The natural sciences are of major importance for present and future issues of humanity, such as climate change and medicine (Eisner et al., 2019, p. 3). Thus, science education is a major concern. Science is currently taught as separate disciplines or in an integrated way in lower secondary education (for Europe see Forsthuber et al., 2011, p. 60). Labudde (2014, p. 15) defines integrated science teaching as the teaching of biology, chemistry, and physics as one subject. In Germany, science is taught integrated, e.g., in lower secondary education in comprehensive schools in Lower Saxony (Niedersächsisches Kultusministerium, 2012) or classes 5 and 6 of grammar schools in Baden-Wuerttemberg (Ministerium für Kultus, Jugend und Sport Baden-Württemberg, 2016). Pre-service teachers usually study two subjects in Germany, and German science teacher education is primarily discipline-specific; teachers are trained to become biology, chemistry, or physics teachers at school (Neumann, Härtig et al., 2017, p. 38). Thus, science teacher education is problematic, as “teachers may be asked to teach a subject they have not formally studied” (Carlson, Daehler, 2019, p. 91) when teaching interdisciplinary science. Teachers demonstrate this issue by assessing teaching unstudied subjects as a problem (Dörges, 2001, p. 230; Fruböse et al., 2011, p. 434). Thus, it is of interest whether (prospective) teachers believe they can teach interdisciplinary science and, subsequently, how they assess their knowledge in unstudied subjects. To investigate such questions in detail and to be able to draw conclusions for science teacher education, there is a need for a measure to gain knowledge on science-related self-rated content knowledge, as content knowledge (CK) is an important part of professional competence (Baumert, Kunter, 2013, p. 29).
2 Theoretical background
2.1 Science teaching
Science can be taught in a number of ways, including different approaches (Metzger, 2010, p. 32; Broggy, O’Reilly, Erduran, 2017, p. 81). While multidisciplinary teaching includes multiple subjects that contribute to a topic separately with their specific identity, interdisciplinary teaching contains subjects that overlap, e.g., with common concepts (Broggy et al., 2017, p. 81). Thus, in multidisciplinary teaching, teachers address their studied discipline-specific subjects, while in interdisciplinary science teaching, one teacher covers multiple subjects, e.g., as integrated science (Metzger, 2010, p.32).
Comparable to Engelmann (2019, pp. 25ff.), we assume that for interdisciplinary science teaching to succeed, the characteristics of biology, chemistry, and physics and the depth of the core ideas of the three subjects are necessary. Interdisciplinary science at comprehensive school underlines this assumption; e.g., in Lower Saxony, interdisciplinary science replaces the three science disciplines in lower secondary education to prepare for subject-specific science in upper secondary education (Niedersächsisches Kultusministerium, 2012). Thus, interdisciplinary science needs to prepare students for discipline-specific subjects and their concepts and methods, too. Jansen et al. (2014, p. 48) also assume that the content in discipline-specific and interdisciplinary science classes largely overlap. In sum, we conclude that a strong knowledge base of each subject is necessary to identify overlaps between the subjects and teach interdisciplinary science in-depth.
2.2 Professional competence for interdisciplinary science teaching
Because German teacher education for secondary schools focuses primarily on CK (Kleickmann et al., 2013, p. 93), CK is of great importance as background for teaching knowledge of unstudied subjects in interdisciplinary science. CK represents one domain of professional knowledge of teachers’ professional competence (Baumert, Kunter, 2013, p. 29). Other domains of professional knowledge are pedagogical content knowledge (PCK), pedagogical/psychological knowledge, organizational knowledge, and counseling knowledge (Baumert, Kunter, 2013, p. 29). Empirical studies regarding various subjects relating to PCK and CK revealed that CK is the basis for PCK and that PCK is essential for teaching (e.g., Kleickmann et al., 2018, p. 126; Sadler et al., 2013, p. 1036, 1038, 1043; Baumert et al., 2010, pp. 162-164.; Käpylä, Heikkinen, & Asunta, 2009, p. 1407). Thus, CK of unstudied science education subjects should also be of great importance for interdisciplinary science teaching.
2.3 Science content knowledge in the curricula
In Germany, core ideas and learning outcomes regarding CK are only specified for students but not for teachers. Thus, we define lower secondary students’ learning outcomes as the minimum competency for teachers teaching unstudied science education subjects. CK in science (especially in biology and chemistry) comprises subject-specific core ideas (e.g., biology and chemistry: Niedersächsisches Kultusministerium, 2015, pp. 44, 72). The core ideas are the basic knowledge of the subject and provide a systematic structure of the subject-specific knowledge (e.g., biology: Niedersächsisches Kultusministerium, 2015, p. 72). Core ideas comprise common principles of the specific subject (e.g., chemistry: Niedersächsisches Kultusministerium, 2015, p. 44) and are a guideline for teachers to choose appropriate topics for their teaching (e.g., biology: Niedersächsisches Kultusministerium, 2015, p. 72). The core ideas provide the same explanation patterns at different topics in all classes and provide vertical networking (e.g., biology and chemistry: Niedersächsisches Kultusministerium, 2015, pp. 44, 72). In addition, the core ideas enable horizontal networking between subjects because explanation patterns of one science education subject could be useful in another science education subject (e.g., chemistry: Niedersächsisches Kultusministerium, 2015, p. 44).
The reference for operationalizing CK in Germany is the national education standards in science for lower secondary education (e.g., biology: Kultusministerkonferenz, 2004). The standards contain the same four competence areas for biology, chemistry, and physics (e.g., biology: Kultusministerkonferenz, 2004). The nationwide standards are the basis for the science curricula for biology, chemistry, and physics for all federal states (statewide curricula) (e.g., for Lower Saxony: Niedersächsisches Kultusministerium, 2015, p. 7). One of the competence areas in the nationwide standards and statewide curricula for biology, chemistry, and physics specifies the CK students are expected to achieve. The statewide curricula, e.g., of Lower Saxony, comprise learning outcomes for different educational levels (classes 5/6, classes 7/8, classes 9/10) and the final competencies students have to achieve at the end of lower secondary education for all core ideas (biology, chemistry) or subject areas (physics) (Table 1) (Niedersächsisches Kultusministerium, 2015).
In contrast, the curriculum for comprehensive schools in Lower Saxony is interdisciplinary for science from classes 5 to 10, while remaining grounded in the same nationwide standards(Niedersächsisches Kultusministerium, 2012, p. 12). Instead of 18 discipline-specific core ideas, it includes seven common, crosscutting concepts for science (Niedersächsisches Kultusministerium, 2012, p. 12; Table 1).
Table 1: Core ideas and subject areas of the content knowledge in biology, chemistry, and physics from the Lower Saxony curricula for lower secondary education at grammar schools (i.e., classes 5-10; Niedersächsisches Kultusministerium, 2015, pp. 15, 49, 74) and the core ideas of the content knowledge in science of the Lower Saxony curriculum for lower secondary education at comprehensive schools (Niedersächsisches Kultusministerium, 2012, p. 12).
