling techniques using PRELIS/LISREL.
It is intended that the analytical decomposition of the procedures used around the sequential organization of increasing complexity of contents allows an integrative synthesis for generating and interpreting test results of structural equation models. To support the development of the contents, students are provided with two reading sheets, one for point 1. and another for point 2., with synopses of the syllabus and respective references, and processed data files with structural modeling techniques using PRELIS/LISREL.
Assessment
Attendance at 75% of classes is compulsory to pass the course, and also obtain a minimum mark of 9.5/20 in each of the following components.
(1) Assessment of group work (maximum of 3 students) performed during the semester: production and analysis of a LISREL output (40%). An oral defense of the group work may be requested.
(2) Individual assessment performed during exam periods: an open-book test of acquired knowledge (60%).
Final Assessment Scheme
Individual assessment, whose classification results are based on combining the results obtained in the continuous assessment components with the respective weighting. A mark equal to or greater than 9.5/20 will be considered for the final classification.
(1) Assessment of group work (maximum of 3 students) performed during the semester: production and analysis of a LISREL output (40%). An oral defense of the group work may be requested.
(2) Individual assessment performed during exam periods: an open-book test of acquired knowledge (60%).
Final Assessment Scheme
Individual assessment, whose classification results are based on combining the results obtained in the continuous assessment components with the respective weighting. A mark equal to or greater than 9.5/20 will be considered for the final classification.
Teaching Staff
- António Augusto Pinto Moreira Diniz [responsible]
Certificações
