Dawn Leeder and Raquel Morales

1
Adaptive Statistical Methods Generative Learning
Objects (GLOs)

• Dawn Leeder and Raquel Morales
• UCeL Director Eduserv
  Research Fellow

2
Outline

• UCeL who we are and what we do
• Generative Learning Object project with C-SAP
• Expert seminar what we want from you

3
UCeL partners
Nottingham Manchester UEA (Norwich) Peninsula
(Plymouth/Exeter) Wolverhampton Cambridge
(hub) London Metropolitan (advisory)
4
UCeL working definition of RLO
Reusable Learning Object (RLO) a
self-contained chunk of interactive web-based
learning, comprising a variety of media
ingredients that are combined to support a single
learning objective
5
Why are RLOs reusable?

• Single learning objective means
• Stand-alone
• Self-contained
• Can be joined up in different ways.
• Dont replace face-to-face,
• but enhance it.
• Can be used in different settings
• and across disciplines.

6
How are RLOs made?
First identify topics to be developed (e.g.
Statistics Research Methods Anatomy
Physiology Epidemiology Sociology) Based on
real educational needs Awareness of topics that
generalise
7
RLO production Stage 1 Content creation
unlocking content content created by subject
experts using template - RLO specification
based on single learning objective contains
activity and assessment peer review - quality
assurance
facilitated by UCeL content creation
workshops hands-on, one day workshops brainstorm
learning objective create content in
groups discuss and evaluate digitally archive
results
8
RLO production Stage 2 RLO development
content spec developed into RLO distributed
production media ingredients assembled (images,
audio, animations etc) Macromedia suite -
Dreamweaver, Director Flash MX technical check
- functionality peer review - second stage
facilitated by UCeL residential developer
training programme intensive week-long
course hot-house learning lab environment get
up to speed with templates, tools and
software share skills, tips and tricks RLOs
delivered to web
9
Use and evaluation

• 29 questions to determine
• Baseline
• comfort with computers
• use of web
• Usability
• navigation
• control
• ease of use
• Look and feel
• aesthetics
• Content
• level
• appropriateness

10

• UCeL A model for the production of
  collaborative, multidisciplinary eLearning
  materials
• Identify topics
• - Based on REAL educational needs
• - Broadly generalisable
• RLO content creation
• - Specification
• - Workshops
• - Peer-review
• RLO development
• - Ingredients and templates
• - Training programme
• - Peer-review
• Use and evaluation

11
Summary RLOs are expensive and time-consuming to
make ? collaborative creation and sharing
UCeL A Model for the the production of
collaborative, multidisciplinary eLearning
materials
12
Next generation of RLOs
Will support adaptable modifications to meet a
range of learning demands.
13
UCeL is now designing the next generation of
learning objects
Generative Learning Objects (GLOs) those that
can be customised by users so materials can be
presented in the most appropriate way for their
particular subject speciality
14

• Generative Learning Object project with C-SAP
• Adaptive Statistical Methods
• 18 months January 05 to June 06
• The project will include workshops,
• content development,
• resources development
• dissemination and
• feedback

15
Adaptive Statistical Methods
Statistical issues arise as important natural
parts of the process of reaching
conclusions. From astronomy to psychology
studies, statistical methods are required for
data analysis across disciplines.
16
STEPS Project STatistical Education through
Problem Solving
As part of the Teaching and Learning Technology
Programme phase 1, the STEPS project developed
problem-based teaching and learning materials for
statistics - specific problems in Biology,
Business, Geography and Psychology. Taking the
STEPS materials as our starting point, UCeL will
be using the different statistical tests as the
core for Adaptive Statistical Methods GLOs.
17
HANDLING DATA CYCLE
Specify the problem plan
Evaluate results
Collect data
Interpret and discuss data
Process and represent data
Adapted from Prof. Neville Davis, School of
Computing and Informatics Nottingham Trent
University
18
Handling Data Cycle Stage Tutor Action i)
Specify the problem and plan - Provide
description of the problem determines
plan/method to apply ii) Collect data -
Provide data to analyse iii) Process and
represent data - Select statistical method -
Select representation iv) Interpret and discuss
data - Provide interpretation - Provide
space for discussion
19
(No Transcript)
20
REALITY CHECKQUESTIONS TO YOU
21
i) What are the main problems tutors and
students encounter?
22
ii) What is your strategy for teaching Statistical
Methods? - is it problem-based approach? -
cover methods first and then applications?
23
iii) Which resources (i.e. software package) do
you use/teach at your department?
24
iv) What would be the optimal learning
resource? What would be the students journey?
25

• v) Having chosen a particular method, what do we
  need
• to interrogate our workshops attendees about?
• statistical method name is there an agreement
  on which methods
• are the most widely used?
• NOTE STEPS uses - plotting and summarising
  data,
• - confidence intervals and hypothesis
  test,
• - regression, and
• - the analysis variance.
• description of the method (optional? links to
  it?).
• software package tool used at their department
  any constrains on the
• data used?
• Glossary
• activity/assessment interpretation of results
• discussion of further action