Handbook of psychology volume 7 educational psychology

Research Focus Areas

69

In subsequent research, Zimmerman and Martinez-Pons

(1990) found evidence of developmental trends among 5th,

8th, and 11th graders. Older students reviewed notes more

and texts less compared with younger children. With devel-

opment, students sought more assistance from teachers

and less from parents. Older students also displayed greater

use of record keeping and monitoring, organizing and trans-

forming, and goal setting and planning. The researchers

found that compared with boys, girls made greater use of

record keeping and monitoring, environmental structuring,

and goal setting and planning; they also found that compared

with regular students, gifted students displayed greater orga-

nizing and transforming, self-consequating, seeking peer as-

sistance, reviewing notes, and seeking adult assistance (fifth

grade only).

Various aspects of self-regulation were addressed by

Pintrich and De Groot (1990). Seventh graders were adminis-

tered the Motivated Strategies for Learning Questionnaire

(MSLQ). This instrument includes two categories: motiva-

tional beliefs (self-efficacy, intrinsic value, test anxiety) and

self-regulated learning strategies (cognitive strategy use,

self-regulation). Sample items tapping motivational beliefs

are Compared with other students in this class I expect to do

well and I think I will be able to use what I learn in this class

in other classes; for self-regulation, some sample items are

When I study I put important ideas into my own words and I

ask myself questions to make sure I know the material I have

been studying. Although the authors distinguished between

motivational beliefs and self-regulated strategies, establish-

ing and maintaining positive beliefs about learning is an

effective self-regulatory strategy (Zimmerman, 2000). The

MSLQ categories and those identified by Zimmerman and

Martinez-Pons (1986) show some overlap.

Operation of Self-Regulatory Processes During Learning

In this section we review research on self-regulatory processes

as students are engaged in academic tasks. Although there is

some overlap between areas, the review is organized accord-

ing to Zimmerman’s (1998) forethought, performance con-

trol, and self-reflection phases (Table 4.2).

Goal Setting

Goal setting is an integral component of the forethought

phase. Allowing students to set learning goals can enhance

their commitment to attaining them, which is necessary for

goals to affect performance (Locke & Latham, 1990). Schunk

(1985) found that self-set goals promoted self-efficacy.

Children with learning disabilities in mathematics received

subtraction instruction and practice over sessions. Some

set session performance goals; others had comparable goals

assigned; those in a third condition did not set or receive

goals. Self-set goals led to the highest self-efficacy and

achievement. Children in the two goal conditions demon-

strated greater motivation during self-regulated practice than

did no-goal students. Self-set children judged themselves

more efficacious for attaining their goals than did assigned-

goals students.

To test the idea that proximal goals enhance achievement

outcomes better than do distant goals, Bandura and Schunk

(1981) provided children with subtraction instruction and

self-regulated problem solving over sessions. Some set a

proximal goal of completing one set of materials each ses-

sion; others pursued a distant goal of completing all sets of

materials by the end of the last session; a third group was ad-

vised to work productively (general goal). Proximal goals led

to the most productive self-regulated practice and to the high-

est subtraction self-efficacy and achievement; the distant goal

resulted in no benefits compared with the general goal.

Schunk (1983c) tested the effects of goal difficulty. Dur-

ing a long division instructional program, children received

either difficult but attainable or easier goals of completing a

given number of problems each session. Within each goal

condition, children either were given direct attainment infor-

mation by an adult (i.e., You can do this) or received social

comparative information indicating that other similar chil-

dren had been able to complete that many problems. Difficult

goals enhanced motivation during self-regulated practice and

achievement; direct goal attainment information promoted

self-efficacy.

Schunk and Swartz (1993a, 1993b) investigated how goals

and progress feedback affected achievement outcomes and

self-regulation. Children received paragraph-writing instruc-

tion and self-directed practice over sessions. An adult modeled

a writing strategy, after which children practiced applying it to

compose paragraphs. Process- (learning-) goal children were

told to learn to use the strategy; product- (performance-) goal

children were advised to write paragraphs; general-goal stu-

dents were told to do their best. Half of the process-goal

students periodically received progress feedback that linked

strategy use with improved performance.

The process-goal-plus-feedback condition was the most

effective, and some benefits were obtained from the process

goal alone. Process-goal-plus-feedback students outperformed

product- and general-goal students on self-efficacy, writing

achievement, self-evaluated learning progress, and self-

regulated strategy use. Gains were maintained after 6 weeks;

children applied self-regulated composing strategies to types

of paragraphs on which they had received no instruction.

