Development of attributional beliefs and strategic knowledge
in Years 5 to 9: A longitudinal analysis.
Lorna K S Chan & Phillip J Moore
The University of Newcastle
Australia
Abstract
This paper reports on a three-year longitudinal study of students' attributional beliefs and
strategic knowledge in school learning. Two cohorts of primary and high school students
were followed through for three years from Years 5 to 7 and 7 to 9, respectively. Data
were collected each year on students' attributional beliefs regarding the reasons for their
school success and failure, their knowledge and reported use of learning strategies, and
academic achievement. Intervention programs were implemented in six Year 6 and seven
Year 8 English classes in the second year of the longitudinal project. The intervention
aimed to promote strategic learning in students through combining the teaching of
learning strategies with attempts at changing students' attributional beliefs. In the third
year of the project, the intervention continued in three Year 7 and three Year 9 English
classes as well as in three Year 7 and one Year 9 Mathematics classes. This paper focuses
on the causal influences of prior measures of attributional beliefs, strategic knowledge and
achievement on measures taken in the following year. Results of the differential patterns
of causal influences of these measures for intervention and non-intervention students are
reported.
Acknowledgements
The research project is supported by an ARC Large Grant. The authors wish to thank the
principals, teachers and students of the primary and high schools involved in this project
for their support. The valuable assistance of Hedy Fairbairn and Anne-Marie Youlden is
also greatly appreciated.
Paper presented at the Australian Association for Research in Education Annual
Conference, Brisbane, Australia, December, 1997.
Student learning research has for some time demonstrated the crucial role played
by motivational and strategic processing factors in academic achievement. The study
reported in this paper aims to extend our understanding of the relationship, particularly
the interdependence, among these factors from a developmental perspective. The
motivation construct of causal attributions and its influence on school achievement are
examined through the mediating variable of strategic learning. In addition, in the second
and third years of the research project, strategy and attribution training programs were
implemented in some classes, and the differential patterns of the relationships among
attributional beliefs, strategic learning and achievement for the intervention and non-
intervention students are compared. The paper follows directly from Fairbairn, Moore &
Chan (1994), Rodwell & Moore (1994), Chan & Youlden (1995), Moore & Chan (1995)
and Youlden & Chan (1997) which reported on the development of the assessment scales
and preliminary results from the first year of the three-year longitudinal project.
The motivation construct under investigation is causal attributions which refers to
what students perceive as the cause of their successes and failures in school. Earlier
research in attributions focused on ability, effort, and external factors like task difficulty,
significant others and luck as the major perceived causes of success/failure (Weiner,
1984, 1991). More recently attention has been drawn to the importance of strategy
attributions in enhancing motivation to learn (Borkowski, Chan, & Muthukrishner, in
press; Chan, 1994, 1996b; Chan & Youlden, 1995; Moore & Chan, 1995). According to
Weiner's theory (1984), attributions can be classified along three dimensions: locus,
stability and controllability. For example, ability attributions have an internal locus and
are stable but uncontrollable whereas effort and strategy attributions have an internal
locus and are unstable (therefore can be changed) but are controllable. Students who
perceive their failures as attributable to lack of ability are likely to develop a set for
future failure because whatever they do is perceived to be dependent upon their ability
which is unchangeable and out of their control. Indeed, in extreme instances such an
attribution can lead to learned helplessness (Chan, 1994, 1996b; Reid & Borkowski,
1987). On the other hand, students who perceive they have control over school successes
or failures are likely to have higher expectations of success and are motivated because
they realise that their effort and use of strategies are closely linked with their
performance.
Perceptions of personal control (effort and strategy attributions) have been shown
to relate positively to strategic learning and subsequently, academic performance
(Borkowski, Weyhing & Carr, 1988; Chan, 1994, 1996a). The recent literature abounds
with calls of enhancing student learning by teaching students critical thinking skills or
learning strategies and getting students to become active, strategic, metacognitive,
independent, self-directed or self-regulated learners (e.g. Borkowski & Muthukrishner,
1992; Paris & Winograd, 1990). Regardless of the term that is used, the emphasis is on
how to think, how to learn and taking active control over one's own thinking and learning
processes. Strategic learning is considered to involve two essential features of
metacognition, self-appraisal and self-management of cognition (Paris & Winograd,
1990). Self-appraisal includes personal reflections about one's knowledge states and
abilities while self-management refers to executive cognitive actions that help guide and
coordinate thinking, such as planning, using a variety of strategies, monitoring, evaluating
and regulating. In sum, strategic learning emphasises awareness and executive
management of our own thinking and learning.
