Data Analysis Research Projects in R

Project 1 - Analysis of a research paper(Survey of Autobiographical Memory)

1. Cleaned the existing data to differentiate between the design variables and variables needed for PCA analysis
2. Analyzed the SAM Dataset to implement PCA and bootstraping techniques to explain the existing data on three principal components.
3. Plotted the Factor score, loading and bootstrap ratio graphs to gain more insights.

Conclution in the study were as follows:

1. Component 1 1a. Row: Normal versus High Memory group 1b. Column: Normal versus High Memory scores

So Component 1 mainly distinguishes people with high versus normal memory group

2. Component 2 2a. Column: Spatial versus other memory types

Distinguishes questions relating to spatial memory versus other memory types. Also shows negative correlation between spatial and future memory ratings.

Project 2 - Bama Politics Dataset (Education versus Politics)

1. Cleaned the existing data to differentiate between the design variables and qualitative variables needed for CA analysis in the survey that mainly explained max variance for our hypothesis question: How education is related to the choice of political parties?
2. Analyzed the Dataset to implement CA and bootstraping techniques to explain the existing data on two principal components which were significant after the bootstraping results.
3. Plotted the Biplot (Symmetric & Unsymmetric) and bootstrap ratio graphs to gain more insights.

Conclusion

When we interpret the Biplot and correlation circle plot together, the CA and bootstrap results revealed:

1. Component 1: The latent structure of the Bama Politics data as revealed by CA indicated that the first component characterized Republican & Independent versus Democrat. Also it indicates that educated people prefer Republican & Independent over Democrat.
2. Component 2: Mainly distinguishes people supporting the Republican versus Independent & Democrats.