Combining Virtual and Hands-on Lab Work in a Blended Learning Approach on Molecular Biology Methods and Lab Safety for Lower Secondary Education Students

Abstract

Molecular biology is becoming increasingly important in everyday life. Virtual and authentic hands-on out-of-school labs have proven effective in teaching it, especially for high-achieving older learners. We developed a blended learning approach integrating the advantages of both methods for lower secondary education students. In a quasi-experimental-control group study we accessed its impact on declarative knowledge in laboratory safety and molecular biology methods, along with perceived authenticity, in comparison to teaching through a virtual desktop simulation, a hands-on wet lab, or regular teaching with a worksheet. N = 229 students took part in the pre–post-follow-up data collection. The results showed a significant difference in laboratory safety knowledge between the blended learning and the other intervention groups. The wet lab group differed significantly in molecular biology knowledge and perceived authenticity from the other intervention groups. Learning success was positively correlated with the authenticity in the overall sample. The blended learning group’s reduced authenticity may contribute to lower learning success in molecular biology.

1. Introduction

Molecular biology has not only revolutionized the field of biology in recent decades but is also becoming increasingly important to everyday life. Risk assessment stemming from the application of molecular biology methods, such as in food production, poses a challenge for the population (Frewer et al., 2013). In particular, the COVID-19 pandemic has underscored that deficiency in understanding molecular biology methods (Fieselmann et al., 2022). It is unsurprising that a significant portion of the population struggles to comprehend the workings of a PCR test or an mRNA vaccine without prior exposure to molecular biology in their education. Despite Kary Mullis earning the Nobel Prize for developing the PCR as early as 1993, molecular biology is still perceived as a ‘young’ discipline gradually making its way into the curriculum. In many educational curricula (for example, in the state of Baden-Württemberg, Germany, where this study was conducted), molecular biology methods typically remain the domain of academically stronger learners in upper secondary education, and research predominantly focuses on these students or students in tertiary education (Savvides, 2018; Schäfers et al., 2020; Scharfenberg et al., 2007). However, for a society where every citizen can contribute to knowledge-based decisions in political or personal debates, it is imperative to devise solutions for imparting molecular biology education to students who may not pursue advanced academic paths beyond lower secondary education.

2. Literature Review

In addition to the complexity of the content, practicing teachers face their most significant challenge in the shortage of time and adequate resources when teaching molecular biology (Haberbosch et al., in press). Modern biology education must not only convey knowledge about key areas such as molecular biology to all learners, but also prepare those who wish to pursue careers in these fields, enabling them to participate in the industrialization of biology (Friedman & Ellington, 2015) and gain insights into a rapidly growing job market. To overcome these challenges and achieve these objectives, virtual laboratories and out-of-school labs have been a focal point of research in recent years.

2.1. Advantages and Disadvantages of Virtual Laboratories in Science Education

In particular, in vocational training professions within the field of molecular biology (such as a biotechnical assistant), the safe execution of experiments plays a significant role. Experimenting proficiency encompasses not only the effective planning and evaluation of experiments but also the responsible and, above all, safe execution of them. Therefore, all students engaging in experiments in classrooms or learning labs must receive an appropriate safety introduction. However, traditional safety briefings, in the form of a presentation of the key laboratory rules through a teacher lecture without opportunities for interaction, are often perceived as boring (Gublo, 2003). Wu et al. demonstrated that science students generally have very limited knowledge of laboratory safety and, for instance, do not recognize important hazard pictograms according to the Globally Harmonized System (GHS) (Wu et al., 2021). Secondary school students also demonstrate deficiencies in their comprehension of laboratory safety, coupled with a lack of confidence in securely carrying out laboratory tasks (Caymaz, 2021). Savvides investigated the impact of varying levels of immersion in digital laboratory safety introductions for science students (Savvides, 2018). A 10-item scale was utilized to evaluate declarative knowledge of lab safety. The items were collaboratively developed with Labster, a provider of virtual simulations, and validated in partnership with university professors. Participants were classified into three immersion levels: high immersion, utilizing virtual reality glasses; medium immersion, involving a desktop simulation; and low immersion, comprising a LabSafety video accompanied by a handout illustrating lab safety procedures. Significant differences in knowledge of lab safety were observed in both the post-test and follow-up test between the high and medium immersion levels compared to the low immersion setting. Notably, no discernible disparities were found in knowledge (and perceived presence) between the use of VR glasses and desktop simulations. The virtual reality simulation also significantly enhanced interest in laboratory safety and self-efficacy for working safely in the laboratory (Makransky et al., 2020).

Moreover, for many schools, the acquisition of real molecular biology equipment to perform experiments within the classroom is too expensive. A possible alternative in this case is the use of well-researched virtual simulations. Virtual simulations offer the advantage of avoiding the use of harmful reagents (e.g., ethidium bromide or SDS), allowing learners to focus solely on experimentation (Lynch & Ghergulescu, 2017). They often visualize in vitro processes, enabling a deeper understanding of the subject matter (Strømme & Mork, 2021) and can positively impact content-related interest (Thisgaard & Makransky, 2017). A systematic review of virtual laboratory simulations in the field of biotechnology showed that virtual labs are more effective for learning basic facts about biotechnology compared to traditional teaching methods, such as lectures (Mercado & Picardal, 2023). Gauthier (2024) demonstrated that undergraduate biology students particularly benefit from virtual simulations because they allow for making mistakes. This process of encountering and addressing errors encourages students to question everyday notions and facilitates a shift toward scientific understanding in the area of molecular emergence. Yu et al. (2022) further demonstrated that biology master’s students, when faced with unexpected results from simulation-generated experiments rather than their own experimentally generated data, are more likely to reflect on biological reasons for why their hypothesis might be wrong, rather than doubting their own experimental skills. In the comparison between digital labs and hands-on lab activities, the results regarding the acquisition of declarative knowledge are not conclusive. Virtual labs also have drawbacks, including handling and gaining experience with uncharacteristic data, and some students may feel inadequately prepared for dealing with real laboratory equipment (Lynch & Ghergulescu, 2017). In the constructivist framework, learning is viewed as a social process facilitating collaborative knowledge construction (Lave & Wenger, 1991). However, Schäfers et al. (2020) found that the use of digital labs diminishes social interactions among learners.

