H. P. A. Boshuizen, R. Bromme, & H. Gruber (Eds.), Professional learning: Gaps and transitions on the
way from novice to expert, 1-16.
©2004 Kluwer Academic Publishers. Printed in the Netherlands.
Empirical findings on expert teachers and teacher development
1. RESEARCH ON THE NATURE OF EXPERTISE
Research on teacher expertise can be advocated from at least two frameworks. The
first is the concern and dissatisfaction with the schools and quality of school
education. Policy makers, particularly in the United States, have often complained
about the quality of teacher education and called for changes in the training and
certification of teachers (e.g. The Holmes Group, 1986; U.S. Department of
Education, 2002). These statements have often been criticized by educational
researchers, who argue that much of the criticism is politically motivated and is not
supported by the research evidence (Berliner & Biddle, 1995; Darling-Hammond &
Youngs, 2002). From this point of view it is important to understand, for instance,
what kind of knowledge successful teachers have and how teacher education can
promote the acquisition of this knowledge.
Another motive is the theoretical interest in the phenomenon of expertise. It is
important to understand expertise to be able to promote its development in the work
environments and further education.
In the following I will focus on the latter point of view. I start by reviewing
empirical studies on expertise and particularly teacher expertise and summarize the
results into a few propositions. My second purpose is to review research theorizing
the phenomenon of teacher expertise. I will also discuss the future research on
Although the focus in the chapter is in expertise it is important to notice that
there are several other approaches to studying teacher knowledge (see e.g. Munby &
al., 2001). For instance, Shulman (1986) has described experienced teachers’
knowledge base by dividing it into several content categories and forms of knowing.
That knowledge is represented in the mental structures of one’s knowledge base.
Bromme and Tillema (1995, p. 263) refer to the activity-oriented knowledge of
practitioners as professional knowledge. This knowledge includes representations of
theoretical rules and statements but also images, metaphors, and attitudes for
2 EERO ROPO
successful practice. They also include teachers’ beliefs and orientations into the
professional knowledge base (Bromme & Tillema, 1995).
Different approaches to studying professional knowledge are needed for many
reasons. One of those is that the domain areas differ. For instance, the most recent
research on expertise has suggested that expertise between domain areas may differ
in such amounts that even the research paradigms are hard to transfer from one area
to another (e.g. Boshuizen & al., 1995, Bromme & Tillema, 1995).
Studies on expertise have made an important contribution to our thinking about
professional knowledge and skills. The beginning of research on the nature of
expertise can be traced to the 1960s, when de Groot (1966), among others, studied
the playing skills of chess masters and their information processing during a game.
Those studies proved that chess masters did not function as predicted, but rather
used a high level of intuition. Computer programmes developed at the time,
although designed to emulate the way people played chess, functioned quite
differently. De Groot’s studies (1978) showed that world class chess players
accessed the best moves during their initial perception of the game board situation
rather than searching for possible moves and analysing their consequences for the
situation. It was later estimated that master level chess players have at least 100,000
different game situations in store and these can be recalled rapidly and intuitively in
the game situation (Chase & Simon, 1973). According to these researchers the
development of expertise requires more than ten years of experience and full-time
Since the late 1960s and early 1970s, the nature of expertise has been studied in
many professional fields. In addition to chess players, for example, physicists,
radiologists, computer programmers (Chi, Glaser, & Farr, 1988) and social scientists
(Voss, Greene, Post, & Penner, 1983) have been studied. The first studies were
conducted in the domain areas that can be called knowledge-rich; rich referring, for
instance, to well defined and developed theories in the domain area. According to
these studies, novice physicists, for instance, who possessed the necessary
knowledge to solve a physics problem, tried to solve it backwards from the question,
whereas expert physicists retrieved a solution plan the other way round, as part of
the normal comprehension process. Chi, Feltovich, and Glaser (1981) showed that
the physics experts had more knowledge than the novices and that it was better
organised in their memory. The experts could therefore represent the problems in
terms of relevant physics theories, whereas the novices’ representations were mainly
based on salient surface features of the problems.
On the basis of the above studies and many others, it is possible to draw at least
two conclusions on how research has approached expertise (Ericsson & Smith,
1991). First, research efforts have been focused on observing outstanding
performance in relatively standardized conditions. In these studies, expertise has
usually been defined as the ability to successfully execute problem-solving tasks
related to one's professional field.
The second conclusion is that the theoretical concern of the studies has been with
the analyses and descriptions of cognitive processes related to expert performance.
One of the results of this approach has been the development of methods not only
for measuring but also for eliciting expert knowledge and describing its structure
and organization in specific domain areas (Cooke, 1994; Hoffman, 1992, 1995). In
TEACHING EXPERTISE 3
many cases the processes studied are the same as those described in the theories of
skill acquisition (Anderson, 1983, 1993).
The efforts of the first two decades in finding out the "secrets" of expertise have
led to fairly clear conclusions on the nature of experts’ knowledge bases. Experts
seem to possess better-organized and more specific knowledge structures that they
can access almost intuitively in the problem situations. Nevertheless, it is not clear
what the mechanisms of the development of this knowledge are and how they are
related, for instance, to the amount and quality of experience or the innate
characteristics of a person. Typically experts show superior performance in the
domain areas in which there are well-developed theories that support reasoning,
such as medicine, physics, chess, bridge and so on (Ericsson & Lehmann, 1996). In
the earlier expertise studies performance and the structure and content of one’s
knowledge base are shown to be closely related.
Experience has also been linked to expertise in several studies. For instance,
Leprohan and Patel (1995) found that the number of years of experience correlates
positively with nurses’ performance in screening emergency calls for medical help.