2.4 Measuring self-rated content knowledge
Previous measures of CK are relatively time-consuming. For example, measuring CK only in biology takes 45 to 60 minutes (Großschedl, Mahler, & Harms, 2018, p. 15). One possible proxy indicator of prospective teachers’ CK performance is self-rated content knowledge (srCK). Table 2 summarizes relevant existing measures for science-related srCK in other contexts. Some measures can be considered to measure srCK, although they indicate to measure (academic) self-concept or beliefs (see Section 2.5).
Table 2: Measures regarding science-related self-rated content knowledge, respectively, (academic) self-concept and beliefs. Specificity: Is the measure only subject-specific or topic-specific? Curriculum: Does a curriculum underlie the development of the instrument? No instrument integrates core ideas explicitly. Purpose: Does the measure ask for teaching or understanding knowledge? We refer to the grey markings to point out the research gap. (A)SC = (academic) self-concept, srCK = self-rated content knowledge. The design of some measures reflects only the level of scientific subjects, such as biology, chemistry, physics, and technology (Jansen et al., 2014, pp. 45f. after Ramm et al., 2006; Velthuis et al., 2014, p. 454). Other measures consider the curricula and focus on specific concepts or topics but in relation to the primary education context, yet not in the German context (Yangin, Sidekli, 2016, pp. 56f.; Yilmaz-Tuzun, 2008, pp. 188, 201). Two instruments take the curricula into account by focusing on self-rated topics in secondary education (Bröll, Friedrich, 2012, p. 182; Hardy, 2014, pp. 560, 573f.). Hardy (2014, p. 555) focuses only on British secondary school students. In general, we defined students’ CK as the minimum benchmark for teachers teaching unstudied science education subjects in lower secondary education. Thus, measures for students are considered at this stage in addition to measures for teachers. Regarding the integration of normative guidelines of content knowledge, Hardy’s (2014, pp. 560, 573f.) measure includes only topics from the British curriculum (e.g., key stage 3; Department for Education, 2013), which are not comparable with the core ideas of German/Lower Saxony curricula (at least in biology and chemistry). Bröll and Friedrich (2012, p. 182) measure teachers’ understanding of topics of a curriculum in the German context (Ministerium für Kultus, Jugend und Sport Baden-Württemberg, 2004). These topics are different from the core ideas in the Lower Saxony curricula (at least for biology and chemistry). Also, the topics are only designed for secondary modern schools from classes 5 to 9, and the instrument contains 62 items (Bröll, Friedrich, 2012, p. 182).
In sum, none of the presented measures consider (prospective) teachers’ understanding of the core ideas in the science discipline-specific curricula of lower secondary education in Germany and Lower Saxony. Thus, there is a need fora discipline-specific measure of (prospective) teachers’ self-rated understanding of the core ideas in relation to the Lower Saxony science curricula.
As a first step to developing this instrument, we conducted an exploratory factor analysis (EFA; i.e., principal factor analysis) on 20 items linked with the core ideas of the curricula (see Section 3.2.1). We generated hypotheses regarding the factor structure with a sample of 114 pre-service and trainee teachers (description of the sample, see Handtke, Bögeholz, 2019, p. 6). Three factors emerged and excluded no items. Sorted by Eigenvalues, factor 1 contained srCK of physics (seven items, Cronbach’s α = 0.95, λ = 0.728 to 0.903), factor 2 contained srCK of biology (eight items, α = 0.92, λ = 0.640 to 0.881), and factor 3 contained srCK of chemistry (five items, α = 0.94, λ = -0.727 to -0.902). The factor structure needs confirmation via an independent sample (Conway, Huffcutt, 2003, p. 149). Therefore, as the main aim of this paper, we present the development and validation (convergent, divergent, factorial, concurrent, retrospective, content validity) of a measure for science-related srCK that integrates the core ideas of the curricula.
2.5 The relation of self-rated content knowledge with self-concept, the studied subjects, and grades
One of the closest constructs to srCK is self-concept. Shavelson, Hubner, and Stanton (1976, p. 411) define self-concept as “a person’s perception of himself.” One component of self-concept is academic self-concept (Shavelson et al., 1976, p. 412). Academic self-concept contains all perceptions of one’s abilities in academic situations (Dickhäuser et al., 2002, p. 394). It can be “divided into subject matter areas” (e.g., science, mathematics) and even more differentiated “into specific areas within a subject matter” (Shavelson et al., 1976, p. 412). Jansen et al. (2014, p. 48) empirically supported the discipline-specific structure of academic self-concept differentiated as biology, chemistry, and physics. SrCK is the self-assessment of one’s CK (Oberle, 2012, p. 50). According to the definition of academic self-concept (Dickhäuser et al., 2002, p. 394), Oberle (2012, p. 27) argues that srCK can be seen as a specific part of academic self-concept. Following the model of Shavelson et al. (1976, p. 412f.) as well, srCK could be a part of academic self-concept on a specific level within a specific subject. Thus, both constructs are closely related.
Research regarding these constructs highlights relations with different variables. Regarding srCK, Bröll and Friedrich (2012, pp. 182f.) found that in-service teachers had lower srCK of topics from science education subjects they did not study and, consequently, higher srCK of topics from science education subjects they studied. Also, in-service teachers completing a degree with two science education subjects had higher srCK in those areas of study than in the unstudied science education subjects (Bröll, Friedrich, 2012, pp. 182f.). These examples illustrate the impact of the studied science education subjects on the corresponding srCK of in-service teachers. Furthermore, in regard to meta-synthesis, researchers identified a positive, moderate relationship between the self-assessment of ability and external criteria for ability, such as grades (r = 0.29) (Zell, Krizan, 2014, p. 116). Specific self-assessments correlated higher with external criteria for ability than global self-assessments (Zell, Krizan, 2014, p. 116). These correlations demonstrate the relation between perceived ability, such as srCK, and an external criterion for ability.
2.6 Research questions and hypotheses
From the need for an instrument to measure srCK regarding the core ideas of biology, chemistry, and physics that integrates the corresponding Lower Saxony curricula for lower secondary science education, we derived the following research question:
Research question 1:
In what way can the factors of biology, chemistry, and physics resulting from the EFA (see Section 2.4) be empirically supported?
In addition, we wanted to identify the potential for interdisciplinary science teaching based on science-related srCK. Thus, we examined the currently unknown intercorrelations of science-related srCK factors and the thus-far unidentified impact of studying non-corresponding science education subjects on science-related srCK factors.
Research question 2:
Do the intercorrelations between the factors of science-related srCK and the influence of the studied science education subjects on non-corresponding factors of science-related srCK reveal potential for interdisciplinarity?
Coming back to research question 1, we formulated five hypotheses for validation purposes. The structure of the three hypothetical factors resulting from the EFA and corresponding with the science education subjects needed assessment. This examination could provide conclusions regarding divergent (intercorrelations of the factors; Hartig, Frey, & Jude, 2012, p. 158; Moosbrugger, Kelava, 2012, p. 17) and factorial validity (Hartig, Frey, & Jude, 2012, p. 162)
The confirmatory factor analysis (CFA) regarding science-related srCK confirms the three identified discipline-specific factors of the EFA: biology, chemistry, and physics.
