Chris Evans, University of Winchester

1
Self-Efficacy and Academic Performancethe
effect of self-evaluation interventions
PLAT June 2006

• Chris Evans, University of Winchester
• Dr Paul Redford, UWE

2
Self-Efficacy and Academic Performancethe
effect of self-evaluation interventions

• Research Aims and Context
• Theoretical background
• Experimental design
• Initial Hypotheses
• Three studies and their results
• Year 1 Statistics
• Year 1 Psychological Thinking
• Year 2 Statistics
• Discussion

3
Aim Context

• Aim to explore the effect of self-evaluation
  interventions during a module on
• Self efficacy for the module
• Final module marks
• Context
• Study 1- Year 1 Statistics module
• Study 2 - Year 1 Statistics Psychological
  Thinking modules
• Study 3 - Year 2 Statistics module

4
Theoretical Background

• Self-efficacy is a personal judgement of
  performance capability that is
• domain specific
• context sensitive
• dependent upon mastery criteria
• Self-efficacy influences level of effort
  persistence
• Self-efficacy influences self-monitoring actions
  and self-evaluative standards
• Higher self-efficacy is linked with improved task
  performance
• Zimmerman, 2000

5
Self-efficacy for new students

• Self-efficacy has three sources
• mastery experiences
• vicarious experiences
• verbal persuasion
• New students initially lack knowledge of their
  subjects and of the standards expected
• So their self-efficacy is initially derived from
  similar prior experiences
• It is then modified through self-evaluation of
  actual experience
• Which influences future study attitudes and
  behaviour
• Bandura, 1997
• might regular self-evaluation of progress help
  enhance self-efficacy and performance?

6
Might regular self-evaluation of progress help
enhance self-efficacy and performance?

• Schunk Ertmer (1999)
• A self-evaluation intervention, administered to
  half the class in 3 week computer studies module
• Students were given clear goals for the module
• Self-efficacy and self-evaluations were measured
  against the same set of performance tasks -
  closely matched to the final performance test
• The intervention was shown to increase
  self-efficacy
• Actual self-evaluation levels were related to
  self-efficacy (r.64, plt0.05)
• Performance was not related to the presence of
  self-evaluation or to self-efficacy

7
Experimental Design - Study 1

• Year 1 (Semester 1, 2003) psychology students
• 12 lectures on introductory statistics using SPSS
  - with 10 weekly practice exercises
• Assessed by computer-based, open-book exam
• Intervention
• Experimental Group complete 10-item
  self-evaluations of progress in weeks 4,7,10
• Control Group answer questions on university
  adaptation

8
Measures

• Self-efficacy - 2 measures on 7 point scale
• Self-Efficacy (S-E) modelled upon Finney
  Schraw (2003)
• Perceived Competence from Deci Ryan (2003)
  Intrinsic Motivation Inventory (IMI)
• Performance - Final examination mark
• Effort - Number of completed weekly worksheets
• Other measures included
• Approaches to learning (ASSIST)
• 5-Factor Personality (NEO-PIR)
• Work Preference Inventory (Amabile et al, 1994)

9
Study 1 hypotheses

• Statistics self-efficacy (S-E) will increase over
  the module
• Post-module S-E will correlate with examination
  mark
• Self-evaluation ratings during the module will be
  related to post-module S-E
• Those completing three self-evaluations will
  have higher self-efficacy and examination marks
• Finney Schraw (2003) Schunk Ertmer
  (1999)

10
Study 1 key findings

• Significant findings supporting hypotheses
• Statistics S-E increased significantly over the
  module (from 2.7 to 4.0)
• t(45)6.62
• Post-module S-E correlated to exam mark
• r.26
• Mean self-evaluation score correlated to
  post-module S-E
• r.79
• plt.01, plt.05

11
Study 1 key findings

• Hypotheses that were not supported
• S-E was higher (4.2) for the experimental group
  (3 evaluations) than for the control (3.8) - but
  not significantly
• t(45)1.47, p.08
• Exam marks were higher (66) for the experimental
  group (3 self-evaluations) than for the control
  (55) - but not significantly
• t(45)1.51, p.07
• But this is a large increase in exam marks -
  perhaps a larger sample should be tested as
  sample size offers low power?