The amount of auditors’ experience was found to relate poorly to the accuracy of
their performance in the study by Bonner and Pennington (1991). Similar results
were obtained in the studies by Rosson (1985) and Doane, Pellegrino, and Klatzky
(1990). All these studies have shown that experts do not always demonstrate
superior performance in the activities representing their domain area tasks.
Sometimes the reason seems to be in the nature of expertise, which is highly
specialized and restricted to a narrow domain area.
2. THE NATURE OF TEACHER EXPERTISE
In contrast to the domain areas described above, in which the problems given to
subjects are often specific and isolated from the social or cultural context, teaching
is very different. The effectiveness of individual performance can be relatively
easily described in such domain areas as physics or chess, but the phenomenon
seems to be different in the school classroom. Instead of being good at a specific and
well-defined problem, expert teachers have to be performers in the problems situated
in socially and culturally complex contexts.
The first problem is to find the expert teachers for the studies. One of the criteria
that have been used is student learning in academic subjects. Expert teachers are
those whose students perform well in the achievement tests. For instance, Leinhardt
has used teachers' long-term success in their profession as a criterion for expertise,
measuring success by the students' results in academic achievement tests (Leinhardt
& Greeno, 1986).
However, there are problems in this kind of criterion because we do not know
enough about the functions teachers have in student learning and in their test results.
Teachers are supposed to make a difference in student learning, but this relation is
not simple. Teaching as a domain field can be described as a knowledge-lean
domain area because of the complexity of the instructional context and the lack of
comprehensive theories explaining it. The theories of teaching and instruction are
vague and the school context is too complex to be described with only those
4 EERO ROPO
theories. This all makes it difficult to explain how a teacher’s expertise influences
student achievement or to assume that the students whose achievements are
excellent must have expert teachers.
However, teachers have important roles in organizing studying, directing the
conversation and affecting the students’ lives in many other ways in schools.
Therefore, we may say that teacher expertise in such partly intuitive and artistic,
partly learned interventions, can be assumed to make a difference in student learning
If we adopt such a conception of the nature of teacher expertise, the perspective
on teacher knowledge and performance becomes very complicated. Artistic and
intuitive decisions may be based on knowledge, but their origins may be more in the
situation. It is, therefore, important to approach teacher expertise from a more
contextual point of view than in some other fields. One example of this kind of
theorizing is the stage model first proposed by Dreyfus and Dreyfus (1986).
According to this the development of expertise proceeds in five stages: Novice,
advanced beginner, competent performer, proficient performer, and expert (Dreyfus
& Dreyfus, 1986). Berliner (1988) has described a novice, advanced beginner and a
competent performer as rational and the proficient performer as intuitive in decision-
making, for instance, in a problem situation. Berliner (1988) describes experts as
irrational. Experts possess intuition with which they can create an overall
representation of a situation. Their actions are flexible and fit the situation at hand.
Experts seem to know what to do without necessarily being able to describe to an
outsider the grounds for their action or how their thinking proceeds. Thinking seems
to be rational in the sense that there is logic behind it. However, the rules of
decision-making are hidden and intuitive, or tacit if we want, even for an expert
himself/herself. A person is an expert because he or she seems to understand the
requirements of the situation better and is able to fit his/her own decision, actions
and interaction into the context.
A popular research strategy in describing expertise in domain fields similar to
teaching has been to identify typical differences between the experts and novices.
Those studies have produced a list of factors or characteristics that seem to
differentiate expert teachers from novices. I summarize those results in the following
2.1. Expertise develops in only a narrow field of knowledge
and the knowledge base is tightly bound to a context
Expertise can only emerge after long experience. It has been estimated that chess
masters have spent between 10,000 and 20,000 hours at the game, expert
radiologists during their active career have studied about 100,000 X-rays and expert
teachers have taught at least 10,000 contact hours, prior to which they have spent at
least 15,000 hours in the classrooms as students (Berliner, 1990). Thus it does not
seem possible for an individual to obtain thorough knowledge in many different
areas. Expertise cannot be particularly wide-ranging, either.
Berliner (1990) describes the situation-specific nature of knowledge in a study in
which a group of expert and novice teachers as well as beginners were asked to
TEACHING EXPERTISE 5
teach a 30-minute unit to high-school students. The subjects were given half-an-hour
to plan the contents of the unit, after which they taught the lesson. The lesson was
videotaped and afterwards, while watching the video, the subjects were asked to
describe what they had been thinking in the original situation. To the surprise of the
researchers, the expert teachers were very emotional about the situation. None of the
experts liked the task they had been given, though they seemed to possess the skills
to handle the situation more effectively than the beginners or novices. One of the
expert subjects dropped out of the study, one began to cry while watching the lesson,
and all the rest expressed hostility towards the researchers. The reason for their
hostility seemed to be that the experts did not feel they did well in a situation where
they had been removed from their own class and from familiar situations. While
they had been given 30 minutes to prepare, some wanted three hours or even a
whole week to prepare the lesson. In addition, they were unhappy about not
personally knowing the students in the class.
The studies on the nature of teacher knowledge show that teachers seem to
develop situation-specific action patterns for classroom instruction. This may
indicate that they have a related, situation specific knowledge base, too. I take two
examples from the research findings. The first deals with teachers’ questioning
during the lessons and the second teachers’ methods of scaffolding student learning.
Questioning has been studied extensively in the past. For instance, Evertson,
Emmer, and Brophy (1980) have found that more successful teachers ask more
questions. Their average number of questions during a successful mathematics
period was 24 whereas less successful teachers asked only 8.5 questions per period
on average. In a related study Ropo (1990a) found that experienced mathematics
teachers asked 32.6 questions and less experienced student teachers 17.1 questions
per period on average. For experienced teachers of English as a second language the
average number of questions was 32.2 and for less experienced novice teachers 24.3
questions per period (Ropo, 1991). The number of questions seems to be
systematically greater for experienced teachers in typical face-to-face instruction.