SrCK is a specific part of academic self-concept (Dickhäuser et al., 2002, p. 394; Oberle, 2012, p. 27). Based on this conceptualization as a hierarchical structure, we expected a close relationship and a strong correlation between the general level of academic self-concept and specific srCK in the context of science as well. This relationship could provide indicators of convergent validity (Moosbrugger, Kelava, 2012, p. 17).
Science-related srCK strongly correlates with the academic self-concept relating to the same science education subject.
Comparable to Bröll and Friedrich (2012, pp. 182f.), we expected an advantage for prospective teachers in the srCK of the science education subject they study or teach. The resulting relation could be an indicator of concurrent validity as the criterion and the construct are measured simultaneously (Moosbrugger, Kelava, 2012, p. 18).
Studying biology, chemistry, or physics predicts the corresponding srCK.
In regard to the positive relationship between the self-assessment of ability and external criteria for ability (Zell, Krizan, 2014, p. 116), we concluded that the (reported) external criterion for ability (grades) predicts srCK, as it is a type of ability self-assessment. The resulting relation could be an indicator of retrospective validity (Schmiemann, Lücken, 2014, p. 113), as past grades should predict current srCK.
Final grades in biology, chemistry, and physics in secondary school predict the corresponding srCK.
In addition, Bröll and Friedrich (2012, pp. 182f.) identified an advantage of studying multiple science education subjects for in-service teachers’ corresponding subject-specific srCK. Thus, prospective teachers studying two science education subjects should have a higher srCK in (at least) two subjects, indicating concurrent validity (Moosbrugger, Kelava, 2012, p. 18).
Studying two science education subjects results in higher srCK in at least two science-related subjects.
The sample consisted of primarily pre-service and trainee teachers (n = 552, Table 3).
Table 3: Description of the sample. If all forms of a variable in total include >/< 100%, the difference is missing values/“no indication” or due to rounding (e.g., 99.99% or 100.1%).
The test participants completed the cross-sectional survey between December 2017 and December 2018. About three-quarters of the test participants came from Lower Saxony. Almost 90% studied to teach in secondary schools. A majority of the sample was female. Each test participant studied at least one of the three subjects of biology, chemistry, or physics. The sample consisted of more biology and chemistry than physics pre-service teachers. Furthermore, physics was rarely studied with other science education subjects. Nearly half of the sample comprised undergraduate students, while a quarter pursued a Master of Education and around an eighth comprised trainee teachers. In German upper secondary education, grades range from 1 to 15 points, 15 to 13 indicating a grade 1 (very good), 12 to 10 indicating a grade 2 (good), 9 to 7 indicating a grade 3 (satisfactory), and so on. We asked the test participants to report their remembered final grades they received in secondary school for biology, chemistry, and physics. Five hundred test participants stated a final grade for biology (M = 11.71, SD = 2.23), 487 for chemistry (M = 11.20, SD = 2.38), and 465 for physics (M = 10.53, SD = 2.67).
3.2.1 Self-rated content knowledge
We designed the measure as a time-saving proxy indicator for CK in biology, chemistry, and physics. In this paper, we describe the development and validation of a measure of science-related srCK due to the desideratum for a discipline-specific measure of (prospective) teachers’ self-rated understanding of the core ideas in relation to the science curricula of Lower Saxony (see Section 2.4). We do not test its suitability as a proxy indicator for CK in this paper due to the time-restrictions of the study. It would be necessary to measure the CK of biology, chemistry, and physics to test suitability. We investigated self-efficacy beliefs in addition to the srCK (Handtke & Bögeholz, 2019). Thus, we focus on the development and validation of a measure for srCK in this paper.
The newly designed measure for science-related srCK comprised the core ideas of the CK in the German curricula for lower secondary science education (see Table 1, grammar school) in biology, chemistry, and physics in Lower Saxony (Niedersächsisches Kultusministerium, 2015, pp. 15, 49, 74). This CK contains the learning outcomes for the students. Teachers teaching unstudied science education subjects should have, at a minimum, the CK required of students. Therefore, we defined it as the bare minimum CK for (prospective) teachers teaching science in lower secondary education (see Section 2.3). For operationalization, we used the discipline-specific scientific curricula instead of the interdisciplinary curriculum, as they account for the requirements of all three science education subjects (biology, chemistry, and physics). In addition, the discipline-specific curricula include the core ideas of the interdisciplinary curriculum (see Section 2.3), and we assume the three subjects to be the base for interdisciplinary science teaching (e.g., Engelmann, 2019, pp. 25ff.; see Section 2.1). We used the curricula of only lower secondary education, as interdisciplinary science is predominantly taught in lower secondary education (see Section 1). Because these curricula for lower secondary education and their core ideas are based on Germany’s nationwide standards (e.g., biology: Kultusministerkonferenz, 2004) (Niedersächsisches Kutusministerium, 2015, p. 7), we assume that our measure is applicable to lower secondary education (grammar and comprehensive schools) throughout Germany. The core ideas from biology, chemistry, and physics for self-rating (“I know very much about…”) were illustrated with examples (see Appendix). This approach enables a more standardized interpretation of the items.
We consulted eight experts to estimate the relevance of many examples to specify the core ideas we derived from the curricula (Niedersächsisches Kultusministerium, 2015) to ensure content validity (Hartig, Frey, & Jude, 2012, pp. 148ff.; Moosbrugger, Kelava, 2012, p. 15). The experts were researchers from biology, chemistry, and physics education and were teacher trainers or teachers in at least one of the three subjects. We obtained two expert assessments from teachers or teacher trainers for each subject and per subject one expert assessment from a researcher of biology, chemistry, and physics education. All experts had at least ten years of experience in their subject areas and only rated examples of their subject(s). Additionally, we encouraged experts to suggest their own examples. We chose examples with content-related agreements from at least two of the three experts for the final measure. We reserved the right to formulate or shift examples based on at least one assessment so that each core idea was specified by not less than two examples. Furthermore, three items were generated regarding the general self-rating of CK for teaching the biological, chemical, and physical parts of interdisciplinary science. Only the subject changed between these three items.
In sum, the measure contained seven specific items for biology, four for chemistry, six for physics, and three general items, and overall, twenty items (see Appendix for wording). We did not include the core idea structure and function in the measure for biology due to psychometric issues (more balanced item composition regarding the subjects) and because it is a core idea that needs to be more specified by others (i.e., substance and energy transformation, control and regulation, information and communication, reproduction, variability and adaptation) (Qualitäts- und UnterstützungsAgentur - Landesinstitut für Schule [QUA-LiS], 2011, p. 1). Consequently, structure and function is less specific than the other biological core ideas and, thus, rather dispensable. The four-point response scale included “Do not agree at all”, “Do rather not agree”, “Do rather agree”, and “Fully agree”. We examined the measure with an EFA (see Section 2.4) and utilized the results as the basis for our analyses.
