in a system of performance management characterized by adaptive aspirations with
a field study of aspiration level setting at a large American multinational.
We believe that aspiration levels may be a plausible vehicle for understanding how
companies go about implementing the kinds of dynamic performance measures that
have attracted much recent attention. Furthermore, we are motivated by the observa-
tion that empirical exploration of how aspirations in organizations are actually for-
mulated has been limited. We have collected data on goals and budgets for individu-
als, groups, departments, and divisions over time at several sites in multiple
countries: We use data from these sites to estimate and compare one model of aspi-
ration level updating.^35 Patterns over time such as those highlighted in the experi-
mental literature might emerge, or some new finding might be uncovered. In addi-
tion, by comparing these goals and budgets across hierarchical levels and within
hierarchical levels over time, we hope to shed some light on the dynamics of goal set-
ting in complex organizations. The present study tests a specific model of aspiration
adaptation, the attainment discrepancy model.^36
According to the attainment discrepancy model, individuals and groups adapt their
aspiration levels using a simple decision rule of adjustment to feedback on perform-
ance compared with aspiration level. The model can be summarized as:
(1)where ALi,tis aspiration level for theith person or units in the current period, ALi,t–1
is the aspiration level for that entity in the previous period, and Pi,t–1is its perform-
ance in the previous period.
The attainment discrepancy model predicts that aspiration levels will be a function
of at least two variables: previous aspiration level and actual performance during the
previous period. The specific functional form shown in equation (1) was found by
Lant to provide the best description of aspiration-level adaptation. It models aspira-
tion level as a function of previous aspiration level and attainment discrepancy, with
a single period lag. In equation (1), attainment discrepancy showns as Pi,t–1ALi,t–1is
equal to the actual performance of the ith unit or person minus the aspiration level
for that entity. Thus, the model can be restated as follows:
(2)where 1 * 1 – 2 and 2 2 from equation (1). In order to avoid having ALi,t–1
enter the stimated equation twice, the parameter estimates in this study are developed
using the model in equation (2). For purposes of comparison with previous results,
we will restate the estimated parameters from equation (2) in terms of the parameters
in equation (1) in the discussion. We test the attainment discrepancy model using
field data on group goals, determined in the context of an American financial service
organization’s retail operations.
Our focus in reporting results from this model will be to provide examples of the
very different kinds of processes of aspiration-level adaptation that may take place in
multiple sites in different nations that are all part of a single division of a large Amer-
ican financial services company.
ALi,tb 0 b 1 *ALi,t 1 b 2 Pi,t 1ALi,ta 0 a 1 ALi,t 1 a 23 Pi,t 1 ALi,t 1426.3 A FIELD STUDY OF ADAPTIVE ASPIRATIONS 26 • 7(^35) Lant, 1992; Glynn, Lant, and Mezias, 1991.
(^36) Lant and Montgomery, 1987; Glynn, Lant, and Mezias, 1991; Lant, 1992.