Scaffolding student learning is a metaphor derived from building construction
(Wood, Bruner, & Ross, 1976). Scaffolds provide support, they extend the range of
a worker and they allow the worker to accomplish tasks not otherwise possible.
Palincsar and Brown (1984) applied this concept in their study of reciprocal
teaching. The extent to which children need scaffolding seems to vary. Teachers also
differ in the extent to which they provide scaffolds for their students. Ropo found
that experienced mathematics teachers applied 8.2 scaffolds per period whereas
novice teachers had only 4.3 scaffolds per period on average. There was also a
qualitative difference between the scaffolds provided. Experienced math teachers
divided the original questions into more simple partial questions more often than
novices, if a student could not answer the question correctly. Experienced teachers
also applied more a strategy in which a student was given a series of sub questions
or specific questions that aimed at leading his/her thinking to the ideas or thoughts
needed to answer the original question correctly (Ropo, 1990a).
Experienced teachers of English applied 6.9 scaffolds per period and novices 5.7
on average. This difference is not as large as that found in mathematics teaching.
However, it is similar to mathematics teachers. One qualitative difference between
the scaffolds of experienced and inexperienced teachers was found in the use of
6 EERO ROPO
examples. Experienced teachers gave examples 2.5 times per period whereas
novices did not use this at all (Ropo, 1991). The samples in the above studies were
small and the results cannot be generalized in a larger population. However, they
indicate that expert teachers also differ from novice teachers in their teaching
2.2. Experts have automatic ways of reacting to frequently recurring situations
Expertise studies have focused on describing the development of cognitive skills,
particularly, subjects’ automatic patterns or responses in specific task situations.
Glaser (1987), for instance, notes that the quick and automatic comprehension of
written text typical of skilful readers frees some of their working memory capacity
for processing other aspects of the situation. This same phenomenon is well known
in the area of motor skills. Processing capacity increases because the verbal
mediation in the performance of a task disappears and the procedure becomes more
and more automated and rapid (Anderson, 1985).
In teaching, the automatisation of instructional actions allows teachers to direct
their attention elsewhere, thus enabling them to better manage the teaching period as
a whole. One example of the automatisation of teachers’ skills can be found in
Leinhardt and Greeno's study (1986) in which they compared the ways experts and
novices started a mathematics lesson in elementary school. The results showed that
on average an expert teacher took about a third less time to start a lesson than a
novice. Second, during the start of a lesson an expert was able to observe the activity
of the students, find out who had done their homework, and assess who would need
help later during the lesson. Third, as a result of the automatic routines that both
they and their students had learned, experts maintained a fluid control of their
In the same study novices’ lessons could be described nearly the opposite.
Novices were not in full control of the progress of the lessons. They had difficulties
in getting the students to be attentive and in ascertaining who had done their
homework. The questions novices asked about homework were not as clear as those
of the experts, and this led the novices to judge the difficulty of the homework
assignments incorrectly (Leinhardt & Greeno, 1986).
An interesting example assumed to be related to the level of automaticity was
found in the language usage of experienced and novice history teachers. In a study
of Finnish teachers of history it was found that experienced teachers did not use their
native accent during the lessons but used the standard, written language expressions
and accent while novices having the same background did not switch their language
to standard usage during the history lessons. Both groups used their native accent
during the interviews (Ropo, 1992). This result may be interpreted to show that the
automaticity of the lesson routines allowed the experienced teachers to monitor their
language while novices had to direct most of their information processing capacity
to controlling the flow of instruction.
2.3. Compared to novices, experts are more sensitive to individual students
in class situations and the characteristics of task situations
TEACHING EXPERTISE 7
In the study of problem solving it has been found that experts are more able than
novices to take into account the specific characteristics of a given context and the
limitations inherent in a specific problem (Chi, Glaser, & Rees, 1982; Glaser, 1987).
Housner and Griffey (1985) studied this issue by comparing experienced and novice
physical education teachers. The subjects of the study were given the task of
planning and implementing a relatively short teaching unit. In the planning stage,
teachers in both groups asked similar numbers of questions concerning areas such as
numbers of students, their age and gender distribution. However, the experts asked
more questions than the novices about the abilities, experience and background of
the students as well as about the features of the space and equipment available. Five
of the eight experienced teachers even wanted to see the space they were to teach in
before they started planning, while none of the novices asked to do so. Later when
they were doing the teaching, the experienced teachers adapted their teaching
according to the situation more than the novices, thereby deviating from their initial
Another interesting difference between expert and novice teachers has been
found in the teachers’ knowledge of their students. For instance, Carter, Sabers,
Cushing, Pinnegar, and Berliner (1987) compared expert, novice and postulant
teachers’ processing and use of information about their students in a simulation of
taking over a class. Postulant teachers were people working in business or industry
and having an interest in teaching. However, they had no formal teacher education.
The study showed that expert teachers seemed to have deeper knowledge of the
students and classroom problems than novices or postulants. Experts made richer
analyses of the students’ earlier experiences than the other groups. They also seemed
to rely more on their own information gathering about the students than the two
other groups (Carter et al., 1987).
In a related study I compared experienced (at least 10 years of experience) and
novice (2-3 years of experience) elementary school teachers’ knowledge of their
own students (Ropo, 1990b). The teachers were asked to describe four randomly
selected students from their class. The results showed that overall experienced
teachers gave longer descriptions than novices. Experienced teachers’ average
protocols were 559 words per student compared to the novices’ 414 word
descriptions per student. Typically both groups described almost 30 different
characteristics per student. Qualitative analyses showed that experienced teachers
seemed to know more than novices about the past or current family events of their
students. They also made more connections between the student’s family
background and his/her school behaviour or problems. Experienced teachers had
more explanations for the origins or reasons for students’ performance in different
school subjects. However, the novices seemed to know more about the students’
hobbies outside school than the experienced teachers. The experienced teachers
discussed only the school-related hobbies whereas the novices listed at least three
other hobbies for each student. There were also differences in the time span of
describing students’ performance or difficulties in the school. The time span was
longer for experienced teachers who seemed to use the past as the basis for
predicting the future success of the students at school.