3.2.2 Science subject-specific academic self-concept and the survey
We measured academic self-concept in biology, chemistry, and physics utilizing the scale from Jansen et al. (2014, pp. 45f.), following Ramm et al. (2006), with four items for each of the three subjects (in total: 12 items). The scale measured the general level of academic self-concept (e.g., “I am just not good at biology/chemistry/physics,” Jansen et al., 2014, p. 45). The four-point response scale ranged from “Does not apply at all” to “Does fully apply” (Jansen et al., 2014, p. 46). We reversed the polarity of the negatively formulated items before analysis.
Overall, the test participants completed the personal data first (e.g., studied subjects or final grades in science subjects), then the measure of academic self-concept, followed by the measure of science-related srCK and the self-efficacy beliefs of interdisciplinary science teaching (SElf-ST) instrument (Handtke, Bögeholz, 2019). A subgroup of pre-service teachers (n = 135) received monetary compensation because they participated beyond regular coursework or during course breaks (e.g., in practical laboratory course), while the other test participants filled out the questionnaire during a course and thus received no compensation. All trainee teachers received monetary compensation for their participation.
We used R with R-Studio (version 1.1.463) and the lavaan package (version 0.6-3) to analyze the data. We conducted a CFA to test hypothesis 1. To determine the correlations of hypothesis 2, we specified the measurement models (CFA). Thereby, we automatically computed the latent correlations of hypothesis 2. Due to the directional hypothesis, the corresponding three p-values were halved. In addition, we applied an individual structural equation model (SEM) to hypotheses 3, 4, and 5 each. In comparison to the EFA (n = 114), we used an independent sample (n = 552) for the CFA in the second step of the development of the measure in this paper (Conway, Huffcutt, 2003, p. 149). To be more precise, we completed the data computation of the Likert scales in ordinal values. In addition, due to non-normality, we used the WLSMV estimator (Brown, 2006, pp. 76, 388f., 404, 409) and pairwise deletion due to less than 1% of missing values. Following Little (2013, pp. 120f., 126), the sample size was appropriate. To rate the fit of our models, we followed the more liberal guidelines of Little (2013, pp. 109, 115) and Wheaton et al. (1977, p. 99) for the lower limits: χ2 /df ≤ 5, CFI > 0.90, TLI > 0.90, RMSEA < 0.10
To investigate the influence of the individual subject(s) studied, we had to create a new variable for each subject and so used effects coding (-1 and 1) (Cohen et al., 2003, pp. 320ff.). For example, the variable for studying biology was coded as 1 if the test participant’s subject of study was biology, regardless of the other subject(s). If a test participant did not study biology, we coded the variable with -1. Similarly, we created the variables for chemistry and physics. To examine the influence of the number of studied science education subjects, we used dummy coding for another new variable (0 = one science education subject, 1 = two science education subjects) (Cohen et al., 2003, pp. 303ff.).
Table 4 shows the measurement model factors of the confirmatory factor analysis (CFA) (hypothesis 1) with their Cronbach’s α values, items, and item loadings.
Table 4: Summary of the science-related self-rated content knowledge (srCK) factors. It includes the names of the factors, number of items, Cronbach’s α, item content [abbreviation], and standardized factor loadings of the items (see Appendix) on the three factors (sorted by Eigenvalues of the exploratory factor analysis). B = biology, C = chemistry, P = physics. 0 = general item, ≥ 1 = specific items. * = item automatically fixed to 1 before standardization for the metric of the latent factors.
The goodness-of-fit indicators displayed a good fit: CFI = 0.977, TLI = 0.974. The RMSEA indicated a mediocre fit: RMSEA = 0.092, with a confidence interval of 90% = 0.086-0.097. The χ2-statistic was significant: χ2 (167) = 937.711, p < 0.001. The ratio of chi-square to degrees of freedom was slightly too high: χ2/df = 5.62. However, considering the relatively large sample and the test for an exact fit, the chi-square statistic can be neglected (Little, 2013, p. 108). Cronbach’s α of the three factors ranged from 0.93 to 0.94. In combination with the high standardized factor loadings of the items in the CFA (from 0.799 to 0.938), the results show good reliability and fit of the measurement model.
To test the relationship between science subject-related self-rated content knowledge (srCK) and science subject-specific academic self-concept (ASC) (hypothesis 2), two measurement models were specified (n = 552, ratio = 3.59, CFI = 0.973, TLI = 0.971, RMSEA = 0.069), which produced the correlations of the latent factors in Table 5. The factors of the science subject-related srCK strongly correlated with the corresponding academic self-concept (rbiology = 0.86, p < 0.01, rchemistry = 0.83, p < 0.01, rphysics = 0.87, p < 0.01). In addition, srCKbiology correlated negatively with srCKphysics (r = -0.35, p < 0.01) and not at all with srCKchemistry. SrCKphysics correlated positively with srCKchemistry (r = 0.34, p < 0.01). Academic self-concepts of all three science education subjects showed nearly the same relationship with one another.
Table 5: Correlations of science subject-related srCK and academic self-concept in biology, chemistry, and physics. srCK = self-rated content knowledge, ASC = academic self-concept.
To check the effect of the studied science education subject(s) on science subject-related srCK (hypothesis 3), we conducted a structural equation model (n = 552, ratio = 2.98, CFI = 0.962, TLI = 0.967, RMSEA = 0.060) with regressions of the three subject-variables on each of the factors of science subject-related srCK. These regressions controlled for the effect of the other subjects, respectively. Table 6 a, b, and c shows that it was positively associated for each of the three factors of science subject-related srCK, if the corresponding subject was studied (βbiology = 0.63, p < 0.01, βchemistry = 0.59, p < 0.01, βphysics = 0.71, p < 0.01). In addition, all impacts – except studying chemistry on srCK in physics (βchemistry = 0.16, p < 0.01) – were negative or not significant. The strongest negative influence was that of studying biology on chemistry (βbiology = -0.22, p < 0.01).
Table 6: Regression of the subjects studied (biology/chemistry/physics) as predictors of science subject-related srCK in a) biology, b) chemistry, and c) physics. β = standardized regression coefficient, SE = standard error.
To test the influence of the final science grades in secondary school on science subject-related srCK (hypothesis 4), we conducted one structural equation model (n = 448, ratio = 3.67, CFI = 0.964, TLI = 0.969, RMSEA = 0.077) with a regression of the final grade in secondary school on the corresponding factor of srCK for each subject. The results revealed substantial impacts of the grade in biology for srCK in biology (βbiology = 0.50, p < 0.01, SE = 0.04), in chemistry for srCK in chemistry (βchemistry = 0.31, p < 0.01, SE = 0.05), and in physics for srCK in physics (βphysics = 0.49, p < 0.01, SE = 0.04).
To examine the influence of studying two science education subjects on science subject-related srCK (hypothesis 5), we conducted one structural equation model (n = 552, ratio = 5.07, CFI = 0.976, TLI = 0.975, RMSEA = 0.086) with a regression of the number of science education subjects studied on the factors of science subject-related srCK. Table 7 shows that studying two science education subjects impacted the srCK in biology (βbiology = 0.24, p < 0.01) and in chemistry (βchemistry = 0.32, p < 0.01).