The interdependence of cognitive, metacognitive and motivational factors in
academic learning has recently been highlighted. The most recent extension of
metacognitive theory proposes that personal-motivational factors energize the self-
regulating executive functions necessary for strategy selection, implementation and
monitoring (Borkowski, Carr, Rellinger, & Pressley, 1990; Borkowski & Muthukrishna,
1992). As proposed in Borkowski, Chan, & Muthukrishner (in press), such planning,
evaluating and regulating processes require effort, initiation, willingness to try as well as
persistence. Further, there has to be some expectation of success before a student is
prepared to try, marshall appropriate effort, and persist when encountering difficulties. If
there is little expectation of success, students will expend little effort in learning. Indeed
students may even avoid tasks that are likely to lead to failure.
Before students, regardless of ability, are prepared to deploy effort in planning,
evaluating and regulating strategy use, they must believe in: (1) the value of good
performance on the given task, that is, they have to want to do well and to care whether
they get a good result; (2) personal control over task outcomes, that is, they must be
convinced that success or failure on the task depends on themselves; (3) usefulness of
strategy use, that is, they must have the knowledge that use of specific strategies will lead
to better performance on the task; and (4) their ability to use strategies effectively and
successfully, that is, they have to perceive themselves as capable and competent. In other
words, students who are committed to do well on a given task, who are confident in their
own capabilities, who have well-developed specific strategy knowledge, and who believe
that their effortful use of strategies will lead to successful task performance are likely to
be active in strategy selection, monitoring and regulation.
The close relationship between attributional beliefs, strategic
learning and achievement has been demonstrated empirically. Importantly, the pattern
of relationships appears to change across the years of schooling, as observed in cross-sectional studies.
(Chan, 1994; Clayton-Jones et al., 1992). In the Clayton-Jones' et al study, students from
Grades 4, 6, 7, 9, 11 and TAFE classes were administered a general attribution scale
incorporating ability, luck, effort and strategy attributions for success and failure. For the
primary aged children, effort attribution for success was positively related to achievement
in Maths and English (a combined score) but at Grade 9, strategy attribution for success
emerged as a positive predictor of achievement. Ability attribution for failure, however,
was a pervasive negative influence across all grades, except Grade 4. In the Chan (1994)
study, students in Grades 7 and 9 who believe that they have personal control over
learning outcomes, who are not inclined to feel helpless in their learning, and who have
high self-perceptions of cognitive competence, were found to be more likely to use
strategies in their learning; while for Grade 5 students, it is only the perceived
competence measure that was found to influence strategic learning. When reading
achievement is taken into consideration, the hierarchical regression analyses reveal that
while the motivation variables have a more important role (relative to strategic learning)
in explaining reading achievement variance in the younger grades, in Grade 9 the role of
strategic learning variables is as important, if not more important, as the motivation
variables. Indeed, the path analyses results have further clarified such relationships.
Strategic learning was found to mediate between the effects of motivation on reading
achievement only for Grade 9 but not the younger grades.
Metacognition, and similarly strategic learning, is considered to be "embedded in
cognitive development and represents the kind of knowledge and executive abilities that
develop with experience and schooling" (Paris & Winograd, 1990, p.19). Indeed,
research into the causal relations between phonological processing and acquisition of
reading skills has provided strong support for a theory of reciprocal causation. It has been
found that certain phonological processing abilities (involving awareness and executive
control of cognitive processes, that is, strategic learning) play a causal role in learning to
read, and learning to read plays a causal role in the subsequent development of certain
phonological processing abilities. Evidence supporting this reciprocal relationship has
come from longitudinal and training studies of children in early primary grades (Juel,
Griffith & Gough, 1986; Perfetti, Beck, Bell & Hughes, 1987; Stanovich, Cunningham &
Feeman, 1984; Wagner, 1988). Given the recent recognition of the interdependence of the
cognitive, metacognitive and motivational aspects of academic learning (Borkowski, 1992;
Borkowski, Carr, Rellinger & Pressley, 1990; Chan, 1991, 1994, 1996a; Paris &
Winograd, 1990), the same reciprocal causation can be hypothesised for the relationship
between attributional beliefs, strategic learning and school achievement. Empirical support
for reciprocal causation requires longitudinal data, not cross-sectional data. The main
objective of the present study, therefore, is to examine empirically the reciprocal causal
relationship between motivational and metacognitive factors in school learning, and more
importantly, the developmental pattern of motivational and metacognitive influence on
achievement across years of schooling from a longitudinal perspective.
Method
Subjects
The subject sample included a primary and a high school cohort of students
followed through for three years from 1994 to 1996. The primary school cohort started
with 391 Year 5 students (183 boys and 208 girls) from 12 primary schools in a
metropolitan school district in New South Wales, Australia. They were followed through
the last two years of primary school (Years 5 and 6) and first year of high school (Year
7). The high school cohort consisted of 803 Year 7 students (419 boys and 384 girls)
from four high schools in the same school district and were followed through to Year 9.
