STAT 113 Scatterplots and Correlation Colin Reimer Dawson Oberlin - - PowerPoint PPT Presentation

stat 113 scatterplots and correlation
SMART_READER_LITE
LIVE PREVIEW

STAT 113 Scatterplots and Correlation Colin Reimer Dawson Oberlin - - PowerPoint PPT Presentation

STAT 113 Scatterplots and Correlation Colin Reimer Dawson Oberlin College September 15, 2017 1 / 8 The Scatterplot With two quantitative values from the same source (e.g. from the same person), we can represent each case (e.g., person) as


slide-1
SLIDE 1

STAT 113 Scatterplots and Correlation

Colin Reimer Dawson

Oberlin College

September 15, 2017 1 / 8

slide-2
SLIDE 2

The Scatterplot

  • With two quantitative values from the same source (e.g. from

the same person), we can represent each case (e.g., person) as a point in 2D space.

  • If we plot all of these points, we obtain a scatterplot

2 / 8

slide-3
SLIDE 3

Scatter Diagrams Show Patterns

Do you see a pattern? 3 / 8

slide-4
SLIDE 4

Correlation

  • You can identify correlations on a scatter diagram by the

distinct patterns they form.

  • The variables are said to have a linear association if the

scatter diagram shows the points clustered around an approximately straight line.

  • (Pearson’s) correlation coefficient measures the strength and

direction of this association. 4 / 8

slide-5
SLIDE 5

Positive Linear Association

5 / 8

slide-6
SLIDE 6

Negative Linear Association

6 / 8

slide-7
SLIDE 7

No Association

7 / 8

slide-8
SLIDE 8

Estimating Relationships

Handout

8 / 8