Research: Science & Education

Concept Maps in Chemistry Education

Alberto Regis and Pier Giorgio Albertazzi
Istituto Tecnico Industriale “Q. Sella” – Via Rosselli 2, 13051 Biella (BI) Italy

Ezio Roletto*
Università di Torino, Dipartimento di Chimica Analitica, via Pietro Giuria 5, 10125 Torino, Italy

During the first half of this century, ideas about the To be successful in learning, students have to take
nature of scientific knowledge radically changed owing possession of knowledge actively, by seeking explicit, con-
to the work of epistemologists and science historians ceptual linkages between new concepts and those they
such as Bachelard, Koyré, Cassirer, Popper, Kuhn, already possess. This process of elaborating personal,
Lakatos, Feyerabend, Laudan, and Putnam. At the be- meaningful knowledge takes place by restructuring the
ginning of this century, epistemologists held an empiri- already existent conceptual frameworks.
cist/positivist conception of science. This latter was con- The concept map (CM) is a tool, based upon the cog-
ceived as a realistic description of the world “as it is”, a nitive psychological theory of constructing meaning, de-
body of established knowledge obtained by uncovering veloped by Novak and Gowin (4) as a convenient and
scientific principles (concepts, laws, and theories) “hid- concise representation of the learner’s concept/proposi-
den” in nature. Scientific knowledge was considered the tional framework of a domain-specific knowledge. The
result of inductive inferences, starting with simple, un- concepts with their linking relationships would be “vis-
prejudiced observations, the secure base from which gen- ible” in a CM as concept labels and verbal connectives,
eralizations may be drawn, leading infallibly to concep- illustrating the organization of the concepts in the
tual explanations. learner’s cognitive structure. It would then be possible,
Science is at the present conceived as a human ac- at least partly, to follow the restructuring and the evo-
tivity, a “fabrication” of scientists, elaborating “models” lution of the cognitive structure by comparing succes-
for interpreting the empirical world and for inventing sive CMs elaborated by the student himself at different
new experiments (1). According to contemporary philo- stages of the teaching/learning process of a given topic.
sophical views, scientific principles do not find their CMs could so reveal:
source in the facts, but they are invented by scientists
to give significance to the facts. Science is not the result • the concepts already present in a student’s mind
of inductive inference, but a hypothetical knowledge fab- (initial concepts);
ricated by human beings in order to understand the • the conceptual linkages between the concepts
world and put some order in it. (context);
These revolutionary changes in the conception of the • the evolution that takes place as a consequence
nature of scientific knowledge (in the field of epistemol- of teaching/learning activities (conceptual
ogy) have been accompanied by radical changes in the change).
conception of how learning occurs (in the domain of edu-
cational psychology): the dominant view is no more the This is the hypothesis on which we based our use of
behavioral psychology but the cognitive one. Learners CMs, being fully aware that CMs can not give a com-
are actively engaged in constructing knowledge: the ac- plete view of the mental structure of a student.
quisition of new knowledge has to be firmly anchored to We report here on our experience with the students
existing concepts, and conceptual frameworks play a key (16–18 years old) enrolled in the final three years of the
role in the acquisition, retention, application of new con- chemistry specialization in a technical school. CMs were
ceptual knowledge and in the problem-solving exercises used as vehicles for visualizing the students’ knowledge
of the school laboratory (2). Epistemology and educa- structures and for documenting and exploring changes
tional philosophy have then a common ground. Science in these structures resulting from learning.
is fabricated by scientists, in order to understand the
world and to make predictions on natural and artificial
phenomena, moving from the scientific principles already Training of Students
defined by the scientific community. Scientific learning
The first class sections are devoted to training the
is constructed by students starting from their “initial”
students in the concept mapping technique by introduc-
conceptions of a subject matter.
ing them to operational definitions of terms applied to
More than twenty years of research on students’ al-
CMs: concept, concept label, context, linking relationship,
ternative frameworks leads to the conclusion that teach-
proposition, cross-link. Some maps are constructed on
ers have to take them into account if they want to help
non-chemical subjects, taken from common sense knowl-
learners to acquire meaningful scientific knowledge, so
edge. Acetate transparencies of students’ constructed
proving the soundness of Ausubel’s fundamental as-
CMs served as models for discussion. The extent of prac-
sumption of cognitive learning:
tice needed for students to acquire proficiency in the con-
The most important single factor influencing struction of CMs depends on several factors, even psy-
learning is what the learners already know. chological ones. On the average, 4 to 6 normal class ses-
Ascertain this and teach accordingly (3). sions (45 minutes each) are required by students to learn
how to construct CMs.

