Lecture 7: Correlational Methods and Statistics

Question 1

Correlational Research

Answer 1

Correlational research designs are used to search
for and describe relationships among measured variables. Correlations do not assess differences, they assess
relationships

Question 2

Correlational designs

Answer 2

Correlational designs are non-experimental methods that allow us to describe the relationship between 2 measured variables. E.g., hours studied and exam score. We call these variables X and Y. They also allow us to make predictions from one variable to the other. E.g. When we know the amount of hours someone studies (X), we can predict
their exam score (Y)

Question 3

Correlation Coefficient

Answer 3

The analysis associated with the correlation is the
correlation coefficient. A † Number that tells us about the relationship between two variables. Represented by r. It tells us about magnitude (strength) and direction of the relationship. Can range from -1 to +1

Question 4

Magnitude

Answer 4

The strength of the relationship. Determined by correlation coefficient. Between -1.00 and +1.00. Weak relationships closer to 0

Question 5

Direction

Answer 5

Direction is indicated by whether the correlation coefficient is positive or negative. A postive relationship is where the two variables go in the same direction (as x increases, y increases). A negative relationship is where the variables go in different directions (as x increases, y decreases.)

Question 6

Curvilinear relationships

Answer 6

A correlation coefficient of 0 could also indicate a

curvilinear relationship. U shape/Inverted U shape relationship.

Question 7

0 relationship (no relationship)

Answer 7

Variables are unrelated and do not move together

in any way

Question 8

Misinterpreting Correlations

Causality and Directionality

Answer 8

The most common error made when interpreting
correlations is assuming the relationship is causal (A change in variable A causes a change in variable B). Just because there’s a relationship between 2 variables does not mean that one is responsible for causing changes in the other. Correlation does not imply causation.

Question 9

Misinterpreting Correlations – Third Variable

Answer 9

When the correlation between two variables is dependent on another (third) variable.

Question 10

Partial correlation

Answer 10

We can control for the third variable problem by using a partial correlation. Partial correlation partials out (controls for) a third
variable

Question 11

Restrictive Range

Answer 11

A restricted range is a range of values that has
been shortened. For example, instead of looking at all depression scores (10 – 60) we only look at moderate depression cut off (40 – 60). If the range changes then r changes.

Question 12

Pearson’s Correlation Coefficient (r)

Answer 12

Used to measure linear relationships between two

interval or ratio level variables

Question 13

Coefficient of Determination

Answer 13

The coefficient of determination (r²) is a measure of the amount of variance explained in one variable by the other. For example, our r for hours studied and exam score is
= .50. Therefore r² is .25 which means that hours studies explains 25% of variance in exam scores.

Question 14

Regression analysis

Answer 14

A procedure that allows us to predict an individual’s score. We can use someone’s score on X to predict their score
on Y. We can use someone’s hours of studying score (X) to
predict their exam score (Y)

Question 15

Line of Best Fit (Regression Line)

Answer 15

The regression line (or line of best fit) is a best-fitting straight line drawn through the centre of a scatterplot
Best fits the data points (least deviation).

Question 16

Simple Regression formula

Answer 16

Y-hat = bX + a

b = slope
X = raw score
a = Y intercept

Question 17

Regression what does a slope (b) of .35 mean

Answer 17

With every one unit change in X (X is on a 1 point scale, so with every 1 point change in X) Y will increase .35 points