L5 - Predictive Analytics in Jamovi

Correlation Analysis

focuses on the relationship between variables.

Linear Regression Analysis

is used to predict the effect of the predictors on an outcome variable.

Logistic Regression

predict the likelihood of something occurring on the basis of possible characteristics across a range of predictor variables.

Correlation Analysis

identifies a linear relationship between two variables (both numerical).

Correlation Analysis

measures the extent to which two variables are related.

Correlation Analysis

correlation coefficient (r): the measure of the linear relationship between two variables

Correlation Analysis

have a bidirectional relationship, meaning no direct independent variable. But this does not imply causation.

Correlation Analysis

Steps:

•Compute the effect: correlation coefficient (r)

•Determine the statistical significance (p value) of the effect

coefficient of determination (r2)

squaring the value of r we get the proportion of variance in one variable shared by the other

Correlation

__________ does not imply Causality

The third-variable problem

in any correlation, causality between two variables cannot be assumed because there may be other measured or unmeasured variables affecting the results.

Direction of causality

Correlation coefficients say nothing about which variable causes the other to change

Linear Regression Analysis

Predicts the value of outcome (numerical DV) on the basis of values on several predictors (numerical or categorical IVs)

Linear Regression Analysis

Steps:

•Assess the goodness of fit of the overall model (R² , F)

•Compute effects of each predictor variable (b1, b2 ....)

•Determine the statistical significance (p value) of each effect

Intercept

The _________ is the value of the Y (Dependent Variable) if all predictors (Independent Variable) is set to 0. The default or else statement.

Coefficients / Estimate

The __________ represent the "Amount of increase in Y for every one unit increase in X"

Linear Regression Analysis

Interpretation:

For every 1 point increase in career growth score, there is a 0.46 (estimate/coefficient) increase in job satisfaction score.

For every 1 point increase in well being scores, there is a 0.30 (estimate/coefficient) increase in job satisfaction score

Linear Regression Equation

Y = intercept + (b x predictor 1) + (b x predictor 2)....

Logistic Regression Analysis

Outcome variable (categorical) is binary (2 categories), predictor variables are either categorical or continuous.

Logistic Regression Analysis

Steps:

•Assess the goodness of fit of the overall model (X² , R²_n)

•Compute effects of each predictor variable (b1, b2 ....)

•Determine the statistical significance (p value) of each effect

Model Chi Square

To test how well the model fits the data (observed data vs. expected data if you used only the model logistic)

Model Chi Square P-Value (sig.)

Tells you whether your model significantly predicts your outcome variable (p value has to be <.05 for the model to be significant)

The Nagelkerke R-square

Gives and estimate of the % variation that we account for in outcome variable.

P value (sig)

Tells us which predictor variable potentially significantly impact the outcome variable

Odds ratios (Exp(B))

Indicates the changes in probability (or odds) of the occurence in the outcome variable for a change in one unit of the predictor variable.

McFadden’s R²

Description | Range | Key Characteristics | Common Use

Cox & Snell’s R²

Description | Range | Key Characteristics | Common Use

Nagelkerke’s R²

Description | Range | Key Characteristics | Common Use

Tjur’s R²

Description | Range | Key Characteristics | Common Use