*Corresponding author.

1084 Journal of Chemical Education • Vol. 73 No. 11 November 1996

 Research: Science & Education

How to Make Concept Labels Available to Students by the same student, are shown. The concept label elec-
tron, which is linked only to current and atom labels in
Three different ways of assigning concept labels are the first map, has four other linking relationships in the
used, leaving the students the choice of the linking re- second one.
lationships. The changes in the links between the labels battery,
current and electrolysis are also interesting. In the sec-
Known Terms CM ond map, it is the battery that generates the electrical
At the first stage, students are invited to produce current and not electrolysis. In one class, after teaching,
what we call a known terms CM. In this case: 19 of 24 CMs linked electrolysis with battery wrongly:
the linking relationship explained that battery worked
• a fixed number of concept labels are assigned; by means of electrolysis. This was interesting since:
• each student constructs his own map by using • electrolysis had never been mentioned in our
only the terms given, choosing the linking rela- lessons, and we had deliberately inserted this
tionships and the concept structure he consid- concept label, which is dealt with in other
ers most suitable. courses (physics, for example);
For example, third-year students who had already • the students had come from different classes
followed a first course on general chemistry were asked with different chemistry teachers.
to develop a CM on the topic “oxidation–reduction” us-
ing the following concept labels: atom, battery, (electri- The very high frequency of the wrong connection
cal) current, electrical conductor, electrolysis, electron, suggested that there must have been some event that
ion, nucleus, oxidation, reduction. Since the objective was had influenced all students in the class. During the dis-
to point out the eventual changes in the cognitive struc- cussion of the maps, we found that the error was the con-
ture following teaching of the topic, the same terms were sequence of the teaching in another course: the teacher
given before and after teaching. The maps were drawn had certainly been successful!
up individually during a normal class session. During Analysis of these “noted terms CMs” gave us the
the drawing up of the postinstruction map, students did possibility of recognizing a first type of cognitive event,
not have their preinstruction CM at their disposal. which we called “Cognitive Event fix” (CEfix). concept
By comparing the two maps drawn up by each stu- that is incorrectly inserted in a student’s conceptual
dent, changes in the structure of the maps were found structure is no longer accepted in the same position af-
in more than three quarters of the students, even if not ter learning (Figure 2). The destabilization of the men-
all the changes were improvements. It is reasonable to tal structure due to meaningful learning has a favorable
think that changes in the maps correspond to similar outcome when the concept in question is stabilized cor-
changes in the conceptual structure of the students due rectly. The connected concepts are partially reordered so
to learning. In Figure 1 two successive maps, produced that the conceptual framework is restructured.
This type of map allows one to follow the evolution
of the cognitive structure after the teaching/learning ac-
tivity. Moreover, since the number of concept labels is
battery
fixed, this kind of map is particularly useful for identi-
works by means of
fying recurrent alternative conceptions held by students.
is based on the phenomena of
electrolysis
nucleus generates

(electrical)
is made
up of current
is movement of battery
is a act as electrical works by means of
atom electron ion
component of conductor
are obtained by
electrolysis
oxidation reduction generates
of
(electrical)
current
works by is movement of
battery electrolysis
means of
generates produces is caused
by
ion
(electrical) works by
current battery electrolysis
means of
oxidation reduction
passes
generates
through is happens happens
when of when
electrical atom (electrical)
conductor with a lack of gives up gains current
or excess
pass
is a flux of electron
orbit nucleus
around
Figure 2. Cognitive Event fix (CEfix). In this case, a concept which
Figure 1. Oxidation and reduction: two successive maps produced is incorrectly inserted in a student’s conceptual structure is no
by the same student. longer accepted in the same position.