Table 7: Regression of the number of science education subjects studied (one or two) as predictors of srCK in biology, chemistry, and physics. β = standardized regression coefficient, SE = standard error.
5.1 Validity and reliability of the discipline-specific measure
Regarding the factor structure of science subject-related srCK (hypothesis 1), the fit-indices of the CFA and Cronbach’s α indicate good reliability and factorial validity (Hartig, Frey, & Jude, 2012, p. 162) of the measure. We argue for divergent validity due to the low or negative correlations of the factors, as we expected three different factors (Hartig, Frey, & Jude, 2012, p. 158; Moosbrugger, Kelava, 2012, p. 17). Despite some valuable existing instruments for different contexts (see Section 2.4), the newly established measure of science subject-related srCK fills in a gap for measuring (prospective) teachers’ (self-rated) understanding of the discipline-specific core ideas of the school curricula of lower secondary German science education. As the interdisciplinary and discipline-specific curricula have the same nationwide foundation, we assume that the measure is applicable to (prospective) teachers of all German grammar and comprehensive schools. Expert-rated examples of relevance illustrated each item. The expert survey ensured content validity (Hartig, Frey, & Jude, 2012, pp. 148ff.; Moosbrugger, Kelava, 2012, p. 15). The CFA confirmed the three subject-specific factors resulting from the EFA. According to expectations, the three srCK factors for biology, chemistry, and physics correspond to the science education subjects of the operationalized core ideas of the underlying curricula (Niedersächsisches Kultusministerium, 2015). Because of these results, we argue for factorial validity (Hartig, Frey, & Jude, 2012, p. 162).
Further indicators of convergent, concurrent, and retrospective validity were gained by testing the hypotheses regarding the relations of science subject-related srCK with science subject-specific academic self-concept (hypothesis 2), the impact of the studied science education subjects (hypotheses 3 and 5), and the grades in science subjects (hypothesis 4). Due to the strong correlations of srCK in biology, chemistry, and physics with the corresponding academic self-concepts (Table 5) we argue for convergent validity, as our science subject-related srCK can be regarded as a part of science subject-specific academic self-concept on a more detailed level (Dickhäuser et al., 2002, p. 394; Oberle, 2012, p. 27). The very similar intercorrelations of the science subject-related srCK factors and science subject-specific academic self-concept factors support this theoretical assumption and, therefore, the convergent validity (Table 5). The intercorrelations of both constructs meet the expectation that academic self-concept in chemistry and physics relate more to each other than to academic self-concept in biology (Jansen et al., 2014, p. 48). Despite the strong correlations between our science subject-related srCK for (prospective) teachers and the corresponding academic self-concepts for students, the new measure is a valuable measure. It is on a more detailed level and provides more specific insights into the content of the science subject-related srCK. Hypothesis 3 provides an argument for concurrent validity, as the expected advantage of studying the science education subject of the srCK to rate (Bröll, Friedrich, 2012, pp. 182f.) is confirmed with pre-service and trainee teachers (Table 6 a, b, c). The predictive power of final grades in secondary school for biology, chemistry, and physics (hypothesis 4) supports retrospective validity, as (past) grades should relate to self-assessment of ability (Zell, Krizan, 2014, p. 116). The impact of the grade in chemistry was lower than in the other subjects. This finding could be explained by the differing structure of chemistry education in lower and upper secondary education (e.g., increasing degrees of abstractness of atomic models). In addition, hypothesis 5 revealed the impact of studying two science education subjects on srCK in biology and chemistry (Table 7). This result indicates that studying more science education subjects increases science subject-related srCK (Bröll, Friedrich, 2012, pp. 182f.) and supports concurrent validity. The sample, which only contained a few test participants studying physics together with biology or chemistry (see Section 3.1), could explain the missing impact of studying two science education subjects on the srCK in physics.
Also, our results show a dissimilarity of srCK in biology, chemistry, and physics, which supports disciplinarity and divergent validity. Based on the correlations (Table 5), we argue for distinct factors (divergent validity), as biology and physics correlated negatively, chemistry and physics correlated positively, but only on a low level, and biology and chemistry did not correlate at all. This finding is comprehensible, as most core ideas differ strongly regarding content. Especially in lower secondary education, for example, there are fewer links between biology and chemistry than in upper secondary education (e.g., neurophysiology, cell respiration). The common core idea of energy (Niedersächsisches Kultusministerium, 2015, pp. 15, 49), which was, excepting the examples, named identically in the questionnaire for both subjects (see Appendix, items C4 and P1), can contribute to explaining the correlation between chemistry and physics. Also, one item of physics had a second loading on chemistry (above 0.3) in the EFA (Table 4, item P2). Biology examines energy, too, but only as one part of one item (Appendix, item B3), probably resulting in a lower effect on the correlations. In addition to the correlations of the factors (Table 5), the impact of the science education subjects on non-corresponding srCK supports the dissimilarity (Table 6 a, b, c; three negative and two not significant effects). For example, studying biology had a negative effect on srCKchemistry (see explanation above and Table 6 a, b, and c). Beyond the corresponding science education subjects, only one positive impact emerged: studying chemistry on srCKphysics. Physical chemistry, which is a mandatory part of chemistry studies, could explain this phenomenon. In sum, the effects of the studied science education subjects support the idea of distinct factors (divergent validity), as studying a science education subject was negatively related or not at all related to non-corresponding srCK in five of six times (Table 6 a, b, c).
Overall, with few explainable deviations (e.g., hypothesis 5), the findings support the hypotheses regarding the relationship between science subject-related srCK and science subject-specific academic self-concept (hypothesis 2), and the impacts of the science education subjects studied (hypotheses 3 and 5) and the final grades in science subjects (hypothesis 4). They provide strong arguments for the convergent, concurrent, and retrospective validity of our measure of science subject-related srCK. In addition, the disciplinarity, factorial validity, and divergent validity (e.g., intercorrelations) of our measure (hypothesis 1) of science subject-related srCK are supported. Consequentially, regarding research question 1, the findings empirically support three reliable and valid factors of srCK, i.e., biology, chemistry, and physics.
5.2 Implications for interdisciplinary science teaching and teacher education
To investigate research question 2, we considered the intercorrelations between the factors of science subject-related srCK and the influence of studying science education subjects on non-corresponding factors of srCK to identify potential interdisciplinarity. Our findings emphasize the disciplinarity of srCK and the dissimilarity of srCK factors in biology, chemistry, and physics (Table 5) in lower secondary education, as they were associated positively, only on a low level, negatively, or not at all. Studying non-corresponding science education subjects underlines these differences by being mostly no advantage for science subject-related srCK (Table 6 a, b, and c). The differences in science subject-related srCK in lower secondary education render interdisciplinary science teaching difficult and indicate that studying one science education subject is insufficient for interdisciplinary science teaching. The results regarding disciplinarity of science subject-related srCK support our assumption of interdisciplinary science teaching needing the knowledge base of all three science education subjects (Engelmann 2019, pp. 25ff.).