The conclusion of the study was that both teacher groups seemed to acquire
knowledge of their students. This acquisition seemed to follow two principles. The
8 EERO ROPO
first principle may be stated as "Collect and store information relevant in helping
students to learn and adapt to life in school". The second principle was "Orient to
individuality and individual features". The results indicated that experienced
teachers seemed to be more advanced in applying those principles than novices
2.4. Expert teachers are faster and more accurate in their observations than novices
This characteristic has been observed in studies of chess masters and experienced
radiologists. The representations that experts make of the situation may not be
judged as correct or incorrect. Even experts differ in their explanations or subjective
theories of a specific situation. Rapid recognition and interpretation of situations is
based on the quantity and quality of knowledge stored in one's memory. Another
prerequisite for fast interpretation is that knowledge structures are organized in a
way that enables rapid recognition without the need for extended processing. Rapid
recognition of situations is useful, since it reduces the processing needed in various
situations, thereby freeing an individual’s processing capacity for measures required
by the situation.
It has been found that expert teachers are typically quicker than novices at
recognizing specific task situations and interpreting them. In instructional
conversations experienced teachers typically follow the students’ learning more
carefully than novices and make focused efforts to correct misunderstandings with
specific scaffolds (Ropo, 1990a, 1991). Experienced teachers also have more
accurate plans for lessons and they even seem to be better at following the flow of
time during a typical 50-minute period (Ropo, 1992).
2.5. Experts take longer to represent a problem to themselves, but they end up
with a better representation of it
Results obtained by analysing problem-solving processes of physicists and social
scientists have shown that experts spend more time forming a representation of a
problem than novices (Chi et al., 1981; Voss et al., 1983). Hanninen (1983)
observed similar differences in problem-solving between expert and novice teachers.
In his study subjects were presented with a problem concerning the teaching of
gifted students. Hanninen measured the time it took for the subjects to start writing
the solution. The novices spent an average of 2.6 minutes whereas experienced
expert teachers of gifted students took an average of 9.8 minutes before they began
to write out their solutions. Experienced teachers of normal students took an average
of 3 minutes to do the same task (Berliner, 1990). From these results we can deduce
that a person's level of experience correlates with the amount of time it takes him/her
to represent and solve a problem. The more an individual knows, the better s/he is
able to take into account the complexity of a situation in his/her representation of the
problem. With physicists and social scientists this was manifested through
incorporating theory into problem-solving (Chi et al., 1981; Voss et al., 1983). With
expert teachers solutions related to situations in classrooms included more lengthy
TEACHING EXPERTISE 9
and thorough analyses of situational factors or more substantial and more logical
arguments for the solutions selected.
2.6. Compared to novices expert teachers’ knowledge is wider concerning
the levels of abstraction and more hierarchically organized
Differences have been found in the ways experts and novices organize their
knowledge. These results suggest that experts are able to organize their knowledge
into more hierarchical levels than novices. In a study by Chi et al. (1981), for
example, expert physicists represented a problem involving an inclined plane by
including the basic laws of physics, which the novices (beginner physics students)
did not mention at all in their representations of the problem. Voss et al. (1983)
found similar differences of hierarchical representation in the ways social scientists
formulated and solved problems.
In my own studies experienced teachers were found to categorize the
instructional goals differently from novice student teachers. Experienced teachers
grouped the objectives hierarchically making a difference between the school levels
(e.g. elementary and middle school), grade levels (e.g. grade levels 7, 8 and 9), and
the generality of the goals and objectives. For instance, one expert expressed her
overall goal for mathematics education that she wants to show students how
beautiful mathematics is. Expert teachers also divided the overall goal into more
specific goals for each grade level. They also seemed to have individual goals and
teaching objectives for particular students in their classes. Novices typically
described their instructional goals at the level of individual lessons without having
the same kind of hierarchies of objectives (Ropo, 1987).
3. A PROTOTYPE VIEW OF TEACHER EXPERTISE
The above propositions and the studies behind them indicate that expert and novice
teachers’ groups differ a lot. The studies have also shown that in addition to the
differences between the groups there is variance within the groups (Ropo, 1990a,
1991). Experts are not similar in every respect and do not constitute a typical
category in the traditional sense of the concept. However, experts seem to be similar
in many respects. This kind of research evidence raises a question on the nature of
expertise. Can it be described and defined by the above type of propositions based
on the lists of differences or do we need different types of theories of expertise?
We have already described the so-called Dreyfus and Dreyfus (1986) model in
theorizing the nature of expertise and its development. Sternberg and Horvath
(1995) have also addressed the question of the nature of expertise and proposed a
model that they call a prototype view. The purpose of this model is to explain the
within group variance among experts. The prototype view serves as middle ground
between a definitional and ad hoc description of teacher expertise. With the
definitional description authors refer to conceptions in which expertise is defined
restrictively in terms of certain characteristics (e.g. reflective practice, or the teacher
as a researcher). The ad hoc fashion refers to the lists of characteristics
differentiating experts and novices (Sternberg & Horvath, 1995).
