SHAPING TODAY THOSE WHO WILL SHAPE TOMORROW

1
SHAPING TODAYTHOSE WHO WILL SHAPE TOMORROW
2
Differentiation in Higher Education Does a
Background in Economics improve performance?
Dr Chris Jones Anne WheelerLSE Developments
in Economics Education Conference Tuesday 6th
September 2011
3
Introduction Background

• Objective of paper
• Does past study in Economics and Math's improve
  exam performance for students who do a 1st year
  undergraduate module?
• This is a large module 400 students.
• 10 Credits, 20 hours of lectures and 5 hours of
  classes.
• If past study of Economics improves performance
  should pedagogy be differentiated by streaming
  students by ability?
• This is controversial at Secondary level have
  people ever considered it for Higher Education?

4
Literature

• Carol Tomlinsons definitions of differentiation
• Differentiation is simply attending to the
  learning needs of a particular student or small
  group of students rather then the more typical
  pattern of teaching the class as though all
  individuals are the same
• There is almost no literature on differentiation
  in Higher Education. Why?
• There are studies that look at the determinants
  of degree performance
• e.g. Smith Naylor (2001) Oxford Bulletin
• Age, Marital Status, Higher A-Level Score,
  social class have a ve impact.

5
Simple Predictions

• H1 Students who have studied Economics at
  A-Level (or an A-Level Equivalent) prior to
  joining Aston University perform better in
  Introductory Economics than students who have not
  studied Economics before.
• H2 Students who have studied Mathematics at
  A-Level (or an A-Level Equivalent) prior to
  joining Aston University perform better in
  Introductory Economics than students who have not
  studied Mathematics before.
• H3 Demographic factors have a significant impact
  on performance in Introductory Economics.

6
Data

• Data Discussion
• All of the data is from Student Information
  System
• We have 3 separate cross sections this is not
  panel data.
• 2007-2009
• On average there are approximately 400
  observations per year.
• We have data on past study in economics and
  maths, sex, age, home vs. overseas, school type
  and parent occupational class.
• Students participate in a number of different
  degrees. None of them study single honours
  Economics.

7
Descriptive Statistics
8
Grade Distributions
9
Econometric Model
Model estimated by OLS and a Tobit specification
because the data is censored at 40 (these are
the people who pass and get given 40). Our
primary interest is beta 1 and beta
2 SchoolType 0,1 Dummy (independent
school vs. the rest) ParOccup 0,1 Dummy
(higher managerial class vs. the rest)
10
Baseline Results
11
Results including Class
12
Discussion

• Results Conclusion
• Past study in Economics improved test scores by
  between 5-8 percent
• Past study in maths improved test score by
  between 1-3 percent
• Age, sex, school type and parental class has
  little impact on performance.
• For 2007 home students performed better by 3 on
  average, in 2008 2009 this disappeared.
• Should class teaching be differentiated by
  ability?

13
Questions?