70

Self-Regulation and Learning

Zimmerman and Kitsantas (1996, 1997) found that provid-

ing process goals (similar to learning goals) raised self-

efficacy and self-regulation during dart throwing. Ninth and

10th-grade girls were assigned to a process-goal condition and

advised to focus on the steps in dart throwing. Others were as-

signed to a product- (performance-) goal condition and told

to concentrate on their scores. Some girls engaged in self-

monitoring by writing down after each throw the steps they

accomplished properly or their throw’s outcome.

In the first study (Zimmerman & Kitsantas, 1996), process-

goal girls attained higher self-efficacy and performance

than did product-goal girls. Self-recording also enhanced

these outcomes. The second study replicated these results

(Zimmerman & Kitsantas, 1997); however, a shifting-goal

condition was included in which girls pursued a process goal,

but after they could perform the steps automatically they

switched to a product goal of attaining high scores. The shift-

ing goal led to the highest self-efficacy and performance.

Social Modeling

Modeling studies provide evidence on how information

conveyed socially can be internalized by students and used

in self-regulation to produce greater learning. In addition to

their benefits on learning, models convey that observers can

succeed if they follow the same sequence. Students who be-

lieve they know how to perform a skill or strategy feel more ef-

ficacious and motivated to succeed (Schunk, 1987).

An important means of acquiring self-evaluative standards

is through observation of models. When children observe

modeled standards, they are more likely to adopt these stan-

dards, and model similarity can increase adoption of standards

(Davidson & Smith, 1982).

Zimmerman and Ringle (1981) found that models af-

fected children’s self-efficacy and achievement behaviors.

Children observed an adult model unsuccessfully try to solve

a wire-puzzle problem for a long or short period; the model

also verbalized statements of confidence or pessimism. Chil-

dren who observed a pessimistic model persist for a long

time lowered their self-efficacy judgments for performing

well.

Schunk (1981) provided children with either adult model-

ing or written instruction on mathematical division, followed

by guided and self-directed practice over sessions. The adult

model verbalized division solution steps while applying these

steps to problems. Both treatments enhanced self-efficacy, per-

sistence, and achievement, but modeling led to higher achieve-

ment and more accurate correspondence between self-efficacy

and actual performance. Path analysis showed that modeling

enhanced self-efficacy and achievement, self-efficacy directly

affected persistence and achievement, and persistence raised

achievement.

Schunk and his colleagues investigated the role of per-

ceived similarity in competence by comparing mastery with

coping models. Coping models initially demonstrate prob-

lems in learning but gradually improve and gain confidence.

They illustrate how effort and positive thoughts can overcome

difficulties. In addition to the modeled skills and strategies,

observers learn and internalize these motivational beliefs and

self-regulatory actions. Coping models contrast with mastery

models, who demonstrate competent performance throughout

the modeled sequence. In the early stages of learning, many

students may perceive themselves more similar in compe-

tence to coping models.

Schunk and Hanson (1985) had children observe models

solving subtraction problems. Peer mastery models solved

subtraction problems correctly and verbalized statements

reflecting high efficacy and ability, low task difficulty, and

positive attitudes. Peer coping models initially made errors

and verbalized negative statements, but then verbalized cop-

ing statements and eventually verbalized and performed as

well as mastery models did. After observing a peer mastery

model, peer coping model, adult mastery model, or no

model, children received instruction and self-regulated prac-

tice over sessions. Peer mastery and coping models in-

creased self-efficacy and achievement better than did adult

and no models; adult-model children outperformed no-

model students.

Schunk, Hanson, and Cox (1987) further explored mastery-

coping differences and found that observing peer coping

models enhanced children’s self-efficacy and achievement

more than did observing peer mastery models. Unlike the

Schunk and Hanson (1985) study, this project used fractions—

a task at which children previously had not been successful.

Coping models may be more effective when students have lit-

tle task familiarity or have had previous learning difficulties.

Schunk et al. also found that multiple peer coping or mastery

models promoted outcomes as well as did a single coping

model and better than did a single mastery model. With multi-

ple models, learners are apt to perceive themselves as similar

to at least one model.