2.2. Authentic Real Laboratory Experiences Foster Scientific Interest

Despite the mentioned advantages of the use of virtual simulations for teaching laboratory safety rules, students suggest that the operation of safety facilities can be better learned directly from the object (Poensgen et al., 2021). Digital laboratory introductions (which do not precisely replicate the real on-site laboratory environment) cannot completely replace on-site instruction anyway. The location of crucial points, such as fire extinguishers, eye showers, chemical waste disposal, or emergency gathering points, must always be specifically indicated for each room. Moreover, on-site safety briefings can also be effective if they lead to a high level of engagement through hands-on breakout safety activities (Nephew & Sunasee, 2021). In order to provide students (especially those in upper secondary education) with opportunities to gain experience in molecular biology, numerous out-of-school labs have been established in recent decades. These labs aim, among other things, to counteract the declining interest in natural sciences (Glowinski, 2007). Schüttler et al. (2021) compared a school- and a university-based student laboratory, both utilizing high-end and low-cost equipment, to examine the influence of authentic learning environments and laboratory equipment on situational interest. They understand interest as an interaction between a person and an object, with situational interest referring to the state of being interested in something within a specific situation, which can either be brief or serve as the foundation for long-lasting individual interest. The study demonstrated that the type of equipment used—rather than the learning location—played a key role in fostering situational interest. Itzek-Greulich et al. (2015) also found no difference in students’ achievement between an out-of-school laboratory and its corresponding implementation in a school setting. The anticipated benefits of out-of-school laboratories, such as fostering interest in natural sciences of the next generation, do not arise automatically. Instead, it is essential to create authentic learning environments in these settings as well, because authenticity correlates positively with the interest (Glowinski, 2007; Pawek, 2009) and the subjective assessment of learning success (Glowinski, 2007) of upper secondary education (grammar school) students. However, there is little information on the effects of out-of-school molecular biology laboratories on younger, lower-achieving students at the lower secondary level (comprehensive schools). Gericke et al. (2023) conducted a systematic review on laboratory work in secondary schools and concluded that hands-on activities can also positively impact learning outcomes. However, the teacher’s guidance is identified as the crucial factor in the success of laboratory work. Hofstein and Lunetta (2004) also showed that hands-on experiences in the laboratory enable a better understanding and application of abstract concepts, while also promoting critical thinking, problem-solving skills, and teamwork.

2.3. Exploring Blended Learning Concepts in Science Education

Teaching with both virtual and wet labs has the advantages and disadvantages described above. Therefore, it is not the (digital) medium itself, but the way it is used, that is crucial for the success of science teaching. The RAT model (Hughes et al., 2006) provides a framework to assess the added value of using digital media in the classroom. It categorizes whether a digital medium merely substitutes an analog one (replacement), improves the effectiveness of teaching (amplification), or creates new learning possibilities (transformation). Blended learning is an educational concept that combines the benefits of virtual teaching and face-to-face instruction (Hafer, 2021). The advantages of this combination include greater flexibility in time management, as students can spend more or less time engaging with the online content depending on their needs. Additionally, they have the option to either spend more time asking questions on-site or independently search for answers during the online phase before continuing their learning (Caravias, 2018). However, blended learning particularly benefits students with strong self-regulation skills, as they tend to engage more deeply with the online learning materials (Tempelaar et al., 2009). The level of digital literacy is also related to the success of using blended learning systems (Tang & Chaw, 2016).

The RAT model can be used to decide for each teaching phase whether virtual or practical laboratory work offers greater added value for conveying the necessary declarative knowledge on laboratory safety and molecular biology. The concept of declarative knowledge is based on Anderson et al.’s (2001) adaptive control of thought theory, which distinguishes between declarative and procedural knowledge. Declarative knowledge (‘knowing what’) includes facts, concepts, and theories, while procedural knowledge (‘knowing how’) relates to the practical execution of actions. While procedural knowledge is crucial for laboratory work (e.g., performing and adjusting PCR protocols) (Sahdra & Thagard, 2003), understanding concepts and facts is important for the general population to assess the potential and risks of molecular biology applications.

The combination of virtual and practical lab work in a blended learning concept could lead to a synergistic integration of the best elements of both approaches and thus improve the quality of teaching and helping students to build greater declarative knowledge. On the other hand, the transition between virtual and wet labs could potentially have a negative impact on the authenticity of the learning environment, which is a key factor influencing the success of learning labs.

Additionally, the transition between virtual and real lab environments could pose a challenge for the cognitive load of the learners. Cognitive load can be divided into three types: intrinsic cognitive load, extraneous cognitive load, and germane cognitive load (Sweller et al., 1998). Extraneous cognitive load refers to the burden caused by the instructional design of the learning task and the learning conditions. It is determined by the presentation and organization of the learning material (Sweller et al., 2019). (Scharfenberg & Bogner, 2010) conducted a study comparing two sets of instructions in an out-of-school molecular biology learning lab with grammar school students. Their goal was to reduce cognitive load and analyze its impact on learning outcomes. They divided the cohort into two groups. In the intervention group, students engaged in four focused discussions of about 5 min before the experimental work in the laboratory to share their ideas and understand the relevance of the upcoming experimental phase. In contrast, the control group received optimal solutions regarding the relevance of each practical step of the experiment on worksheets from the teacher. The intervention group demonstrated lower cognitive load during the discussion phase of the experiment and achieved better long-term learning outcomes on the follow-up test conducted six weeks later compared to the group with higher cognitive load.

People with a higher level of education generally exhibit better cognitive performance. Furthermore, individuals with a high school diploma as their highest level of education exhibit optimal cognitive performance at the age of 17 (Guerra-Carrillo et al., 2017). Students in the 9th and 10th grades of lower secondary education are not yet mature enough to achieve their peak cognitive abilities and also aspire to a lower educational level compared to the relatively older graduates of grammar schools. Consequently, taking cognitive loads into account becomes especially crucial when designing learning interventions for students in lower secondary education.

3. Research Question

Although the combination of virtual and wet labs is often suggested in the literature (Mercado & Picardal, 2023), we are not aware of any study that has investigated a blended learning approach to lab safety or molecular biology teaching in lower secondary education. Our research questions in this regard are as follows:

• (i). Are blended learning concepts in out-of-school labs more effective in building declarative knowledge of laboratory safety and molecular biology compared to working solely in virtual or wet labs, or compared to regular teaching?

• (ii). Does combining virtual and wet labs in a blended learning concept reduce perceived authenticity?

• (iii). Is there a correlation between the perceived authenticity and knowledge gain in the areas of molecular biology and laboratory safety among lower secondary education students?

4. Methodology

In addressing our research questions, significant emphasis was placed on solely varying the medium through which students worked in class while ensuring effective control over other variables that could influence learning outcomes (e.g., prior knowledge, physical environment, or teacher’s theory explanations). To ensure optimal participation of all learners, a 60-min preparatory lesson on the molecular structure of DNA was conducted in schools for all interventions, which also ensured comparable prior knowledge across groups. In all interventions, the theory of molecular biology methods was explained through instructional videos to minimize potential bias from the teacher, who is also the first author of this study. These videos were developed at a level suitable for lower secondary education students in preparation for the study. All interventions were carried out in an out-of-school learning laboratory. On the one hand, this allowed us to control for important factors such as differences in the equipment (and authenticity) of the classrooms. On the other hand, as the control group also moved to an alternative learning environment (although this was not strictly required), we could ensure that the location of the instruction (in-school versus out-of-school) did not influence the learning outcomes.

