Issue: Experimental-specific Characteristics of Quality - A Video Study in Chemistry Education

During the last years the discussion about quality of instruction has been moved into the centre of attention. General quality of instruction is defined “as a bundle of characteristics describing instruction (process quality) which have a positive effect on teaching outcomes (product quality)” (Einsiedler, 2002). Thus, ‘good instruction’ can be defined as something from which we can measure positive effects concerning the output of the students.

This definition, in particular the focus on process and product quality, can be found in models of instructional quality (e.g.: Reusser and Pauli, 2003; Helmke, 2009). Most characteristics of instructional quality are for general instruction and not for a specific subject (Brophy & Good, 1986; Ditton, 2000; Helmke, 2003; Tobin & Fraser, 2003; Helmke, 2009). Helmke (2002) and Ditton (2000) state that it is “out of date” to look only at the general quality of instruction, so they call for specific characteristics of quality in specific subjects. Rakoczy & Pauli (2006) speak about the “need of subject and domain specific processes to measure subject specific quality in education”. Experimentation has a central and distinctive role in chemistry instruction and can be seen as such a subject-specific operation (Pfeifer, Häußler & Lutz, 2002; Bader & Schmidkunz, 2002; Barth, 2005; Hofstein & Lunetta, 1982; KMK, 2004).

Based on the theoretical background, the aim of the project was to translate the results of the research on general instructional quality to specific characteristics in chemistry education (in particular to the experimental phases) to identify experimental-specific characteristics of quality that enhance the learning process. This is achieved by observational research (video-analysis) based on a category system and student questionnaires. The study consists of two parts. In the first part, regular chemistry lessons have been observed to test first ideas about experimental-specific quality criteria.

In the second part, the significance of these criteria was proven in an intervention-study. To describe experimental phases in chemistry education, 18 regular lessons with experimental phases have been videotaped in 2008/2009 (N = 264; 10th grade; mid-school; topic alcohols). These video-data were analyzed based on a category system. This category system, consisting of a low-inferential and a high-inferential component, has been conducted and evaluated before (Schulz & Walpuski, 2009). In both studies students’ characteristics (learning achievement, motivation and interest) have been acquired in a pre-post testdesign. Additionally, the cognitive abilities (Heller & Perleth, 2000) and the knowledge of scientific procedures (Klos et al., 2008) have been collected, so the test data could be analyzed more precisely. The second study (intervention study) has been conducted one year later in 2009/2010 (N = 322; 10th grade; mid-school; topic alcohols). The same teachers as in the first study participated in the intervention study and the same test instruments were used. In the intervention study, the experimental phases were enriched with the possible quality criteria from the first study using a short teacher training. The students of the first study formed the control-group and the students of the second study the intervention group.

In the first study (2008/2009) significant correlations could be found between the students’ characteristics and the quality criteria of experimental phases (e.g.: clear question/problem at the beginning of the lesson, pre-knowledge of the students, statements of students in small groups, and the traceability of the experiment). These and other criteria were tested regarding their importance.

The results show that the intervention was successful. The descriptive comparison of the intervention-group and the control-group (video-analysis) illustrates the increase of quality criteria in the intervention-group. Among others, significant higher values concerning these quality criteria could be shown for the categories statements of students, hypothesis testing, instructional efficiency, problem-solving and traceability of the experiment.

Furthermore, it could be shown that the intervention-group has a higher learning achievement than the control-group. A difference between these groups concerning the motivation and interest of the students could not be found in general, while the situational interest of the intervention-group was significantly higher than the situational interest of the control-group. A mediationanalysis was conducted to prove the situational interest as a mediator for the learning achievement. The result of the mediationanalysis showed a partial mediation.

Termtime

• 2008-2010

Supervisors

• Maik Walpuski & Elke Sumfleth

Dissertation

• Alexandra Schulz