10 EERO ROPO
In the prototypical view the main idea is to postulate a central or “prototypical”
category member that serves as a summary representation of the category (Rosch,
1978; Sternberg & Horvath, 1995). According to this view members of a category
may resemble the prototype member to differing extents in different features. If this
were true, then two members of a category would not necessarily be similar in a
given respect, even if they belong to a category because of overall similarity in other
respects. The second aspect of the prototype view is the differential weighting of
features in the computation of overall similarity to the prototype (Sternberg &
Horvath, 1995). The third characteristic of the model is that the features making up
the category may be correlated, which means that they may occur together in a
category member at a level greater than chance.
Sternberg and Horvath (1995) use their model to sketch an outline of features
important for a prototype expert teacher. They use the same literature base we have
referred to in this article to identify the features that differentiate experts from
novices. The first feature Sternberg and Horvath (1995) choose is knowledge.
Experts bring more knowledge to problem situations and as a result solve them more
effectively. The second feature is efficiency of problem-solving. Experts are faster
and more efficient in their problem-solving than novices. The third feature is insight.
This has been included in the prototype model because experts seem to be more
likely to arrive at novel and appropriate solutions to problems within their domains
One implication of the prototype view is that it seems to respect the naturally
fuzzy nature of expertise found in empirical studies. Experts are different from each
other, although they are at the same time similar. One teacher may have wide
knowledge of subject matter, another a lot of pedagogical knowledge on teaching of
a subject matter, and a third is insightful about students. All may be categorized as
expert teachers. Sternberg and Horvath (1995) argue that the prototype view
broadens the picture of teacher expertise without making every experienced teacher
Another implication the authors mention for the model is that it makes it possible
to describe an expert with a smaller set of features than other theorizations offer.
Bereiter and Scardamalia (1993) give an example of this by saying that an expert
can be defined as one who works on the leading edge of his or her knowledge and
skill. This means that a real expert complicates even simple problems to the edge
whereas a nonexpert seeks to reduce the problem to fit available methods into it
(Sternberg & Horvath, 1995).
The last implication of the model refers to the perception of expertise in the
social context. If we define teacher expertise as a broad and rather fuzzy prototype
then this may enlarge the view people have about an expert teacher. In particular
policy makers who may have a restricted picture of teacher expertise may find that
the prototype model broadens their own view significantly. This may have
implications for ways of evaluating teachers or for the recruitment and training of
Overall, Sternberg and Horvath have formulated an interesting proposition that
helps in understanding the nature of teacher expertise. It is related to the earlier
Dreyfus model in the fuzziness of the category of expert teacher. However,
TEACHING EXPERTISE 11
Sternberg and Horvath take more detailed account of the later empirical research
than the Dreyfus model.
4. DEVELOPING EXPERTISE IN NATURAL CONTEXTS
So far we have reviewed research focusing on the typical characteristics of expert
teachers and the nature of the concept. We have found that there are experts and
novices who differ from each other in many crucial ways. In addition to knowing
this, it is also important to understand the processes of acquiring expertise.
The literature on the acquisition of expertise can be divided into at least three
rather separate perspectives. First, the acquisition of expertise may be viewed from
the standpoint of individual giftedness, intelligence, or exceptional abilities that
develop through experience or interaction between heredity and experience with the
environment. The second perspective is the so-called cognitive view that emphasizes
the role of acquired knowledge (both declarative and procedural) in the process of
developing expertise. The third framework in the current literature deals with the
social theory of learning in which the acquisition of knowledge and expertise is
typically seen as a kind of side effect of gaining membership of a social network.
The last two frameworks both emphasize the role of experience in the process of
acquiring expertise. We may argue that one of the key issues in understanding the
acquisition of expertise is to understand better the functions experience has in the
process. Studies addressing expert performance in open and ill-defined tasks have
shown that the nature of subjects’ experience explains the performance better than
the amount or length of experience (Sonnentag, 1995; Waltz, Elam, & Curtis, 1993).
Bereiter and Scardamalia (1993) have suggested that the central determinant of
high-level expertise is the subject’s ability to surpass his or her previous level of
knowledge and competence. Consequently, we may ask if the experts are those who
have had qualitatively exceptional experiences or if their experiences and individual
characteristics, such as giftedness, have interacted in an appropriate manner
resulting in the development of expertise. I will discuss this notion in the following.
4.1. Expertise and giftedness
The concept of expertise has started to attract growing attention among researchers
of intelligence and giftedness. For instance, Sternberg (2001) has made a proposition
in his recent article that relates expertise and giftedness in an interesting way. He
asks if giftedness could be seen as developing expertise. Sternberg’s argument is
that by the concept of developing expertise we can integrate two theories of
giftedness –static and dynamic. The static conception states that intelligence is a
relatively stable entity. Although certain kinds of intelligence areas may increase or
decrease with age, rank orders remain fairly stable over time (Sternberg, 2001). An
alternative, dynamic, view is that giftedness is to be found within a zone of proximal
development. This means that it is an ability itself ‘to advance from abilities that are
ready to be developed to those that are developed’ (Sternberg, 2001, p. 159).
From the expertise research point of view we may turn Sternberg’s idea the other
way round and ask if expertise could be conceptualised as developing giftedness. If
12 EERO ROPO
so, the abilities and giftedness underlying expertise would be seen as developing
from novice level to full blown expertise. Expertise develops because of "the
ongoing process of the acquisition and consolidation of a set of skills needed for a
high level of mastery in one or more domains of life performance" (Sternberg, 2001,
Sternberg’s idea of the close relation between abilities and expertise is an
interesting one. Although the existence of this kind of relation has not been denied,
the discrepancy between the two research traditions has been a paradigmatic one.
Typically expertise researchers have emphasized the role of knowledge in the
acquisition of expertise and left such concepts as abilities and gifts to researchers
interested in the nature of intelligence. Theorizing on giftedness and intelligence has
most often emphasized the static view of human performance in which the more or
less inherited abilities have been seen to be in the main role.
