Testing Two Explanations for GoalSetting Effects: A Persistent Question

1
Testing Two Explanations for Goal-Setting
Effects A Persistent Question

• Charles A. Scherbaum
• Jeffrey B. Vancouver
• Ohio University

2
Abstract

• It is widely accepted that goal setting works.
  Why it works is controversial. This paper
  examines two theoretical explanations
  (goal-setting theory and control theory) for
  goal-setting effects. Computational models based
  on each explanation were developed and compared
  to data collected from participants. The results
  supported the control theory account.

3
Introduction

• The effects of goal setting have been well
  documented (Locke Latham, 1990).
• The explanation for why it works has been more
  controversial (e.g., Naylor Ilgen, 1984).
• We think much of this controversy relates to
  different levels of theorizing.
• The goal-setting theory account uses a static
  process theory and molar concepts of causality to
  explain what happens.

4

• Control theory (Powers, 1973), a contender for
  explaining goal-setting effects, uses a dynamic
  process theory and micromediational concepts of
  causality to explain how it happens.
• Locke (1991) contents that goal-setting theorys
  elaboration of the goal-setting effect, which
  involves the motivational mechanisms of effort,
  persistence, and direction, serve as an adequate
  explanation of why specific, difficult goals have
  their effect on performance.
• Others have contented that control theory should
  serve as the explanation of goal-setting effects
  (Lord Hanges, 1987).

5

• Yet, finding empirical ways to separate these
  explanations have been difficult, with each side
  claiming the existing data supports their
  position.
• Thus, our intention was to present an empirical
  context in which predictions from the two
  theories could be differentiated. This was
  achieved via computational models derived from
  each theory and the comparison of the behavior of
  the computational model, to the behavior of
  participants performing the same task.

6
Method

• Participants
• Participants were 96 undergraduates.
• Empirical Context
• To examine the two explanations, a scheduling
  task called SimNurse (see Figure 1) was used.
• Computational models representing each theory and
  participants were assigned a goal for the cost of
  the personnel schedules. When the goal was met,
  the schedule was submitted and a new schedule to
  revise was received.

7

• Goal-Level Manipulations
• The goal for the cost of the schedules changed
  three times depending on the schedule being
  worked on. We utilized a within-subjects ABBA
  design with a random, between-person order
  assignment.
• Cost of Shift Manipulation
• Within the A and B trials, the cost of an
  employee shift was manipulated. Specifically,
• the cost was kept either the same, reduced, or
  increased throughout the trial.
• In the increasing and decreasing conditions the
  cost of a shift changed exponentially as the
  number of shifts deleted increased.
• A Latin square determined the order of
  presentation.
• This manipulation distinguished between whether
  actions or perceptions were being controlled (see
  computational model results).

8
Figure 1The SimNurse Program Interface
9
Goal-Setting Computational Model

• Given the linear conceptualization, and the
  articulation by Locke (1994) that acts, not
  perceptions, are controlled, we had the model
  calculate the number of shifts that would need to
  be deleted to close the distance between the
  initial cost of each schedule and the goal.
• If goal-setting theory is correct, then there
  should be a main effect of the cost of shift
  manipulation on the posted costs of schedule
  (see Figure 2 for the results).

10
Control Theory Computational Model

• For the control theory model, a conditional
  inequality tested if the cost of the schedule at
  any point in time was greater than the goal.
• If true, the response was to delete a shift. The
  result of deleting a shift was a decrease in the
  cost of the schedule. This continued until the
  cost was under the goal.
• If control theory is correct, there should be no
  effect of the cost shift manipulation on posted
  cost (see Figure 2 for the results).

11
Figure 2Comparing Computational Models
12
Results from the Participants

• Using a repeated measures MANOVA, we found a main
  effect of goal level on posted cost, Pillais
  Trace .89, F(1, 132) 1018.84, p lt .001, ?2
  .885, which was predicted by both models.
• However, there was no effect for the cost of
  shift manipulation, Pillais Trace .03, F(1,
  132) 2.23, p .11, ?2 .033, which was
  only predicted by control theory.

13
Discussion

• The results of this study clearly show support
  for the control theory explanation of the
  goal-setting effect.
• The ability of the study to differentiate
  predictions from the two theories is attributed
  to the dynamic protocol.
• We think the current study shows that control
  theory explanations are a piece of the puzzle
  that can plug into goal-setting theory at the
  level of explanation of relationships, as opposed
  to the level of describing the relationships.

14
References

• Locke, E. A. (1991). Goal theory vs. control
  theory Contrasting approaches to understanding
  work motivation. Motivation Emotion, 15, 9-28.
• Locke, E. A. (1994). The emperor is naked.
  Applied Psychology An International Review, 43,
  367-370.
• Locke, E. A., Latham, G. P. (1990). A theory of
  goal setting and task performance. Englewood
  Cliffs, NJ Prentice-Hall.
• Lord, R. G., Hanges, P. J. (1987). A control
  system model of organizational motivation
  Theoretical development and applied implications.
  Behavioral Science, 32, 161-178.
• Naylor, J. C., Ilgen, D. R. (1984). Goal
  setting A theoretical analysis of a motivational
  technology. In B. M. Staw L. L. Cummings
  (Eds.), Research in Organizational Behavior (Vol.
  6, pp. 95-140). Greenwich, CT JAI Press.
• Powers, W. T. (1973). Behavior The control of
  perception. Chicago Aldine.