Introduction

The process of science learning is intricate and interconnected, encompassing various objectives. The National Academy of Sciences has outlined a comprehensive framework for science learning goals and practices, comprising six interrelated strands: fostering interest in science, comprehending scientific knowledge, engaging in scientific reasoning, reflecting on science, participating in scientific practice, and identifying with the scientific enterprise (National Research Council 2012). Of these, science identity stands out as the pinnacle goal, emerging from sustained engagement in scientific activities and representing a synthesis of the other five dimensions, encapsulating an individual’s holistic science literacy. Science identity refers to an individual’s ability to perform scientific practices, understand scientific content, recognize oneself as a “science person,” and be recognized by others as a “science person” (Carlone and Johnson 2007).

Scholars underscore the pivotal importance of science identity in the training of scientific and technological talents, as well as in students’ scientific literacy (e.g., Archer et al. 2010a; Hazari et al. 2022; Vincent‐Ruz and Schunn 2019). As well as an essential factor in explaining and predicting various aspects of science learning, such as interest in science, science learning engagement, science self-efficacy, science sense of belonging, science learning motivation, science value beliefs, science academic performance, academic career intentions, and science achievement (e.g., Archer et al. 2010; Hazari et al. 2022; Vincent‐Ruz and Schunn 2019; Fraser and Ward 2009; Merolla and Serpe 2013; Seyranian et al. 2018; Starr 2018; Chen et al. 2021; Lee and Mun 2023; Tang et al. 2023; Trujillo and Tanner 2014; White et al. 2019). Notably, the Program for International Student Assessment (PISA) science team has introduced a fresh dimension of “science identity” to its evaluation framework, complementing the existing dimensions of “scientific knowledge” and “scientific ability” (Osborne et al. 2020).

The development of students’ science identity has taken center stage in contemporary research on science education. Scholars emphasize that engagement in science practices, both formal and informal, contributes significantly to science identity development (Kolodner and Clegg 2010; Fajardo 2015; Robinson et al. 2018; Seymour et al. 2004; Williams et al. 2018). Inquiry-based learning, a highly effective form of scientific practice, facilitates student-centered learning by actively involving students in the inquiry process, mirroring scientists’ approach, and fostering skills, understanding, and the process of inquiry learning (Rivera Maulucci et al. 2014; Zhai et al. 2014).

The advent of information technology has provided educational opportunities, with the current generation labeled as “digital natives” due to their immersion in a digital environment since birth. Digital products have become integral to their lives, necessitating a focus on enhancing the science identity of these “digital natives”. Some researchers propose that integrating information technology into inquiry learning can effectively cultivate students’ science identity (Asbell-Clarke et al. 2012; Clarke and Dede 2009; Johnson-Glenberg et al. 2023; Liu and Hannafin 2010; Reilly et al. 2021). However, existing research has paid scant attention to whether web-based inquiry environments can proficiently promote the development of students’ science identity. Addressing this gap is crucial for fully harnessing the potential of information technology in science education. Investigating how web-based inquiry fosters science identity among digital natives can guide the design of effective learning environments that capitalize on the unique affordances of information technology. Therefore, understanding how to enhance the science identity of “digital natives” in the digital age remains a critical inquiry, necessitating further research to unravel the potential of web-based inquiry in fostering science identity within this population.

Building upon this rationale, the current study employed a web-based inquiry learning environment as the experimental platform, engaging Chinese fifth-grade elementary school students as participants. Utilizing an adapted science identity scale, the study aimed to explore the impact of web-based inquiry on students’ science identity through comparative experiments. The research sought answers to the following questions:

  1. (1)

    Does web-based inquiry have a discernible impact on the science identity of elementary school students?

  2. (2)

    What are the differences between web-based inquiry and traditional inquiry in fostering science identity among elementary school students?

Literature review

Science identity

The exploration of students’ scientific literacy from an identity perspective offers a unique analytical lens (Carlone and Johnson 2007; Gee 2000). Science identity has been described as a person’s ability to work like a scientist and to be recognized as a “science person” by others (Carlone and Johnson 2007). NRC (2009) argued that science identity involves thinking of oneself as a science learner and developing an identity as someone who knows about, uses, and sometimes contributes to science. Vincent‐Ruz, Schunn (2019) argued that an individual’s identities are a constant negotiation of their goals, cultural expectations, and the different social identities they choose to endorse. From the definitions listed by the researchers above, it is apparent that science identity is essentially a student’s perception of themselves as a science person.

