Scribd
Upload
2 views

When to use_

The document outlines three statistical methods: correlation, factor analysis, and regression. Correlation measures the linear association between two quantitative variables without assuming causation, while factor analysis identifies underlying patterns among multiple variables for dimensionality reduction. Regression explores the relationship between a dependent variable and one or more independent variables, often for prediction, assuming a causal and linear relationship.

Uploaded by

thihinisme
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
2 views

When to use_

The document outlines three statistical methods: correlation, factor analysis, and regression. Correlation measures the linear association between two quantitative variables without assuming causation, while factor analysis identifies underlying patterns among multiple variables for dimensionality reduction. Regression explores the relationship between a dependent variable and one or more independent variables, often for prediction, assuming a causal and linear relationship.

Uploaded by

thihinisme
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

1.

Correlation
- To measure the degree and direction of the linear association between
two variables.
- Variables' Characteristics:
+ Both variables should be quantitative (numerical), either interval or
ratio scale.
+ The relationship between the variables should ideally be linear.
+ Both variables should be independent observations (no
dependent/independent variable distinction, as you are only
examining association).
+ No causation is assumed—correlation only looks at the strength of
the relationship, not how one variable influences another.
Example:
+ Studying the relationship between temperature and ice cream sales.
+ Finding the association between hours studied and test scores.

2. Factor Analysis
- To identify underlying patterns or latent factors in the data by examining
relationships among multiple variables. It reduces dimensionality and
identifies groups of related variables.
- Variables' Characteristics:
+ It requires multiple variables that are continuous (or ordinal that
can be treated as continuous).
+ These variables should exhibit some degree of correlation with
each other (high correlation indicates that factor analysis may be
appropriate).
+ Often assumes that the variables are normally distributed,
especially for some methods of factor extraction (e.g., principal
component or maximum likelihood methods).
+ Used for dimensionality reduction—grouping related variables that
measure the same underlying concept (e.g., survey items measuring
"job satisfaction").
Example:
+ Identifying key factors that explain students’ academic performance (e.g.,
intelligence, motivation, and study habits) from multiple variables.
+ Grouping survey questions into categories such as satisfaction, loyalty,
and experience.

3. Regression
- To explore and model the relationship between a dependent variable and
one or more independent variables. Often used for prediction and
determining the extent to which independent variables influence the
dependent variable.
- Variables' Characteristics:
+ Dependent Variable:
● Must be quantitative (interval or ratio scale) for linear
regression.
● Can be categorical for logistic regression and other variants.
+ Independent Variable:
● Can be quantitative or categorical (categorical variables
typically need to be dummy-coded for linear regression).
+ Assumes a causal relationship between the independent and
dependent variable(s).
+ Assumes a linear relationship between the dependent variable and
the predictors (for linear regression).
+ Observations must be independent, and multicollinearity (high
correlation between predictors) should be avoided.
Example:
+ Predicting house prices based on features like square footage, number of
bedrooms, and location.
+ Analyzing the effect of education level, gender, and work experience on
salary.

You might also like