Another Paper on Goals, SelfEfficacy and Performance

1
Another Paper on Goals, Self-Efficacy and
Performance
But with a very Different set of Findings

• Jeffrey B. Vancouver
• Amy A. Williams
• Charles M. Thompson

2
Self-Regulation

• Social Cognitive Theory (Bandura, 1997)
• discrepancy reduction clearly plays a central
  role in any system of self-regulation (p. 131).
• however
• self-regulation via negative discrepancy tells
  only half the story and not necessarily the more
  interesting half (p. 131).
• More interesting half is positive effect of
  cognitive variables (e.g., SE) on motivation

3
The Paradox

• Powers argued that when individuals have high
  self-efficacy (as compared to when they have low
  self-efficacy) they may
• have overly optimistic assessments of the degree
  to which they are meeting their goals
• hence, lower discrepancies,
• hence, apply fewer resources to meeting goals
• Hence, self-efficacy would negative relate to
  subsequent performance

4
Evidence to Resolve Paradox

• Stajkovic Luthans (1998) meta-analysis of the
  relationship between self-efficacy and
  work-related performance found
• 28 increase in performance due to
  self-efficacy (p. 252)
• based on 109 studies, 148 correlations, 16,441
  participants
• Causality cannot be inferred
• Bandura acknowledges past performance as one of
  the antecedents to self-efficacy

5
Different Method

• Within-person, longitudinal design
• Several games (Gi), followed by observations of
  performance (Op) and self-efficacy(Ose)
• Lagged regressions

6
Hypotheses

• Positive correlation between performance and
  subsequent self-efficacy
• accounts for between-person positive correlation
• Across time, within an individual, self-efficacy
  will negative relate to performance
• i.e., the complacency effect

7
Personal Goals

• Although central to SCT and control theory
• Not sure self-report actually measures level of
  relevant internally desired state
• Instead, expect observations of personal goal to
  parallel self-efficacy observations
• positively related to past performance
• negatively related to subsequent performance

8
Graphic Representation


Performance
Self-Efficacy
Personal Goal
-
9
Method

• 56 undergraduates
• 10 games (2 practice, 8 trials)
• Observations
• performance row in which solution is found
• self-efficacy
• magnitude predicted row
• strength average of confidence rating across all
  possible levels of performance (reverse coded)
• personal goal for next game

10
Descriptive Statistics and Between-Person Level
of Analysis
aLower scores represent better performance p lt
.01 p lt .001
11
better
performance
worse
low
high
Self-efficacy
12
Within-Person SE Results
p lt .01 p lt .001
13
Within-Person SE Results
p lt .01 p lt .001
14
better
performance
worse
low
high
Self-efficacy
15
Within-Person Goal Results
p lt .01 p lt .001
16
Discussion

• Uncomfortable conclusions
• self-efficacy and high personal goals undermine
  performance
• Paradox resolved by eliminating discrepancy
  creation horn
• New paradox, personal goal key to discrepancy
  reduction horn

17
Study 2

• Assuming personal goal is measurement problem, we
  manipulated goal
• Assuming adoption of difficult goal is a decision
  making process, predict self-efficacy will
  positively relate to adoption (and therefore
  performance) in difficult goal condition

18
Method

• Identical to Study 1 except
• assigned goal after 3rd trial
• easy (find solution by 7th row)
• hard (find solution by 4th row)
• control (no additional instructions)
• played 3 more games
• 3 level HLM model

19
Results

• Replicated Study 1 within-person results
• Goal manipulation had typical effect
• hard goal individuals performed better than easy
  goal individuals (after controlling for baseline
  performance)
• Self-efficacy positively affected goal and
  performance in difficult goal condition

20
Self-Efficacy in Difficult Goal Condition (n 50)
apartial correlation p lt .05 p lt .01
21
General Discussion

• Self-efficacys relation to motivation (and
  hence, performance) is more complex than
  previously thought
• Self-reported personal goals may not reflect the
  internally represented desired states against
  which we attempt to control our perceptions

22
Conclusion

• Both halves of self-regulation process are
  interesting