This conclusion reveals a major issue for interdisciplinary science teaching and teacher education. Discipline-specific teacher education seems to be judged sufficient for multidisciplinary science teaching (Table 6 a, b, and c: corresponding subjects) yet insufficient for interdisciplinary science teaching (Table 6 a, b, and c: non-corresponding subjects). This finding reflects the fact that German teachers receive training to become biology, chemistry, or physics teachers but not science teachers (Neumann, Härtig et al., 2017, p. 38).
The fact that prospective teachers believe they do possess substantially less CK in science education subjects not studied during teacher training reveals a need for further qualification. Studying two science education subjects, thus having more qualification in science education, indicated higher srCK in more science education subjects (Table 7) and supports this demand. The need for qualification becomes enhanced as a lack of CK in teacher training seems to impair the development of necessary PCK (for mathematics, see Baumert et al., 2010, p. 158, 167). Therefore, teaching science education subjects that a teacher is unprepared for seems to be a problem in interdisciplinary science teaching and a primary reason why teachers (e.g., Fruböse et al., 2011, p. 436) and scientists (e.g., Bröll, Friedrich, 2012, p. 185) understandably demanded an adjustment of teacher education for science teaching.
Thus, interdisciplinary science teaching remains a challenge at the moment. However, some universities are already implementing special studies (e.g., in Karlsruhe for science and technology) to address this need. Göttingen is taking another approach, where pre-service teachers can complete a certificate focusing on interdisciplinary science teaching with 16 ECTS required in two not regularly studied science education subjects (for the optional area of their studies) (Eggert et al. 2018, pp. 52f.). The certificate includes a module with one course each regarding CK and PCK for each missing science education subject and a final teaching practice integration module (Eggert et al. 2018, p. 53). It follows the concept of building upon the specific knowledge bases of all relevant science education subjects and integrating them afterward (Eggert et al. 2018, p. 55; see Section 2.1).
Thus, regarding research question 2, interdisciplinary science teaching needs support from teacher education as the factors of science subject-related srCK (for lower secondary education) are judged as different by the test participants. In addition, the current discipline-specific teacher education does not seem to be adequate for this major additional challenge of interdisciplinary science teaching at grammar and comprehensive schools in certain classes of lower secondary education.
5.3 Limitations and future research
Regarding science subject-related srCK as a proxy indicator for the corresponding CK, it is a limitation that it does not fully equate to CK, and we were unable to test the relationship between our srCK and CK. This is a task to solve in future research. Thus, srCK allows more cautious conclusions than actual CK, but its assessment is possible within shorter periods. Further research is desirable to investigate the role of science subject-related CK for interdisciplinary science teaching. Nevertheless, our measure considered the curricula and is a proper solution for measuring science subject-related srCK. It allows a parsimonious integration of science subject-related srCK in studies in contrast to factual science subject-specific CK (Großschedl et al., 2018, p. 15), e.g., for validation purposes, if CK is not the only examined construct in a survey or if the CK of various science education subjects is required. It should further be considered that the sample included some pre-service teachers from universities where one subject is studied more in-depth (major) in undergraduate studies than the other subject (minor) and inversely in their Master of Education programs. This condition could affect the results to some degree. An effect of the monetary compensation can be denied. A structural equation model for the effect of the monetary compensation (controlled for the single science education subjects studied) only showed standardized regression coefficients smaller than 0.1. Thus, the reimbursement seems not to affect our results. Also, only a few prospective teachers with physics and a second science education subject could be integrated into our studies. This fact could be an open point for future research.
Mid-term, comparable to self-efficacy beliefs (Forsthuber et al., 2011, p. 106), the measure can contribute to evaluating and developing teacher education in science education based on evidence. It can provide hints on the effectiveness of specific interventions. The presented results can be applied to science education and its specific requirements. The potential for interdisciplinarity of other subjects has to receive an individual examination.
Besides science, other iteaching tasks without a specific interdisciplinary teacher training exist, e.g., “Gesellschaftslehre” (i.e., interdisciplinary teaching of geography, politics, and history as one subject), and Education for Sustainable Development (i.e., part of and in many subjects, such as biology, geography, and politics, but no independent subject). These interdisciplinary teaching tasks consist of different disciplines as well. Considering German teacher education (i.e., studying two subjects), it is probable, that teachers of these interdisciplinary teaching tasks have to teach subjects not studied in teacher training (e.g., for “Gesellschaftslehre”: Eggert et al. 2018, p. 52), associated with the same problems of interdisciplinary science teaching. The CK of these interdisciplinary teaching tasks can be derived from the curricula for students’ CK or possible additional normative guidelines as well.
The approach for the development of a measure for srCK in science education subjects with school curricula can be transferred to measuring srCK in “Gesellschaftslehre” and Education for Sustainable Development. For “Gesellschaftslehre” the core ideas of CK in politics (e.g., interactions and decisions), the core topics of CK in geography (e.g., changing towns) and the structuring aspects of CK in history (e.g., reign and statehood) of the discipline-specific school curricula could be used to operationalize a questionnaire of the srCK. The core topics and structuring aspects are similar to core ideas as they are recurring motifs and ensure vertical networking as well. A measure for srCK in Education for Sustainable Development could be designed similarly. The core ideas of biology (e.g., substance and energy conversion) and politics (e.g., motives and incentives) and the core topics of geography (e.g., global challenges of the 21. century) of the discipline-specific curricula could be used to operationalize this measure. Such measures could allow insights into the srCK of (prospective) teachers regarding further interdisciplinary teaching tasks without specific teacher training to identify particular potentials or problems of these teaching tasks.
We thank all the experts for their effort and all the test participants for responding to our questionnaire. We want to express special thanks to Prof. Monika Oberle for her valuable input during the development of the measure and to Prof. Sascha Schroeder for his valuable advice during manuscript preparation.
This project is part of the “Qualitätsoffensive Lehrerbildung”, a joint initiative of the Federal Government and the Länder, which aims to improve the quality of teacher training. The programme is funded by the Federal Ministry of Education and Research (reference number 01JA1617). The authors are responsible for the content of this publication.
Baumert, J., & Kunter, M. (2013). The COACTIV Model of Teachers’ Professional Competence. In M. Kunter, J. Baumert, W. Blum, U. Klusmann, S. Krauss, & M. Neubrand (Eds.), Mathematics teacher education: Vol. 8. Cognitive Activation in the Mathematics Classroom and Professional Competence of Teachers: Results from the COACTIV Project (pp. 25–48). New York: Springer.
Baumert, J. et al. (2010). Teachers’ Mathematical Knowledge, Cognitive Activation in the Classroom, and Student Progress. American Educational Research Journal, 47(1), 133–180. https://doi.org/10.3102/0002831209345157
Bröll, L., & Friedrich, J. (2012). Zur Qualifikation der Lehrkräfte für den NWA-Unterricht: - eine Bestandsaufnahme in Baden-Württemberg -. MNU Journal, 65(3), 180–186.
Broggy, J., O’Reilly, J., & Erduran, S. (2017). Interdisciplinarity and Science Education. In K. S. Taber & B. Akpan (Eds.), Science Education: An International Course Companion (pp. 81–90). Rotterdam: SensePublishers.
