This review has been accessed times since January 1, 2005
King, Patricia M. and Kitchener, Karen S. (1994).
Developing Reflective Judgment: Understanding and Promoting
Intellectual Growth and Critical Thinking in Adolescents and
Adults. San Francisco: Jossey-Bass Publishers
351 pp.
$37.95 ISBN: 1-55542-629-8w99
Reviewed by Jan van Aalst, Simon Fraser University
Steven Katz, OISE/University of Toronto
August 11, 1999
The intended audiences of this book are educators who
work with college students and who attempt to promote
critical thinking among high school and non-college
students, those who teach about college students in higher
education programs, faculty and administrators with
responsibility for the assessment of college outcomes, and
developmental psychologists who seek to understand the
nature of cognitive development (pp. xviii-xx). The book
discusses more than 15 years of theory building and research
on what King and Kitchener call the Reflective Judgment
Model (RJM)--a stage model of how assumptions about
knowledge and concepts of justification develop throughout
adolescence and (young) adulthood. Chapters 1 and 2
describe reflective judgment and previous work on it,
distinguishing it from critical thinking, and provide a
preliminary account of the model; a fuller description of
the epistemic assumptions and concepts of justification for
each stage of the model is given in chapter 3. In chapter 4
a set of criteria is proposed for assessing reflective
judgment; the authors argue that extant practice in the
assessment of critical thinking and postformal thinking does
not meet these criteria. In chapter 5 the Reflective
Judgment Interview (RJI) is described as an alternative
assessment instrument, and validation studies are discussed.
Chapter 6, which the authors view as the centerpiece of the
book, discusses research on reflective judgment. This
chapter was written in collaboration with Phillip K. Wood.
Chapter 7 examines the relationship of reflective judgment
to critical thinking, and chapter 8 to character
development. The final chapter describes some possible
educational strategies. The book has a number of extensive
appendices which provide further details on the RJI and the
studies discussed in the main text.
King and Kitchener begin their argument (after a brief
interview transcript excerpt) with a discussion of John
Dewey's writing on reflective thinking:
[Dewey] observed that true reflective thinking is
initiated only after there is recognition that a real
problem exists. Such real problems, he argued, cannot be
solved by logic alone. Rather, they are resolved when a
thinking person identifies a solution to the problem that
temporarily closes the situation. True reflective
thinking...is uncalled for in situations in which there
is no controversy or doubt, no concern about the current
understanding of an issue, or in which absolute,
preconceived assumptions dominate. (p. 6)
Thus the development of thinking King and Kitchener are
interested in has more to do with such issues as the ability
to recognize that there is a problem and that some problems
may not be stated with certainty, than with logical
argument. Problems that require reflective judgment--they
call these ill-structured problems--go beyond such
problems as "figuring out the circumference of a circle,"
"translating a set of instructions into a computer language,
or playing a game of chess," which, the authors claim, can
be solved "quite adequately" on the basis of mathematical
formulae, logic, or rules of play (pp. 6-7). The choice of
examples here is unfortunate. For example, any design
problem involves an evaluation of competing design
constraints which cannot be done on the basis of logic
alone; nor can one know with certainty with which move,
consistent with the current board configuration and the
rules of the game, a player's opponent will respond.
(Figuring out the circumference is a well-defined
problem, but does not seem to require formal thought.) The
authors seem to oversimplify problem solving. Nevertheless,
there are many problem solving situations in which the
uncertain nature of knowledge is not evident, such as items
on the Cornell Critical Thinking Test (CCTT) and the Watson-
Glaser Critical Thinking Appraisal (WGCTA). These tests,
they claim, invoke cognitive or metacognitive processes
rather than epistemic assumptions of the respondents. (p.
12).
