Lab 10 - Introductory Statistics, Correlation, and Bivariate Regression

The much dreaded first statistics module was not so dreadful :) In this week's lesson we covered quite a bit! We reviewed simple descriptive statistics, inferential statistics, regression statistics (specifically the meanings of Adjusted R-Square and p-value), population vs sample data; determined correlation coefficients,  produced a correlation matrix; and tested relationships between variables using bivariate regression analysis. All of these functions were conducted in Excel by learning how to use individual formulas as well as learning how to use the Data Analysis ToolPak.

For the last section in our lab assignment (Part C), we used regression results to predict annual precipitation data (rainfall) for a station that had missing information from 1931 to 1949. To do so, I first calculated the intercept coefficient (a = 162.34) and the slope (b = 0.85) of the regression using known data from weather Stations A and B (as values Y and X, respectively) from years 1950 to 2004. The intercept coefficient represents where x = 0 and/or where the regression line crosses the y-axis. This is helpful because linear regression is all about fitting the best line between two variables that describes a relationship, and can assist in predicting values. By plugging these known values into the Linear Regression Formula: Y = bX + a, I was able to fill-in the missing values for Station A with rainfall estimates.

Below is a screenshot of various functions covered in the lab assignment.