However, it is possible also to view expert performance from the perspective of
exceptional skills and abilities. Particularly, if our goal is to find out why some
individuals develop into experts and others do not, it may be fruitful to analyse the
individual characteristics interacting with the environmental factors. Sternberg’s
idea of seeing giftedness as developing expertise may narrow the gap between the
research on expertise and on intelligence. Experts may have had skills and abilities
that have ‘reacted’ well to the practice in a constructive environment. If the research
on intelligence and giftedness adopts Sternberg’s idea, this may lead to advances in
the research on expertise as well.
4.2. Expertise and knowledge
The cognitive approach to expertise research has dominated the field since its
beginning. It hypothesizes that exceptional performance is due to well organized
knowledge that experts can access rapidly in a problem situation. Empirical
evidence supports this view. How this knowledge is acquired is an important
question. Cognitive research has typically looked for explanations for exceptional
performance either in the information processing skills or in the contents of an
expert’s knowledge base. The cognitive view typically assumes the existence of
knowledge structures in memory.
In the current research on learning the so-called situative, sociohistoric, or social
view has gained popularity (Greeno, Collins, & Resnick, 1996). This view
emphasizes the notion that learning is inherently a social process. Learning is a way
of becoming a member of and participant in the culture and social networks and
acquisition of expertise is part of that process (Lave & Wenger, 1991; Wenger,
1998). Therefore, expertise is not as much as the cognitive view assumes, an
individual characteristic of one’s knowledge structures, but an outcome of being a
member of the social and communicational networks of individuals and groups.
Expertise from this point of view is a kind of side effect of acquiring membership of
and legitimate rights in a social network. This perspective views knowledge
differently from cognitive framework. Knowledge is not a static schematic structure,
stored in one’s head, but rather a way of relating and participating in the immediate
TEACHING EXPERTISE 13
social networks around oneself (Agnew, Ford, & Hayes, 1997; Brown & Duguid,
1994; Wenger, 1998).
The concept that is widely applied among social theorists of learning is identity.
Identity has had at least two different meanings in the studies. First, it has been
applied to refer to the membership aspect of the social network. It is assumed that in
the process of becoming a group member a person’s growing identity as a legitimate
member having more expertise and power supports the acquisition of knowledge.
Second, a person’s identity as an active and responsible learner is also assumed to
support the acquisition of knowledge (Greeno et al., 1996).
It is assumed that identity processes and knowledge construction are related,
although we are only at the beginning of empirical research on this issue. There is,
however, interesting literature on the issue that can direct future studies. For
sociologists identity is a concept related to one’s relation to the historical, cultural
and social environment. In the changing world an individual identity is "a reflexive
project" that is under continuous change and development (Giddens, 1991).
Although sociologists have typically discussed the development of identity and self
in a broad context such as the global post-modern society, there is also literature that
may offer more concrete frameworks for empirical studies on the relations of
identity and expertise. For instance, van Langenhove and Harré (1999) have
proposed the so-called positioning theory, which argues that the human self
develops by taking positions in specific contexts. A position is related to a concept
of role and it is conversational in nature. A person takes a position and changes it as
the conversation develops (van Langenhove & Harré, 1999) From this perspective
expertise might develop if a person makes deliberate decisions of taking a
developing expert’s position in a specific situation. For instance, a child might take a
position on becoming a mathematician at school. This position will lead to personal
identity and the acquisition of mathematical knowledge if it persists long enough.
However, situational positioning can only persist if the environment supports the
positioning. The will and motivation to surpass one’s level of earlier performance
may be supposed to come from the processes related to the construction of personal
identity (Bereiter & Scardamalia, 1993). At the moment we have very little
empirical research on this issue. The existing empirical research suggests that
contextual factors are important in developing expertise and that it may have
different roles at different stages of acquisition process (Eteläpelto & Collin, 2001,
this volume; Schmidt & Boshuizen, 1993).
In this final section I discuss a few conclusions emerging from the past theorizing on
teacher expertise. First, I argue that teacher expertise is an important research area
that should be studied in the future, too. Although empirical research on teaching
expertise has decreased during the last decade these studies have important
implications in understanding the development of excellence. Being a pedagogical
expert is at the same time similar and different from expertise in other domain fields.
Pedagogical expertise can be regarded as a domain area having its own specialties in
the same sense as engineering or medicine. Expert teachers have knowledge, skills,
14 EERO ROPO
and social networks that are typical for them alone. The core of school teachers’
expertise, for instance, is in a person’s performance in a classroom with a group of
students teaching a specific and complex mixture of values, knowledge, and skills.
Teachers as experts are professionals whose work is accomplished in a social
situation in which they are typically alone with their students. The concept of an
expert teacher and expertise in teaching is relevant for the current political
discussion on the quality of education or teacher education. For instance, the recent
results of the OECD organised PISA studies on comparing student performance in
the OECD countries has shown that good teaching makes a difference in the
students’ performance in the school (PISA 2000). It is therefore necessary that we
continue research on the nature and development of teaching expertise.
The second conclusion of the research on teacher expertise (or pedagogical
expertise if you like) is that it is necessary to study expertise in authentic contexts.
Educational practice benefits most of the studies describing the characteristics of
expert teachers in relation to the social context of their authentic work or duties.
Studies on expertise in other fields indicate that expertise develops as long as the
individuals are exposed to situations in which they have to overcome the restrictions
of their earlier knowledge. The work context has, consequently, a major role in
directing or affecting the person’s development in expertise. This notion is not only
important for our conceptions of teacher expertise but our conceptions of teachers’
work and working conditions. We need to ask if the changes taking place in the
work and its context support the continuous development of teacher expertise.
