Quiz 3

Question 1

Overall characteristics of correlational research

Answer 1

• The goal is to find a relationship between two variables and then to describe the nature of that relationship
• Relationships can be described but cannot be explained
• No attempt to manipulate, control, or interfere with the variables
  * does not determine a cause-and-effect relationship
• an experiment is required to determine a cause

Question 2

Scatter plots

Answer 2

• a way in which data can be presented as

- each subject’s score is shown as a single dot with a horizontal coordinate (X) and a vertical coordinate (Y)

Question 3

Positive and Negative relationships/correlations. What are each and how do they work? Be able to identify within examples.

Answer 3

• positive and negative relationships/correlations indicate the direction of a relationship
• in a positive relationship, two variables change in the same direction. As one variable increases, the other variable increases.
• in a negative relationship, two variables change in opposite directions. As one variable increases, the other variable decreases.

Question 4

Values of a correlation coefficient (between -1.0 and +1.0). What do the numbers and signs (+ or -) represent for the relationship between variables? What does a value of 0.0 mean? Be able to translate to what the data would look like in a scatter plot.

Answer 4

• a correlation coefficient measures and describes the relationship between two variables
• the sign (+/–) indicates the DIRECTION of the relationship
• numerical value (0.0 to +/-1.0) indicates the STRENGTH or consistency of the relationship
• a correlation of +1.00 (or -1.00) indicates a perfectly consistent linear relationship while 0.0 indicates no consistency whatsoever
• see figure 12.3 in textbook

Question 5

In terms of values (between -1.0 and +1.0) what is a strong or weak relationship?

Answer 5

• +0.9 would be a strong positive correlation
• -0.4 would be a relatively weak negative correlation
• -1.00 would a perfect negative correlation
• a correlation of would show no linear trend

Question 6

Coefficient of determination

Answer 6

• the coefficient of determination is the squared value of a correlation
  - Pearson correlation = r
  - Coefficient of determination = r^2
• measures how much of the variability in one variable is predictable from its relationship with the other variable

Question 7

What does a significant correlation indicate?

Answer 7

• that it is very unlikely to have been produced by random variation
• and represents a real relationship that exists within the population

Question 8

Situations when to use Pearson or Spearman correlations

Answer 8

• the type of correlation (Pearson or Spearman) indicates the form of the relationship
• A Pearson correlation describes linear relationships when both variables are numerical scores from interval or ratio scales
• If one score is numerical and the other score is non-numerical, then:
  
  • use the non-numerical variable to organize the scores into separate groups.
  • Non-numerical value with two categories
    - Numerically code the categories as 0 and 1
    - Calculate the Pearson correlation
• A Spearman correlation describes monotonic relationships when both variables are ranks from an ordinal score
  
  • a monotonic relationship is one that appears to be consistent and predictable but not linear. for example, the relationship between practice and performance. During the first few weeks of practice, the increases in performance are large. However, after years of practice, one more week produces a hardly noticeable change in performance.

Question 9

-Criterion and Predictor variables

Answer 9

• one variable (predictor) is used to predict the other (criterion)
  
  • ex. GRE score (predictor) predicts grad school success (criterion)
• predictor variable (X value) is the first variable. it is relatively simple and well defined.
• criterion variable (Y value ) is the second variable (being explained or predicted)
  - relatively complex and unknown

Question 10

Using correlations for reliability and validity (test-retest reliability and concurrent validity)

Answer 10

• recall reliability evaluates the consistency or stability of a measurement
  
  • correlation can determine the test-retest reliability of a test
• validity evaluates the extent to which the measurement actually measures what it attempts to measure
  
  • correlation can determine concurrent validity if there is a relationship between scores on new test and scores on an established test

Question 11

Problems with using correlations (directionality problem and third-variable problem)

Answer 11

• the directionality problem: correlational research does not establish a relationship of cause-and-effect
  
  • ex. can’t determine if the television content is influencing behavior, or whether the behavior is influencing the choice of television programs
• third-variable problem states that just because two variables vary together does not mean that there is a direct relationship between the variables
  
  • ex. ice cream consumption and crime rate, a third variable such as temperature may be responsible