Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.mso.anu.edu.au/pfrancis/HarleyWood.pdf Дата изменения: Mon Nov 4 03:46:23 2013 Дата индексирования: Thu Feb 27 21:46:31 2014 Кодировка:

How to teach people astrophysics so that they still remember it more than a week after the end of their last exam

Paul Francis, ANU

!1


Many of you will go into university teaching
· Many are probably doing it now. · But is it a critical problem?

!2


The amount of knowledge doubles every ~ 15 years

!3


So if a physics degree took three years in 1900...

· ·

Learning things with the same thoroughness today would take 507 years. Unless... we can teach it better.

!4


But despite big increases in spending, how much students are learning isn't increasing

!5


Net result, increasing specialization
· Arguably this is slowing scientific progress

!6


So is it possible to get knowledge into heads faster?
· And if so, how much faster? · Or is there some fundamental limit to how fast knowledge can be inserted into heads?

!7


What's the problem?
· · · · · Low recall? Misconceptions? Plug and chug? Teaching the wrong stuff? Transfer

!8


How much you remember?
100 90 80 70 % Recalled 60 50 40 30 20 10 0 0 5 10 15 20 25 30 35 40 45

Time in lecture
!9


Misconceptions
· If you throw a ball across a room... · What forces are acting on it as it flies through the air?

!10


Plug and Chug
· Rote-learn every possible equation · Use key-word recognition or timetabling to work out which equation to use. · Do algebra, plug in numbers.

!11


Are we teaching the right things?
Solving equations Statistics Physics Teaching Writing Programming

My degree

What I actually do

!12


Conventional wisdom amongst economists
· Nobody learns anything useful. · Higher education is about "signalling" and "credentialism"

!13


Can we do better?
· Yes, so it seems

!14


Force Concept Inventory Gain
100

Percentage of misconceptions fixed

90 80 70 60 50 40 30 20 10 0 2008 2009 2010 2011

Year of Teaching
!15


What works?
· · · · · Active not passive Know your goals Social, not solitary Student centred Deliberative practice and cognitive load

· Experimental with short feedback cycles

!16


Clickers

!17


Deliberative practice
· Research into real expert sports-people, chess players, scientists, doctors, musicians etc reveals that the real secret of elite performance is around 10,000 hours of "Deliberative Practice" · Very intense practice, constantly varying your approach, with instant feedback.

!18


Cognitive Load Theory
· What's hard about physics? · Working memory · Need to automate basic tasks

!19


Most important of all
· Keep an experimental, trial-and-error approach. · They are not like us · All educational theory is suspect due to publication bias
Number of Students 80

60

We are here

40

20

20

40

60

80

100

Total Mark

!20


Conclusions
· I don't know the answer - how fast people can be taught. · I suspect that it is a factor of ~3 faster than we currently achieve. · The situation is so complex that trial and (lots of) error is the only way to figure this out.

!21