Motivation and interest play a key role in student learning and choice of profession (Bennett and Hogarth, 2009; Gottfried et al., 2009; Archer et al., 2010; Krapp and Prenzel, 2011; Organisation for Economic Co-operation and Development (OECD), 2016, 2019; Ryan and Deci, 2017). However, Programme for International Student Assessment studies have demonstrated that students lose their science-related motivation and interest through the course of their education (Krapp and Prenzel, 2011; Organisation for Economic Co-operation and Development (OECD), 2016, 2019). With regard to an increasing demand for graduated scientists and employees in the fields of science, technology, engineering, and mathematics (STEM; Organisation for Economic Co-operation and Development (OECD), 2016; Bureau of Labor Statistics and Department of Labor, 2017; Statistik der Bundesagentur für Arbeit, 2019), this negative trend in student motivation is a cause of concern and needs to be addressed in science education.

As a result, many studies have focused on out-of-school learning in science educational outreach programs in recent years (e.g., Salmi, 2012; Schütte and Köller, 2015; Clark et al., 2016; Roberts et al., 2018; Tsybulsky et al., 2018; Tal and Dallashe, 2019; Tsybulsky, 2019; Maiorca et al., 2021; Solis et al., 2021). To complement formal science learning at school, informal and non-formal science learning can make valuable contributions to student learning due to, among other things, authentic hands-on learning experiences (Tal and Dallashe, 2019; Maiorca et al., 2021). Science learning in informal and non-formal contexts can be beneficial for student engagement in science classes (Roberts et al., 2018), the acquisition of skills and knowledge (Berg et al., 2021; Solis et al., 2021), and the promotion of student motivation and interest in the fields of STEM (Mohr-Schroeder et al., 2014; Schütte and Köller, 2015; Clark et al., 2016; Tal and Dallashe, 2019). Against this backdrop, outreach science labs (OSLs) have been established in Germany. In these out-of-school learning environments, students can perform experiments in an authentic science laboratory (Scharfenberg et al., 2019; Euler and Schüttler, 2020). It has been assumed that performing experiments at an OSL can be beneficial to students' science-related motivation and interest (Hofstein and Lunetta, 2004; Scharfenberg et al., 2019; Euler and Schüttler, 2020).

Previous studies have generally painted single visits to an OSL in a positive light (for an overview see Guderian and Priemer, 2008; Nickolaus et al., 2018; Scharfenberg et al., 2019). For instance, the characteristics of an OSL, including its authenticity, the laboratory work, and the supervision, have been found to have a positive impact on students' situational interest (Pawek, 2009) and situational competence experience (see also Glowinski, 2007; Glowinski and Bayrhuber, 2011; Itzek-Greulich and Vollmer, 2017). Students' situational competence experience was also found to be affected by their self-concept of abilities (Glowinski and Bayrhuber, 2011; Damerau, 2012) and contextual competence perception (i.e., a more habitual competence perception; Kowal and Fortier, 2000; Milyavskaya et al., 2013). However, the effects of these more habitual motivational variables on situational motivational outcomes may depend on the learning environment (see Glowinski and Bayrhuber, 2011; Itzek-Greulich and Vollmer, 2017; Scharfenberg et al., 2019).

Situational competence experience is an important antecedent of self-determined motivation (Reeve, 2015; Ryan and Deci, 2017; Ryan and Moller, 2017). Therefore, investigating predictors of students' situational competence experience at OSLs and at school is of unique importance for the understanding of both motivational processes during laboratory work and the effectiveness of OSLs in terms of student motivation (see Itzek-Greulich and Vollmer, 2017; Nickolaus et al., 2018). However, the current research in this field lacks studies that consider a school treatment under more realistic conditions (e.g., using equipment that is usually available at schools) and the same pedagogical staff to conduct the same educational program in both settings (see Itzek-Greulich and Vollmer, 2017; Nickolaus et al., 2018; Tsybulsky, 2019; Röllke et al., 2020). The current study addresses this research gap. The same educational program was conducted by the same pedagogical staff at an OSL and at school using equipment that is usually available in the respective learning environments. In these learning environments, we investigated the moderating effect of the learning environment on the relationship between students' contextual competence perception and situational competence experience during experimentation.