The 12 primary schools belong to the cluster of feeder primary schools for the four high
schools. All participating schools are comprehensive and co-educational. Students
attending these schools come from both lower and middle socio-economic backgrounds.
The study reported in this paper is part of a large research project examining
primary and high school students' attributional beliefs, strategic learning and achievement
in English, Mathematics, Social Studies and in general school learning from both a cross-
sectional and longitudinal perspective. In the second year of the project, an intervention
program (further details will be given below) was implemented in some Year 6 and 8
English classes. In the third year, intervention programs were implemented in some Years
7 and 9 English and Maths classes. The data analysis reported in this paper involves only
subjects with complete data in the second and third year of the longitudinal follow-up.
The subjects were classified into two groups. The no intervention group comprises those
subjects who have not participated in any of the intervention classes, neither in the second
nor in the third year of the project. The intervention group comprises subjects who have
participated in an intervention class in the second or third year of the project. While we
were able to follow up and retain a reasonable proportion of the Year 7 subjects through
to Year 9 (166 in the intervention group and 312 in the no intervention group), we were
not as successful with the primary school cohort (88 in the intervention group and 96 in
the no intervention group). Due to limitations of resources, we were only able to follow
up those subjects in the primary cohort who enroled in the four participating high schools
in the third year.
Assessment Instruments
The instruments used to assess attributional beliefs and strategic learning are
described below. Two parallel versions were constructed for each scale, one for primary
school students and the other, high school students. Language use in the statement in each
item is simplified for the primary school version.
Self-Regulated Learning Strategies (General) Scale. The Self-regulated Learning
Strategies (General) Scale was designed to assess students' awareness and reported use of
self-regulated learning strategies. This scale was developed from an earlier version
constructed for previous research (Youlden, 1993; Youlden and Chan, 1994). It consists
of 20 items, each describing a student using one of the ten self-regulated learning
strategies identified by Zimmerman & Martinez-Pons (1988), such as, "John asks himself
questions on what he has been studying to make sure he can remember it". After each
description, students are required to rate the strategy on two separate four-point scales in
terms of how helpful they consider the strategy to be and how often they study that way.
Four subscales were developed using one-factor congeneric measurement models
(Joreskog & Sorbom, 1989), including knowledge and use of two sets of strategies. Each
of the Knowledge of and Reported Use of General Study Strategies subscales consists of
10 items, describing 2 goal-setting and planning, 2 seeking information, 2 self-evaluating,
2 seeking social assistance, and 2 environmental structuring strategies. Each of the
Knowledge of and Reported Use of Metamemory Strategies subscales consists of 7 items,
describing 2 keeping records and monitoring, 1 reviewing records, 2 organising and
transforming, and 2 rehearsing and memorising strategies. (2 self-consequating and 1
environmental structuring strategy items were discarded because of poor measurement
properties.) Each subscale score represents a maximally weighted composite of the items
for the subscale.
Causal Attributions (General) Scale. Students' attributional beliefs were assessed
using the Causal Attribution Scale initially developed by Chan (1994) and revised for this
study. The Causal Attribution (General) Scale is a ten-statement scale designed to assess
students' tendency of attributing their school success and failure experiences to the four
likely reasons of effort, ability, strategy use and luck. Five statements describe success
incidents and five describe failure incidents. For each statement, four different reasons
were listed and students were required to rate each reason on a four-point scale to
indicate how true they consider that particular reason to be for them. Hence the scale
consists of eight subscales, (ability, effort, strategies and luck for success and failure),
each with five items. One-factor congeneric measurement models (Joreskog & Sorbom,
1989) were employed to obtain subscale scores for the eight success and failure
attributions. Each subscale score represents a maximally weighted composite of the five
items for the subscale.
Procedures
Scale administration. Data collection took place during the end of Term Three
and the beginning of Term 4 each year during 1994-96. Since this study is part of a
bigger research project investigating relationships between attributional beliefs, strategic
learning and achievement in different subject domains, the questionnaires were
administered over four different occasions. On each visit, a set of three questionnaires
was administered to each grade - Learning Process Questionnaire, Causal Attributions
Scale and Self-Regulated Learning Strategies Scale. Due to the repetitive nature of the
questionnaires, care was taken to avoid an order bias. This was achieved by randomising
the order of the items for the questionnaire in each subject-domain, randomising the order
in which the subject domains were presented over the four visits; and randomising the
order in which the three questionnaires were given on each visit. Procedures were
standardised for administering the questionnaires. Students were read the instructions and
then completed the practice items for the questionnaire. The items were read to the
students as they completed the questionnaires.