Brown, T. A. (2006). Confirmatory Factor Analysis for Applied Research. New York, NY: The Guilford Press.
Carlson, J., & Daehler, K. R. (2019). The Refined Consensus Model of Pedagogical Content Knowledge in Science Education. In A. Hume, R. Cooper, & A. Borowski (Eds.), Repositioning Pedagogical Content Knowledge in Teachers’ Knowledge for Teaching Science (pp. 77–92). Singapore: Springer Singapore.
Cohen, J. et al. (2003). Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences (3. ed.). Mahwah, NJ: Lawrence Erlbaum.
Conway, J. M., & Huffcutt, A. I. (2003). A Review and Evaluation of Exploratory Factor Analysis Practices in Organizational Research. Organizational Research Methods, 6(2), 147–168. https://doi.org/10.1177/1094428103251541
Department for Education (2013). Science programmes of study: key stage 3: National curriculum in England. Retrieved from https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/335174/SECONDARY_national_curriculum_-_Science_220714.pdf
Dickhäuser, O. et al. (2002). Die Skalen zum akademischen Selbstkonzept: Konstruktion und Überprüfung eines neuen Instrumentes. Zeitschrift für Differentielle und Diagnostische Psychologie, 23(4), 393–405. https://doi.org/10.1024//0170-1718.104.22.1683
Dörges, A. (2001). Erfahrungen mit dem integrierten naturwissenschaftlichen Unterricht. MNU Journal, 54(4), 230–232.
Eggert, S. et al. (2018). Herausforderung Interdisziplinäres Unterrichten in der Lehrerbildung: Das Göttinger Zertifikatsmodell. journal für lehrerInnenbildung, 18(3), 51–55.
Eisner, B. et al. (2019). Gemeinsamer Referenzrahmen für Naturwissenschaften (GeRRN): Mindeststandards für die auf Naturwissenschaften bezogene Bildung. Ein Vorschlag. Retrieved from https://www.mnu.de/images/publikationen/GeRRN/MNU_GeRRN_3.pdf
Engelmann, P. (2019). Fächerübergreifende Naturwissenschaften in der Lehrerfortbildung: Eine Didaktische Rekonstruktion. Dissertation. Friedrich-Schiller-Universität Jena. Retrieved from https://www.db-thueringen.de/servlets/MCRFileNodeServlet/dbt_derivate_00045048/dissengelmann.pdf
Forsthuber, B. et al. (2011). Science Education in Europe: National Policies, Practices and Research. Brussels: Education, Audiovisual and Culture Executive Agency. https://doi.org/10.2797/7170
Fruböse, C. et al. (2011). Unterricht im integrierten Fach Naturwissenschaften: Erfahrungen aus gymnasialer Sicht. MNU Journal, 64(7), 433–439.
Großschedl, J., Mahler, D., & Harms, U. (2018). Construction and Evaluation of an Instrument to Measure Content Knowledge in Biology: The CK-IBI. Education Sciences, 8, 145. https://doi.org/10.3390/educsci8030145
Handtke, K., & Bögeholz, S. (2019). Self-Efficacy Beliefs of Interdisciplinary Science Teaching (SElf-ST) Instrument: Drafting a Theory-based Measurement. Education Sciences, 9(4), 247. https://doi.org/10.3390/educsci9040247
Hardy, G. (2014). Academic Self-Concept: Modeling and Measuring for Science. Research in Science Education, 44(4), 549–579. https://doi.org/10.1007/s11165-013-9393-7
Hartig, J., Frey, A., & Jude, N. (2012). Validität. In H. Moosbrugger & A. Kelava (Eds.), Testtheorie und Fragebogenkonstruktion (pp. 143–171). Berlin: Springer.
Jansen, M. et al. (2014). Interdisziplinäre Beschulung und die Struktur des akademischen Selbstkonzepts in den naturwissenschaftlichen Fächern. Zeitschrift für Pädagogische Psychologie, 28(1-2), 43–49. https://doi.org/10.1024/1010-0652/a000120
Käpylä, M., Heikkinen, J.‐P., & Asunta, T. (2009). Influence of Content Knowledge on Pedagogical Content Knowledge: The case of teaching photosynthesis and plant growth. International Journal of Science Education, 31(10), 1395–1415. https://doi.org/10.1080/09500690802082168
Kleickmann, T. et al. (2013). Teachers’ Content Knowledge and Pedagogical Content Knowledge: The Role of Structural Differences in Teacher Education. Journal of Teacher Education, 64(1), 90–106. https://doi.org/10.1177/0022487112460398
Kleickmann, T. et al. (2018). Teacher Knowledge Experiment: Conditions of the Development of Pedagogical Content Knowledge. In D. Leutner, J. Fleischer, J. Grünkorn, & E. Klieme (Eds.), Competence Assessment in Education: Research, Models and Instruments (pp. 111–129). Cham: Springer International Publishing. https://doi.org/10.1007/978-3-319-50030-0_8
Kultusministerkonferenz (2004). Bildungsstandards im Fach Biologie für den Mittleren Schulabschluss: Beschluss vom 16.12.2004. Retrieved from https://www.kmk.org/fileadmin/Dateien/veroeffentlichungen_beschluesse/2004/2004_12_16-Bildungsstandards-Biologie.pdf
Labudde, P. (2014). Fächerübergreifender naturwissenschaftlicher Unterricht – Mythen, Definitionen, Fakten. Zeitschrift für Didaktik der Naturwissenschaften, 20(1), 11–19. https://doi.org/10.1007/s40573-014-0001-9
Little, T. D. (2013). Longitudinal Structural Equation Modeling. New York: The Guilford Press.
Marsh, H. W., & Shavelson, R. (1985). Self-Concept: Its Multifaceted, Hierarchical Structure. Educational Psychologist, 20(3), 107–123. https://doi.org/10.1207/s15326985ep2003_1
Metzger, S. (2010). Die Naturwissenschaften fächerübergreifend vernetzen. In P. Labudde (Ed.), Fachdidaktik Naturwissenschaft: 1.-9. Schuljahr (pp. 29–44). Bern: Haupt.
Ministerium für Kultus, Jugend und Sport Baden-Württemberg (2004). Bildungsplan 2004: Realschule. Retrieved from http://www.bildungsplaene-bw.de/site/bildungsplan/get/documents/lsbw/Bildungsplaene/Bildungsplaene-2004/Bildungsstandards/Realschule_Bildungsplan_Realschule_Gesamt.pdf
Ministerium für Kultus, Jugend und Sport Baden-Württemberg (2016). Biologie, Naturphänomene und Technik (BNT): Bildungsplan des Gymnasiums. Retrieved from http://www.bildungsplaene-bw.de/site/bildungsplan/get/documents/lsbw/export-pdf/depot-pdf/ALLG/BP2016BW_ALLG_GYM_BNT.pdf
Moosbrugger, H., & Kelava, A. (2012). Qualitätsanforderungen an einen psychologischen Test (Testgütekriterien). In H. Moosbrugger & A. Kelava (Eds.), Testtheorie und Fragebogenkonstruktion (pp. 7–26). Berlin: Springer.