During the introduction to laboratory safety, all lower secondary school students learned to recognize chemical hazard pictograms, understand how to handle corrosive chemicals, and familiarize themselves with protective clothing and suitable emergency measures, among other safety protocols. The focus then shifted to molecular biology methods such as PCR and gel electrophoresis. In a mock murder case, students were asked to solve the mystery using DNA evidence from the crime scene and three suspects based on their genetic fingerprints.

The design of all four interventions focused on quality criteria for effective biology teaching (Schaal et al., 2022) derived from the literature. The emphasis was on cognitive activation—encouraging students to think critically by formulating their own hypotheses and developing ideas on how to test them—and fostering communication skills, which involved discussing their thoughts and learning to deal with the many technical terms used in molecular biology. In the regular teaching group (RT), students were introduced to laboratory rules through a teacher lecture, supported by videos of experiments that all other groups conducted independently in the virtual or wet lab. To solidify their knowledge, students answered multiple-choice questions presented by the teacher. They had to agree on an answer in small groups and then justify it in the plenary. The application of molecular biology methods to solve crimes was conducted through a worksheet, which students also completed in small groups. In this process, students assisted the police officer ‘Judy Riddle’ in solving the case by applying PCR and gel electrophoresis. Subsequently, they had to convince a judge of their competence in court by flawlessly explaining the molecular biology methods. In case of comprehension difficulties, additional tasks were provided with step-by-step explanations of the molecular biology methods. In the virtual lab group (VL), a 3D desktop simulation provided by Labster was used for safety training, providing a gamified approach to learning laboratory rules while virtually solving a crime using PCR and gel electrophoresis. The simulation included animations of the in vitro processes and also provided theory via a virtual tablet, including multiple choice questions for assessment. The wet lab group (WL) carried out hands-on experiments such as pH determination and acid neutralization experiments during their safety briefing. To solve the crime, they independently initiated a PCR. Due to time constraints, pre-prepared PCR products were provided for subsequent separation and visualization of the DNA mixture in agarose gel electrophoresis. In the blended learning group (BL), the decision between hands-on and virtual laboratory work was guided by the RAT model. Whenever the use of the digital lab enabled an amplification or transformation of the lesson, the digital lab replaced the hands-on experimentation. The students started the safety training in the virtual lab to virtually experience the consequences of not following the safety rules (in this specific example, experimenting without lab goggles) (an opportunity they would not have in the real world and therefore a transformation of teaching). The subsequent practical experiments on acid neutralization could also be carried out using safe reagents. According to the RAT model, this would only be a replacement, which is why they were carried out in the wet lab instead of the virtual. In the molecular biology lab, students performed the PCR virtually to benefit from animations illustrating the rather complex in vitro processes and direct feedback to enhance their understanding through multiple-choice questions (an amplification of teaching). Following the virtual PCR, real PCR products were provided to allow students to perform gel electrophoresis in the wet lab. The virtual lab would only be a replacement, while the practical execution offered the advantage of making the otherwise abstract molecular biology ‘visible’ in the experimental results. Table 1 provides an overview of the sequence of the four interventions.

4.1. Research Design

A quasi-experimental intervention study was conducted using a pre-post-follow-up design. After the preparatory lessons, all learners completed the pre-test as online questionnaire. Subsequently, all learners visited the learning lab and participated in one of the four interventions. Immediately after the interventions, learners completed the post-test, and six weeks later at the schools, they completed a follow-up test to assess the long-term effects. The questionnaire was conducted on iPads at all time points. Prior to participation, all learners (and their legal guardians in the case of minors) gave written consent that they had been informed that participation was voluntary and anonymous and that they agreed to the collection, processing, analysis and publication of data.

4.2. Study Sample

A total of 16 classes from the ninth or tenth year of lower secondary education (comprehensive schools) in a large German city (>500.000 inhabitants) participated in the study. Participants were randomly assigned to different intervention groups within their class cohort. A total of N = 327 students participated in the intervention, and N = 229 completed questionnaires across all measurement points were used for analysis. Participant allocation was carried out using an anonymous research code. The high dropout rate can be attributed to illness during the winter months of 2021 and 2023, in which the study was conducted. The mean age was 15.5 years (SD = 0.9). 49.2% of the participants identified as male, 45.0% as female and 5.2% did not identify within the binary gender system.

4.3. Measurement Instruments

The declarative knowledge of laboratory safety was assessed using a scale consisting of ten single-choice items. Savvides (2018) demonstrated a test–retest reliability correlation coefficient of r = 0.72. One item from the original scale had to be removed because it was not clearly solvable in our opinion. The questions were partially derived from Labster’s simulation on laboratory safety. To minimize any potential advantage of the virtual lab group in solving these items after the intervention, these questions were also integrated into the safety briefing of the other groups.

We could not find an appropriate scale for declarative knowledge of molecular biology methods, in particular PCR and gel electrophoresis, that met the learning objectives of lower secondary education. Therefore, in preparation for the main study, we constructed a scale in a two-step process. First, 25 single-choice items on PCR, gel electrophoresis, and genetic fingerprinting were developed. These items were tested with N = 157 teaching in secondary schools bachelor students at the University of Education Ludwigsburg. From the item pool, 4 items on PCR, 5 items on gel electrophoresis, and 1 item on genetic fingerprinting were selected based on their discriminatory power. Subsequent reliability estimation using Cronbach’s alpha yielded α = 0.76, indicating acceptable reliability (Blanz, 2021; Cronbach, 1951). The scale was then tested with N = 405 secondary school students after they had completed a molecular biology workshop in a learning laboratory. The item on genetic fingerprinting had to be excluded, as not all classes attended the workshop on this topic. Also, with secondary school students, an acceptable reliability of α = 0.77 was observed. In the confirmatory factor analysis, the items for ’PCR’ and ’gel electrophoresis’ were assessed for their fit to the previously identified factors. The results showed an acceptable model fit (χ² = 122.57, df = 26, p = 0.51, CFI = 0.98, RMSEA = 0.03). At all measurement points, both declarative knowledge about laboratory safety and molecular biology were assessed. For both variables, a maximum score of 10 points could be achieved. Within the post-test only, authenticity of the teaching was assessed using a five-point Likert scale, constructed by Engeln (2004) and further developed by Pawek (2009) (α = 0.80).