Intervention programs. In the second and third years of the project, intervention
programs were implemented in some classes. The intervention aimed to promote strategic
learning in students through combining the teaching of cognitive and metacognitive
strategies with attempts at changing students' attributional beliefs. In the second year of
the project, intervention programs were implemented in the subject domain of
English/reading in six Year 6 and seven Year 8 classes. In the third year of the project,
the intervention in English/reading continued in three Year 7 and three Year 9 classes
while intervention programs in the subject domain of mathematics were introduced in one
Year 9 and three Year 7 classes. In the English intervention program, strategy lessons
were conducted in each class once a week for 12-15 weeks by a trained graduate research
assistant. Comprehension and summarising strategies were taught in the Years 6 and 8
classes and writing strategies were taught in the Years 7 and 9 classes. Collaborating
class teachers were requested to emphasise using strategies and to employ feedback
statements highlighting strategy attributions in their classes to support the weekly
intensive strategy lessons. In the Mathematics intervention program, the collaborating
class teachers taught the strategy lessons themselves using instructional and learning
materials prepared by the researchers. The focus of the Mathematics intervention program
was on strategies for solving word problems.
Achievement measures. End-of-year results in English, Mathematics, and Science
(for high school only) were obtained from the school for each student for each of the
three years. In schools where ability streaming was used, some within-school scaling was
carried out on the end-of-year results with the assistance of the head teachers concerned.
For each year level in each subject, the scores provided by the schools were converted to
z-scores (standardised across schools) for the statistical analyses.
Results and Discussion
Tables 1 and 2 presents the means and standard deviations of the achievement
measures and scores from the Causal Attributions (General) and Self-regulated Learning
Strategies (General) Scales. For the achievement measures, raw score means are reported
to facilitate interpretation.
It has been argued by Borkowski, Chan & Muthukrishna (in press) that to make
inferences as to the consequences of particular motivational orientations, it is not
sufficient to know students' tendency of making any one attribution for success or failure.
Instead we need to examine the relative pattern of a student's tendency to attribute success
and failure to the various internal or external, consistent or inconsistent, controllable or
uncontrollable causes. Latent variables were formed for primary and high school students'
attributional beliefs, using confirmatory factor analysis on the causal attributions variables
(see Figures 1-8). For the high school cohort, three latent variables were formed. The
first refers to an adaptive attributional beliefs construct "Belief in Personal Control over
Success" (PC), comprising tendency to attribute success to personal ability, effort and use
of strategies. The second, "Self-blame for Failure" (SB), involves a tendency to attribute
failure to lack of ability, insufficient effort, and not using strategies, in other words,
blaming one's own inadequacies for failure. The third is the "Learned Helplessness" (LH)
maladaptive attributional beliefs construct and consists of a tendency to attribute
success/failure to luck and failure to lack of ability. For the primary school cohort, only
the first two latent variables, "Belief in Personal Control over Success" and "Self-blame
for Failure", were observed in the data.
Apart from the attributional beliefs constructs, two other latent variables were
formed. One was "Strategy Knowledge and Use" (Str), consisting of the four knowledge
and use of general study and metamemory strategies subscale scores from the Self-
Regualted Learning Strategies (General) Scale. The other was "Achievement' (Ach),
consisting of English and Maths scores for the primary cohort, and English, Maths and
Science scores for the high school cohort.
Intercorrelations among Latent Variables
Pairwise correlations among latent variables for each of the three years in the
primary and high school cohorts are reported in Tables 3 and 4. Several key results can
be observed.
Firstly, the correlations between Personal Control over Success and Strategy
Knowledge & Use were the highest, ranging from 0.41 to 0.74 across the primary and
high school years for the intervention groups and 0.38 to 0.76 for the no intervention
groups. Further, a substantial increase (of about 0.2) in the magnitude of the correlation
coefficients across the three years was observed in the intervention group for both
cohorts, though that was not the case for the no intervention groups.
Secondly, results indicated moderate to low negative correlations between Self-
blame for Failure and Strategy Knowledge & Use in the first and second year (ranging
from -0.13 to -0.32) and virtually non-existent (near zero) in the third year in the
intervention group for both cohorts. For the no intervention groups, however, the
negative correlations between Self-blame and Strategies were consistently low in the
primary cohort (ranging from -0.07 to -0.11), and consistently moderate in the high
school cohort (-0.17 to -0.28).
Thirdly, correlations between Personal Control over Success and Achievement
were positive but relatively low for the Intervention Groups in both cohorts, except for
Year 9 (r=0.36 compared to 0.15 in Year 7). For the no intervention groups, the
correlations were moderate in both cohorts, except for Year 9 (r=0.20 compared to 0.35
in Year 7).
