Associative Analysis Flashcards
Types of Relationships between two variables
Nonmonotonic relationship `
General pattern - Presence of one due to presence of other
`
Monotonic Relationship
General direction - positively correlated relationships. X up, Y up. X down, Y down
Linear Relationship
Precise association. If X increases by z units, Y will increase by m*z units consistently
Curvilinear relationship
Positively then negatively correlated (or opposite) as z increases
Characteristics of relationships
Presence: Presented when Ho is rejected (Ho: no relationship / correlation)
Direction: + or -
Strength: Correlation coefficient (-1, 1, strong, 0 weak)
Cross-Tabulations & Chi-square Analysis (only for nominal Variables)
1) Cross Tabulations
2) Chi Square Analysis
3) Chi Square Value
4) Critical Chi Square Values
1) Cross Tabulations
Frequency: Actual raw numbers
Raw Percentage: Raw numbers divided by the grand total
Column Percentage: Raw numbers divided by the column total
Row Percentage: Raw numbers divided by the row total
2) Chi Square Analysis
Null Hypothesis: No association between the two variables
Difference between observed frequencies and expected frequencies
Expected cell frequency =
{cell column total * cell row total} / grand total
3) Chi Square Value
X^2= {Observed i - Expected i}^2 / Expected i
Observed = Actual number in cell
Expected: Check formula
(o-e)/e
4) Critical Chi Square Values
Degrees of freedom:
r-1)*(c-1
Correlation Analysis
-Covariation (co movement)
What happens to y when x changes?
- Null Hypothesis; Ho: no correlation
- Sign: +/-
- Coefficient (b in a+ xb)
3 assumptions of Pearson Product Moment Correlation
1) Do not take into account interaction with other variables
2) No causal relationships
3) Only linear relationships
Rank order correlation
Ordinal / Ranking variable
Monotonic relationship
Spearman rank order correlation