4.4. Data Analysis

Differences in declarative knowledge of laboratory safety and molecular biology among the intervention groups and across the measurement time points were identified conducting a repeated measures analysis of variance (rmANOVA). Prior to analysis, we assessed the assumption of variance homogeneity using Levene’s test. Additionally, we checked for sphericity through Mauchly’s test. In the dataset related to molecular biology knowledge, we observed a violation of sphericity (W = 0.946, p = 0.002), necessitating the application of a Greenhouse–Geisser correction. In cases where significant within-subject and between-subject effects were observed, post hoc tests were conducted, and we applied a Tukey correction to account for multiple comparisons. Cohen’s d was calculated to determine the effect size.

As the Shapiro–Wilk test revealed non-normal distribution of the data on the authenticity of the instruction (p < 0.001), a Kruskal–Wallis test was conducted to examine whether a significant difference exists in the perception of authenticity among the intervention groups. Subsequently, pairwise comparisons were carried out using the Dwass–Steel–Critchlow–Fligner method to investigate which specific groups differed significantly.

The relationship between perceived authenticity and knowledge gain (calculated as the post- or follow-up test score minus the pre-test score) in the realms of molecular biology and laboratory safety were examined using a two-tailed partial correlation. Without prior knowledge, the mean in the pre-score was calculated to be 2.5, considering the random probability of guessing (0.25) over 10 items. Since the obtained scores in the pre-test (M = 2.93, SD = 1.41) only slightly deviated from this (and did not show a significant difference between the intervention groups), they are primarily attributable to guessing. Therefore, in the partial correlation, we controlled for the influence of the pre-score. The interpretation of effect sizes is based on Döring and Bortz (2016). Since we identified significant differences in laboratory safety knowledge between the groups in the pre-test, we conducted an additional ANCOVA following the rmANOVA. This analysis accounted for prior knowledge (score at the pre-measurement time point) as a covariate to identify differences between the treatments.

5. Results

Table 2 presents the means and standard deviations of the experimental groups across the three measurement time points for declarative knowledge of lab safety, molecular biology, and perceived authenticity.

The development of declarative knowledge of laboratory safety in the four interventions is illustrated in Figure 1. Declarative knowledge of laboratory safety differed significantly among the interventions (F(3, 225) = 6.67, p < 0.001, η²p = 0.082). The interaction between the interventions and the repeated measures factor is also significant (F(6, 450) = 2.16, p = 0.047, η²p = 0.028). The BL, VL, and WL groups show highly significant differences (all: p < 0.001) between pre- and post-test in the declarative knowledge of laboratory safety with large effect sizes (BL: Cohen’s d = −1.59; VL: Cohen’s d = −1.745; WL: Cohen’s d = −1.31). The differences between pre- and follow-up tests in these three intervention groups are also highly significant (all: p_Tukey < 0.001), with large to medium effect sizes (BL: Cohen’s d = −0.73; VL: Cohen’s d = −0.999; WL: Cohen’s d = −0.870). The regularly teaching group also shows a highly significant difference between the pre- and post-test with a large effect size (p_Tukey < 0.001, Cohen’s d = −1.354). However, in the RT group, the difference between the declarative knowledge of lab safety between pre- and follow-up test is significant but corresponds only to a small effect size (p_Tukey = 0.022, Cohen’s d = −0.448). The analysis also revealed a significant difference between the pre-test of the BL and VL group (t(225) = 3.85, p = 0.008). The ANCOVA, which accounted for prior knowledge as a covariate, revealed a significant effect of the treatment, F(3, 224) = 17.23, p < 0.001, η²p = 0.088, as well as prior knowledge, F(1, 224) = 16.50, p < 0.001, η²p = 0.069. The overall model was significant, F(4, 224) = 11.53, p < 0.001, explaining a substantial portion of the variance in post-test scores. Subsequent post hoc tests indicated significant differences between the BL group and the RT group (p < 0.001), the WL group (p < 0.001), and the VL group (p = 0.016). Additionally, the VL group differed significantly from the RT group (p = 0.047).

The declarative knowledge of molecular biology is depicted in Figure 2. Significant differences were observed between interventions (F(3, 225) = 3.81, p = 0.011, η²p = 0.048). Subsequent post hoc tests showed that only the WL group differed significantly from the RT group (t(225) = 3.15, p_Tukey = 0.010). There was also a significant interaction effect between interventions and the repeated measures factor after a Greenhouse–Geisser correction (F(5.69, 426.76) = 2.28, p = 0.038) with an effect size of η²p = 0.03. The subsequent post hoc tests revealed highly significant differences in molecular biology knowledge between pre- and post-tests in all experimental groups (all: p_Tukey < 0.001), with the WL group exhibiting the largest effect size (BL: Cohen’s d = −0.967, VL: Cohen’s d = −0.874, WL: Cohen’s d = −1.265, RT: Cohen’s d = −0.867). Between pre- and follow-up tests, (highly) significant results were found with a small effect size in the BL (p_Tukey < 0.001, Cohen’s d = −0.36) and the VL group (p_Tukey = 0.037, Cohen’s d = −0.449). Only the WL group showed a highly significant difference in declarative knowledge of molecular biology with a large effect size between pre- and follow-up test (p_Tukey < 0.001, Cohen’s d = −0.903). No significant difference was found in declarative knowledge of molecular biology between pre- and follow-up tests in the RT group (p_Tukey = 0.291).

The interventions significantly influenced the perceived authenticity of the lessons (χ²(3) = 28.4, p < 0.001), as Figure 3 illustrates. Significant differences were observed between the WL and the BL group (W = −3.74, p = 0.041), the VL group (W = −5.32, p < 0.001), and the RT group (W = −6.81, p < 0.001). Additionally, the perceived authenticity also differed significantly between the BL and the RT group (W = −3.70, p = 0.044).

The perceived authenticity correlates positively with knowledge gain after the intervention (τb = 0.087, p = 0.05) and knowledge retention over 6 weeks (τb = 0.179, p < 0.001) in the field of molecular biology within the overall sample when controlling for the influence of guessing in the single-choice test. There was no significant correlation observed between the gain of declarative knowledge of laboratory safety and the perceived authenticity, both immediately after the intervention (p = 0.185) and at the 6-week follow-up-test (p = 0.863).

6. Discussion and Conclusions

The study conducted was a quasi-experimental design in which students were assigned to the respective treatments as intact class groups. Randomized assignment of individual students to the interventions was not feasible due to organizational constraints. Therefore, care was taken to select classes with comparable profiles (e.g., all classes were from the same metropolitan area). However, it became evident, for example, that the group had prior knowledge of laboratory safety, making the interpretation of results in this case more challenging. In preparation for the main study, a scale was developed to assess declarative knowledge of the molecular biology methods PCR and gel electrophoresis (provided in the Appendix A). This measurement instrument has proven effective in capturing the knowledge development among lower secondary education students. The scale for declarative knowledge of lab safety, developed by Savvides (2018) for science students, also proved effective for lower secondary students as well.