Schunk and Hanson (1989) investigated self-modeling, or

cognitive and behavioral changes brought about by observ-

ing one’s own performances (Dowrick, 1983). Children were

videotaped while solving mathematical problems and then ob-

served their tapes, after which they engaged in self-regulated

practice. These children displayed higher self-efficacy, motiva-

tion, and self-regulated strategy use than did children who had

been taped but did not observe their tapes and children who

had not been taped.

Research Focus Areas

71

Social Comparisons

Social comparisons provide normative information for as-

sessing one’s capabilities during the performance control

phase. During long-division instructional sessions, Schunk

(1983b) gave some children performance goals; the others

were advised to work productively. Within each goal con-

dition, half of the students were told the number of prob-

lems that other similar children had completed—which

matched the session goal—to convey that the goals were at-

tainable; the other half were not given comparative informa-

tion. Goals enhanced self-efficacy; comparative information

promoted self-regulated problem solving. Students receiving

goals and comparative information demonstrated the highest

mathematical achievement. These results suggest that the

perception of progress toward a goal enhances motivation for

self-directed learning and skill acquisition.

Attributional Feedback

Self-regulation is facilitated by providing learners with attri-

butional feedback, or information linking performance with

one or more causes. Providing effort feedback for prior suc-

cesses supports students’ perceptions of their progress, sus-

tains motivation, and increases self-efficacy for learning.

Feedback linking early successes with ability (e.g., That’s

correct. You’re really good at this.) should enhance learning

efficacy. Effort feedback for early successes may be more

credible when students lack skills and must expend effort to

succeed. As they develop skills, switching to ability feedback

sustains self-efficacy and self-regulation.

Schunk (1982a) found that linking children’s prior

achievements with effort (e.g., You’ve been working hard.)

led to higher self-directed learning, self-efficacy, and achieve-

ment than did linking future achievement with effort (e.g., You

need to work hard.). Schunk (1983a) showed that ability feed-

back for prior successes (e.g., You’re good at this.) enhanced

self-efficacy and achievement better than did effort feedback

or ability-plus-effort feedback. Children in the latter condi-

tion may have discounted some ability information in favor of

effort. Schunk (1984b) found that providing children with

ability feedback for initial learning successes led to higher

ability attributions, self-efficacy, and achievement than did

effort feedback for early successes.

Schunk and Cox (1986) gave children with learning dis-

abilities effort feedback during the first or second half of a

subtraction instructional program or no effort feedback. At-

tributional feedback promoted self-efficacy, achievement,

and effort attributions better than did no feedback. Students

who received effort feedback during the first half of the

program judged effort as a more important cause of success

than did learners who received feedback during the second

half. Over a longer period, effort feedback for successes on

the same task could lead students to doubt their capabilities

and wonder why they still have to work hard to succeed.

Collectively, these results suggest that the credibility of at-

tributional feedback may be more important than the type.

Feedback that students believe is likely to enhance their self-

efficacy, motivation, and achievement. When feedback is not

credible, students may doubt their learning capabilities, and

motivation and achievement will suffer.

Strategy Instruction and Self-Verbalization

Learners’ verbalizations of self-regulatory strategies can guide

their learning during the performance control phase. Schunk

(1982b) provided modeled instruction on long division and

self-directed practice to children with low mathematical

achievement. Adult models verbalized strategy descriptors

(e.g., multiply, check) at appropriate places. During self-

directed practice, some children verbalized the descriptors,

others constructed their own verbalizations, those in a third

group overtly verbalized strategies and self-constructions, and

children in a fourth group did not verbalize.

Self-constructed verbalizations yielded the highest self-

directed practice and mathematical achievement. Children

who verbalized strategies and self-constructions judged self-

efficacy the highest. Self-constructions typically included

the strategies and were oriented toward successful problem

solving.

Schunk and Cox (1986) examined the role of verbalization

during learning of subtraction problem solving strategies

among children with learning disabilities. While solving

problems, continuous-verbalization students verbalized aloud

problem-solving operations. Midway through the instruc-

tional program, discontinued-verbalization children were

asked to no longer verbalize aloud. No-verbalization children

did not verbalize aloud.

Continuous verbalization led to the highest self-efficacy

and achievement. When instructed to discontinue verbalizing

aloud, these students may have not continued to use the ver-

bal mediators to regulate their academic performances. For

verbal mediators to become internalized, students may need

to be taught to fade overt verbalizations to a covert level.

