Causality and Correlations

Causality

a chance in one variable produces a change in the second variable

4 criteria essential to establishing a causal relationship between 2 variables (RATN)

1. Rationale (logic)

2. Association

3. Time sequencing

4. Non-spuriousness

Rationale (logical explanation)

qualitative criteria, ex. how hours studied would correlate to higher exam grades

Association

statistical relation; when one variable changes, the other does too

Time sequencing

IV must occur before DV

Non-spuriousnesness

relationship cannot be explained by a third variable

Correlations (r_xy)

• Examines the relationship between variables with values ranging from -1 to +1

• How the value of one variable changes in relation to changes in another variable

• does not imply causation, just a numeric relationship

2 things to look for in Correlations

1. Sign

2. Magnitude

Sign

1. Positive Correlation (+)

   • direct correlation

   • when one variable increases the other variable increases too

2. Negative Correlation (-)

   • indirect correlation

   • when one variable increases the other variable decreases

Magnitude

0.2 - 1.0; weak to strong relationship on a scale

Correlation Matrix

Determines set of data to calculate variable correlation, producing a new set of data with knowledge of X and Y

operationalization

defining an abstract concept that is specific enough to be measured

variable

measurable trait or characteristic subject to change under different conditions

DV

variable that depends on the influence of the independent variable (the effect)

IV

variable that is hypothesized to cause or influence another (the cause)