1. Home
  2. NOVA School of Social Sciences and Humanities
  3. Bachelor in Sociology
  4. Multivariate Data Analysis

Multivariate Data Analysis

Objectives

a) Understanding of how to use systematic, statistical and computer processing of data in relation to the general research strategy; b) Knowledge of and ability to apply mediator operations between the research dossier and analytical protocol; c) Ability to define the technical specifications for data analysis; d) Knowledge of and ability to use the principal statistical solutions for purpose of multivariate data analysis in Sociology; e) Knowledge and ability to use the analyses and measures of association for purposes of multivariate processing; f) nderstanding of modes of integration in analytical protocols, factor analysis (simple and multiple) and post factor analysis techniques; g) Understanding of algorithms of simple and multiple multivariate analyses; h) Ability to use the main computer programmes for statistical processing on the market; i) Ability to communicate the results of multivariate statistical analyses and their interpretation in a rigorous and significant manner.

General characterization

Code

711081059

Credits

6.0

Responsible teacher

Ana Lúcia Albano Teixeira

Hours

Weekly - 4

Total - 168

Teaching language

Portuguese

Prerequisites

Available soon

Bibliography

Benzécri, J. P. (1973). L'analyse des données: L'analyse des correspondances. Dunod (pp. 3-17).
Carvalho, H. (2008). Análise multivariada de dados qualitativos - Utilização da ACM com o SPSS. Sílabo.
Ghiglione, R., & Matalon, B. (1992). O Inquérito – Teoria e Prática. Celta.
Guimarães, R. C., & Cabral, J. S. (1997). Estatística. McGraw Hill (pp. 13-16).
Hill, M. M., & Hill, A. (2000). Investigação por Questionário. Sílabo.
Marôco, J. (2021). Análise Estatística com o SPSS Statistics. Report Number.
Pestana, M. H., & Gageiro, J. N. (2000). Análise de Dados para Ciências Sociais – A complementaridade do SPSS. Sílabo.
Tabachnick, B. G., & Fidell, L. S. (2013). Using multivariate statistics. Pearson.

Teaching method

Theoretical classes, predominantly expository, where the systematization and structuring of theoretical and technical knowledge is carried out, with students being asked to contribute to the construction of reasoning.
Practical classes where practical exercises are discussed and solved.

Evaluation method

Two written tests (45% each): 90%
Practical exercises: 10%

Subject matter

1. Methodological and statistical assumptions for multivariate data analysis
1.1. The process of research and measurement of social phenomena - data analysis in the context of the scientific process
1.2. From concepts to variables; measurement scales and types of statistical analysis
1.3. Introduction to Multivariate Analysis: Building and Reading Contingency Tables, Chi2 Metrics, and Residuals
2. Multiple Correspondence Analysis
2.1. Framework of MCA in the context of multivariate data analysis
2.2. MCA as a theory building tool in Sociology – an application example
2.3. The MCA algorithm: theoretical and operational framework – the geometry of quantifications
2.4. The development of an MCA: dimensions, variables and categories
2.5. Complementary analysis: the importance of supplementary variables
3. Principal Component Analysis
3.1. PCA framework in the context of multivariate data analysis
3.2. Development of the PCA: objectives and application conditions
3.3. PCA development: component selection criteria; communalities; analysis of results and validation
4. Classification Analysis – Cluster Analysis
4.1. Framework of cluster analysis in the context of multivariate data analysis and objectives
4.2. Cluster analysis as a set of object classification methods: presentation of hierarchical, non-hierarchical and mixed methods and agglomerative and divisive methods
4.3. Development of cluster analysis through the agglomerative hierarchical method: selection of input variables; distance matrices; grouping criteria; definition, justification and validation of a solution
5. Introduction to linear regression – a simple linear regression analysis
5.1. Linear regression framework in the context of multivariate data analysis
5.2. The simple linear regression model: regression coefficients and prediction
5.3. Evaluation of the quality of the regression model: determination coefficient