Let me refer to a related discussion in the area of school education. According to
Apple and Jungk (1990), many factors indicate that the nature of teachers' work has
in the last decades taken a turn for the worse from the viewpoint of professional
development. According to him, teachers’ work has become more one-sided and less
rich, teachers' abilities to manage their work as a whole have declined, and the
intensity and quantity of work has increased. This has resulted in increased stress
among the teachers. At the same time there are developments that emphasize the
accountability of teachers’ work assessed by simple measurements of student
The last implication deals with the nature of theorizing behind the empirical
studies on teacher expertise. It seems fruitful to adopt a multidisciplinary framework
in the research on expertise. Our discussion has already shown that many of the
novel ideas in the current research are multidisciplinary, coming from various
directions, such as sociology, philosophy, psychology as well as education. In the
future research we have a lot of new ideas to explore. For instance, autobiographical
theory and the processes related, studied in literature, philosophy, psychology, and
sociology has not yet influenced on the analyses of expertise and professional
Agnew, N. M., Ford, K. M., & Hayes, P. J. (1997). Expertise in context: Personally constructed, socially
selected, and reality relevant? In P. Feltovich, K. M. Ford, & R. F. Hoffman (Eds.), Expertise in
context: human and machine (pp. 219-244). Menlo Park: MIT Press.
Anderson, J. R. (1983). The architecture of cognition. Cambridge: Harvard University Press.
TEACHING EXPERTISE 15
Anderson, J. R. (1985). Cognitive psychology and its implications. New York: Freeman.
Anderson, J. R. (1993). Problem solving and learning. American Psychologist, 48, 35-44.
Apple, M. W., & Jungk, S. (1990). "You don't have to be a teacher to teach this unit." Teaching,
technology, and gender in the classroom. American Educational Research Journal, 27, 227-251.
Bereiter, C., & Scardamalia, M. (1993). Surpassing ourselves – An inquiry into the nature and
implications of expertise. Chicago: Open Court.
Berliner, D. C. (1988). Implications of studies of expertise in pedagogy for teacher education and
evaluation. In New directions for teacher assessment. Proceedings of the 1988 ETS Invitational
Berliner, D. C. (1990, September). Characteristics of experts in the pedagogical domain. Paper presented
at the International Symposium "Research on Effective and Responsible Teaching". University of
Fribourg, Fribourg, Switzerland.
Berliner, D. C., & Biddle, B. J. (1995). The manufactured crisis. Reading: Addison-Wesley.
Bonner, S. E., & Pennington, N. (1991). Cognitive processes and knowledge as determinants of auditor
expertise. Journal of Accounting Literature, 10, 1-50.
Boshuizen, H. P. A., Schmidt, H.G., & Custers, E. J. F. M., & Van De Wiel, M. W. (1995). Knowledge
development and restructuring in the domain of medicine: The role of theory and practice. Learning
and Instruction, 5, (4), 269-289.
Bromme, R., & Tillema, H. (1995). Fusing experience and theory. Learning and Instruction, 5, (4), 261-
Brown, J. D., & Duguid, P. (1994). Patrolling the border: A reply. Human – Computer Interaction, 9,
Carter, K., Sabers, D., Cushing, K., Pinnegar, S., & Berliner, D. C. (1987). Processing and using
information about students: A study of expert, novice, and postulant teachers. Teaching & Teacher
Education, 3, 147-157.
Chase, W. G., & Simon, H. A. (1973). The mind's eye in chess. In W. G. Chase (Ed.), Visual information
processing (pp. 215-281). New York: Academic Press.
Chi, M. T. H., Feltovich, P. J., & Glaser, R. (1981). Categorization and representation of physics
problems by experts and novices. Cognitive Science, 5, 121-152.
Chi, M. T. H., Glaser, R., & Farr, M. J. (Eds.). (1988). The nature of expertise. Hillsdale: Erlbaum.
Chi, M. T. H., Glaser, R., & Rees, E. (1982). Expertise in problem solving. In R. Sternberg (Ed.),
Advances in the psychology of human intelligence (pp. 1-75). Hillsdale: Erlbaum.
Cooke, N. J. (1994). Varieties of knowledge elicitation techniques. International Journal of Human –
Computer Studies, 41, 801-849.
Darling-Hammond, L., & Youngs, P. (2002). Defining “highly qualified teachers”. What does
“scientifically-based research” actually tell us? Education Researcher, 31 (9), 13-25.
De Groot, A. D. (1966). Perception and memory versus thought: Some old ideas and recent findings. In
B. Kleinmuntz (Ed.), Problem solving: Research, method and theory (pp. 19-50). New York: Wiley.
De Groot, A. D. (1978). Thought and choice in chess. The Hague: Mouton.
Doane, S. M., Pellegrino, J. W., & Klatzky, R. L. (1990) Expertise in a computer operating system:
Conceptualization and performance. Human – Computer Interaction 5, 267-304.
Dreyfus, H., & Dreyfus, S. (1986) Mind over machine. Oxford: Blackwell.
Ericsson, K. A, & Lehmann, A. C. (1996). Expert and exceptional performance: Evidence of maximal
adaptation to task constraints. Annual Review of Psychology, 47, 273-305.
Ericsson K. A, & Smith, J. (1991). Prospects and limits of the empirical study of expertise: An
introduction. In K. A. Ericsson & J. Smith (Eds.), Toward a general theory of expertise: Prospects
and limits (pp. 1-38). Cambridge: Cambridge University Press.
Eteläpelto, A., & Collin, K. (2001, August/September). From individual cognition to communities of
practice: Theoretical underpinnings in analysing professional learning and expertise. Paper presented
at the 9th Biennial Conference of the European Association for Research on Learning and Instruction
(EARLI), Fribourg, Switzerland.
Evertson, L., Emmer, E., & Brophy, J. (1980). Predictors of effective teaching in junior high mathematics
classrooms. Journal for Research in Mathematics Education, 11, 167-178.