Description of the Reflective Judgment Model
The model is based on Perry's (1970) work on reflective
thinking (Note 1), as well as works by a variety of
philosophers (e.g., Popper, Lakatos, Dewey), and has
undergone further development since the authors' first study
of reflective judgment (Kitchener & King, 1981). According
to the current model, there is a progression of seven
distinct sets of epistemic assumptions about knowledge and
how knowledge is acquired; each set has its own logical
coherency, and is called a stage. Each successive
stage is "posited to represent a more complex and effective
form of justification, providing more inclusive and better
integrated assumptions for evaluating and defending a point
of view. ...The more advanced sets allow greater
differentiation between ill-structured and well-structured
problems and allow more complex and complete data to be
integrated into a solution" (p. 13). The stages, they
claim, are traversed in an invariant order, but "homogeneity
associated with stagelike development decreases with age"
(p. 22), and people at any time operate in a range of
stages, depending on, for example, the problem context. The
stages can be mapped onto the seven levels of Fisher's (
1980) skill theory. In this theory, seven levels, marked by
qualitative differences in the skill structures an
individual can control, occur between ages 2-30. Further, a
person's best performance in a familiar domain improves
sharply when a new level is emerging. Optimal performance
requires that performance is induced and supported by the
environment (p. 34). During the plateau the new skill
structure is elaborated and applied in new areas. King and
Kitchener hypothesize that the assumptions associated with
each RJM stage develop concurrently with the corresponding
skill level in Fisher's theory.
We summarize (the authors' account of) the seven stages
of the RJM. The stages fall into three groups indicative of
pre-reflective thought (Stages 1-3), quasi-
reflective thought (Stages 4 and 5), and
reflective thought (Stages 6 and 7).
- Stage 1: Students use single, undifferentiated
categories; no justification concept is needed because
there is assumed to be an absolute correspondence between
what is believed to be true and what is true. Students
do not see discrepancies between two views or see that
two people disagree on an issue (pp. 48-50).
- Stage 2: There is a true reality that can be known
with certainty but is not known by everyone. (p. 51).
Certain knowledge is seen as the domain of authorities,
and those who disagree with authorities are wrong.
Defending one's point of view is not done to explain the
reasons for beliefs but rather to show (by stating them)
that one's own beliefs are right and those who believe
otherwise are wrong" (p. 52). Underlying the Stage 2
belief system is the ability to relate two or more
instances of something, but the individual "considers
knowing in relationship to concrete issues and not as an
abstraction" (p. 53). Knowledge is assumed to be
available directly through the senses or via authority
figures. Students also often try to align evidence with
their own views by distorting evidence. When they do
perceive differences, they assume these can be resolved
simply.
- Stage 3: Although in some areas even authorities may
not have the truth, at some point in the future there
will be knowledge from concrete data. In cases where
authorities do have answers, beliefs continue to be
justified on the basis of the word of an authority. This
stage involves further differentiation of stage 2
categories into simple concrete systems. Where
authorities do not know, people can believe what they
want (p. 55). Assumptions do not reflect an
understanding of evidence as an abstraction, a relational
concept (p. 56).
- Stage 4: This is the first of the quasi-reflective
stages. One cannot know with certainty, and there is the
emergence of knowledge as an abstraction, not limited to
concrete instances. However, knowledge and justification
remain poorly differentiated from each other. (p. 58).
Although students may acknowledge that opinions do not
form sufficient basis for developing an argument, "they
are not consistent in their use of evidence for this
purpose" (p. 58). Students are likely to be unable to
differentiate a theory from evidence for the theory, thus
"they cannot perform the necessary mental operations that
would allow them to evaluate the theory on its own
merits" (p. 60).
- Stage 5: While people may not know with certainty or
directly, what is known is always limited by the
perspective of the knower--a position sometimes referred
to as relativism. The underlying concept of this
stage is the ability to relate two abstractions. In this
stage, students can differentiate an event from the
interpretation of the event. However, knowledge remains
context-bound--the individual has not yet developed the
ability to relate several abstractions into an abstract
system. Students do recognize that there are alternate
theories and that some evidence does not support one
theory, but what is missing is "the ability to coordinate
the two in a well-reasoned argument" (p. 64).
Nevertheless, while students are unable to perform the
above coordination, they are "trying on one frame of
reference after another" (p. 66), and this provides the
basis for synthesis.
- Stage 6: Knowing is a process that requires action
on the part of the knower, and that knowledge must be
understood in relationship to context and evidence.
Students are now able to coordinate the subtle
similarities and differences of abstract relationships
into intangible systems; they make decisions on the basis
of compelling evidence rather than for idiosyncratic
reasons.