Fourthly, correlations between maladaptive attributional beliefs (Self-blame for
Failure in the primary cohort and Learned Helplessness in the high school cohort) and
Achievement were negative and moderate (ranging from -0.18 -0.48) for both groups in
both cohorts. Further, a decrease in the magnitude of the negative correlation coefficients
across the three years was indicated between maladaptive attributional beliefs and
achievement, again for both groups in both cohorts.
Lastly, correlations between Strategy Knowledge & Use and Achievement
increased in magnitude across the three years in both cohorts for the intervention groups
(from 0.11 and 0.06 to 0.18 and 0.28 for the primary and high school cohorts,
respectively). For the no intervention groups, the correlation increased from the first to
the second year, then dropped back to the first year level or lower in the third year.
To sum up, results indicated that Strategy Knowledge & Use and Achievement
were positively and moderately related to adaptive attributional beliefs (Personal Control
over Success), but negatively related to maladaptive attributional beliefs. Relationship
between Achievement and Strategy Knowledge & Use, however, was low. Further, the
intervention seems to have resulted in reducing the negative relationship of maladaptive
attributional beliefs with Achievement and with Strategy Knowledge & Use, but
increasing the positive relationship of adaptive attributional beliefs with Achievement and
with Strategy Knowledge & Use, as well as the positive relationship of Strategy
Knowledge & Use with Achievement.
Stability of Individual Differences in the Latent Variables across Time
To examine the stability of individual differences in the latent variables, estimates
of the correlation of each latent measure with itself assessed at consecutive occasions
were calculated. These correlations for both the intervention and no intervention groups
are presented in Table 5. Overall, the correlations are moderate to high, indicating that
individual differences in attributional beliefs, strategy knowledge and use and achievement
are quite stable over time. Individual differences in achievement, in particular, are
extremely high across Years 5 to 7 and 7 to 9 for both groups (r>0.88), apart from the
transitional period of the last year of primary school to first year of high school (r=0.75
and 0.73). Significant decrease in stability of individual differences during this transitional
period was also observed in attributional beliefs and strategy knowledge and use for the
intervention group.
Correlations between Latent Variables Assessed at Consecutive Occasions
The correlations between latent measures assessed at consecutive occasions are
presented in Table 6. These correlations examine the relationship of a prior measure (T1)
with a subsequent measure (T2). Hence the correlation of PC1 with STR2 for Years 5-6
refers to the correlation between Personal Control over Success in Year 5 and Strategy
Knowledge & Use in Year 6 while the correlation of STR1 with PC2 for Years 5-6 refers
to the correlation between Strategy Knowledge & Use in Year 5 and Personal Control
over Success in Year 6.
For the no intervention group, the associations between prior Personal Control
over Success and subsequent Strategy Knowledge & Use, and between prior
Strategy Knowledge & Use and subsequent Personal Control over Success are the
highest, with those observed for the high school cohort (r=0.45 to 0.64) higher than
those for the primary cohort (r=0.27 to 0.40). The associations between prior
Personal Control and subsequent Achievement are also higher for the high school cohort
(r=0.39 and 0.31) than for the primary cohort (r=0.26 and 0.11). However, the
associations between prior Achievement and subsequent Personal Control indicate the reverse
pattern, being higher for the primary cohort (r=0.36 and 0.47) than for the high school
cohort (r=0.23 and 0.21). The negative correlations between prior maladaptive attributional beliefs, Self-
Blame for Failure and Learned Helplessness, and subsequent Strategy Knowledge & Use,
and between prior Strategy and subsequent maladaptive attributional beliefs are also
higher for the high school cohort (r=-0.17 to -0.27) than for the primary cohort (r=-0.07
to -.18). The same pattern of higher association for high school than for primary school
was observed for the negative correlation between prior maladaptive attributional beliefs
and subsequent achievement and between prior achievment and subsequent maladaptive
attributional beliefs. Results suggest that the negative association of maladaptive
attributional beliefs with strategy knowledge and use and with achievement is greatly
increased in high school than in primary school.
For the intervention group in the high school cohort, the negative relationships of
Learned Helplessness with Strategy and with Achievement are much lower than those
observed for the no intervention group. In the primary school cohort, on the other hand,
the negative relationships of Self-blame for Failure with Strategy and with Achievment
are higher in the intervention group than in the no intervention group. However, the
relationship between prior Strategy and subsequent Achievement is higher in the
intervention group than in the no intervention group. While these correlations between
latent variables assessed at consecutive occasions provide useful indications of the trend in
causal relationships, they are not accurate estimates of the magnitude of
causal relations because other important causal factors have not been taken intoconsideration.