Gage, N. L. (1978). The scientific basis of the art of teaching. New York: Teachers College Press.
Giddens, A. (1991). Modernity and self-identity: Self and society in the late modern age. Stanford:
Stanford University Press.
Glaser, R. (1987). Thoughts on expertise. In C. Schooler & W. Schaie (Eds.), Cognitive functioning and
social structure over the life course (pp. 81-94). Norwood: Ablex.
16 EERO ROPO
Greeno, J. G, Collins, A. M., & Resnick, L. B. (1996). Cognition and learning. In D. C. Berliner & R. C.
Calfee (Eds.), Handbook of educational psychology (pp. 15-46). New York: MacMillan.
Hanninen, G. (1983). Do experts exist in gifted education? Unpublished manuscript, University of
Arizona, College of Education.
Hoffman, R. R. (Ed.). (1992). The psychology of expertise. Cognitive research and empirical AI. New
Hoffman, R. R. (1995). Eliciting knowledge from experts: A methodological analysis. Organizational
Behavior and Human Decision Processes, 62, 129-158.
Housner, L. D., & Griffey, D. C. (1985). Teacher cognition: Differences in planning and interactive
decision making between experienced and inexperienced teachers. Research Quarterly for Exercise
and Sport, 56, 44-53.
Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge:
Cambridge University Press.
Leinhardt, G. (1988). Situated knowledge and expertise in teaching. In J. Calderhead (Ed.), Teachers'
professional learning (pp. 146-168). London: Falmer.
Leinhardt, G., & Greeno, J. G. (1986). The cognitive skill of teaching. Journal of Educational
Psychology, 78, 75–95.
Leprohan, J., & Patel, V. L. (1995). Decision making strategies for telephone triage in emergency medical
services. Journal of Medical Decision Making 15, 240-253.
Munby, H., Russell, T., & Martin, A. K. (2001). Teachers’ knowledge of how it develops. In V.
Richardson (Ed.) Handbook of research on teaching. (pp. 877-904). Washington, DC: AERA.
Palincsar, A. S., & Brown, A. L. (1984). Reciprocal teaching of comprehension-fostering and
comprehension-monitoring activities. Cognition and Instruction 1, 117-175.
PISA 2000. The OECD Programme for international student assessment. Retrieved November 24, 2003
Ropo, E. (1987, April). Teachers’ conceptions of teaching and teaching behavior: Some differences
between expert and novice teachers. Paper presented at the Annual Meeting of the American
Educational Research Association, Washington.
Ropo, E. (1990a). Teachers' questions: Some differences between experienced and novice teachers. In H.
Mandl, E. De Corte, S. N. Bennett, & H. F. Friedrich (Eds.), Learning and instruction. European
research in an international context (pp. 113-128). Oxford: Pergamon.
Ropo, E. (1990b, September). What teachers know about their students: Some differences between
experienced and novice elementary school teachers. Paper presented at the Conference of Effective
and Responsible Teaching, Fribourg, Switzerland.
Ropo, E. (1991). Expert and novice English teaching: Differences in the lessons of experienced and
novice English teachers. In M. Carretero, M. Pope, P. R.-J. Simons, & J. I. Pozo (Eds.), Learning and
instruction. European research in an international context (pp. 539-560), Oxford: Pergamon.
Ropo, E. (1992, June). Expert and novice history teaching: Eight differences between experienced and
beginning history teachers. Paper presented at the First European Conference of Educational
Research, Enschede, The Netherlands.
Rosch, E. (1978). Principles of categorization. In E. Rosch & B. Lloyd (Eds.), Cognition and
categorization (pp. 27-48). Hillsdale: Erlbaum.
Rosson, M. B. (1985). The role of experience in editing. Proceedings of INTERACT 84 IFIP Conference
on Human – Computer Interaction (pp. 45-50). New York: Elsevier.
Schmidt, H. G., & Boshuizen, H. P. A. (1993). On acquiring expertise in medicine. Educational
Psychology Review, 5, 205-221.
Shulman, L. S. (1986). Those who understand: Knowledge growth in teaching. Educational Researcher,
15 (2), 4-14.
Sonnentag, S. (1995). Excellent software professionals: Experience, work activities, and perceptions by
peers. Behavior and Information Technology, 14, 589-599.
Sternberg, R. J. (2001). Giftedness as developing expertise: A theory of the interface between high
abilities and achieved excellence. High Ability Studies, 12, 159-179.
Sternberg, R. J., & Horvath, J. A. (1995). A prototype view of expert teaching. Educational Researcher,
24 (6), 9-17.
The Holmes Group (1986). Tomorrow’s teachers: A report of the Holmes Group. East Lansing: The
U.S. Department of Education (2002). Meeting the highly qualified teachers challenge: The Secretary’s
annual report on teacher quality. Washington: U.S. Department of Education, Office of
Postsecondary Education, Office of Policy, Planning, and Innovation.
TEACHING EXPERTISE 17
van Langenhove, L., & Harré, R. (1999). Introducing positioning theory. In R. Harré & L. van
Langenhove (Eds.), Positioning theory: Moral contexts of intentional action (pp. 14-31). Oxford:
Voss, J. F., Greene, T. R., Post, T. A., & Penner, B. C. (1983). Problem-solving skill in the social
sciences. In G. H. Bower (Ed.), The psychology of learning and motivation: Advances in research
theory (pp. 165-213). New York: Academic Press.
Waltz, D., Elam, J., & Curtis, B. (1993). Inside a software design team. Communication of the ACM, 36,
Wenger, E. (1998). Communities of practice: Learning, meaning and identity. Cambridge University
Wood, D. J., Bruner, J. S., & Ross, G. (1976). The role of tutoring in problem solving, Journal of Child
Psychology and Psychiatry, 17, 89-100.