Based on the aforementioned rationale, students' situational competence experience during an experimental workshop may be affected by more habitual perceptions of competence, such as their contextual competence perception (section Perceived Competence). However, the place of learning may have an effect as well. The support provided and further characteristics of OSLs may facilitate positive experiences of competence when performing experiments (section Non-formal Learning at Outreach Science Labs and The Effects of an Outreach Science lab on Students' Perceived Competence). Previous research suggests that a visit to an OSL may be more effective for less motivated and low achieving students (e.g., Damerau, 2012; Itzek-Greulich and Vollmer, 2017). Since these variables are related to perceived competence on a contextual level (Hughes et al., 2011; Marsh et al., 2017), students who generally feel less competent in experimentation may also benefit from visiting an OSL. Specifically, we assume that the students' situational competence experience might be less dependent on their contextual competence perception at an OSL than at school. That is, the influence of students' contextual competence perception on their situational competence perception might be moderated by the learning environment (Figure 1).

In this study, we assessed students' perceived competence during two phases of their laboratory work—namely, conducting and evaluating experiments (see Klahr, 2000; Schreiber et al., 2009). Therefore, the moderating effect of the learning environment was investigated in two hypotheses, each based on one of the two phases.

Preliminarily, no differences in students' situational competence experiences between the groups taught by different supervisors at each location were found (OSL: F_{(2, 57)} = 0.08, p = 0.927; school: F_{(4, 112)} = 0.31, p = 0.872). Correlations between the independent and dependent variables were tested and can be found in Table 2.

To test whether the effect of students' experiment-related contextual competence perceptions on their situational competence experiences during experimentation was moderated by the learning environment, we performed two moderation analyses: one regarding conducting the experiments and the other regarding evaluating the experiments. Both models were found to be suitable for identifying predictors of students' situational competence experience (Table 3). Specifically, the interaction effects of contextual competence perception and the treatment were found to be significant. The coefficients indicate that the effects of students' contextual competence perception on their situational competence experience were weaker at the OSL than at school.

To investigate the moderating effect of the treatment in more detail, simple slope analyses were performed. Regarding conducting experiments, students' contextual competence perception did not predict their situational competence experience at the OSL (Table 4). In contrast, the prediction was significant in school (Table 4). The higher students' contextual competence perception, the higher their situational competence experience. Regarding evaluating experiments, students' contextual competence perception predicted their situational competence experience in both learning environments (Table 4). The higher students' contextual competence perception, the higher their situational competence experience.

Figure 3 visualizes the interaction effects. The slope of the regression lines for the school treatment were steeper than the slopes of the regression lines for the OSL treatment. This pattern illustrates a stronger correlation between students' contextual perceptions and situational experiences of competence at school than at the OSL (see also Table 4). Moreover, Figure 3 illustrates that students' situational competence experiences differed between the treatments when their respective contextual competence perception was lower (Mean – SD). Indeed, these students felt more competent during experimentation at the OSL than at school (Table 4). On the other hand, when contextual competence perceptions were higher (Mean + SD), the students felt competent during experimentation to the same extent at both places of learning (Table 4).

In summary, these findings indicated that students' situational competence experience was more strongly predicted by their contextual competence perception at school than at the OSL. Additionally, the students with a lower contextual competence perception had a higher level of situational competence experience at the OSL than at school.

In the current study, we hypothesized that the learning environment moderates the effect of students' experiment-related contextual competence perceptions on their situational competence experiences during experimentation. The findings of the analyses were consistent with our expectations. We found significant interactions between the investigated predictors. These findings support the moderating role of the learning environment in the relationship between contextual perception and situational experience of competence during conducting (H₁) and evaluating experiments (H₂). For both phases of experimentation, the effect of students' contextual competence perception on their situational competence experience was stronger at school than at the OSL. In the school treatment, there was a significant effect for both conducting and evaluating experiments. In the OSL treatment, we found different relationships. When conducting experiments, there was no effect. That is, students with a higher contextual competence perception did not experience themselves as more competent when conducting the experiments than students with a lower contextual competence perception. However, regarding the evaluation of experiments, the students' contextual competence perception predicted their situational competence experience. Compared to the rather cognitive activities in the evaluation of experiments (see Schreiber et al., 2009), the practical activities when students conducted the experiments may have provided intense experiences of competence (see Hofstein and Lunetta, 2004; Tal and Dallashe, 2019; Tsybulsky, 2019), particularly at the OSL (see Itzek-Greulich and Vollmer, 2017). This circumstance might have contributed to a lower correlation between students' contextual and situational perceived competence in conducting experiments over evaluating experiments. Accordingly, Damerau (2012) found a stronger correlation between students' self-assessed and experienced experiment-related competencies in evaluating experiments than in conducting experiments.

