SelfEfficacy and Resource Allocation Planning: A Discontinuous Model

1
Self-Efficacy and Resource Allocation Planning A
Discontinuous Model

• Jeffrey B. Vancouver
• Karen L. Scherbaum
• Ohio University

2
Abstract

• A discontinuous segmented model of the
  relationship between self-efficacy and resource
  allocation was tested. The non-monotonic model
  was largely supported based on 138 undergraduate
  students participating in a computer game.
  Discussion focuses on the implications regarding
  the distinction of goal processes and the
  interpretation of experimental research with
  self-efficacy.

3
Introduction

• Self-efficacy has been imbued with many positive
  qualities with regard to performance (e.g.,
  Bandura, 1977 Locke, 2001).
• choice of goal level,
• persistence in challenging goal pursuit, and
• a direct positive effect on performance.
• Bandura (1977) has argued that self-efficacy is a
  general mediator towards all performance hence,
  the higher the self-efficacy, the higher the
  performance, although with some caveats (Bandura,
  1997).
• Yet, early on, researchers complained that it was
  not clear how self-efficacy had these effects
  (see Rachman, 1978).

4

• We argue that the addition of the goal construct
  to self-efficacy theory (Bandura, 1997) helped
  explain the self-efficacy effect substantially.
• Specifically, self-efficacy positively influences
  goal choice, which has a positive effect on
  effort toward a task.
• However, it also likely to negatively influence
  effort allocation within a goal choice.
• Combining these processes, we suggested that
  resource allocation, which can be assessed in a
  planning context, would vary as a function of
  self-efficacy (see Figure 1).
• For very low self-efficacy (i.e., point a),
  individuals will not choose to accept the goal,
  so they will put no resources toward it.
• At point b, the individual has enough
  self-efficacy to accept the goal, but anticipates
  that substantial resources will need to be
  applied to accomplish it.
• Finally, at point c, the individual perceives
  that the goal should be accepted and will be
  relatively easy for him or her to achieve.

5
Figure 1 Discontinuous Model
6
Method

• Participants
• 138 undergraduate students (68 female, average
  age was 19.4 years).
• Task
• The Hurricane Game (see Figure 1).
• Participants were instructed to click on squares
  that move rapidly and randomly about in a
  specific area on a computer screen. The goal of
  the game was to hit as many boards (i.e.,
  squares) as possible in a three-minute period of
  time (i.e., a trial).

7

• Manipulations and Measures
• Self-Efficacy Manipulation
• Board sizes from 0 (the smallest) to 5 (the
  largest).
• Manipulation check participants were asked how
  likely they felt they were to hit each board size
  given each of the possible time limits (1 to 10
  s) using a scale from 1 (not at all likely) to 5
  (very likely).
• Resource Allocation Measure (DV)
• The participants had 3 s to allocate between 0
  (pass) and 10 s to try to hit the board for the
  given board size.
• Procedure
• Six, 10-sec. practice rounds for each of the 6
  board sizes, progressing from largest size to the
  smallest.
• Two practice trials,
• 3 min. each
• Each trial consisted of rounds where board sizes
  were randomly presented.
• Self-efficacy manipulation check measure
• Four experimental trials (same as practice
  trials)
• Self-efficacy manipulation check measure

8
Figure 2 Hurricane game screen shot.
9
Results

• Reliabilities and Manipulations Checks
• Participants allocation decisions were stable
  across trials (ICC2 .92 for Trials 1-4).
• Board size was highly related to the
  self-efficacy manipulation check (r .89, p lt
  .001).
• Assessing the Discontinuous Model (Figure 1)
• Figure 3 shows the frequency of seconds allocated
  for each of the board sizes.
• The probability that a board is rejected (i.e.,
  skipped) was a function of the board size.
• Correlation between the probability of endorsing
  0 s and the board size was -.95 (i.e., the
  smaller the board, the more likely the board was
  skipped).

10

• Correlation between the probabilities of
  endorsing the highest amount of time available
  (i.e., 10 s) was also highly negatively
  correlated (r -.94).
• Thus, the two boards that were perceived as most
  difficult had bimodal distributions.
• After removing the 10 s response, Figures 4 5
  describe the dummy codes and results of the
  regression models used to test the discontinuous
  model.
• Regressions were run for each participant
  individually. The median results are presented.

11
(No Transcript)
12
6
-.558 (-.317)
2.66 (.586)
Resources (seconds allocated)
Median R2 .601
.134 (.041)
0
Board size 0 1 2 3 4 5 a -1 1 0
0 0 0 b 0 0 1 1 1 1 c 0 0 -2 -1 1
2
Dummy codes
Figure 4 Median regression coefficients (and
betas) for Model 1 dummy codes.
13
6
1.474 (.321)
-1.210 (-.336)
Resources (seconds allocated)
1.632 (.413)
Median R2 .633
0
Board size 0 1 2 3 4 5 a -1 0 1
0 0 0 b 0 0 0 1 1 1 c 0 0 0 -1 0
1
Dummy codes
Figure 5 Median regression coefficients (and
betas) for Model 2 dummy codes.
14
Discussion

• The findings confirmed the hypothesis that the
  nature of the relationship between expectancies
  and resource allocation is non-monotonic in a
  planning context.
• The discontinuous model reflects the key role
  goal acceptance is believed to play in motivated
  behavior (Austin Vancouver, 1996).
• Beliefs about the ability to realize the goal,
  which in this case was clicking on the rapidly
  moving board, were expected to largely determine
  the likelihood that the individual would try for
  the goal. Indeed, it accounted for 90 of the
  variance in this choice.

15

• On the other hand, we expected that if the person
  chooses to pursue the goal, beliefs about the
  difficulty of reaching the goal would largely
  determine the resources applied.
• When just considering the probability of putting
  the maximum amount of resources (10 s) to the
  goal, difficulty accounted for 88 of the
  variance, with greater difficulty predicting
  probability of maximum resource allocation.

16
References

• Austin, J. T., Vancouver, J. B. (1996). Goal
  constructs in psychology Structure, process, and
  content. Psychological Bulletin,120, 338-375.
• Bandura, A. (1977). Self-efficacy Toward a
  unifying theory of behavioral change.
  Psychological Review, 84, 191-215.
• Bandura, A. (1997). Self-efficacy The exercise
  of control. New YorkFreeman.
• Locke, E. A. (2001). Self-set goals and
  self-efficacy as mediators of incentives and
  personality. In M. Erez, U. Kleinbeck, H.
  Thierry (Eds.), Work motivation in the context of
  a globalizing economy (pp. 13-26). Mahwah, NJ
  Lawrence Erlbaum.
• Powers, W.T. (1991). Commentary on Banduras
  Human Agency. American Psychologist, 46,
  151-153.
• Rachman, S. (Ed.) (1978). Special issue of
  perceived self-efficacy. Advances in Behaviour
  Research and Therapy, 1, 137-269.