- Stage 7: While "reality is never a given,
interpretations of evidence and opinion can be
synthesized into epistemically justifiable conjectures
about the nature of the problem under consideration" (p.
70). People in Stage 7 take on the role of inquirers;
they are agents involved in constructing knowledge (p.
70); they recognize that their knowledge claims may later
be superseded by others. There also is an ability to
integrate several Stage 6 systems (knowledge and
justifications about scientific issues, social science
issues, etc.) into a general framework about knowing and
justification. This allows for a generalization of
assumptions and a clarity of judgment that were not
present at Stage 6. This stage goes beyond Perry's model--the
idea of being committed to a point-of-view does not
fit in this stage (p. 71).
One example may illustrate how the stages map onto the
Fisher levels. In Level 4, two or more representational
systems are related to produce a new way of thinking--single
abstractions. This is the first of the abstract levels, and
represents the ability to construct and operate on
intangible categories; Fisher claims this occurs early in
adolescence. In a single abstraction about knowledge, in
the RJM, a person can coordinate concrete observations that
different people draw different conclusions on the basis of
their own circumstances and biases (p. 32).
Evaluating Reflective Judgment
Chapter 4 turns to evaluating reflective judgment. The
authors present a set of seven "desirable features of a
measure of reflective thinking:" (a) Ill-structured tasks
should be the focus of the assessment task; (b) assessment
should elicit information about the rationale for a response
as well as its content; (c) thinking should be sampled
across a variety of issues; (d) the content should be
generally familiar to a wide range of individuals; (e) the
content should not constrain the instrument's use to
individuals in educational settings; (f) the reading level
should be such that it can be used with a wide range of
potential test-takers; and (g) if the measure is based on a
model that purports to describe the development of
reflective thinking, the theoretical model on which it is
based should be validated. As already mentioned, on the
authors' view tests of critical thinking such as the Watson-
Glaser test do not have the desired focus on ill-structured
tasks.
The Reflective Judgment Interview (RJI) discussed
in chapter 5 typically consists of four ill-structured
problems and a set of standard follow-up questions. Each
question focuses on at least one major concept of the
Reflective Judgment Model (i.e., the assumptions about
knowledge and concepts of justification). The four standard
problems cover a range of issues; each problem is defined by
two contradictory points of view and is designed to focus on
problems in the intellectual domain. An example is a
problem on the safety of chemical additives to foods:
There have been frequent reports about the relationship
between chemicals that are added to foods and the safety
of these foods. Some studies indicate that such
chemicals can cause cancer, making these foods unsafe to
eat. Other studies, however, show that chemical
additives are not harmful, and actually make the foods
containing them more safe to eat. (p. 101)
A fifth problem on nuclear waste has also been used. All
five of the standard problems, as well as several
discipline-based problems (in chemistry and psychology), and
instructions for conducting and scoring the interviews are
provided in an appendix.
The second part of chapter 5 addresses the validity and
reliability of the RJI. The authors insist that trained and
certified people must be used for conducting and scoring the
interviews. After a presentation of verbatim discussions of
one standard problem (two interviews), interrater
reliability and agreement, test-retest reliability, and
internal consistency (between the standard problems) are
examined. This part of the chapter provides an informative
account of the kinds of issues one must attend to in test
development; it should be of interest to graduate students.
King and Kitchener consider the scores of two raters to
be in agreement when they are discrepant by less than one
stage (p. 110). Most of the 32 studies that used the
standard RJI problems report (Pearson) correlations between
ratings given by two blind raters, as well as percentage
agreement; Cohen's kappa, which corrects percentage
agreement for chance agreement, is not reported. Almost 40%
of the studies reported an agreement of at least 87%, and
25% an agreement of at least 90%. The authors conclude that
trained raters can score RJIs consistently; from four
studies in which there was less than 70% agreement they
conclude that "some raters may be differentially evaluating
cues in the transcripts from adult students or that older
students may be responding in more diverse ways to the
interview questions" (p. 111). Test-retest reliability was
.71 for a three-month interval, and .87 for a two-week
interval (p. 112). For 14 studies that met the criteria of
using all four problems and reported (Cronbach's) alpha, the
median alphas were .79 and .85. A second measure of
consistency across problems used was the inter-problem
correlations (problem 1 with problem 3, etc.). These
correlations ranged from very low to very high, typically
falling around .40. Problem-Total correlations (scores for
any given problem with the average of the scores for the
other three problems) were typically .60 and varied little
from problem to problem. Thus, the authors conclude, the
problems are remarkably consistent.
