Understanding Correlation and Regression

What is a scatter diagram used for?

• To create a pie chart representation.

• To display data in a tabular format.

• To show the relationship between two variables.

• To calculate the average of a dataset.

Define Karl Pearson’s coefficient of correlation.

• A method to calculate the mean of a dataset.

• Karl Pearson’s coefficient of correlation is a statistical measure that quantifies the degree to which two variables are linearly related.

• A measure of the average of two variables.

• A technique for visualizing data relationships.

How is Spearman’s coefficient of rank correlation different from Pearson’s?

• Spearman's coefficient is for ranked data and non-parametric, while Pearson's is for continuous data and assumes linearity.

• Spearman's coefficient requires a linear relationship between variables.

• Pearson's correlation measures ranked data and is non-parametric.

• Spearman's coefficient is used for continuous data and assumes normality.

What does a correlation coefficient of 1 indicate?

• No relationship at all.

• A perfect negative linear relationship.

• A perfect positive linear relationship.

• A weak positive linear relationship.

Explain the method of least squares in regression analysis.

• The method of least squares is a graphical technique for data visualization.

• The method of least squares is used to calculate the average of a dataset.

• The method of least squares focuses on maximizing the sum of the residuals.

• The method of least squares is a statistical technique used to determine the best-fitting line in regression analysis by minimizing the sum of the squares of the residuals.

What is the purpose of regression analysis?

• To analyze relationships and make predictions between variables.

• To summarize data without any analysis.

• To create visual representations of data only.

• To determine the average of a dataset.

How do you interpret a negative correlation coefficient?

• A negative correlation coefficient means both variables increase together.

• A negative correlation coefficient shows a positive relationship between two variables.

• A negative correlation coefficient indicates no relationship between two variables.

• A negative correlation coefficient signifies an inverse relationship between two variables.

What are the assumptions made in linear regression?

• The assumptions include perfect multicollinearity and non-normality.

• Linearity, independence, and perfect correlation are key assumptions.

• Heteroscedasticity and dependence are assumed in linear regression.

• The assumptions made in linear regression include linearity, independence, homoscedasticity, normality, and no multicollinearity.

What is the significance of the slope in a regression line?

• The slope measures the total number of data points in the dataset.

• The slope indicates the average value of the dependent variable.

• The slope signifies the relationship strength and direction between the variables.

• The slope represents the intercept of the regression line.

How can outliers affect correlation and regression results?

• Outliers only affect the visual representation of data, not the calculations.

• Outliers have no effect on correlation or regression results.

• Outliers can distort correlation and regression results by skewing coefficients and influencing model accuracy.

• Outliers always improve the accuracy of regression models.