Progress Feedback and Self-Evaluation

As learners pursue goals, it is important that they believe

they are making progress. During periods of self-reflection,

learners can evaluate their progress on tasks having clear

72

Self-Regulation and Learning

criteria; however, on many tasks it is difficult to determine

goal progress, especially when standards are not clear or

progress is slow. Feedback indicating progress can substanti-

ate self-efficacy and motivation. As learners become more

skillful, they become better at self-evaluating progress.

Schunk (1996) investigated how goals and self-evaluation

affected self-regulated learning and achievement outcomes.

Children received instruction and self-directed practice on

fractions over sessions. Students worked under conditions in-

volving either a goal of learning how to solve problems or

a goal of merely solving them. Half of the students in each

goal condition evaluated their problem-solving capabilities

after each session. The learning goal with or without self-

evaluation and the performance goal with self-evaluation led

to higher self-efficacy, skill, and motivation than did the per-

formance goal without self-evaluation. In a second study, all

students in each goal condition evaluated their progress once.

The learning goal led to higher motivation and achievement

outcomes than did the performance goal.

Frequent opportunities for self-evaluation of capabilities or

progress raised achievement outcomes regardless of whether

students received learning or performance goals. Conversely,

infrequent opportunities for self-evaluation promoted self-

regulated learning and self-efficacy only among students re-

ceiving learning goals. Under these conditions, self-evaluation

may complement learning goals better than it does perfor-

mance goals.

Schunk and Ertmer (1999) replicated these results with

college students during instruction on computer skills. When

opportunities for self-evaluation were minimal, the learning

goal led to higher self-efficacy, self-evaluated learning pro-

gress, and self-regulatory competence and strategy use; self-

evaluation promoted self-efficacy. Conversely, frequent

self-evaluation produced comparable outcomes when cou-

pled with a learning or performance goal.

Self-Monitoring

The effects of self-monitoring have been studied extensively

(Mace et al., 1989; Zimmerman, Bonner, & Kovach, 1996). In

an early study (Sagotsky, Patterson, & Lepper, 1978), fifth-

and sixth-grade students periodically monitored their work

during mathematics sessions and recorded whether they were

working on appropriate materials. Other students set daily per-

formance goals, and students in a third condition received self-

monitoring and goal setting. Self-monitoring significantly

increased students’ time on task and mathematical achieve-

ment; goal setting had minimal effects. The authors suggested

that children may have needed training on how to set chal-

lenging but attainable goals.

Schunk (1983d) found benefits of monitoring with chil-

dren during mathematics learning. Self-monitoring students

recorded their progress at the end of each session; external-

monitoring students had their progress recorded by an adult;

no-monitoring students were not monitored and did not self-

monitor. Self- and external monitoring enhanced self-efficacy

and achievement equally well, and both produced better re-

sults than did no monitoring. Effects of monitoring did not de-

pend on session performance because the three conditions did

not differ in work completed during self-directed practice. The

key was monitoring of progress rather than who performed it.

Reward Contingencies

Performance-contingent rewards during self-reflection can

enhance self-regulation and learning. During mathematical

division instruction with self-directed practice, performance-

contingent reward children were told they would earn points for

each problem solved correctly and that they could exchange

their points for prizes (Schunk, 1983e). Task-contingent reward

students were told that they would receive prizes for participat-

ing. Unexpected-reward children were allowed to choose

prizes after completing the project to disentangle the effects of

reward anticipation from those of reward receipt. Performance-

contingent rewards led to the highest self-regulated problem

solving, self-efficacy, and achievement. The other two condi-

tions did not differ. In other research, Schunk (1984) found that

combining performance-contingent rewards with proximal

goals enhanced self-efficacy and achievement better than did

either treatment alone.

INTERVENTIONS TO ENHANCE

SELF-REGULATION

Self-regulation does not develop automatically with matura-

tion, nor is it acquired passively from the environment. Sys-

tematic interventions assist the development and acquisition

of self-regulatory skills. In this section we describe in depth

an intervention project.

This project involved strategy instruction in paragraph

writing with elementary school children (Schunk & Swartz,

1993a, 1993b). The interventions used goal setting, progress

feedback, and self-evaluation of progress; the primary out-

come variables were achievement, self-regulated strategy

use, and self-efficacy.

Children received instruction and practice during twenty

45-min sessions over consecutive school days. The format

for each session was identical. The first 10 min were devoted

to modeled demonstration in which the teacher (a member of