Causal Influences of the Latent Variables
Figures 1-8 depict the causal influences of prior attributional beliefs, strategy
knowledge & use, and achievement on these same constructs measured in the following
year for the intervention and no intervention groups in the primary and high school
cohorts. Included in each model are all prior and subsequent motivation, strategies and
achievement latent variables with their respective indicator variables. For each structural
equation model, three sets of paths were tested. One set involves the influences of
attributional beliefs on strategies and achievement, and those of strategies on achievement,
assessed in the same year. The second set involves the autoregressive effect of a latent
variable assessed in a previous year on the same latent variable in the following year. The
third set involves the causal relations of a prior latent variable on another assessed in the
following year. Note that in each model, all prior motivation, strategies and achievement
latent variables were allowed to be simultaneous causes of a subsequent latent variable.
Thus this third set of paths reflects the causal influence of a prior latent variable on
another subsequent latent variable that is independent of the other prior measures and with
the autoregressive effect of the subsequent latent variable partialled out. For example, the
path from Year 6 Strategies to Year 7 Achievement indicates the causal influence of
strategy knowledge and use in Year 6 on Year 7 achievement that is independent of the
influence of prior measures of achievement and motivation.
In Figures 1-8, only significant path coefficients are shown. Paths for
autoregressive effects (e.g., PC1 to PC2) are significant in all models but for clarity are
not depicted in the diagram. Further, autocorrelated errors for each pair of repeated
measures indicator variables (e.g., correlation of Year 5 Success-ability error with Year 6
Success-ability error) are included in the measurement model but for clarity are not
shown in the diagrams.
Primary school cohort. The resultant causal models for the no intervention group
are presented in Figures 1 and 2, while those for the intervention group appear in Figures
3 and 4. For the no intervention group, a similar pattern of relationships is observed for
both Years 5 to 6 and 6 to 7 time periods. Reciprocal causal influences of Belief in
Personal Control over Success with Achievement and with Strategy Knowledge & Use are
observed in both time periods. Further, Self-Blame for Failure was found to have a
negative influence on concurrent Achievement in the first year. Results suggest that for
primary students greater strategy knowledge and use enhance subsequent development of
belief in personal control over success, which further promotes strategy knowledge and
use. In addition, students with greater belief in personal control over success are likely to
achieve at a higher level, and higher levels of achievement allow the further development
of belief in personal control. On the other hand, those who tend to blame their own
inadequacies for failure are likely to achieve at a lower level. This observed pattern of
relationships among motivation, strategic learning and achievement provides empirical
support for theories of causal attributions (Weiner, 1984) and metacognitive development
(Borkowski, Chan, & Muthukrishner, in press; Borkowski & Muthukrishner, 1992).
For the intervention group, the pattern of relationships is different, particularly in
the second time period. The causal influences of Strategy Knowledge & Use and
Achievement in the first year on Belief in Personal Control in the second year was greatly
reduced and become nonsignificant. The concurrent influence of Belief in Personal
Control on Strategy Knowledge and Use, though, was greatly strengthened in the second
year. Further, Self-Blame for Failure was found to have a significant positive influence
on Strategy Knowledge & Use in the second year, and Year 6 Strategy Knowledge & Use
was found to have significant causal influence on Year 7 Achievement. These students
participated in Year 6 and/or 7 in an intervention program which aimed to promote
strategic learning in students through combining the teaching of cognitive and
metacognitive strategies with attempts at changing students' attributional beliefs. The
results suggest that the intervention program succeeded in reducing the influence of
previous strategy knowledge and achievement on adaptative attributional beliefs, while
strengthening the positive influence of adaptive attributional beliefs on strategy knowledge
and use. More importantly for these students, the strategies acquired in Year 6 lead to
higher achievement in Year 7. The finding that the intervention students' strategy
knowledge was significantly influenced by the self-blame construct appears odd.
However, the path coefficient is positive, and examination of the means on the self-blame
scores reveal an increase across Years 5 to 7. It appears that the intervention resulted in
greater tendency for students to attribute failure to insufficient effort, ineffective use of
strategies and lack of ability. Such attributional beliefs may also reflect greater
willingness to accept responsibility for failure, which may not necessarily imply a
maladaptive attributional belief, particularly if strategies are seen as a way of resolving
such problems.