6.1. Blended Learning Concepts for Teaching Laboratory Safety Rules

Even after accounting for the higher prior knowledge of the BL group as a covariate, the BL group showed significant differences from all other experimental groups. However, the VL group also differed significantly from the RT group, which served as the control group. Both experimental groups learned the safety rules completely (VL) or at least partially, when, according to the RAT model, there was an advantage in learning through the virtual simulations. Muenz et al. (2023) found that direct feedback and the visualization of complex processes are crucial for the success of digital educational games in the field of education for sustainable development. These criteria can also be observed in the utilized simulation for lab safety. In this simulation, the direct consequences of disregarding laboratory rules become virtually tangible. This direct feedback through the digital tool could have underscored the importance of safety rules, making it also a critical factor for the success of the simulation in the context of laboratory safety. The development of declarative knowledge of lab safety using virtual simulations is comparable between our lower secondary education students and the results of Savvides (2018) with science students. Virtual lab safety introductions could contribute to ensuring safe experimentation in classrooms in the future. However, it seems that the combination of virtual simulations with content in the actual laboratory provides an additional benefit to learning success. However, our study design and the data collected do not allow for conclusions regarding the reasons behind this.

Nevertheless, it is crucial to also take the training of teachers into account. Cavas & Koç (2022) illustrated a potential avenue in this regard. They examined the progression of expertise in laboratory safety among N = 33 Turkish secondary school science teachers during a six-week professional development seminar. At the beginning of the course, the teachers displayed only moderate knowledge, which significantly increased over the course of the seminar, irrespective of gender or educational background.

6.2. Blended Learning Concepts for Teaching Molecular Biology Methods

All intervention groups exhibited a significant increase in knowledge of molecular biology methods between the pre- and post-test, highlighting the capability of comprehensive schools students to understand complex molecular biology. However, only the groups that conducted the molecular biology methods either virtually or hands-on exhibited a significant learning effect that remained stable over the six weeks until the follow-up test. Better retention effects could possibly be achieved (especially in the RT group) if teachers continued to incorporate the content conveyed in the learning laboratory into their following lessons. Glowinski (2007) underscores the importance of post-lesson reviews after the visit of molecular biology learning laboratories to foster sustained interest.

The literature frequently describes the phenomenon that practical laboratory work promotes long-term learning success (Bryant et al., 2009; Scharfenberg et al., 2007). However, from a learning psychology perspective, it is not trivial. In this study, the primary difference lay in the execution of molecular biology methods, while the delivery of theory was identical across the groups. For instance, the distinction between the WL and RT groups is that WL students pipetted various fluid volumes (primers, DNA template, and polymerase) into an actual PCR tube, whereas the RT group performed these steps on a worksheet. The simple act of pipetting in the laboratory cannot directly contribute to the understanding of PCR and should, theoretically, pose an additional cognitive load (Scharfenberg & Bogner, 2010). Nevertheless, the wet lab group demonstrated the biggest learning effects. This contradicted our assumption that the BL group would benefit more than the wet lab group because of the integration of animations illustrating in vitro processes or direct feedback through multiple choice questions in the simulation. However, the combination of virtual and hands-on laboratory work in the blended learning concept reduced the perceived authenticity of the teaching session. The authenticity of learning environments has been shown to positively influence the content-related interest of secondary school students (Pawek, 2009). Interested students should exhibit higher motivation to engage with the subject matter because they learn autonomously (Deci & Ryan, 1993), which should have a positive impact on their learning success. We cannot test this hypothesis through our data, but we can demonstrate that authenticity correlates positively with learning success, particularly when correcting for the pre-test score, which is presumably influenced by guessing effects. Additionally, the transition between the virtual and real learning environments could pose an additional burden on the extraneous cognitive load, potentially explaining the reduced learning success in the BL group. However, we are unable to verify this hypothesis, as cognitive load was not measured in our study.

However, this does not necessarily mean that blended learning concepts are inherently unsuitable for molecular biology education in lower secondary schools. Rather than combining virtual and wet lab activities at the expense of authenticity in a single teaching session, it might be more effective to prepare secondary students for wet lab work through virtual labs. Civil engineering students showed significantly faster work execution and less need for assistance in conducting their experiments when prepared for lab work through virtual labs compared to a textual lab guide (Vahdatikhaki et al., 2024). Navarro et al. (2024) investigated the perceptions of Labster simulations from the perspectives of first-year undergraduate students and their teachers in a cell biology course. Their findings highlighted that the primary strength of virtual labs lies in their ability to prepare students for traditional labs and also positively influencing learning attitudes. However, they also emphasized that using these simulations from home introduces technical challenges (such as the need for a stable internet connection and a capable device). These issues must be considered, particularly in the context of educational equity.

It is crucial to highlight that, alongside the advantages of hands-on laboratory work, the use of virtual lab simulations has proven effective in fostering declarative knowledge of molecular biology among lower secondary education students. This provides an opportunity for schools lacking proper laboratory equipment and those with remote access to learning labs. Furthermore, virtual labs facilitate the conduct of experiments that are impractical in both in-school and out-of-school learning labs due to cost and safety considerations.

Author Contributions

The conceptualization and methodology were developed and conducted by M.H. under the supervision of S.S. Teaching was done by M.H. Data collection was done by M.D. und M.H. M.D. also assisted in the organization of the study and during teaching. Data analysis was done by M.H. Funding was secured by S.S. The first draft of the manuscript was prepared by M.H. and reviewed by all other authors. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

The authors declare that this work is entirely their own, and all sources used have been properly cited. The study was ethically reviewed by the Doctoral Committee of the Faculty Council II of the University of Education Ludwigsburg. The implementation of the study was approved by the executive school principal of the Stuttgart schools in the “Sekundarstufe 1” (secondary level 1) sector. All students participated in the study voluntarily and were informed that choosing not to participate would not result in any disadvantages.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study. All students participated in the study voluntarily and were informed that choosing not to partic-ipate would not result in any disadvantages.

Data Availability Statement

The data presented in this study are available on request from the corresponding author. While a clear identification of the cooperating schools is not unequivocally possible, the dataset could potentially allow inferences about the performance of schools in the city where the study was conducted.

Conflicts of Interest

The authors declare no conflicts of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. Development of declarative knowledge of laboratory safety. The median is depicted by the black line, and the mean is represented by the black cross. Each individual data point is color-coded based on its corresponding intervention. The asterisks indicate the level of statistical significance: p ≤ 0.05 (*), p ≤ 0.01 (**), p ≤ 0.001 (***).

