##### 311 Unit 7 Study Guide

Set Details
Share

updated 4 years ago by wouhib

Show:

1

Correlational research strategy

- Two or more variables are measured to obtain a set of scores (usually 2) for each individual (or source)
- Measurements are examined to identify patterns that exist between the variables and to measure the strength of the relationship

2

What correlations describe

- The direction of the relation: Positive or negative relations
- The form of the relation: Linear (using Pearson correlation) or Monotonic (using Spearman correlation)
- The consistency or strength of relation: How close the relation is to a perfect linear or a perfect monotone relation

3

Monotonic vs. Linear

Monotonic correlation examples

- Memorizing a list of 25 words (number of repetitions required)
- Practicing a skill (like knitting)
- Impact of sleep deprivation

4

Uses of correlational studies

- Predict: Determine relation between variables, if you know the value of one variable, can you predict the value on another variables?
- Evaluate: If you have a theory about a relation between variables, correlation can be the first step in assessing the quality/accuracy of that theory
- Assess validity: For example, to establish convergent and divergent validity, researchers assess relation between score on the new instrument against a score on another instrument
- Assess reliability: For example, test-retest reliability is the calculation of the correlation between one test score and and another, when you re-take the test a second time

5

When manipulation is unethical

- Correlation is useful when variables cannot be
*ethically*manipulated for purposes of an experiment - Examples: Do vaccines cause autism? Does smoking cause cancer? How many people die from texting and driving?
- Also, more ethical early in a line of research, when you don’t know if something might help or hurt

6

Third variable problem and causation

7

Directionality and causation

8

Direction of relations

Positive relations

- Tendency for two variables to change in the
*same*direction - As one variable increases, the other also tends to increase

Negative relations

- Tendency for two variable to change in
*opposite*directions - As one variable increases, the other tends to decrease

9

Correlation coefficient

- A numerical value that measures or describes the relation
between two variables, typically represented by
*r* - The direction of relation: Indicated by the correlation’s sign (+ -)
- The form of relation: Indicated by type (Pearson or Spearman)
- The consistency or strength of relation:
- Indicated by correlation numeric value
- Range = -1.0 to 1.0
- 0 = no relation

10

Strength of correlation

11

Significance of a correlation

- A
*significant*correlation means that the correlation is unlikely to have been produced by random variation - Small samples (i.e., 2 people) can produce significant correlations when no relationship is actually present
- As sample size
increases, likelihood of significant
*actual*correlations becomes stronger

12

Null hypothesis of correlation

- Null hypothesis:
*r*= 0 - Alternative
hypothesis:
*r*≠ 0

13

Reporting correlation