Vol. 73 No. 11 November 1996 • Journal of Chemical Education 1085

Research: Science & Education

wave function in
AufBau's principle spin
Hund's rule
CM1 atom
wave indicates the direction between there there is a
CM2
exclusion principle of rotation of are
hybridization
quantum numbers are found orbit around
nucleus
uncertainty electrons
spatial geometry in shell the
principle of
corpuscle
a sharing corresponds to
spin
of various
orbit
shell necessary
ionisation chemical
orbital
uncertainty principle
energy to break bond
energy levels
ionisation energy
electron
chemical bond
energy level
atom
nucleus
uncertainty
0 2 4 6 8 10 12 principle spin
with opposite
Figure 3. The X-axis indicates the number of students who used holds true for wave
the various concept labels that are indicated on the Y-axis. The considered
darker bars show the original choice and the lighter ones repre- electron corpuscle
Pauli's as
sent the choice made at the end of the unit. principle
are found
says that an around the
Guided Choice Terms CM
orbital following AufBau's
A second way of assigning concept labels is to give nucleus principle
describes the
the students a number of terms greater than what is re- state of only 2
i. e.
according to
quested to construct the map. The criteria to follow is:
energy levels
• a fixed number of concept labels related to the
knowledge being investigated are assigned—
e.g., 20; Figure 4. Atomic structure: an example of two successive maps
constructed by the same student.
• each student must choose only a fixed part of
these labels—e.g., 10—to construct his own
map using the linking relationships and struc-
ture that he considers most suitable. We can identify two types of cognitive events in
these maps. The first, which we call “Cognitive Event in”
For example, third-year students were invited to (CEin), occurs when a concept label, in this case “orbital”,
construct a guided choice CM on atomic structure dur- enters the conceptual framework (Figure 5). The second,
ing a normal class session. Before teaching the topic, the which we call “Cognitive Event out” (CEout), occurs when
following twenty terms were supplied to assist the sub- a concept label, in this case “orbit”, of the first map is
jects with the task: atom, Aufbau principle, chemical excluded from the second one (Figure 6). As can be seen,
bond, corpuscle, electron, energy level, exclusion prin- the concept “orbit” is not simply replaced by the concept
ciple, Hund’s rule, hybridization, ionization energy, “orbital”, since the latter occupies a different position
nucleus, orbital, orbit, quantum numbers, shell, spatial with completely different links. It can reasonably be as-
geometry, spin, uncertainty principle, wave, wave func- sumed that the student’s cognitive structure is under-
tion. The students were then asked to choose ten terms going a conceptual reconstruction.
and develop a CM. This procedure was repeated at the
end of the teaching/learning activities, leaving the stu- Concept Stimulus CM
dents the possibility of using ten terms they considered
most suitable—not necessarily the same terms used for In the preceding approaches, the teacher gives the
the first map, which was not at their disposal. students the concept labels referring to a specific domain
Figure 3 summarizes the number of students who of chemical knowledge. But it is also possible to ask the
used the different concept labels in the CMs. The darker students to identify the most suitable or important con-
bars refer to the choices made for the first map (before cept labels and to construct what we call a Concept
teaching) and the lighter ones concern the choices made stimulus CM. In this case:
for the second map (after teaching). • only one concept label (stimulus) is assigned;
Analysis of the maps shows that before teaching,
seven students out of ten chose the concept label “or- • the number of labels the students can add is
bital”, but no one associated it with the label “wave func- fixed;
tion”. Four of these seven use the labels “orbit” or “shell” • each student elaborates his own map using the
to structure their knowledge. The concept “orbital” does concept labels, the relationships, and the struc-
not seem to be an alternative one for “orbit”, but rather ture that he considers most suitable.
complementary. After teaching, the label “orbital” is used
by ten out of twelve students; three of these ten still as- Fifth-year (final-year) students were invited to con-
sociate this concept with the label “orbit”, while the la- struct this kind of map on a very vast subject, thermo-
bel “wave function” is still not used by any student. dynamics. The aim of the teacher was to give the stu-
Figure 4 shows two maps that were constructed by dents, after studying this topic, the possibility of dem-
the same student before and after teaching. They are onstrating the meaningfulness of what they had learned.
quite different. Starting from the concept stimulus “thermodynamics”,