Preliminary data on three domain-specific problems are
also discussed. Average scores on these were "almost
identical" to averages on the standard problems done by the
same respondents, leading the authors to conclude that the
RJI taps people's underlying assumptions about knowledge,
and not knowledge about the discipline (p. 118). A test of
the hypothesis based on Fisher's theory that spurts in
performance in cognitive skills occur between ages 14-15,
18-20, and between 23-25 is reported. Respondents were
presented with two prototypical responses to a problem for
each of Stages 2-7 and asked to summarize the statements in
their own words. The study (Kitchener, Lynch, Fisher, and
Wood, 1993) found evidence for spurts and for improved
scores due to practice, as well as a ceiling effect--
students younger than age 23, even with support and
practice, did not accurately paraphrase the statements that
reflected Stage 7 reasoning (p. 122).
Research on Reflective Judgment
We focus our review on chapter 6, describing the findings
reported in chapters 7 and 8 very briefly. Chapter 6 asks
six questions about reflective judgment (pp. 124-25):
- Does reflective judgment develop between late
adolescence and middle adulthood?
- How do high school, college, and graduate students
reason about ill-structured problems?
- Does student reasoning improve with additional
exposure to and involvement in higher education?
- Do adult learners differ from traditional-age
students in their reflective thinking?
- How does the reasoning of adults who have not
participated in higher education compare to the reasoning
of those who have earned a college degree?
- Are there gender or cross-cultural differences in
the development of reflective thinking?
The first part of the chapter deals with a 10-year
longitudinal study which traced the development of
reflective judgment of 80 individuals between 1977 and 1987;
interviews were conducted in 1977, 1979, 1983, and 1987. At
the first three testings, the four standard problems of the
RJI were used; the problem on nuclear waste management was
added at the fourth testing (p. 129). The authors are
careful to point out limitations of the studies that arise
from the fact that not all individuals were interviewed at
each of the testings (p. 131). They analyze three groups of
individuals: (a) 38 people who participated in all four
testings, (b) 13 who missed only one testing, and (c) 80 who
participated in any one testing. Consistent patterns of
increasing RJI mean scores are reported for all three groups
at each subsequent testing (Note 2); RJI mean scores of former high
school junior students showed the largest gains, 2.9 to 5.5.
Examining this gain in more detail, the authors conclude
that the greatest amount of growth--3.9 to 5.3--was over a
four-year period between 1979 and 1983, and that no other
group came close to this "spurt." The period in question
coincides with college attendance for 75% of the
participants, and is predicted by Fisher's theory; thus,
King and Kitchener conclude, the spurt may be due to a
combination of education and development (p. 136).
Also discussed is a study of the variability of the
scores over stages, that is, of the extent to which
individuals in the 10-year study were operating in a spread
of stages. The study (Wood, 1983) re-analyzed data from 15
studies, calculating the proportion of time each stage was
represented in an individual's rating across four standard
problems, referring to this as the 'stage utilization
score'. Wood performed a spline regression of these scores
against the overall reflective judgment score. As King and
Kitchener discuss, the analysis showed that the spread over
stages increases with the dominant stage, the stage
most frequently assigned at scoring. (For example, it is
smaller for scores with the highest percentage utilization
score at Stage 3 than for scores for which it is at Stage
6.) The authors point out that for Stages 2 and 7, the
slopes of the curves may be distorted by the fact that only
one adjacent stage is used (p. 140), which is another
example of the care the authors take in qualifying their
claims. The section concludes with a discussion of two
examples of verbatim RJI transcripts, which helps to
illustrate the various changes in RJI scores over time (p.
141-146).
To examine the influence of age on RJI scores, all data
in the 10-year study were pooled to calculate the modal (in
some cases bimodal) frequencies for each stage over 5-year
intervals. For Ages 16-20 the model stage was 3, for Ages
21-25 Stage 4, for Ages 26-30 Stage 5, for Ages 36-40 Stage
6, and above Age 40 Stages 6 and 7. There was a strong
relationship between age and the stage of development, but
the authors caution that the subjects were in a variety of
educational pursuits at the times of the testings (p. 150).
