ance. By definition, outcome goals, such as readers’ reactions to a journalist’s
article, provide the ultimate criterion by which process attainments can be
measured, and, as such, outcome goals can motivate moderately successful
learners to continue the quest toward higher levels of mastery (Locke &
Latham, 2002). At level four, learners can practice a skill with minimal proc-
ess monitoring because of automaticity, and their attention can be shifted to
performance outcomes without detrimental consequences, such as when stu-
dent journalists can shift their attention from their writing techniques to de-
sired audience reactions regarding an article for the school newspaper. How-
ever, like level three self-controllers, level four self-regulators will attribute
unsuccessful performance outcomes to ineffective processes or techniques be-
cause level three learning focused on the importance of the quality of these
processes (or means) to an expert model’s success (or ends). As a result of
their process attributions, level four students will display higher self-efficacy
for positive outcomes and minimal reversals in self-efficacy for setbacks,
which can greatly reduce pernicious swings in emotions.
A self-regulated level of skill is reached when learners can adapt their per-
formance successfully to changing personal conditions and outcomes, Thus,
a multilevel analysis of the development of self-regulatory competence begins
with most extensive social guidance at the first level, and this social support is
reduced systematically as learners acquire underlying self-regulatory skill.
This systematic reduction in social assistance to enhance students’ develop-
ment of self-regulation is similar to apprenticeship formulations advocated
by Vygotskian researchers (Rogoff, 1990). However, learners’ level four func-
tioning continues to depend on social resources on a self-initiated basis, such
as when a journalist seeks advice from a colleague about whether a draft of
an article is compelling. Because the effectiveness of one’s task skills depends
on variations in contexts, new performance tasks can uncover limitations in
existing processes and can require additional social learning experiences. This
multilevel formulation does not assume that learners must advance through
the four levels in an invariant sequence as developmental stage models as-
sume, or that once the highest level is attained, it will be used universally. In-
stead, this multilevel model assumes that students who master each skill level
in sequence will learn more easily and effectively. Although level four learn-
ers have the competence to perform self-regulatively, they may not choose to
do so because of low levels of self-motivation (Bandura, 1997).
Evidence of Levels in Self-Regulatory
Development of Skill
There is a growing body of evidence indicating that the speed and quality of
learners’ self-regulatory development and self-motivation are enhanced sig-
nificantly if learners proceed according to a multilevel developmental hierar-
- SELF-REGULATION 337