Statistical Modelling for University Administrators using R (online)

Description:

Overview of 2-day course
Are you working in Student Analytics?
Ever been asked if the average mark is changing significantly over academic years, or if the difference between the rate of change for females and males is statistically significant? 
Or which factors are associated with non-continuation?
Or which factors are associated with accepting an offer?
Or if the chance of achieving a first class honours degree is associated with tariff points on entry? 
This two-day course provides participants with hands-on experience of analysing their own type of records for data-driven planning and confidently interpreting numerical results for reports to policy makers and committees. The focus of the course is on the use of two statistical modelling techniques:

• Linear regression

• Logistic regression

Linear regression is used to examine how the mean of a numerical outcome, like final year mark, might be associated with different characteristics. If the outcome is binary, such as drop-out, logistic regression is used to investigate how the chance of failing to continue to the second year is associated with different characteristics.  Logistic regression is a popular modelling technique, for example it is advocated by the Office for Students in their Financial support evaluation toolkit.
The course also illustrates how these modelling techniques may be used for one-step-ahead forecasting into next year.
Presentations, demonstrations and hands-on computer practicals are based around the free statistical software R; see https://www.r-project.org/. Formulae are kept to a minimum; instead, we concentrate on results, their interpretation and reporting in plain language.

Cost
£498 (inclusive of 20% VAT) 

Delivery Mode
All training is online and will be delivered live on each day between 10:00 and 16:30 (GMT+1). The delivery platform is Zoom, which may be freely accessed. Questions may be asked using Zoom's chat box. Note we are a team of two presenters, so one of us is always available to provide additional support.  We also use Zoom meetings rather than webinars to encourage further interaction during an online course.​

Who Should Attend?
Administrators in educational establishments working in Policy, Planning and Strategy units; Data and Insight units; Business Intelligence units; those involved in extracting actionable insights from student records and in reporting to policy makers or committees. Anyone in these positions needing to answer questions around how student outcomes may be associated with different factors will benefit greatly from this course.
It is assumed that participants will, prior to the course, have:

• An understanding of mathematical functions and equations. In particular, the natural logarithmic and exponential functions (loge() and exp() respectively), the equation of a straight line and its geometrical representation

• Attended the one-day course Statistics for University Administrators, or Statistics for University Administrators using R, or have equivalent knowledge.

No previous experience of the R software is required; a brief introduction for the purpose of the course will be given.

How You Will Benefit
By the end of the course you will be familiar with two common statistical modelling methods for investigating associations and extracting actionable insights, be able to report the results in plain language, and be able to perform analyses using free statistical software. You will also be able to follow official guidance on the use of such models, e.g. the Office for Students’ guidance on the use of binary logistic regression for investigating the effectiveness of financial support with respect to student outcomes.

What Do We Cover?

• Introduction to the R software·

• Simple linear regression for relating a numerical outcome to a numerical explanatory variable

• Extending the linear regression model to incorporate categorical explanatory variables and interactions to allow for effect modification

• Using binary logistic regression in place of linear regression when modelling binary outcomes

• One-step-ahead forecasting.

Software
Practical work will be done in R.
Note:

• For practical work, participants must download and install the R software prior to the start of the course

• Practical work is based on the Windows operating system.

Extra Information
The R software is used on the course for two reasons:

• It is a free dedicated statistics package and can be used for other analyses

• It is a widely used software which will be maintained by the R Foundation for many years to come.

Cost:

£498

Website and registration:

Register for this course

Region:

International

Keywords:

Quantitative Data Handling and Data Analysis, Regression Methods, Linear regression, Logistic regression, Student analytics, Student drop-out rate, Offer acceptance, Student attainment, One-step-ahead forecasts, Odds ratio

Related publications and presentations from our eprints archive:

Quantitative Data Handling and Data Analysis
Regression Methods
Linear regression
Logistic regression

Back to the training database