Figure 2. Development of declarative knowledge of molecular biology. The median is depicted by the black line, and the mean is represented by the black cross. Each individual data point is color-coded based on its corresponding intervention. The asterisks indicate the level of statistical significance: p [greater than] 0.05 (ns), p ≤ 0.05 (*), p ≤ 0.01 (**), p ≤ 0.001 (***).

Figure 3. The interventions demonstrate significant differences in perceived authenticity. The black line represents the interquartile range, and the white line represents the median. The asterisks indicate the level of statistical significance: p ≤ 0.05 (*), p ≤ 0.01 (**), p ≤ 0.001 (***).

Table 1

Overview of the sequence of the four interventions..

	BL	VL	WL	RT
Preparation Lesson
Getting to know the structure of DNA by building a three-dimensional model in small groups	x	x	x	x
Lab Safety
Putting on lab coats	x	-	x	-
Teacher emphasizes lab safety for everyone and its impact on experiment outcomes	x	x	x	x
Getting to know hazard pictograms	x (as in VL)	x (DS)	x (TL)	x (TL)
Determining pH and neutralization of acids	x (as in WL)	x (DS)	x (HO)	x (TL)
Proper handling of safety equipment (goggles, lab coats, etc.)	x (as in VL)	x (DS)	x (TL)	x (TL)
Safety measures for unwanted chemical reactions	x (as in WL)	x (DS)	x (HO)	x (TL)
Introduction to important locations (and their functions) (e.g.,: fire extinguishers)	x (TL)	x (DS)	x (HO)	x (TL)
Molecular biology
Presenting the fictional murder case	x	x	x	x
Presentation of theory of PCR through an explanatory video	x	x	x	x
PCR application	x (as in VL)	x (DS)	x (TL)	x (WS)
Presentation of theory of gel electrophoresis through an explanatory video	x	x	x	x
Gel electrophoresis application	x (as in VL)	x (DS)	x (TL)	x (WS)

Note: x = the group participated in this phase; - = the group did not participate in this phase; TL = teacher’s lecture; DS = desktop simulation; HO: = hands-on; WS = worksheet.

Table 2

Mean and standard deviations of the experimental groups across the three measurement time points.

Treat-ment	Lab Safety Knowledge Pre	Lab Safety Knowledge Post	Lab Safety Knowledge Follow-Up	Molecular Biology Knowledge Pre	Molecular Biology Knowledge Post	Molecular Biology Knowledge Follow-Up	Authen-ticity
BL	5.58 ± 1.76	8.58 ± 1.23	7.02 ± 1.95	2.90 ± 1.46	5.06 ± 2.19	4.62 ± 1.94	4.06 ± 0.618
WL	5.03 ± 1.39	7.56 ± 1.81	6.68 ± 2.07	3.03 ± 1.34	5.81 ± 2.18	5.10 ± 2.07	4.35 ± 0.635
VL	4.45 ± 1.54	7.71 ± 1.58	6.59 ± 1.80	2.93 ± 1.39	5.14 ± 2.24	3.98 ± 2.11	3.90 ± 0.694
RT	4.80 ± 1.43	7.20 ± 1.70	5.86 ± 2.12	2.80 ± 1.51	4.96 ± 2.24	3.65 ± 2.07	3.66 ± 0.783

Appendix A

Table A1

Overview of the Items in the Scale: Declarative Knowledge of Molecular Biology Methods.

Scale	Declarative knowledge of molecular biology methods
Sample size	N = 405
Sample	Secondary school students after their visit of a molecular biology out-of-school learning lab
Reliability	α = 0.77
Item	Possible answers	Mean	Item-Rest correlation	α without item
1. What is the aim of PCR?	(a) [Image omitted. Please see PDF.]	0.71	0.39	0.75
	(b) [Image omitted. Please see PDF.]
	(c) [Image omitted. Please see PDF.]
	(d) [Image omitted. Please see PDF.]
2. What is the function of a DNA template?	(a) It serves as a blueprint to produce millions of copies of DNA.	0.61	0.31	0.77
	(b) It is an enzyme that cuts DNA into small fragments.
	(c) It helps to denature DNA.
	(d) It transports DNA from the cell nucleus to the mitochondria.
3. What is the function of a primer?	(a) Primers bind to DNA and serve as a starting point for DNA polymerase.	0.62	0.43	0.75
	(b) Primers denature DNA and allow the DNA to be degraded.
	(c) Primers are the building blocks of DNA and are assembled by DNA polymerase.
	(d) Primers prevent DNA from denaturing and thus maintain the DNA double strand.
4. What is the function of DNA polymerase?	(a) DNA polymerase creates double-strand breaks in the DNA molecule.	0.48	0.45	0.75
	(b) DNA polymerase synthesizes the complementary DNA strand.
	(c) The DNA polymerase provides the DNA with a negative charge.
	(d) DNA polymerase serves as a scale for determining the length of unknown DNA molecules.
5. Which drawing describes the migration of a DNA molecule in gel electrophoresis best?	(a) [Image omitted. Please see PDF.]	0.59	0.57	0.73
	(b) [Image omitted. Please see PDF.]
	(c) [Image omitted. Please see PDF.]
	(d) [Image omitted. Please see PDF.]
6. How does gel electrophoresis work?	(a) The DNA in the gel is amplified by the electric current.	0.54	0.38	0.76
	(b) The electrical voltage stimulates cell division in the gel.
	(c) The electric current removes impurities from the sample.
	(d) The electric current causes DNA molecules to migrate through a gel.
7. What is the function of the gel in gel electrophoresis?	(a) It gives the DNA its negative charge on the phosphate residues.	0.65	0.43	0.75
	(b) It acts as a molecular sieve and separates the mixture of DNA molecules according to size.
	(c) It enables the electrical current to be passed on.
	(d) It removes interfering contaminants from the gel electrophoresis chamber.
8. Which statement about the travelling speed of DNA molecules in an electric field is correct?	(a) Large DNA molecules move faster to the positive pole than small DNA molecules.	0.54	0.60	0.72
	(b) The size of a DNA molecule is not decisive for its travelling speed.
	(c) Small DNA molecules run to the positive pole, large DNA molecules run to the negative pole.
	(d) Small DNA molecules run faster to the positive pole than large DNA molecules.
9. What does the gel of a DNA gel electrophoresis consist of?	(a) Agarose.	0.73	0.60	0.72
	(b) Fats.
	(c) Ordinary hair gel.
	(d) Proteins.
10. Consider the following lane pattern and decide which suspect is the perpetrator. [Image omitted. Please see PDF.]	(a) Suspect A is the perpetrator.	N.A.	N.A.	N.A.
	(b) Suspect B is the perpetrator.
	(c) Suspect C is the perpetrator.
	(d) None of the suspects is the perpetrator.

