Teaching Statistics in Psychology

1
Teaching Statistics in Psychology

• Martin Le Voi

2
Statistics is hard to understand

• At least for psychologists
• Many psychology departments struggle with
  explaining it
• Students struggle
• Every lecturer has a pet way of teaching
• Is the problem with one-size-fits-all

3
How do you understand the mean

• The central tendency of a group of numbers
• The expected value of a random variable
• As an equation

4
In pictures?
5
As a simulation?
6
As a song?!!

• http//resources4statistics.typepad.com/blog/2010/
  03/mean-median-and-mode-song.html

7
Students learn in different ways

• Well, they claim some things are easier than
  others
• Some text books are easier than others
• And it never seems unanimous
• Certainly some concepts can be explained in
  different ways

8
The Matching Hypothesis

• If you teach a student in concordance with their
  learning style, learning will be optimised
• But what is Learning Style?
• And how do you teach to it?
• Masser Mayer (2006). ATI hypothesis fails!

9
The ATI Hypothesis

• Masser Mayer (2006).
• Visual student, visual instruction GOOD
• Visual Student, verbal instruction BAD
• Verbal student, visual instruction BAD
• Verbal Student, verbal instruction GOOD
• Fails

10
Dodge the problem!

• Find and evaluate existing OERs for teaching
  statistics
• Classify by teaching approach
• Invite students to self-assess by sampling
  quality materials in different approaches
• Automate the process!

11
Search for psychometrically valid learning style
assessment

• Mayer Masser 2003
• Factor Analysed 14 measures together
• Best are
• MLPQ Cronbachs alpha 0.8
• Loads 0.979/0.541 on Learning Preference factor
• VVLS loads 0.831 on Cognitive Style factor
• Coded both into OpenLearn (demo)

12
Finding resources the problem

• http//psych.hanover.edu/aps/teaching.htmlstatist
  ics
• http//it.stlawu.edu/rlock/tise98/onepage.html
• http//www.stat.duke.edu/sites/java.html
• Problem of evaluation
• Merlot http//www.merlot.org/

13
Merlot

• Merlot has a systematic peer review system
• http//www.merlot.org/
• Peer review example (file)
• Clear winner for Visualisation teaching
• http//www.seeingstatistics.com/

14
OpenLearn courses in development

• Learning Statistics by visualisation
• Learning Statistics by graphics (maybe too
  similar to 1)
• Learning Statistics by algebra and text
• Not in OpenLearn
• Learning Statistics the non-linear approach (to
  be prepared in Prezi http//prezi.com/)