Data Science · Intermediate level

Intermediate level · Module 4 of 6 · 13 guided hours

Relationships and Regression for Analysis

Describe relationships with correlation and regression, control for other factors and recognise confounding.

Learning objectives

  • Measure and visualise correlation
  • Fit and interpret multiple linear regression for explanation
  • Recognise confounding and the limits of observational data

Topics

  • correlation coefficients
  • simple and multiple regression
  • statsmodels output
  • dummy variables
  • confounding
  • regression diagnostics

Lessons

  1. Lesson 1 Correlation

    Measure the strength and direction of relationships and understand what correlation does not show.

    Pearson correlationSpearman correlationscatter plotcorrelation vs causation

  2. Lesson 2 Multiple Regression for Explanation

    Estimate the association of several factors with an outcome at once and interpret regression output.

    statsmodels OLScoefficient and confidence intervaldummy variablesholding other factors constant

  3. Lesson 3 Confounding and Limits of Observational Data

    Recognise confounding and selection bias and describe what observational analyses can and cannot conclude.

    confounderselection biasomitted variablecareful wording

Practical activity

Model school attendance as a function of distance, school feeding and teacher ratio, interpret the coefficients and discuss confounders.

Expected outcomes

  • Careful explanatory models
  • Awareness of what regression cannot prove

Assessment

Regression analysis marked on interpretation, diagnostics and discussion of confounding.

Taught live by an instructor — online or in person. Free course lessons are for self-study and do not lead to a certificate — see registration options.