The second part of chapter 6 deals with data found across
six other studies that use the RJI, using 12 different
samples and 214 individuals. In all samples tested, scores
either stayed the same or increased with time; and with two
exceptions, the mean scores increased significantly for all
groups tested in 1-4 year time intervals. (The exceptions
were adult college freshmen and advanced doctoral students.)
Another finding across the studies is that the amount of
change over time is strongly related to the duration of time
between testings (p. 146). The authors examine not only
increases with age, but also "regressions," (decreases with
time). The conclusion from the analysis of reversals is
that "stability rather than development characterized the
thinking of some participants at some of the longitudinal
testings suggest[ing] that the rate of change across stages
varies across individuals" (p. 157).
Next, several cross-sectional studies are discussed--in
different regions and across 10 educational levels. The aim
of their analyses was to discover how individuals at
different age and educational levels think about and resolve
ill-structured problems, but as the authors point out, age
and educational level are often confounded--age brings with
it access to a greater range of educational opportunities.
(p. 160). The mean score for undergraduate students
(traditional-age and non-traditional) is 3.8; for graduate
students it is 4.8. Scores increase, but so does
variability. As with the 10-year study, scores are not
representative of US high schools--of 11 the samples
examined, 7 were from academically talented students (p.
162). An analysis of 20 cross-sectional studies of
traditional-age college students that used standard RJI
scoring techniques showed that only one sample of 44
averaged above 4.0. The authors comment that the prevalence
of stage 3 reasoning "means that like the high school
students, many of the college freshmen articulated the
belief that absolute truth is only temporarily inaccessible,
that knowing is limited to one's personal impressions about
the topic (uninformed by evidence), and that most if not all
problems are well structured (defined with a high degree of
certainty and completeness)" (p. 165). Nevertheless, it is
apparent that 2/3 of freshmen reasoned between stages 3 and
4, "suggesting that other freshmen were beginning to accept
the concept that uncertainty may be an ongoing
characteristic of the knowing process. Consequently, they
are beginning to accept the idea that some problems are
truly ill structured" (p. 166). In addition, although there
is only a small difference during college (3.6 for Freshmen
to 4.0 for seniors), the stage 4 reasoning prevalent among
the senior samples is "clearly more adequate and defensible"
than the reasoning it replaced.
An analysis of reflective judgment of 196 graduate
students distinguished between students in the first two
years of graduate studies and those beyond that (labeled
'advanced doctoral'). The authors observe that graduate
students were the first to consistently demonstrate the use
of Stage 5 assumptions on the RJI; moreover, there was a 0.7
stage increase from beginning to advanced doctoral levels,
and only at the latter level was there there evidence of the
consistent use of Stage 6 assumptions on the RJI (p. 172).
They point out that graduate admissions committees may
select students who have already demonstrated advanced
reflective thinking skills, and that it may be no accident
that students who scored high were in programs that required
a thesis (p. 173). One study showed disciplinary
differences (King, Wood, & Mines 1990): Graduate students in
social science disciplines scored significantly higher than
did their counterparts who were enrolled in the mathematical
sciences (p. 173).
King and Kitchener discuss gender effects for the 10-year
study as well for the studies in the second part of the
chapter. At the first two testings (1977 and 1979) of the
10-year study, no statistically significant effects were
found, although a Time by Gender effect "approached
significance" (p. 147); in the fourth testing, men scored
higher than women. Of the 14 studies discussed in the
second part of the chapter, 6 reported higher scores for men
and 6 reported no statistically significant effects. The
authors state that the available research that used the RJI
is inconclusive about gender effects.
In chapter 7, two studies are discussed that examined if
development of reflective judgment could be due to increases
in logical reasoning skills. From the first (Brabeck,
(1983a), King and Kitchener conclude that "while only those
scoring high in critical thinking scored above a Stage 4 in
reflective judgment, high WGCTA (Watson-Glaser Critical
Thinking Appraisal )scores did not uniformly lead to high
RJI scores." (p. 191). There also was more variability
among RJI scores for the high critical thinking group. The
second study (Brabeck & Wood, 1990) retested 25 high school
students from the original Brabeck (1983a) sample on both
the RJI and WGCTA. No differences were found on the WGCTA
scores, although differences were found for the RJI scores.