References

Anderson, J. R.; Grabowski, J.; Graf, R. Kognitive psychologie (3. Auflage, Spektrum Lehrbuch); Spektrum Akademischer Verlag: 2001; ISBN 978-3-8274-1024-5

Blanz, M. Forschungsmethoden und statistik für die soziale arbeit: Grundlagen und anwendungen; 2nd ed. W. Kohlhammer GmbH: 2021; [DOI: https://dx.doi.org/10.17433/978-3-17-039819-1]

Bryant, A.; Gieskes, K.; McGrann, R. The role of active learning through laboratory experimentation pertaining to memory retention in first-year engineering programs. 2009 39th IEEE Frontiers in Education Conference; San Antonio, TX, USA, October 18–21; 2009; pp. 1-6. [DOI: https://dx.doi.org/10.1109/FIE.2009.5350597]

Caravias, V. Management Association. Teachers’ conceptions and approaches to blended learning: A literature review. Online course management: Concepts, methodologies, tools, and applications; IGI Global: 2018; pp. 974-997. [DOI: https://dx.doi.org/10.4018/978-1-5225-5472-1]

Cavas, B.; Koç, S. The effects of laboratory safety professional development seminars implemented on science teachers: Laboratory safety knowledge levels. Science Education International; 2022; 33, 4 pp. 438-448. [DOI: https://dx.doi.org/10.33828/sei.v33.i4.12]

Caymaz, B. Secondary school students’ knowledge and views on laboratory safety. Journal of Science Learning; 2021; 4, 3 pp. 220-229. [DOI: https://dx.doi.org/10.17509/jsl.v4i3.30752]

Cronbach, L. J. Coefficient alpha and the internal structure of tests. Psychometrika; 1951; 16, 3 pp. 297-334. [DOI: https://dx.doi.org/10.1007/BF02310555]

Deci, E. L.; Ryan, R. M. Die Selbstbestimmungstheorie der Motivation und ihre Bedeutung für die Pädagogik. Zeitschrift für Pädagogik; 1993; 39, 2 pp. 223-238. [DOI: https://dx.doi.org/10.25656/01:11173]

Döring, N.; Bortz, J. Forschungsmethoden und Evaluation in den Sozial-und Humanwissenschaften; Springer: 2016; [DOI: https://dx.doi.org/10.1007/978-3-642-41089-5]

Engeln, K. Schülerlabors: Authentische, aktivierende Lernumgebungen als Möglichkeit, Interesse an Naturwissenschaften und Technik zu wecken; Logos: 2004; Vol. 36.

Fieselmann, J.; Annac, K.; Erdsiek, F.; Yilmaz-Aslan, Y.; Brzoska, P. What are the reasons for refusing a COVID-19 vaccine? A qualitative analysis of social media in Germany. BMC Public Health; 2022; 22, 1 846. [DOI: https://dx.doi.org/10.1186/s12889-022-13265-y]

Frewer, L. J.; Van Der Lans, I. A.; Fischer, A. R. H.; Reinders, M. J.; Menozzi, D.; Zhang, X.; Van Den Berg, I.; Zimmermann, K. L. Public perceptions of agri-food applications of genetic modification—A systematic review and meta-analysis. Trends in Food Science & Technology; 2013; 30, 2 pp. 142-152. [DOI: https://dx.doi.org/10.1016/j.tifs.2013.01.003]

Friedman, D. C.; Ellington, A. D. Industrialization of biology. ACS Synthetic Biology; 2015; 4, 10 pp. 1053-1055. [DOI: https://dx.doi.org/10.1021/acssynbio.5b00190]

Gauthier, A. Game and simulation stimulate conceptual change about molecular emergence in different ways, with potential cultural implications. Education Sciences; 2024; 14, 4 366. [DOI: https://dx.doi.org/10.3390/educsci14040366]

Gericke, N.; Högström, P.; Wallin, J. A systematic review of research on laboratory work in secondary school. Studies in Science Education; 2023; 59, 2 pp. 245-285. [DOI: https://dx.doi.org/10.1080/03057267.2022.2090125]

Glowinski, I. Schülerlabore im Themenbereich Molekularbiologie als Interesse fördernde Lernumgebungen; Christian-Albrechts-Universität: 2007.

Gublo, K. I. A Laboratory Safety Trivia Game. Journal of Chemical Education; 2003; 80, 4 425. [DOI: https://dx.doi.org/10.1021/ed080p425]

Guerra-Carrillo, B.; Katovich, K.; Bunge, S. A. Does higher education hone cognitive functioning and learning efficacy? Findings from a large and diverse sample. PLoS ONE; 2017; 12, 8 e0182276. [DOI: https://dx.doi.org/10.1371/journal.pone.0182276] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28832590]

Haberbosch, M.; Vick, P.; Schaal, S. (in press) Navigating molecular biology education at the intermediate level: Content relevance & teachers’ challenges, self-efficacy, and knowledge. IJSE; in press

Hafer, J. Wollersheim, H.-W.; Karapanos, M.; Pengel, N. Auf der Suche nach dem Präsenzgen in der Universitätslehre. Eine Spurensuche in den Präsenzdiskursen der letzten Dekade. Medien in der Wissenschaft-Bildung in der digitalen Transformation (Issue 78, pp. 47–58); Waxmann: 2021; [DOI: https://dx.doi.org/10.25656/01:26618]

Hofstein, A.; Lunetta, V. N. The laboratory in science education: Foundations for the twenty-first century. Science Education; 2004; 88, 1 pp. 28-54. [DOI: https://dx.doi.org/10.1002/sce.10106]

Hughes, J.; Thomas, R.; Scharber, C. Crawford, C.; Carlson, R.; McFerrin, K.; Price, J.; Weber, R.; Willid, D. Assessing Technology Integration: The RAT–Replacement, Amplification, and Transformation–Framework. Proceedings of SITE 2006–Society for Information Technology & Teacher Education International Conference; 2006; pp. 1616-1620.

