PSYC 2260 Lecture Notes - Lecture 11: Coefficient Of Determination, Effect Size, Standard Score
26 Jun 2018
School
Department
Course
Professor
PSYC 2260 Introduction to Research Methods in Psychology Chapter 11
Chapter 11 Correlation
Correlation: a measure of the strength of linear association between 2 equal interval variables
- Association: if variables “go together” or are “related”
- Correlation is not cause
Equal interval variables:
- It’s ok if variables on different scale
Important:
- Both variables have underlying continuity
- Assumed to follow a normal curve
Linear correlation: pattern of dots suggests a “straight line” through the data
Curvilinear correlation: pattern of dots suggests a “curve line” through the data
- Important look at pattern first, as calculation assume linear relationship
No correlation: no pattern in the scatterplot
- This is null hypothesis Ho: r = 0
r
Ranges between -1 to +1
- Positive r: as X increase, Y increase
- Negative r: as X increase, Y decrease
- Zero r: no relationship between X and Y
Strength of correlation: the “circle” as narrow indicate as high r
- Eyeballing often difficult
- Thus need to determine r via a formula
Formula for r
1. r = denominator called “cross product” of deviation
- see handout Mar13
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Correlation: a measure of the strength of linear association between 2 equal interval variables. Association: if variables go together or are related . Linear correlation: pattern of dots suggests a straight line through the data. Curvilinear correlation: pattern of dots suggests a curve line through the data. Important look at pattern first, as calculation assume linear relationship. No correlation: no pattern in the scatterplot. This is null hypothesis ho: r = 0 r. Zero r: no relationship between x and y. Strength of correlation: the circle as narrow indicate as high r. Thus need to determine r via a formula. 1. r = denominator called cross product of deviation see handout mar13. 2. r = numerator called the sum of z score cross product. Correlation and causality correlation measures co-variation does not show causality. X and y both caused by and unknown 3rd variable. Experiments controls for unknown variables: not so with correlation.