In addition, comparing students' situational competence experiences between the two learning environments, the investigated effects are in favor of the students who had a lower contextual competence perception and then visited the OSL. These students experienced themselves as more competent than students with a lower contextual competence perception at school. Those with a higher contextual competence perception had a comparable situational experience of competence in both learning environments. Compared to students who experimented at school or had a higher contextual competence perception, this finding reveals that students who perceived themselves as generally less competent benefited most in terms of their situational competence experience through the performance of experiments at the OSL. Previous studies have shown similar findings on student motivation. For instance, they have revealed that less interested students (e.g., Pawek, 2009; Glowinski and Bayrhuber, 2011; Damerau, 2012) and low achieving students (Itzek-Greulich and Vollmer, 2017) benefited most from attending an OSL in terms of effects on their motivational outcomes. In general, low achievers exhibit low interest (Hong and Lin, 2011), which is strongly correlated with students' self-concept of abilities, especially in higher grades (Marsh et al., 2005; Möller and Trautwein, 2015). They rate their skills low and feel less competent at a contextual level or within a specific domain (Hughes et al., 2011). For these students, support can be particularly helpful (see Vygotsky, 1978; Schmidt-Weigand et al., 2008; Itzek-Greulich and Vollmer, 2017, 2018; Großmann and Wilde, 2019). In the current study, the support was provided by the tutors and teachers (section Educational Program and Treatments). Their support might have helped the students with a lower contextual competence perception to feel more confident in experimentation and to perceive themselves as effective or successful throughout that process (see Schmidt-Weigand et al., 2008; Arnold et al., 2017; Großmann and Wilde, 2019). The students received support to operate the laboratory equipment successfully during the OSL visit. The laboratory equipment (e.g., microliter pipettes) is of higher quality than the equipment that is usually available at schools (e.g., one-way pipettes). It is possible that successfully operating the better laboratory materials might have led to a higher experience of effectiveness in the OSL than successfully operating the materials at school (see Pawek, 2009; Schüttler et al., 2021). In addition, the support at the OSL was provided more frequently than at school. The students at the OSL might have received more individual attention than the students at school, especially in situations where several groups might have needed support from their supervisor at the same time. These circumstances might explain why students with a lower contextual competence experienced themselves as more competent at the OSL than the students at school with only one teacher instructing the whole class (see Jang et al., 2010; Reeve, 2015). On the other hand, the situational competence experience of the students with a higher contextual competence perception did not differ between the treatments although there was more support available at the OSL. It may be that these students might have required less support than students who felt insecure and less competent in experimentation when working on the same task (see Vygotsky, 1978; Schmidt-Weigand et al., 2008; Großmann and Wilde, 2019). This circumstance might also explain why there were lower correlations between contextual perceptions and situational experiences of competence at the OSL compared to those in school.

Moreover, the findings could be explained by differences in the formality of learning within the two learning environments. The non-formal environment of the OSL might have contributed positively to students' social interactions (see Hofstein and Lunetta, 2004; Schwan et al., 2014; Tal and Dallashe, 2019). In addition to the tutors' support, peer assistance among the students might have helped those with a lower contextual competence perception to master the experimental tasks (see Donato, 1994; Levine and Thompson, 2004; Reicher et al., 2006). In turn, these students might have experienced themselves as more competent at the OSL than at school (see Wentzel, 2017). Furthermore, the OSL was visited as a supplement to the regular biology class, and the students' performance was not graded (section Educational Program and Treatments). Therefore, the activities at the OSL might have been less associated with grading performance than the activities at school (see Pawek, 2009) and might have been perceived “as a fresh start” (Itzek-Greulich and Vollmer, 2017, p. 18). Since OSLs emphasize the acquisition of new competencies in an assessment-free space with special respect paid to students' individual needs (see Pawek, 2009; Euler and Schüttler, 2020), the less performance-oriented atmosphere of this environment might have positively affected students' motivational experiences as well as their perceived autonomy and competence (see Meece et al., 2006; Itzek-Greulich and Vollmer, 2017; Alp et al., 2018). These results from previous studies support our findings on students' situational competence experience. Additionally, as the students' performance was not graded, asking for or receiving support would not be associated with a bad grade, but rather with informative feedback and constructive support (see Pawek, 2009). To prevent the students from feeling too closely observed, the tutors were instructed to not express controlling behaviors (see Reeve, 2015) and to ensure that the students conducted and evaluated the experiments predominantly on their own (section Educational Program and Treatments). We also explained to the students why support can sometimes be helpful and that it could be requested (e.g., in using the micropipettes). Therefore, they may have experienced being controlled less but rather supported at the OSL. If the students perceived themselves as being controlled by their tutors, their situational competence experience might have been impaired (Jang et al., 2010; Su and Reeve, 2011; Reeve, 2015; Großmann et al., 2020). However, as our analyses revealed, the students at the OSL did not have had a lower competence experience than the students at school. Therefore, the possible negative effect of the tutoring did not lead to a lower situational competence experience at the OSL compared to the school treatment.