12
Study 2 - Aims and Context

• To increase the sample size by adding a second
  years statistics data
• To test a second subject to see whether any
  effects can be generalised beyond Year 1
  Statistics
• Participants were Year 1 (Semester 1, 2004)
  psychology students studying
• The same Statistics module
• Psychological Thinking - an introduction to
  psychological written argument assessed by two
  short essays and an examination

13
Study 2 - key findings for 2004 Statistics Module

• In this sample the effect of the 3 interventions
  was significantly related to higher exam marks
  (75 vs 64)
• t(58)2.40, plt.05
• Post-module S-E was not significantly affected by
  experimental group (4.5 vs 4.6)
• Post-module S-E also was not significantly
  correlated to exam mark

14
Combined Statistics Findings

• For the combined results (N107) S-E was
  correlated to exam mark (r.25)
• Other factors with significant 1-tailed
  correlations to achievement included
• Perceived Competence (r.34)
• Interest Enjoyment (r.30)
• Pressure Tension (r-.28)
• Number of completed worksheets (r.23)
• plt.05

15
Combined experimental result

• The effect of the self-evaluation intervention
  was also significant
• Mean exam marks were 61 for control and 71 for
  the experimental group
• t(105)2.62, p.01
• In a linear regression including all factors
  correlated to exam results, the following factors
  were extracted (plt.05)
• Experimental Group
• Perceived Competence
• Neuroticism
• Combined R².23

16
Psychological Thinking Findings

• The results for the Psychological Thinking module
  showed no effects of the self-evaluation
  intervention and little relationship between S-E
  and achievement
• The Perceived Competence measure had a similar
  relationship to exam marks as for Statistics
  (r.34) this may be a better predictor than
  the S-E scale
• This is an unconventional module, so may be a
  poor comparator
• But the Statistics experimental finding doesnt
  generalise to this module
• plt.01
• Perhaps it will generalise to a 2nd year
  Statistics module?

17
Study 3 - Aims and Context

• To see if the Statistics findings can also be
  detected in a similar Year 2 module
• Year 2 (Semester 1, 2005) psychology students
• Identical course structure to first year course
• Reduced set of measures

18
Study 3 - Findings

• These results also show no effect of the
  experimental group on achievement
• Perceived Competence is related to exam marks
  (r.35)
• The strongest predictor of achievement is the
  marks gained in the Year 1 exam (r.55)
• The small number (33) of participants is not
  ideal (power estimated at .25)
• However this does not provide evidence for
  generalising the Year 1 findings to this other
  Statistics module
• plt.01

19
Possible interpretation of findings

• There is a real and significant effect of the
  self-evaluation intervention on year 1 Statistics
  module
• Is this a Year 1 effect?
• Is is a subject effect?
• Is this due to the nature of the module?

20
A possible explanation?
self-reflection
understanding assessment standards
understanding of personal capability
development of study behaviour
results

• Self-evaluation intervention promotes
  self-reflection in a non self-regulating student

• This only influences the development of good
  study practice where self-reflection is accurate

• Accurately calibrated understanding of personal
  capability needs (a) an understanding of
  assessment standards
• (b) some relevant information about personal
  capability

• First-years are more likely to benefit from such
  an intervention - if their course makes
  assessment standards clear and gives relevant
  formative feedback

20
21
References

• Amabile, T. M., Hill, K. G., Hennessey, B. A.,
  Tighe, E. M. (1994). The Work Preference
  Inventory - assessing intrinsic and extrinsic
  motivational orientations. Journal of Personality
  and Social Psychology, 66(5), 950-67.
• Bandura, A. (1997). Self-Efficacy The Exercise
  of Control. NY, NY WH Freeman.
• Deci, E.L. Ryan, R.M. (2003). Intrinsic
  Motivation Inventory. Retrieved December 8, 2003
  from the World Wide Web
• http//www.psych.rochester.edu/SDT/measures/intri
  ns_scl.html
• Finney, S. J., Schraw, G. (2003). Self-efficacy
  beliefs in college statistics courses.
  Contemporary Educational Psychology, 28 (2),
  161-86.
• Schunk, D. H., Ertmer, P. A. (1999).
  Self-regulatory processes during computer skill
  acquisition Goal and self-evaluative influences.
  Journal of Educational Psychology, 91 (2),
  251-60.
• Zimmerman, B. J. (2000). Self-efficacy An
  essential motive to learn. Contemporary
  Educational Psychology, 25 (1), 82-91.