King and Kitchener suggest that "while the development of
critical thinking skills may not continue into the early
college years, development in reflective judgment clearly
does so" (p. 192).
The chapter continues with two longitudinal studies of
the relationship between reflective judgment and verbal
reasoning. In the first study, three groups of students
were tested at two-, six- and ten-year intervals, using the
RJI and Terman's (1973) Concept Mastery Test (CMT). There
was development over time on both measures, but the amount
of development differed by group, the youngest group (Age
16) showing the largest increase on both measures. Changes
in RJI scores could not be accounted for statistically by
changes in CMT scores. The second study led to a similar
conclusion. Finally, one study (King, 1977) is discussed
that examines the relationship between formal reasoning and
reflective judgment. King's sample included graduate
students, college juniors, and high school students who were
matched on scholastic aptitude. While there were
significant differences between the three groups' scores on
the RJI, there were no differences in the scores on the
formal operational tasks; the correlation between the two
measures was also "extremely low" (p. 201).
Chapter 8 examines the relationship between reflective
judgment and character development. Here, King and
Kitchener argue that moral questions are different from
other questions requiring reflective judgments:
Epistemological issues that would correspond to the moral
question of whether to discontinue life supports include
evaluating the prognosis for recovering and estimating
the long-term financial and emotional effects on the
family. Part of the ill-structured nature of moral
problems derives from different values or conceptions of
the good rather than from epistemological issues about
the nature of knowledge. (p. 206)
But still there is a structural similarity in development of
the two kinds of thinking. In a longitudinal study that was
concurrent with Kitchener and King's (1981), a moderate
correlation was found between RJI scores and the Defining
Issues Test (DIT; Rest, 1979). The data is summarized by
the claim that the development of reflective judgment is
necessary but not sufficient for the development of moral
thought, as measured by the DIT (p. 210). King and
Kitchener suggest that the structural similarity implies
that "there may be mechanisms underlying development in both
domains...that may affect how the individual operates on
both sets of tasks" (p. 207).
In the final chapter, King and Kitchener make several
suggestions based on the research they discuss that
instructors can use to help students question their
assumptions about knowledge (p. 223). They detail each of
the proposals in several exhibits (pp. 231-255). The
proposals range from tolerance for and awareness of the
epistemelogical assumptions students bring to learning
tasks, to proposals for acting on such knowledge. For
example, in proposal 2 the authors observe that students
differ in regard to their epistemic assumptions, and in
proposal 7 that students may view learning tasks differently
as a result. Naturally, the proposed teaching strategies
involve practice with considering ill-structured problems
from different viewpoints. These proposals repeat, at
another level, similar calls for taking students
preconceptions into account in teaching (e.g., Ausubel,
1968; Diver & Easley, 1978).
Discussion
We end with some observations about this book and the
research it discusses. First, it is an excellent account of
a large body of research. It is well written, and the
authors have clearly made much effort not over-generalize
from the studies. The authors conclude each chapter with a
succinct summary, which is useful for readers who need a
quick reference. Moreover, readers of the full text will
benefit from reading the summaries prior to immersion in the
chapter specifics, a practice that the authors and we
especially advocate for the detail-rich chapter 6.
Although King and Kitchener argue in chapter 2 that the
RJM is noticeably distinct from other models of
epistemological development, we suggest that such
differences lie in the details of the models, the
most obvious of which deal with terminology. In their
review of models of epistemological development in late
adolescence and adulthood, Hofer and Pintrich (1997) point
out the structural similarities which exist between the RJM
and the work of other theorists such as Baxter Magolda
(epistemological reflection), Belenky et al. (women's ways
of knowing), and Perry (intellectual and ethical
development). We reference this not to criticize King and
Kitchener for a lack of originality, but to validate what is
emerging from diverse arenas as a cogent and important
domain of human development. Indeed, Kuhn (1999) has
recently argued that the popular critical thinking movement
has much to gain by considering the development of
epistemological meta-knowing in relation to its subject
matter, a lacuna that King and Kitchener's work can help to
fill.