1086 Journal of Chemical Education • Vol. 73 No. 11 November 1996

 Research: Science & Education

defines and
spin 1st law ∆U ∆H
indicates the direction
of rotation of say nothing
places about
limitations on

uncertainty electrons spontaneity affinity

principle of 2nd law
determines the ∆G corresponds
from which to
derives

uncertainty studied from which

principle spin ∆S 3rd law
determines
ST
with opposite
holds true for

electron heat and work

standard chemical
state makes the meaning phenomena
cleares of
is used in
interprets
orbital
state to define the chemical allow a thermodynamic
describes the functions properties of the thermodynamics definition of system
state of only 2
in must be can undergo
state are the defined to
basis of study
variables
Figure 5. Cognitive Event in (CEin), occurs when a concept label, changes
reversible and are studied to
in this case “orbital”, enters the conceptual framework. irreversible laws
formulate
processes

Figure 7. Thermodynamics: an example of two successive maps

constructed by the same student.
spin
indicates the direction The labels “1st law”, “2nd law”, and “3rd law” in the first
of rotation of
map have been changed into “laws” in the second one;
electrons
are found orbit the labels “∆U”, “∆H”, “∆G, “∆S” in the first map are sum-
uncertainty shell
principle of
in marized in the term “state functions” in the second one.
This can also be seen for the labels “chemical thermody-
namics” and “chemical phenomena” in the second map,
which derive from the terms “affinity” (understood as
chemical affinity) and “spontaneity” (spontaneity of a
uncertainty reaction) in the first map.
principle spin The vastness of the topic caused some difficulty for
with opposite
holds true for
the students, who resorted to more general concepts only
in the second map, thus summarizing the large variety
electron of terms used in the domain of thermodynamics. It must
be remembered that the way of giving the students the
concept labels directly influences the types of cognitive
events to be found in the CMs they develop. “Noted terms”
Figure 6. Cognitive Event out (CEout) occurs when a concept la- maps never show “in” or “out” cognitive events, while
bel, in this case “orbit”, of the first map is excluded from the sec- “guided choice terms” or “concept stimulus” maps nor-
ond one. mally show all three events. These obviously do not rep-
resent the complexity of a human’s mental processes: they
are extreme cases, which should be looked for in students’
the students were assigned the following task for the CMs, as they are signs of meaningful learning.
first map: the ten terms that they considered most im- Moreover, a concept map is an idiosyncratic repre-
portant for representing the basics of thermodynamics sentation of a domain specific knowledge. Consequently,
were to be chosen and a CM constructed from them. The the maps reproduced are not representative or typical
maps were collected by the teacher, who kept them for a of our students. A CM strictly reflects the conceptual or-
fortnight, during which time the subject was discussed ganization of the single student who has produced it, giv-
in class. The maps were then given back, and each stu- ing evidence to a specific level of conceptual understand-
dent had the opportunity to restructure his own map ing.
accordingly. The three cognitive events (CEfix, CEin,
CEout) can be identified in all the maps. In Figure 7 the Conclusions
first and second maps of one of the students are reported.
Initially, it seems very difficult to link the changes It is now four years since we introduced concept
between the two maps to the cognitive events cited be- maps into our chemistry courses, as participants in an
fore. Since the maps are of the type “concept stimulus”, action-research project concerning the improvement of
any term considered suitable is accepted. In the maps chemical education in secondary schools. At the begin-
reported, all the concepts labels have been changed be- ning, we tried to use them as assessment devices: scores
tween the first and second maps. But, have they really? were assigned to postinstruction maps for the number