Conceptual frameworks for understanding science identity vary, including two-dimensional, three-dimensional, and four-dimensional framework (Creighton 2007; Hazari et al. 2010; Carlone and Johnson (2007). Creighton (2007) proposes a two-dimensional framework, “science self” and “science identity in social practice.” In contrast, Carlone and Johnson (2007) propose a comprehensive three-dimensional framework for understanding science identity. The first dimension, recognition, encompasses an individual’s self-perception as a “science person” and the acknowledgment of others toward them as such. The second dimension, competence, is defined as possessing scientific knowledge and the ability to comprehend scientific content. Lastly, the third dimension, performance, refers to one’s capability to engage in relevant scientific practices. Hazari et al. (2010) extended this to a four-dimensional framework, incorporating “recognition-competence-performance-interest” highlighting the role of scientific interest in shaping a robust science identity. These frameworks offer valuable insights for designing questionnaires and promoting students’ science identity development.

Situated Learning Theory, Symbolic Interactionism, and Identity theory significantly influence science identity (Gunstone 2016). Wenger (1999) asserts in her book “Communities of Practice: Learning, Meaning, and Identity” that identity is not only based on external perceptions but also evolves through practice, where students represent themselves intentionally and become aware of the meaning of their identity through interaction with the situation and others. This process allows them to generate and reinvent themselves in practice and gradually develop from novice to expert identity (Wenger 1999). Identity theory focuses on the influence of an individual’s role in a social structure on activating an identity rather than other identities. It examines how social structures affect self-structures (Stets and Burke 2000). Symbolic Interactionism holds that society shapes the self and social behavior. Social roles depend on role expectations in relational networks, and identities are internalized role expectations (Mead 1934). The process of identity formation is influenced by both external social structures and the development of ego structures (Wenger 1999).

Social identity theory posits identity as a contextual and socialized self-concept formation, influenced by time, place, and background (Abrams and Hogg 1990). The literature review reveals factors affecting science identity categorized into social, teaching, personal, and other factors (Beeton 2007; Fajardo 2015; Kim 2017; Le et al. 2019). Studies identify three main approaches to cultivating science identity. First, through social interactions within school, classroom, family, teachers, and peers. Second, through formal or informal science learning (Chapman 2013; Kolodner and Clegg 2010; Cook 2011; Dawson et al. 2019; Huvard et al. 2020; Olitsky 2006; Ong et al. 2011; Riedinger 2011; Zhai et al. 2014). Third, through individual determinants like gender, race, scientific interest, self-efficacy, sense of belonging, and academic achievements (Chang 2015; Chapman and Feldman 2016; Dawson et al. 2019; Huffmyer et al. 2022; Jackson and Suizzo 2015; Nguyen 2015; Rosa 2018).

Fostering students’ science identity through inquiry learning

In the domain of science education, inquiry learning stands out as a widely adopted strategy for fostering scientific practice (Gunstone 2016). This approach empowers students to think independently and draw conclusions from data, aligning to cultivate scientific thinking (Minner et al. 2009). Streamlining the implementation of inquiry learning, researchers have proposed structured procedures, such as the five inquiry phases: orientation, conceptualization, investigation, conclusion, and discussion (Pedaste et al. 2015). Furthermore, Linn et al. (2003) introduced a knowledge integration (KI)-oriented inquiry learning model, emphasizing various facets, including diagnosing problems, critiquing experiments, planning investigations, revising views, researching conjectures, and constructing models. This model underscores the integration of multiple knowledge domains, fostering a comprehensive understanding of scientific concepts (Linn et al. 2003).

Inquiry learning, mimicking scientists’ work, plays a pivotal role in shaping and developing science identity (Archer et al. 2010; Carlone and Johnson 2007; Schön 2015). Empirical evidence attests to the positive impact of inquiry learning on science identity development. Meyer et al. (2014) propose that authentic science inquiry, particularly in citizen science projects, provides students with rich environments for inquiry learning, enhancing their science identities. Schon (2015) reported increased interest in science and identity development among students participating in outdoor science school inquiry learning experiences. Similarly, Richmond and Kurth (1999) investigated high school students’ understanding of scientists’ work, finding that inquiry learning provided students with insights into the life of a scientist. Rivera Maulucci et al. (2014) discovered that authentic science inquiry projects contribute not only to academic achievement but also to expertise awareness and the development of science identity.