From the standpoint of applied educational research, the
RJM, like other recent models of epistemological
development, makes its greatest contribution in offering a
way of knowing that bridges the apparent gap between
absolutism and relativism. We refer here to stages 6 and 7,
to what the authors call "reflective judgment." Olson and
Katz (in press) document the ways in which practicing
teachers are challenged to both meet the demands of a fixed
school curriculum and the individual needs of children with
their varied interests, backgrounds, and understandings that
make up a class. Such bipolar concerns, they argue, are
problematic because the poles tend to collapse onto the
orthogonal epistemological positions of absolutism and
relativism respectively. Stage 7 offers a potential
solution to this pedagogical and epistemological paradox
because it purports that knowledge, although constructed,
can be constructed from multiple viewpoints, and that
insight gained from each perspective must be synthesized
into a knowledge system (see also diSessa, 1996; van Aalst &
Scardamalia, in preparation). Olson and Katz explain that
the "known" and the "knower" can be recognized and related
this way; they believe that is what Dewey (1902/66) had in
mind in The Child and the Curriculum.
King and Kitchener point out that the data they discuss
are drawn from a predominantly white population in the Mid-
West, with high school students matched to participants who
enrolled in doctoral programs on scholastic aptitude;
therefore, this population is not representative of American
college students. Further, in relation to generalizability,
the emphasis on knowledge revision and abstract knowledge in
Stages 6 and 7 is an artifact of Western culture, and may be
less prevalent in other cultures (Bidell & Fisher, 1992).
Although the authors report that average scores for three
discipline-based problems (in chemistry, business, and
psychology) have been "almost identical to the scores earned
by the same participants on the standard RJI problems" (p.
118), there is some suggestion from other sources, at least
in science and mathematics, that assumptions about knowledge
do vary with problem context and subject matter domain (Roth
& Roychoudhury, 1994; Schoenfeld, 1992; van Aalst & Key,
under review). We suggest that there is sufficient reason
to reconsider the emphasis on epistemological issues in
isolation from subject matter and problem solving contexts
that runs through King and Kitchener's work.
An important issue that King and Kitchener leave open is
the relevancy of reflective judgment to solving (ill-structured)
problems. Although the claim that reflective
judgment requires ill-structured problems is persuasive, it
must also be demonstrated that developing reflective
judgment benefits problem solving ability. Thus we would
like to see longitudinal studies which document people's
struggles with complex issues such as those used for the
standard problems over a period of time, studies that can
elaborate which aspects of reflective judgment enhance
problem solving ability--or the converse. It could be
argued that a single measure of reflective reasoning, such
as the RJI provides, confounds issues that are better
examined separately in connection with problem solving. For
example, knowledge as authority-based versus uncertain
knowledge is a separate issue from knowledge as isolated
pieces versus knowledge as a coherent system (Hammer, 1994).
Further, like all stage theorists, King and Kitchener are
faced with the problem of having to explain the mechanism(s)
by which individuals advance through the stages. Without a
clear understanding of such mechanisms, little progress can
be made in designing the interventions that the authors say
are necessary for "optimal performances" (p. 34). In
particular, the seemingly inviting effects of formal
education await empirical investigation in a way that will
untangle the maturation-education interaction. (This would
require studies that involve younger children). Only then
will the instructional recommendations that constitute the
substance of chapter 9 move beyond the realm of rhetoric.
Notes
- Perry observed a progression of nine points of view, which
he called positions (p. 36). These positions traced out the
evolution of students' thinking about knowledge, truth, and
values. Initially, students used "discrete, absolute, and
authority-based categories..." and "people did not perceive
legitimate conflicts about knowledge" (pp. 36-37). In
positions 3-4, which he called 'multiplicity', students began
to acknowledge multiple answers to complex questions. He
identified major changes at positions 5 and 6, which he called
'relativism', and hypothesized that students begin to see
knowledge as contextual and not necessarily true, leaving
students with the need to make a decision. Perry called
positions 7-9 'commitment to relativism.'
- One exception were former doctoral students, who remained
constant after 1989.
References
Ausubel, D. (1968) Educational psychology: A cognitive
view. New York: Holt, Rinehart and Winston.