Vol. 73 No. 11 November 1996 • Journal of Chemical Education 1087

Research: Science & Education

and correctness of the relationships portrayed, for the ics involving the students and the teacher, who acts as
levels of hierarchy, and for cross-linking. We also tried the chairman. Von Glaserfeld (6) has emphasized the im-
to assign scores for the convergence of the students’ maps portance of social interaction in the construction of new
to teacher-constructed maps. But we soon had to recog- meanings, and we have found that by using the CMs we
nize that students’ CMs are highly idiosyncratic repre- can effectively act both on the interpersonal and the
sentations of a domain-specific knowledge, and the intrapersonal aspects of learning.
interindividual differences displayed among them were As Novak says “concepts map... is no “magic bullet”,
far more striking than the similarities. Taking into ac- no “quick fix” for classroom where rote learning predomi-
count the critical opinions on the scoring of concept maps nates” (7) and use of the maps can be successful only by
(5), this observation led us to shift emphasis and focus adopting a constructivist approach to chemical educa-
on changes in content and organization of CMs over time, tion. Many teachers object that turning to a
and on helping the students to become aware of and criti- constructivist approach and using concept maps is very
cize their own frames and those of the others. Such use time-demanding, and their objection is both true and
is consistent with a constructivist theory of learning, false. It is true if we limit our attention to the begin-
which is the foundation of our teaching strategy, viewing ning of the course, when students have to become ac-
knowledge as being actively constructed by the learner. quainted with the idea that they construct their own
During the four years of practical experience in us- knowledge and learn to develop concept maps. It is false
ing CMs, we have grown more and more impressed by if we go further along the course, since the meaningful
the potential of this metacognitive tool to help chemis- learning of the first fundaments of chemistry (the par-
try teachers and learners to improve teaching and learn- ticulate model and the conceptions of pure substance and
ing. Concept maps are useful for teachers, since they give chemical reaction) gives the students powerful instru-
them information on what students know, showing the ments to construct further meaningful knowledge more
concepts already present in their minds (initial concepts), easily and more quickly than in traditional teaching.
how they are related to one another (context), and how
learners reorganize their cognitive structure after a spe- Acknowledgments
cific teaching activity. In this way, teachers can be aware
of the presence of misconceptions that are potential “ob- Work carried out with the financial contribution of
stacles” to the construction of new, meaningful knowl- MURST (Ministero dell’Università e della Ricerca
edge. Moreover, CMs give the teacher the possibility to Scientifica e Tecnologica. Fondi 40%—Progetto:
check the influence of teaching on the cognitive struc- Insegnamento e Apprendimento della Chimica) and with
tures of students, since this tool is especially valuable the support of IRRSAE Piemonte.
in documenting and exploring the restructuring of con-
ceptual frameworks. Literature Cited
But concept maps are also judged a very useful
1. Chalmers, A. F. Science and its Fabrication; Open University: Buckingham,
metacognitive tool by students. In fact, many of those 1990.
trained to develop CMs in chemistry have spontaneously 2. Hodson, D. Sci. Educ. 1988, 72 (1), 19–40.
adopted them to represent knowledge in other disciplines, 3. Ausubel, D. P. Educational Psychology: a cognitive view; Holt, Rinehart and
Winston: New York, 1968.
such as history and Italian, claiming that CMs are pow- 4. Novak, J. D.; Gowin, D. B. Learning How to Learn; Cambridge University:
erful helps in meaningful learning of new subject matter. 1984.
5. Stuart, H. A. Eur. J. Sci. Educ. 1985, 7, 73–81.
Concept maps are also currently used in our classes 6. Von Glaserfeld, E. Synthese 1989, 80, 121–140.
as the starting points for discussions on chemistry top- 7. Novak, J. D. J. Res. Sci. Teach. 1990, 27, 937–949.

1088 Journal of Chemical Education • Vol. 73 No. 11 November 1996