Survey methods and factorial analysis

Description

This course provides an in-depth exploration of survey methodology and factorial analysis using R CRAN. Students will learn how to collect, analyze, and interpret survey data, as well as how to perform various factorial analysis techniques. Practical applications using data sets will be emphasized throughout the course.

Factorial analysis offers a way of identifying patterns in data using association-based methods. The idea is to reduce the dimensionality of a data set by plotting all the observations on 2D graphs depending on how close the observations are with respect to their characteristics. Observations can then be divided into groups according to their proximity. These groups serve to identify individuals with similar characteristics, e.g., the beneficiaries of a particular policy intervention.

For instance, the approach can be used to create a typology of jurisdictions according to their socio-economic characteristics to better explain differences in public spending. In education, one can identify profiles of students and bring to light the determinants of success and failure at school. The approach can be used in health to differentiate types of patients. Those typologies may in return be used to inform policy-makers about particular needs and possible interventions.

The method is also particularly when dealing with survey data, when the purpose is to show evidence of associations between qualitative or quantitative variables.

Objectives:

By the end of this course, students will be able to:

Utilize R CRAN for data analysis and visualization.
Design and conduct surveys, including sample size determination and questionnaire construction.
Collect and code survey data effectively.
Apply measures of association to examine relationships between variables.
Perform factorial analysis techniques, including Principal Components Analysis (PCA), Hierarchical Clustering, and Multiple Correspondence Analysis (MCA).
Interpret and communicate the results of survey method and factorial analysis effectively.

Content (18 hours)

Part 1: Short introduction to R CRAN

Installation and functioning
Main commands
Main graph options in R-CRAN
Simple graphics
Simple regressions
Additional packages and functions to be used in this lecture

Part 2: Sampling and construction of variables

Introduction
The sample size
Conception of the questionnaire
Data collection and coding

Part 3: Measures of association

Introduction
Testing for correlation (application on R-CRAN: mydata1.csv)
Chi-square test of independence (application on R-CRAN: mydata1.csv)
Tests of difference between means (application on R-CRAN: mydata1.csv)

Part 4 : Factorial analysis

Introduction
Principal components analysis (PCA) (application on R-CRAN: mydataPCA.csv)
Hierarchical clustering on principal components (application on R-CRAN: mydataPCA.csv)
Multiple correspondence analysis (application on R-CRAN: mydataMCA.csv)

Bibliography

Chapters 2 and 4 of ‘STATISTICAL TOOLS FOR PROGRAM EVALUATION: Methods and Applications to Economic Policy, Public Health, and Education’ by Josselin and Le Maux, Springer.
Statistics Canada (2010). Survey Methods and Practices.
Giudici, P. (2005). Applied Data Mining: Statistical Methods for Business and Industry. John Wiley & Sons.
Tufféry, S. (2011). Data Mining and Statistics for Decision Making. John Wiley & Sons.
Statistics Canada (2010). Survey Methods and Practices.

Examination

A one-page statistical study using factorial analysis.

Topic: Economic Growth and the Environment
Group size: 2/3 students;
Data source: OECD countries, http://stats.oecd.org/.
Table of content (suggestion) : problematics/literature review/methodology/data description/factorial analysis/conclusion/references.

See, e.g., Grossman, Gene M., and Alan B. Krueger. “Economic Growth and the Environment.” The Quarterly Journal of Economics, vol. 110, no. 2, 1995, pp. 353–377.