As the regular teachers were present during the intervention, they might have had an influence as well. In both learning environments, the regular teachers observed but did not intervene in the conduction of the workshop. According to the concepts of non-formal and formal learning, the students in the OSL treatment were not graded, but the students in the school treatment were graded by their regular teachers (section Educational Program and Treatments). Grades can exert pressure on students that may have a negative effect on their perceived competence (see Su and Reeve, 2011; Reeve, 2015). The negative effect may be stronger when students' contextual competence perception is low as these students may think they are more likely to receive a lower grade (see Skaalvik and Skaalvik, 2005). It is possible that grading in the school treatment might have contributed to our results.

With regard to the interpretation and explanation of the effects found, it is essential to consider a few aspects. First, the investigated learning environments had more differences in characteristics apart from their location outside of or inside schools, such as the provision of support and the formality of learning. Therefore, the effects found cannot be attributed to a single characteristic, but rather to the sum of the characteristics that shape experimentation at an OSL or in school. Future studies could investigate such variables in more detail, for example by varying the supervisor-student-ratio at an OSL. However, this study was conducted in two authentic settings that could yield ecological validity (Lewkowicz, 2010) and provide evidence regarding the value of practical laboratory work at OSL in terms of students' situational competence experience.

Second, the sample size was not very large. Nevertheless, it was sufficiently large for the performed statistical analyses (Field, 2018). In addition, the students only came from one type of secondary schools in North Rhine-Westphalia. School type might have an impact on the effect of the learning environment (see Basten et al., 2014) and education may differ between the federal states, for instance, regarding the curriculum (Standing Conference of the Ministers of Education Cultural Affairs of the Länder in the Federal Republic of Germany, 2019). Consequently, the applicability of our results to other school types and other federal states may be limited. In future studies, a larger sample should be examined and additional types of secondary schools such as a ‘Gymnasium' (i.e., the highest track) in other federal states should be included.

Third, other than conducting and evaluating experiments, the planning of an experimental setup is also a phase of experimentation (Klahr, 2000; Schreiber et al., 2009). As the investigated OSL focused on conducting and evaluating experiments and the students did not plan their own experiments in the workshop, we did not investigate students' perceived competence in planning experiments. It may be interesting to investigate whether similar results can be found for planning experiments, in particular, with regard to students with a higher contextual competence perception in planning. As students generally rate their skills in planning the lowest compared to other phases of experimentation (Damerau, 2012; see also Franken et al., 2020), the provision of support during this phase of experimentation may also be important for students with a higher contextual competence perception in terms of their situational competence experience. Moreover, further variables, such as prior knowledge, may have an influence on students' perceived competence (see Jang et al., 2010; Reeve, 2015; Ryan and Deci, 2017). However, in a previous study (Großmann et al., 2020), students' prior knowledge could not have been confirmed as a predictor of their perceived competence during biology class. In addition, as the current study took place before the regular teachers had started to teach their students in enzymology, all participants might have shared similar prior knowledge (section Educational Program and Treatments). Nevertheless, we cannot rule out that the students might have acquired knowledge on enzymology in their previous school career.