Itzek-Greulich, H.; Flunger, B.; Vollmer, C.; Nagengast, B.; Rehm, M.; Trautwein, U. Effects of a science center outreach lab on school students’ achievement—Are student lab visits needed when they teach what students can learn at school?. Learning and Instruction; 2015; 38, pp. 43-52. [DOI: https://dx.doi.org/10.1016/j.learninstruc.2015.03.003]

Lave, J.; Wenger, E. Situated learning: Legitimate peripheral participation; 1st ed. Cambridge University Press: 1991; [DOI: https://dx.doi.org/10.1017/CBO9780511815355]

Lynch, T.; Ghergulescu, I. Review of virtual labs as the emerging technologies for teaching stem subjects. 11th International Technology, Education and Development Conference; Valencia, Spain, March 6–8; 2017; pp. 6082-6091. [DOI: https://dx.doi.org/10.21125/inted.2017.1422]

Makransky, G.; Petersen, G. B.; Klingenberg, S. Can an immersive virtual reality simulation increase students’ interest and career aspirations in science?. British Journal of Educational Technology; 2020; 51, 6 pp. 2079-2097. [DOI: https://dx.doi.org/10.1111/bjet.12954]

Mercado, J.; Picardal, J. P. Virtual laboratory simulations in biotechnology: A systematic review. Science Education International; 2023; 34, 1 pp. 52-57. [DOI: https://dx.doi.org/10.33828/sei.v34.i1.6]

Muenz, T. S.; Schaal, S.; Groß, J.; Paul, J. How a digital educational game can promote learning about sustainability. Science Education International; 2023; 34, 4 pp. 293-302. [DOI: https://dx.doi.org/10.33828/sei.v34.i4.5]

Navarro, C.; Arias-Calderón, M.; Henríquez, C. A.; Riquelme, P. Assessment of student and teacher perceptions on the use of virtual simulation in cell biology laboratory education. Education Sciences; 2024; 14, 3 243. [DOI: https://dx.doi.org/10.3390/educsci14030243]

Nephew, S.; Sunasee, R. An engaging and fun breakout activity for educators and students about laboratory safety. Journal of Chemical Education; 2021; 98, 1 pp. 186-190. [DOI: https://dx.doi.org/10.1021/acs.jchemed.0c01109]

Pawek, C. Schülerlabore als interessefördernde außerschulische Lernumgebungen für Schülerinnen und Schüler aus der Mittel-und Oberstufe; Christian-Albrechts-Universität: 2009.

Poensgen, F.; Hohr, A.; Vit, M.; Reiners, C. S. Gefahren virtuell begegnen. Nachrichten Aus Der Chemie; 2021; 69, 10 pp. 11-15. [DOI: https://dx.doi.org/10.1002/nadc.20214118874]

Sahdra, B.; Thagard, P. Procedural knowledge in molecular biology. Philosophical Psychology; 2003; 16, 4 pp. 477-498. [DOI: https://dx.doi.org/10.1080/0951508032000121788]

Savvides, P. Teaching science lab safety: Are virtual simulations effective?; Arizona State University: 2018.

Schaal, S.; Meissner, M.; Schaal, S. Reflexive Unterrichtspraxis in der Lehrkräftebildung—Fachdidaktische Reflexion im Lehr-Lern-Labor fördern. Lehrerbildung auf dem Prüfstand; 2022; 15, 1 pp. 5-25.

Scharfenberg, F.; Bogner, F. X. Instructional efficiency of changing cognitive load in an out-of-school laboratory. International Journal of Science Education; 2010; 32, 6 pp. 829-844. [DOI: https://dx.doi.org/10.1080/09500690902948862]

Scharfenberg, F.; Bogner, F. X.; Klautke, S. Learning in a gene technology laboratory with educational focus: Results of a teaching unit with authentic experiments. Biochemistry and Molecular Biology Education; 2007; 35, 1 pp. 28-39. [DOI: https://dx.doi.org/10.1002/bmb.1]

Schäfers, M. S.; Schmiedebach, M.; Wegner, C. Virtuelle Labore im Biologieunterricht: Auswirkungen von Labster auf die Selbsteinschätzung von Schülerinnen und Schülern. MedienPädagogik: Zeitschrift Für Theorie Und Praxis Der Medienbildung; 2020; pp. 140-167. [DOI: https://dx.doi.org/10.21240/mpaed/00/2020.08.31.X]

Schüttler, T.; Watzka, B.; Girwidz, R.; Ertl, B. Die wirkung der authentizität von lernort und laborgeräten auf das situationale interesse und die relevanzwahrnehmung beim besuch eines naturwissenschaftlichen schülerlabors. Zeitschrift für Didaktik der Naturwissenschaften; 2021; 27, 1 pp. 109-125. [DOI: https://dx.doi.org/10.1007/s40573-021-00128-z]

Strømme, T. A.; Mork, S. M. Students’ conceptual sense-making of animations and static visualizations of protein synthesis: A sociocultural hypothesis explaining why animations may be beneficial for student learning. Research in Science Education; 2021; 51, 4 pp. 1013-1038. [DOI: https://dx.doi.org/10.1007/s11165-020-09920-2]

Sweller, J.; Van Merriënboer, J. J. G.; Paas, F. Cognitive architecture and instructional design: 20 years later. Educational Psychology Review; 2019; 31, 2 pp. 261-292. [DOI: https://dx.doi.org/10.1007/s10648-019-09465-5]

Sweller, J.; Van Merrienboer, J. J. G.; Paas, F. G. W. C. Cognitive architecture and instructional design. Educational Psychology Review; 1998; 10, 3 pp. 251-296. [DOI: https://dx.doi.org/10.1023/A:1022193728205]

Tang, C. M.; Chaw, L. Y. Digital literacy: A prerequisite for effective learning in a blended learning environment?. The Electronic Journal of e-Learning; 2016; 14, 1 pp. 54-65.

Tempelaar, D. T.; Rienties, B.; Giesbers, B. Who profits most from blended learning?. Industry and Higher Education; 2009; 23, 4 pp. 285-292. [DOI: https://dx.doi.org/10.5367/000000009789346130]

Thisgaard, M.; Makransky, G. Virtual learning simulations in high school: Effects on cognitive and non-cognitive outcomes and implications on the development of stem academic and career choice. Frontiers in Psychology; 2017; 8, 805. [DOI: https://dx.doi.org/10.3389/fpsyg.2017.00805] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28611701]

Vahdatikhaki, F.; Friso-van Den Bos, I.; Mowlaei, S.; Kollöffel, B. Application of gamified virtual laboratories as a preparation tool for civil engineering students. European Journal of Engineering Education; 2024; 49, 1 pp. 164-191. [DOI: https://dx.doi.org/10.1080/03043797.2023.2265306]

Wu, K.; Jin, X.; Wang, X. Determining university students’ familiarity and understanding of laboratory safety knowledge—A case study. Journal of Chemical Education; 2021; 98, 2 pp. 434-438. [DOI: https://dx.doi.org/10.1021/acs.jchemed.0c01142]

Yu, A.; Wisinski, J.; Osmundson, T.; Sanderfoot, A.; Cooper, S.; Klein, J. Instructional innovations in college-level molecular bioscience labs during the pandemic-induced shift to online learning. Education Sciences; 2022; 12, 4 230. [DOI: https://dx.doi.org/10.3390/educsci12040230]