Concept Mapping in Chemistry

Prof. Dr. Elke Sumfleth - Research Interests


Concept Mapping in Chemistry


^ top

Concept mapping to structure content knowledge as well as teaching and learning processes in chemistry lessons as a means to foster cumulative learning

As a result of large scale assessments the demand for more connectivity and cumulative learning in science education has increased. Constructing knowledge seems to be an additive rather than a cumulative process. It has been proven that German students have problems in transferring scientific concepts into new situations and conceptual understanding (Baumert et al., 1997).

The German context-oriented approach "Chemie im Kontext" (Chemistry in Context; ChiK) tries to overcome this situation. It claims to strengthen the cognitive network between the everyday and the scientific domain (i.e., vertical linkage). Therefore, two main hypotheses are tested:

  • The degree of linkage in ChiK-instruction is higher than in traditional instruction.
  • The knowledge structure of students is more elaborated after ChiK-instruction than after traditional instruction.

In total, seventeen 10th grade gymnasium classes were analysed. Twelve traditionally taught classes were selected out of a larger pool of videotaped chemistry lessons according to the degree (level) of linkage in instruction during single lessons. To assign such levels, videos have been coded via a highly inferent coding scheme including six 'levels of vertical linkage': one fact, several facts, one relation between facts, several relations, linked relations. Student and teacher statements have been coded separately to minimize a bias in-between groups. Sufficient interrater agreement could be achieved in both groups. The average level of so-called high-linking classes (N=6) is significantly higher in both groups as compared to the level of linkage in low-linking classes (N=6). The remaining five classes have been taught context-based and have been video-coded as well.

In order to compare the groups, cognitive abilities and motivation & interest have been controlled as not to bias any other results. At the end of the school year, performance of students was assessed. To investigate knowledge structures, individual concept maps were constructed by students. The relations of the maps were rated according to correctness and other structural parameters, like e.g. the total number of relations in the map. Additionally, an achievement test was administered (22 Items; α=.68). Both tests covered the 10th grade compulsory topic 'acids and bases'.

For students, the level of vertical linkage in ChiK classes is significantly higher than in low-linking classes and almost significant if teachers are considered. If ChiK-instruction is compared to high-linking classes, students demonstrated a significantly higher level of linkage, while teachers showed significantly lower linkage levels. Results within groups show that the level of vertical linkage in teachers is significantly higher than that of the students in high-linking groups and approaches significance in low-linking classes while there is no difference in ChiK-classes.

According to performance tests, main effects could be found on achievement and the correctness in concept maps. According to these results both hypotheses can be accepted if the teachers' level of linkage is disregarded. As one important result, correlations between knowledge structure and individual linkage in instruction hint at an interdependency between these two aspects. Putting video and performance results together, the question why low and high linking classes do not achieve different results in performance tests arises. An explanation might be that in ChiK-classes teachers' and students' level of linkage match. Therefore, learning processes seem to be more effective. In the traditionally instructed classes teachers offer a higher level than students are able to pick up, therefore learning processes might be less effective which leads to poorer performance results.

Funded by DFG

Further information:
Vertikale Vernetzung (german)
Vertical Linkage (english)
Ina Glemnitz



^ top

Concept mapping as a learning strategy

Concept mapping strategies can also be used in learning with texts if students have been trained to use this method. Otherwise, the cognitive load might be too high so that learning outcomes will not be achieved.


Further information:
Jasmin Neuroth


In order to guarantee proficiency in the concept mapping method, a training program was developed and evaluated. It is based on a student booklet containing learning units and examples.