Neumann, K. et al. (2017). Science Teacher Preparation in Germany. In J. E. Pedersen, T. Isozaki, & T. Hirano (Eds.), Model Science Teacher Preparation Programs: An International Comparison of What Works (pp. 29–52). Charlotte, NC: Information Age Publishing.
Niedersächsisches Kultusministerium (2012). Kerncurriculum für die integrierte Gesamtschule Schuljahrgänge 5-10: Naturwissenschaften. Retrieved from http://db2.nibis.de/1db/cuvo/datei/kc_2012_igs_nws_i.pdf
Niedersächsisches Kultusministerium (2015). Kerncurriculum für das Gymnasium Schuljahrgänge 5-10: Naturwissenschaften. Retrieved from http://db2.nibis.de/1db/cuvo/datei/nw_gym_si_kc_druck.pdf
Oberle, M. (2012). Politisches Wissen über die Europäische Union: Subjektive und objektive Politikkenntnisse von Jugendlichen. Wiesbaden: Springer.
Qualitäts- und UnterstützungsAgentur - Landesinstitut für Schule (2011). Basiskonzept Struktur und Funktion. Retrieved from https://www.schulentwicklung.nrw.de/materialdatenbank/material/download/3129
Ramm, G. et al. (2006). PISA 2003: Dokumentation der Erhebungsinstrumente. Münster: Waxmann.
Sadler, P. M. et al. (2013). The Influence of Teachers’ Knowledge on Student Learning in Middle School Physical Science Classrooms. American Educational Research Journal, 50(5), 1020–1049. https://doi.org/10.3102/0002831213477680
Schmiemann, P., & Lücken, M. (2014). Validität - Misst mein Test, was er soll? In D. Krüger, I. Parchmann, & H. Schecker (Eds.), Methoden in der naturwissenschaftsdidaktischen Forschung (pp. 107-118). Berlin: Springer.
Shavelson, R. J., Hubner, J. J., & Stanton, G. C. (1976). Self-Concept: Validation of Construct Interpretations. Review of Educational Research, 46(3), 407–441. https://doi.org/10.3102/00346543046003407
Velthuis, C., Fisser, P., & Pieters, J. (2014). Teacher Training and Pre-service Primary Teachers’ Self-Efficacy for Science Teaching. Journal of Science Teacher Education, 25(4), 445–464. https://doi.org/10.1007/s10972-013-9363-y
Wheaton, B. et al. (1977). Assessing Reliability and Stability in Panel Models. Sociological Methodology, 8, 84–136.
Yangin, S., & Sidekli, S. (2016). Self-Efficacy for Science Teaching Scale Development: Construct Validation with Elementary School Teachers. Journal of Education and Training Studies, 4(10), 54–69. https://doi.org/10.11114/jets.v4i10.1694
Yilmaz-Tuzun, O. (2008). Preservice Elementary Teachers’ Beliefs About Science Teaching. Journal of Science Teacher Education, 19(2), 183–204. https://doi.org/10.1007/s10972-007-9084-1
Zell, E., & Krizan, Z. (2014). Do People Have Insight Into Their Abilities?: A Metasynthesis. Perspectives on Psychological Science, 9(2), 111–125. https://doi.org/10.1177/1745691613518075
Self-rated content knowledge of biology, chemistry, and physics
Henceforth, we ask you to self-rate your content knowledge relating to teaching science (teaching of biology, chemistry, and physics as one subject). This information is meaningful, as you may find yourself in the situation of teaching all three science education subjects interdisciplinarily (e.g., at a comprehensive school).
„I think that I have enough content knowledge ...
[B0]… to teach the biological parts of the interdisciplinary subject  well.“
[C0]… to teach the chemical parts of the interdisciplinary subject science well.“
[P0]… to teach the physical parts of the interdisciplinary subject science well.“
Biology specific items
“I know very much about the core idea of ...
[B1]… compartmentalization (e.g., cell theory, such as organelles and animal and plant cell, and the functional division of types of tissue and organs in the organism, and the principle of competitive exclusion).”
[B2]… control and regulation (e.g., physiological regulation, such as regarding body temperature or hormones, and ecological interactions such as predator-prey relationship).”
[B3]… substance and energy transformation (e.g., photosynthesis, digestion, and cell respiration and enzymes, and substance and energy flow within an ecosystem such as within food webs).”
[B4]… information and communication (e.g., information paths within the organism, such as stimulus conduction, stimuli, and sensory organs, and information intake and exchange of information such as by social behavior).“
[B5]… reproduction (e.g., sexual and asexual procreation, technical cloning, recombination, and the manifestation of genetic information, such as genes, phenotype, mutation).”
[B6]… variability and adaptation (e.g., diversity of species, annidation, and processes of selection, such as breeding and evolutionary processes).”
[B7]… history and relationship (e.g., groups of vertebrates, pedigree analysis, and homology and analogy).”
Chemistry specific items
“I know very much about the core idea of ...
[C1]… matter-particle (e.g., atomic models, substance properties, such as pH-value or the solubility, separation processes such as chromatography, and pure substances and mixtures of substances).”
[C2]… structure-property (e.g., intermolecular interactions, such as dipole-dipole, van der Waals, and hydrogen bonds, and substance classes and groups, such as metals, non-metals, and salts).”
[C3]… chemical reaction (e.g., characteristics of chemical reactions, such as educts, products, energy conversion, and redox reactions, and acid-base reactions, and formation of reaction equations).”
[C4]… energy (e.g., activation energy, catalyst, ionization energy, and exothermic and endothermic reactions).”
Physics specific items
„I know very much about the subject area of ...
[P1]… energy (e.g., principle of conservation of energy, types of energy).“
[P2]… thermodynamics (e.g., Kelvin scale, ideal gas, cycles, efficiency).”
[P3]… magnetism and electricity (e.g., molecular magnets, generator, Kirchhoff’s first and second law, current, voltage and resistor, pn-junction).”
[P4]… mechanics (e.g., powers, gravitational constant g, uniform movement, p-t- and v-t-graphs).”
[P5]… phenomenological optics (e.g., reflection, dispersion, refraction, lenses, and colored light).”
[P6]… atomic and nuclear physics (e.g., nuclear power, radioactive decay, α-, β-, and γ-radiation, isotopes, and the nuclear atom model).”
In the following, interdisciplinary science also means integrated science (biology, chemistry, and physics as one subject; Labudde, 2014, p. 15).
The reference contains all three curricula for biology, chemistry, and physics. Henceforth, subject areas and core ideas are summarized as core ideas.
Due to the focus on primary education, the topics are not comparable with topics, not to mention core ideas, of secondary education and, thus, not further considered.
As we focus at least on the level of subject-specific academic self-concept, and not the higher-order academic self-concept (e.g., verbal/academic self-concept), more recent models, like that presented by Marsh and Shavelson (1985, p. 120), are not explicitly taken into account.
After confirming the factors with the CFA, we now call srCK science subject-related srCK instead of only science-related.
The subjects of biology, chemistry, and physics are taught as one subject “science“ at the comprehensive school from fifth to tenth class.