Fourth, the novelty of an out-of-school learning environment may have negative effects on student learning (see Eshach, 2007). In our study, the students visited an OSL that was unfamiliar to them and might have perceived the environment as complex at first (see Pawek, 2009; Euler and Schüttler, 2020). Here, the students performed comprehensive experimental tasks using laboratory equipment that they were not familiar with from school (see Scharfenberg and Bogner, 2013; Garner et al., 2014; Affeldt et al., 2015). As a result, the students at the OSL might have considered the experiments initially more challenging, difficult, and extensive than the students at school did (see Pawek, 2009; Euler and Schüttler, 2020) and might have been overwhelmed when executing the experimental tasks (Damerau, 2012; Scharfenberg and Bogner, 2013). In this situation, the students' situational competence experience could have been impaired (see Jang et al., 2010; Reeve, 2015). As discussed, the provided support at the OSL might have prevented the students from being overwhelmed by an unfamiliar and complex environment (see Kirschner et al., 2006; Kersaint et al., 2011; Scharfenberg and Bogner, 2013) and might therefore have reduced a possible effect of novelty on students' situational competence experience (see Vygotsky, 1978; Mohr-Schroeder et al., 2014). In addition, our results reveal that the students at the OSL did not perceive themselves as less competent than the students who attended the workshop in their familiar science classroom at school. This result indicates that the novelty of the new environment did not lead to a lower situational competence experience at the OSL compared to the school.

In addition to novelty effects, some further aspects should be taken into consideration when visiting OSLs. As was the case in this study, most of the investigated effects for OSLs are short-term effects (Guderian and Priemer, 2008; Schütte and Köller, 2015; Nickolaus et al., 2018). In the long term, students may benefit from their experiences at an OSL in terms of habitual forms of motivation, such as individual interest or contextual competence perception, provided that the positive learning experiences of the visit could be maintained through post-visit instruction (see Itzek-Greulich et al., 2015; Itzek-Greulich and Vollmer, 2017). In addition to post-visit instruction, teachers should prepare visits to OSLs, as the effectiveness of the visit depends on the (organizational) pre-visit instruction (Orion, 1993; Wilde and Bätz, 2006; Glowinski and Bayrhuber, 2011; Itzek-Greulich and Vollmer, 2017). Therefore, an OSL visit should not be considered a separate, single event or a substitute for regular science classes (Glowinski and Bayrhuber, 2011; Itzek-Greulich et al., 2015; Euler and Schüttler, 2020). Both science classes in school and science educational outreach programs are mutually complementary ways for students to learn science. That is, linking the lessons at school with OSL visits may be desirable (Orion, 1993; Eshach, 2007; Itzek-Greulich, 2020).

In conclusion, our findings contribute to the current state of research in the field of non-formal learning's impact on student motivation by investigating the interrelationship of students' competence perceptions and experiences during experimentation at an OSL and at school. They highlight great potential for fostering students' situational motivation in the context of out-of-school learning. In particular, the students who previously felt less competent in experimentation could benefit from performing experiments at an OSL in terms of their situational competence experience. Performing experiments at an OSL seems to provide positive situational experiences of one's own competence, almost independently of a more general perception of competence. This finding is in line with the assumption that an OSL visit contributes to the development of positive qualities of student motivation (Garner et al., 2014; Affeldt et al., 2015; Scharfenberg et al., 2019; Euler and Schüttler, 2020); it also aligns with findings from previous studies that less motivated students may especially benefit from attending a workshop at an OSL (e.g., Glowinski and Bayrhuber, 2011; Damerau, 2012; Itzek-Greulich and Vollmer, 2017). Our findings suggest that OSLs offer suitable conditions for addressing student heterogeneity in terms of their perceived competence.

The raw data supporting the conclusion of this article will be made available by the authors, without undue reservation. Requests to access the datasets should be directed to TK, tim.kirchhoff@uni-bielefeld.de.

The studies involving human participants were reviewed and approved by Ethik-Kommission der Universität Bielefeld. Written informed consent from the participants' legal guardian/next of kin was not required to participate in this study in accordance with the national legislation and the institutional requirements.

TK developed the concept of and design for the study, recruited the sample, performed the statistical analysis, and wrote the first draft of all article sections. MW contributed to the study design and reviewed the article. NG contributed to the study design and the statistical analyses and reviewed the article. All authors agree to be accountable for the content of the work. All authors have contributed, read, and approved the submitted manuscript.

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 (Funding Code: 01JA1908). We acknowledge support for the publication costs by the Open Access Publication Fund of Bielefeld University and the Deutsche Forschungsgemeinschaft (DFG). The authors are responsible for the content of this publication.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.