Context-oriented Chemistry Teaching
Prof. Dr. Elke Sumfleth - Research Interests
Context-oriented Chemistry Teaching
- Analyzing Influences of a Real-life Context Compared to a Subject-related Context on Students' Interest and Achievement
Effects of context-oriented learning on student interest and achievement in chemistry education
To face low interest levels in subjects like chemistry teaching, approaches have been introduced which try to cope with the highly theoretical nature of the subjects by introducing everyday contexts. Making the content structure more relevant to students by connecting their everyday life to science concepts was seen as a way to raise interest levels and foster learning. Major context-based courses like ChemCom (American Chemical Society [ACS], 1988) and the Salters approach (Burton et al., 1994) were introduced on the basis of this objective. Considering the wide-spread implementation of these courses it is, however, astounding how little their effects on students' learning outcomes have been evaluated within a discipline as well as cross-disciplinary (Bennett & Holman, 2002; Taasoobshirazi & Carr, 2008). Especially the question of whether the content knowledge is as adequately acquired in a context-based environment and whether this knowledge can be transferred has not been thoroughly addressed. This project investigates the effects of a context-oriented learning environment on students' ability to achieve and transfer knowledge in chemistry and biology and compares effect sizes.
Context-based approaches to teaching science focus on this transfer of knowledge by linking the macroscopic level described as students' everyday experience and the microscopic level described as the general content and conceptual structure of the discipline. By repeatedly applying the content knowledge to everyday phenomena, better student ability to transfer knowledge is expected. As this transfer is supposed to be more difficult in chemistry, higher effect sizes in transfer measures should be found in chemistry as compared to biology by introducing everyday contexts.
In both disciplines, the study is conducted in a two-factorial control-group design resulting in four treatment groups which differ with regard to two variables, namely context-orientation and the respective revision activity. It is assumed that content knowledge can be more adequately acquired if structuring aids like concept maps help connecting the learnt concepts. Thus, two revision activities, concept mapping and written summaries, are compared. The variation in context is embedded in a collaborative and inquiry-based small group task. Students have to solve problems in a specific content area which is either introduced through an everyday context in the experimental group (LifeC) or a subject-related context in the control group (LabC). In chemistry, students work on topics from the content area of acids and bases, while the same students also attend the biology intervention on the blood circulation system. For example, if the students' task is to investigate how to neutralize acid solutions with solid chemicals (e.g. calcium oxide), the concrete task of the LabC group is to neutralize hydrochloric acid with a selection of chemicals, while the LifeC group has to find out which material saves the acidic football soil from degeneration. Solutions and the experimental material in the boxes given to small group are the same in both conditions. In each discipline, students take part in the intervention over the period of one week with one small group session a day. The intervention is conducted in the respective school but with university staff as facilitators to ensure correct implementation.
Achievement levels are measured in a pre-post-test design by two kinds of multiple choice items: while the first set of items retrieves content knowledge (CK), the second set of items requires the students to apply their content knowledge to either an everyday context (LifeC) or a context within the subject domain (LabC). In both disciplines, students show significant learning gains when attending the intervention. Table 1 summarizes the descriptive learning gains as measured by pre- and post-test achievement scores. Test scores clearly indicate that students show higher pre-knowledge in the biology topic (blood circulation system) than in the chemistry topic. This pre-requisite also accounts for the fact that higher overall learning gains can be found in the chemistry environment.
Table 1. Mean achievement test scores [%] in biology and chemistry (N=187)
To measure differences in learning gains between treatment and control group in the respective discipline standardized residuals are computed. With regard to the variable context-orientation, effects are calculated by means of a two-factorial ANOVA with cognitive abilities and interest in chemistry and biology as covariates.
Table 2. Differences between groups in the two disciplines
Table 2 shows that the treatment group outperforms the control group in almost all conditions. Even if differences do not get statistically significant, treatment groups are better on a descriptive basis. Having a closer look at the scales, it becomes obvious that students in the chemistry environment show higher learning gains in content knowledge (CK) if they learn in the treatment condition as compared to the biology environment. Students in that environment do not profit from the treatment with respect to their acquisition of content knowledge.