High school cohort. For the no intervention group (Figures 5 and 6), a clear and
consistent pattern of the adaptive attributional beliefs, Belief in Personal Control over
Success, having a positive influence on Strategy Knowledge & Use and on Achievement
was observed, both concurrently and subsequently. Reciprocal causal influences of
Personal Control with Strategies are also indicated, Year 8 Personal Control having a
positive causal influence on Year 9 Strategies and Year 8 Strategies having a positive
causal influence on Year 9 Personal Control. On the other hand, the maladaptive
attributional beliefs, Learned Helplessness, consistently demonstrated a negative influence
on Achievement, again both concurrently and subsequently. Further, this negative causal
influence of Learned Helplessness with Achievement is reciprocal: higher learned
helplessness beliefs in Year 8 tend to lead to low achievement in Years 8 and 9, further,
low achievement in Year 8 leads to more entrenched learned helplessness beliefs. Finally,
a negative reciprocal causal influence of Strategies with Self-Blame for Failure and
Learned Helplessness was indicated. Greater strategy knowledge and use in Year 7
reduces self-blame and learned helplessness beliefs in Year 8, and reduced self-blame
beliefs is associated with greater strategy knowledge and use. Again, this observed pattern
of relationships among motivation, strategic learning and achievement provides empirical
support for theories of causal attributions (Weiner, 1984) and metacognitive development
(Borkowski, Chan, & Muthukrishner, in press; Borkowski & Muthukrishner, 1992).
For the intervention group (Figures 7 and 8), the negative reciprocal causal
influence between learned helplessness and achievement was greatly reduced and became
nonsignificant. Further, Year 7 Strategy Knowledge & Use was found to exert a
significant causal influence on Year 8 Achievement, and Achievement in Year 8 was
found to have a significant causal influence on Belief in Personal Control over Success in
Year 9. For the high school students, the intervention program seems to have broken the
vicious cycle of low achievement leading to learned helplessness beliefs, which leads to
more failures and subsequently more entrenched learned helplessness beliefs.
General Discussion
The longitudinal design of the study not only allowed the relationships of
attributional beliefs, strategic knowledge and achievement across primary and high school
years revealed in cross-sectional studies to be confirmed, but more importantly, allowed
reciprocal causal relationships among these constructs to be empirically tested. Two key
findings were observed from the results of the several analyses examining the
developmental patterns of relationships among attributional beliefs, strategic knowledge
and achievement.
Firstly, maladaptive attributional beliefs tend to have negative impacts on academic
achievement and in some instances on strategic knowledge and use. These maladaptive
attributional beliefs seem to take the form of attributing failures to one's own
inadequacies in the primary school years, but turn into learned helplessness beliefs in the
high school years. Further, the negative impact of learned helplessness beliefs on strategic
learning and on achievement increases across Years 7 to 9. Maladaptive learned
helplessness beliefs are not prominent in primary school students, but become more and
more entrenched in the high school years as a result of a vicious cycle of low
achievement leading to learned helplessness beliefs, which leads to more failures and
subsequently more entrenched learned helplessness beliefs.
Secondly, intervention designed to promote strategic learning in students through
combining the teaching of cognitive and metacognitive strategies for learning with
attempts at convincing students to attribute success and failure to effort and effective or
noneffective use of strategies, which are factors they have personal control over, can
succeed in breaking the vicious cycle of entrenched learned helplessness beliefs and their
negative impacts on strategic learning and academic achievement. More importantly,
enhanced beliefs in personal control over success and greater strategic knowledge and use
are likely to lead to subsequent higher achievement.
References
Borkowski, J.G. (1992). Metacognitive theory: A framework for teaching
literacy,
writing and maths skills. Journal of Learning Disabilities, 25(4),
253-257.
Borkowski, J.G., Carr, M., Rellinger, L. & Pressley, M. (1990).
Self-regulated
cognition: Interdependence of metacognition, attributions and
self-esteem. In B.J.
Jones & L.Idol(Eds.). Dimensions of thinking and cognitive
instruction (p.53-92).
Hillsdale:Lawrence Erlbaum.
Borkowski, J.G., Chan, L.K.S. & Muthukrishna, N. (in press). A
process-oriented model
of metacognition: Links between motivation and executive
functioning. In G.
Schraw & J. Impara (Eds.), Metacognitive assessment
(Buros-Nebraska Series on
Measurement and Testing). Lincoln, NE: Buros Institute of Mental
Measurements.
Borkowski, J. G. & Muthukrishna, N. (1992). Moving metacognition into
the classroom:
"Working models" and effective strategy teaching. In M. Pressley,
K. R. Harris
& J. T. Guthrie (Eds.), Promoting academic competence and literacy
in school.
San Diego, CA: Academic Press.
Borkowski, J.G., Weyhing, R.S. & Carr, M. (1988). Effects of
attributional retraining on
strategy-based reading comprehension in learning-disabled
students. Journal of
Educational Psychology, 80(1), 46-53.
Chan, L.K.S. (1991). Promoting strategy generalisation through
self-instructional training
in students with reading disabilities. Journal of Learning
Disabilities, 24(7), 427-
432.
