Weekly Summary 2.1

CBSC520.18

Weekly Summary 2.1

This week is the second week of class. This week, in chapter three, we learned about finding relationships among variables. We see that relationship between variables is the primary thing of interest that we need for data analysis.

Examining relationships between categorical variables, the count of categories is done, and then the sum / percentage of the corresponding categories will help in analysis. This data is usually displayed in table called as crosstabs, which is also called contingency table.

We saw the difference between stacked and unstacked variables, out of which stacked variable is the two long variable stacked with each other. The Unstacked variable will have two short variables in an unstacked format.

The study of relation between numerical variables includes plotting the chart called scatterplot, and two summary measures called correlation and covariance. The scatterplot is created by scattering points, where each point denotes value of an observation for the two selected variables.

We see Correlation and Covariance help measuring the strength and direction of a linear relationship between two numerical variables. The relation is found strong if the points on the plot scatter and form like a straight line. And if this straight line rises from left to right, the relationship is positive, otherwise negative. We see Covariance is and average of products of deviations from means.

We see Pivot table which is available in Excel as a tool is used to break down data by categories. Pivot table is easy to sort or filter based on values or categories.

Thank you.