If content knowledge is to be applied in the students' everyday world (LifeC) a high effect size can be found in chemistry only. Compared to the biology environment by an ANOVA with repeated measures the effects between the groups differ from each other in a statistically significant way. This significant effect between the two disciplines can only be found on that scale. Thus, the treatment effects in the disciplines differ mostly with respect to the transfer of knowledge to an everyday context which is relevant to the student.
Variable: Revision Mode
While students revising the learnt concepts by means of a concept map benefit from the structuring method in biology, no main effects can be found in chemistry. Only male students profit from the method with regard to their connected knowledge. Link zu den Bio-Ergebnissen zum CM?!
Funded by DFG
Analyzing Influences of a Real-life Context Compared to a Subject-related Context on Students' Interest and Achievement
In total, it can be stated that context-based teaching tries to integrate science specific topics into students' social or personal environments aiming at kindling students' interest and advancing the quality of science education. The results of Fechner (2009) within the chemistry domain underline that students learning with real-life contexts show higher situational interest and achievement than students learning with traditional contexts. The differences in learning outcome were shown to be completely mediated by an increased situational interest. However, effects vary depending on the respective context.
Therefore, this study aims at investigating the effects of context-oriented learning on students' situational interest and their learning achievement concerning different chemical concepts. The question is: What influences do a subject-specific context and a real-life context have on the learning outcome across two concepts in chemistry education? Concerning this question, it is suggested that students working with a real-life context show higher situational interest and achievement than students learning with subject-related contexts.
A real-life context is meant to be known by students from their everyday lives and is relevant to them. In comparison, a subject-related context should be close to students' regular chemistry classroom experiences. Therefore, "lakes" was taken as a real-life context and "school laboratory" was chosen for the subject-related context. Students were randomly assigned to one of these contexts. Every context-group learned two different chemistry topics embedded in the respective context. The chemistry topics were taken from the German curriculum, namely (1) water as a solvent and (2) structure of the water-molecule. To establish the topics and contexts into the lessons, worked-out examples (Mackensen-Friedrichs, 2004) were developed. Every worked-out example consists of a context-based problem statement (question), professional problem-solving steps (solution steps) and, finally, the actual problem solution (answer). So, the real-life context group will learn with examples embedded in stories about lakes, while the other group will learn about the same topic with the same solution steps embedded in a lab environment. Overall, students were confronted with two worked-out examples towards one chemistry topic. In order to avoid sequence effects, topics were rotated.
The testing phases spanned three successive days. The first day was used for students to fill in paper-pencil tests measuring their interests, beliefs, cognitive abilities (Heller & Perleth, 2000), and prior knowledge. Data from the pre-test were regarded as control variables to ensure that the two context-groups could be balanced.
Over the subsequent two days the intervention phase took place. During the intervention, students learned individually with the worked-out examples for 60 minutes. After the learning phase had ended, a post-test - measuring students' learning achievement and situational interest - was administered to the students (see Fig. 1).
Fifty-seven 9th graders from two different higher track secondary schools participated in the pilot study. The mean age was 14.5 years and 53% were female. However, there is only relatable data from pre-and post-tests for 31 students from the samples. The mean age was 13.7 years and 48.4% were female.
The predominant aim of the pilot study was to test the worked-out examples and test-instruments regarding their quality. In the next step, we take a look on students' learning achievement to analyse if the context-based worked-out examples are conducive. Data show an increased learning achievement from pre-test to post-test. Therefore it is assumed that the worked-out examples are conducive.
Furthermore, an analysis of covariance with context (subject-related / real-life) predicting students' achievement and prior knowledge as a covariate did not show any main effect of context. However, students' in the real-life context show a tendency to score higher in the post-test than students learning with the subject-related context.
In addition, an analysis of covariance with context (subject-related / real-life) predicting students' situational interest shows no main effect of context. However, students learning in the real-life context group show higher situational interest levels than students learning in the subject-related context-group.
In October this year, the main study will commence with approximately two hundred 9th grade students from higher track secondary schools. Regarding the results of the pilot study, we suggest that the reported tendencies will become significant.
Funded by DFG
Investigation of context-effects caused by structural or motivational aspects
Both studies just started
Funded by grants of the Research Training Group nwu-essen (DFG)
Another project dealing with context-oriented learning and the conception of Chemistry in Context (ChiK) is described focusing on concept mapping and cumulative learning.