Chan, L.K.S. (1994). Relationship of motivation, strategic learning and
achievement in
Grades 5, 7 and 9. Journal of Experimental Education, 62(4),
319-342.
Chan, L.K.S. (1996a). Motivational orientations and metacognitive
abilities of
intellectually gifted students. Gifted Child Quarterly, 40(4),
184-193.
Chan, L.K.S. (1996b). Combined strategy and attributional training for
seventh grade
average and poor readers. Journal of Research in Reading, 19(2),
111-127.
Chan, L.K.S. & Youlden, A.M. (1995). Motivational orientations,
strategic learning and
achievement in English in Years 5, 7 and 9 students: Preliminary
findings. Paper
presented at the AARE Annual Conference, Tasmania, November.
Clayton-Jones, L., Rodwell, K., Skehan, J., Archer, J., Chan, L.K.S., &
Moore, P.J.
(1992). Approaches to learning, attributions, goal orientations
and achievement: A
developmental perspective. Paper presented at the 1992 Australian
Association for
Research in Education Conference, Geelong, November.
Fairbairn, H., Moore, P.J., & Chan, L.K.S. (1994). Assessing approaches
to learning,
attributional beliefs and strategic learning in specific subject
domains: Scales
development. Paper presented at the Annual Australian Association
for Research
in Education Conference, Newcastle, November.
Joreskog, K. G. & Sorbom, D. (1989). LISREL 7: A guide to the program
and
applications (2nd ed.). Chicago: SPSS Inc.
Juel, C., Griffith, P.L., & Gough, P.B. (1986). The acquisition of
literacy: A
longitudinal study of children in first and second grade. Journal
of Educational
Psychology, 78, 243-255.
Moore, P.J., & Chan, L.K.S. (1995). Attributional beliefs and strategic
knowledge of
students in Years 5, 7 and 9: Comparison across subject domains.
Paper presented
at the AARE Annual Conference, Tasmania, November.
Paris, S.G. & Winograd, P. (1990). How metacognition can promote
learning and
instruction. In B.J. Jones & L.Idol(Eds.). Dimensions of thinking
and cognitive
instruction (p.15-51). Hillsdale:Lawrence Erlbaum.
Perfetti, C.A., Beck, I., Bell, L. & Hughes, C. (1987). Phonemic
knowledge and
learning to read are reciprocal: A longitudinal study of first
grade children.
Merrill-Palmer Quarterly, 33, 283-319.
Reid, M.K. & Borkowski, J.G. (1987). Causal attributions of hyperactive
children:
Implications for teaching strategies and self-control. Journal of
Educational
Psychology, 79(3), 296-307.
Rodwell, K., & Moore, P.J. (1994). Motivation, strategic learning and
achievement in
social studies in Years 5 to 9. Papaer presented at the AARE
Annual Conference,
Newcastle, November.
Stanovich, K.E., Cunningham, A.E. & Feeman, D.J. (1984). The relation
between early
reading acquisition and word decoding with and without context: A
longitudinal
study with first grade children. Journal of Educational
Psychology, 76(4), 668-
677.
Wagner, R.K. (1988). Causal relations between the development of
phonological
processing abilities and the acquisition of reading skills: A
meta-analysis. Merrill-
Palmer Quarterly, 34(3), 261-279.
Weiner, B. (1984). Principles for a theory of student motivation and
their application
within an attributional framework. In R.E. Ames & C. Ames (eds.),
Research on
motivation in education, vol. 1 (pp. 15-38). Orlando: Academic
Press.
Weiner, B. (1991). Metaphors in motivation and attribution. American
Psychologist, 46,
921-930.
Youlden, A.M. (1993). Motivation, self-regulated learning and academic
achievement in
Years 7 and 9 students. Unpublished Masters' thesis, University of
Newcastle.
Youlden, A.M. & Chan, L.K.S. (1994). Motivation, self-regulated
learning and
achievement in Years 7 and 9 students. Paper presented at the
Annual Australian
Association for Research in Education Conference, Newcastle,
November, 1994.
Youlden, A.M. & Chan, L.K.S. (1997). Developmental differences in
attributional beliefs
and strategic learning in English of students in Years 5 to 9.
Paper presented at
the VIIth European Conference on Developmental Psychology, Rennes,
France,
September.
Zimmerman, B. J. & Martinez-Pons, M. (1988). Consruct validation of a
strategy
model of student self-regulated learning. Journal of Educational
Psychology, 80
(3), 284-290.
Note: Tables 1 to 7 and Figures 1 to 8 can be obtained from:
Associate Professor Lorna K S Chan
Faculty of Education, The University of Newcastle
Fax: (61) (02) 4921 6895
Email: edksc@cc.newcastle.edu.au