PSYC311: Research Design and Statistics III - Week 5 Lecture

Flashcards for key concepts from the PSYC311 Week 5 lecture on regression and statistics.

The method of least squares is used to __ the error of prediction in regression analysis.

minimize

In multiple regression, the goodness of fit of the model is assessed using __ and R².

R

A significant F statistic indicates that the model is significantly better than __.

no model

In regression analysis, the individual contribution of predictors is indicated by __, t, and p.

b

The adjusted R² value __ the R² value based on the number of predictors in the model.

corrects

The regression equation for exam performance includes coefficients for __ and exam anxiety.

time spent revising

In multiple regression diagnostics, __ helps identify extreme positive or negative influences on the model.

outliers

The assumption of __ requires that all cases are independent from each other.

independence

The variance of Y must be constant at all levels of X according to the assumption of __.

homoscedasticity

High collinearity between predictors makes it difficult to assess their __ importance.

relative

Cook’s distance compares the regression outcome with the case __ versus excluded.

included

To assess normality of residuals, one can check the __ of standardized residuals and the Q-Q plot.

histogram

In the context of regression, the Durbin-Watson statistic tests for __ among residuals.

autocorrelation

The semi-partial correlation accounts for the unique contribution of a predictor after controlling for __ variables.

other

The model predicting record sales from attractiveness and plays on the radio was significant with F(2,197) = __, p<.001.

67.73

A significant semi-partial correlation for airplay indicates a unique contribution of __ to the model.

radio plays