Data Analysis with RStudio has 5 learning outcomes:

1. Use statistics to reduce complex data situations to manageable formats in order to describe, explain and model them;

2. Derive descriptive statistics for various data types and manipulate data in RStudio;

3. Perform and critique statistical tests (parametric and non-parametric) on two and more sample data;

4. Use multiple regression with RStudio and other advanced statistical techniques to allow prediction of a score on one variable on the basis of the scores on several other variables;

5. Communicate effectively research findings in a clear concise manner using correct terminology based on output from RStudio.

Introduction to Data Analysis in RStudio

This part of the training provides an introduction to RStudio. The courses addresses LO1, LO2, and L06 by covering:
• A review of statistical terminology;
• Demonstration on how to organise data, create a dataframe, import data, export results, etc.;
• Conclude with compiling descriptive statistics (both numerical and graphical) for various data types.

Data Manipulation with RStudio

An important skill to using RStudio is being comfortable with manipulating your data. Often when it comes to statistical tests, data needs to be in a certain format. This part of the training will focus on common data manipulation techniques encountered in RStudio – i.e., Indexing; Sorting; Ordering a dataframe; Renaming variables; Data types; Grouping a variable; Merging Rows/Columns; Handling missing values; Locating outliers; Filtering. This part of the training addresses LO1 and LO3.
Statistical Inference with RStudio
The final segment to Part I assumes that the delegate will have an understanding of hypothesis testing. The focus of this part of the training is to demonstrate how to perform a proportion of inferential statistics in RStudio, with a focus on how to line up and analyse various data sets properly in both a parametric and non-parametric way. The course focusses on LO3, LO4 and L06 by covering:

• Worked examples of testing for normality/differences between two measurements;
• In the case of non-normal data, the workshop will discuss suitable transformations;
• One-way Analysis of Variance (ANOVA) with suitable posthoc testing;
• Within-Subjects Designs (Repeated Measures);
• Between- and Within-Subjects Designs (Mixed Factorial Experiments);
• Between-Subjects Designs (Two-way ANOVA).

Part II
Correlation and Multiple Linear Regression in RStudio
Correlation analysis is used to determine if there is a statistically significant relationship between two variables. Linear regression analysis is used to make predictions based on the relationship that exists between two variables. However, most things are too complicated to “model” them with just two variables - Multiple Regression is needed. Multiple regression is a statistical technique that allows us to predict a score on one variable based on the scores on several other variables. The course focuses on LO2, LO3, LO5 and L06 by covering:
• Scatterplots and partial regression plots;
• Correlation and regression analysis;
• How detect for multi-collinearity [using Variance Inflation Factor (VIF)] and outliers (using leveraging);
• How to check that the residuals (errors) are approximately normally distributed and random;
• How to interpret regression equations and how to use them to make predictions.

Not applicable to this module.

• Feedback individually to students, on an activity or draft prior to summative assessment
• Feedback individually to students, post-assessment

Not yet recorded.