Bidell, T.R., & Fisher, K.W. (1992). Beyond the stage
debate: Action, structure, and variability in Piagetian
theory and research. In R.J. Sternberg & C.A. Berg (Eds.),
Intellectual development (pp. 100-140). New York: Cambridge
University Press.
Brabeck, M. (1983). Critical thinking skills and
reflective judgment development: Refining the aims of higher
education. Journal of Applied Developmental Psychology, 4
(1), 23-34.
Brabeck, M., & Wood, P.K. (1990). Cross-sectional and
longitudinal evidence for differences between well-
structured and ill-structured problem solving abilities. In
M.L. Commons, C. Armon, L. Kohlberg, F.A. Richards, T.A.
Grotzer, & J.D. Sinnott (Eds.), Adult development 2: Models
and methods in the study of adolescent and adult thought.
New York: Praeger.
Dewey, J. (1902/1966). The child and the curriculum.
Chicago: University of Chicago Press.
DiSessa, A. (1996). What "just plain folk" know about
physics. In D. Olson & N. Torance (Eds.), Handbook of human
development and education (pp. 709-730). Cambridge:
Blackwell.
Driver, R., & Easley, J. (1978). Pupils and paradigms: a
review of literature related to concept development in
adolescent science students. Studies in Science Education,
5, 61-84.
Fisher, K.W. (1980). A theory of cognitive development:
The control and construction of hierarchies of skills.
Psychological Review, 87 (6), 477-531.
Hammer, D. (1994). Epistemological beliefs in
introductory physics. Cognition and Instruction, 12, 151-
183.
Hofer, B., & Pintrich, P. (1997). The development of
epistemological theories: Beliefs about knowledge and
knowing and their relation to learning. Review of
Educational Research, 67 (1), 88-140.
King, P.M. (1977). The development of reflective
judgment and formal operational thinking in adolescents and
young adults. Dissertation Abstracts International, 38,
7233A.
King, P.M., Wood, P.K., & Mines, R.A. (1990). Critical
thinking among American college and graduate students.
Review of Higher Education, 13 (3), 167-186.
Kitchener, K.S., & King, P.M. (1981). Reflective
judgment: Concepts of justifications and their relationship
to age and education. Journal of Applied Developmental
Psychology, 2 (2), 89-116.
Kitchener, K.S., Lynch, C.L., Fisher, K.W., & Wood, P.K.
(1993). Developmental range of reflective judgment: The
effect of contextual support and practice on developmental
stage. Developmental Psychology, 29 (5), 893-906.
Kuhn, D. (1999). A developmental model of critical
thinking. Educational Researcher, 28 (2), 16-25, 46.
Olson, D., & Katz, S. (in press). The fourth folk
pedagogy. In B. Torff & R. Sternberg (Eds.), Understanding
and teaching the intuitive mind. Hillsdale, NJ: Erlbaum.
Perry, W.G., Jr. (1970). Forms of intellectual and
ethical development in the college years:a scheme. Troy,
MO: Holt, Rinehart & Winston.
Rest, J. (1979). Development in judging moral issues.
Minneapolis: University of Minnesota Press.
Roth, W.-M., & Roychoudhury, A. (1994). Physics
students' epistemologies and views about knowing and
learning. Journal of Research in Science Teaching, 31 (1),
5-30.
Schoenfeld, A. (1992). Learning to think mathematically:
Problem solving, metacognition and sense-making in
mathematics. In D.A. Grouws (Ed.), Handbook of research on
mathematics teaching and learning (pp. 334-370). New York:
Macmillan.
Terman, L.M. (1973). Concept mastery test manual. New
York: Psychological Corporation.
Van Aalst, J., & Key, T. (under review). Physics
expectations of pre-professional students. Submitted to
American Journal of Physics.
Van Aalst, J., & Scardamalia, M. (in preparation). A
comparison of radical constructivism and knowledge building.
To be submitted to Science & Education.
Wood, P.K. (1983). Inquiring systems and problem
structure: Implications for cognitive development. Human
Development, 26, 249-265.
About the Reviewers
Jan van Aalst, Simon Fraser University, Burnaby, B.C.,
CANADA, V5A 1S6
Steven Katz, OISE/University of Toronto, Toronto, ON,
CANADA, M5S 1V6
|