# Ffynone House is a Private Secondary School located in Swansea, Wales and They have contacted Buxton Analytics: Business Analytics Case Study, SU, UK

 University Swansea University (SU) Subject Business Analytics

Case Study

Ffynone House is a Private Secondary School located in Swansea, Wales. They have contacted Buxton Analytics to create an analysis report on students’ Welsh scores from 2014-2021. As class sizes vary the number of observations for each year also varies. Ffynone House has collected a large amount of data on a number of students’ attributes over the course of their time at the school. This data along with the grades each student received for their Welsh tests at three different test points.

Questions

1. Using the Glossary table within the provided excel file determine the type of
data used by each variable: Time, Discrete Numerical, Continuous Numerical,
Ordinal Categorical, Binary, Nominal Categorical

2. Using appropriate measures of descriptive analytics analyze the numerical and categorical variables comparing them in the graphical or tabular format against G1, G2, and G3.

3. Complete a multiple regression with G3 as the dependent variable with all
relevant variables as independent variables.

a. Undertake a correlation to determine which variables are correlated with each other.

i. Which variables are highly correlated with G3?

ii. Are there any independent variables that are highly correlated
with each other?

b. Convert categorical variables to dummy variables if you are including
them in the analysis

c. Undertake the multiple regression

d. Determine from the multiple regression which variables are insignificant

e. Rerun the regression with the insignificant variables removed

f. Provide an interpretation of the final regression model remember to
discuss any issues of multicollinearity.

4. You are expected to create and compare two forecasting methods with G3 as the dependent variable.

a. In order to successfully achieve this you will need to average the grades
for all students in each year group before undertaking your forecast.

b. Choose from the following 4 forecasting methods and remember to make
where required.

i. Method 1 – Naïve Forecast
ii. Method 2 – Moving Averages Forecast
iii. Method 3 – Exponential Smoothing Forecast
iv. Method 4 – Regression based Forecast

c. The average accuracy measures for each model also need to be
calculated.

5. After completing a thorough analysis of the Welsh scores the teachers at
Ffynone House would like to know:

a. What four variables have the most impact on student Welsh Scores 