Cultural and environmental factors play crucial roles in shaping science identities within the context of inquiry learning. Liu and Hannafin (2010) explored how students shape their identities during cross-cultural collaborative inquiry learning, identifying four intertwined identities: natural, institutional, affinity, and dialogic. Hanauer (2010) highlighted that students’ scientific competence, laboratory experience, and family background influence their science identity in laboratory inquiry learning. In summary, current research underscores the pivotal role of inquiry learning, the primary form of scientific practice, in shaping students’ self-concept as scientists through real-world activities, cultural influences, and environmental contexts, thereby contributing to the development of their science identity.

Web-based scientific inquiry environment and web-based inquiry

The web-based scientific inquiry environment integrates web resources into a structured series of inquiry activities (Slotta 2002). Characterized by interactivity, visualization, virtualization, openness, and autonomy (Linn and Eylon 2011; McBride 2017), this environment employs web and visualization technologies to offer a multisensory, interactive learning experience. It provides near-real computer simulations, facilitating virtual experiments unattainable in traditional settings (McBride 2017; Svihla et al. 2018; Vitale and Linn 2018), making it an effective instructional tool (Omarchevska et al. 2021). While some learners may not entirely embrace it as a substitute for traditional inquiry (Erdosne Toth et al. 2009; Kapici et al. 2019; Pyatt and Sims 2011; Winkelmann et al. 2014), it extends the reach of laboratory lessons to a broader and more diverse student population (Burkett and Smith, 2016), garnering positive attitudes from certain students (Pyatt and Sims 2011).

The Web-Based Inquiry Science Environment (WISE), developed by Marcia C. Linn’s team at the University of California, Berkeley, stands out as a platform for enabling inquiry-based learning across various science domains. WISE leverages web technology, simulation, and visualization to provide students with tools such as online discussions, data collection, automatic graphing, free-form argumentation, resource sharing, and simulations (Wiley et al. 2019). The STEM learning unit within WISE is designed to align with the KI-oriented inquiry learning model, involving eliciting, adding, distinguishing, and reflecting on ideas to structure inquiry activities (Wiley et al. 2019). Topics like Photosynthesis, Global Climate Change, Plate Tectonics, and Self-Propelled Vehicles empower students to engage in inquiry by diagnosing problems, evaluating experiments, making choices, speculating, gathering information, constructing explanations, debating with peers, and forming coherent arguments (McBride et al. 2018). Although WISE deviates from traditional inquiry by using virtual labs and equipment, both inquiry methods emphasize active student participation, hands-on learning, and experiencing the inquiry process for scientific knowledge acquisition, comprehension of scientific ideas, and methodological learning. Guided inquiry in WISE enhances its overall effectiveness (Linn and Eylon 2011).

The advantage of web-based inquiry is its integration of technology-enhanced inquiry with global networked information resources, sociocognitive research, and technological advancements in responsive learning environments (Linn et al. 2013). Empirical studies validate WISE as an efficacious tool for inquiry-based learning, bolstering student interest, engagement, inquiry skills, conceptual understanding, and facilitating effective self-directed learning (Boda et al. 2021; Matuk et al. 2019; Petra et al. 2016; Slotta 2013; Ulus and Oner 2020). Over the past decade, Chinese science education researchers have examined the effectiveness of WISE, grounded in the KI theoretical framework, in enhancing students’ science learning (Pei et al. 2020). However, research literature addressing the impact of web-based inquiry on science identity remains lacking.

Theoretical framework and research hypothesis

Previous literature indicates that web-based inquiry within the STEM learning unit on the WISE platform enhances inquiry learning through technology. This inquiry method significantly boosts students’ interest, engagement, inquiry skills, and conceptual understanding, facilitating self-directed learning (Boda et al. 2021; Matuk et al. 2019; Petra et al. 2016; Slotta 2013; Ulus and Oner 2020). Importantly, these elements are crucial for strengthening students’ science identity. Web-based inquiry integrates technology and practical elements, offering a comprehensive learning experience. Technological aspects include digital tools such as multimedia, simulations, and collaboration tools, which create an interactive learning environment (Donnelly et al. 2012; Toth et al. 2008; Kapici et al. 2019; Pyatt and Sims 2011). Additionally, these tools support personalized learning that adapts to individual needs and interests. Practically, web-based inquiry focuses on applying scientific principles to address real-world challenges, promoting authentic investigations, experiment design, data analysis, and effective communication (Donnelly et al. 2012; Meyer et al. 2014). This method cultivates critical thinking, problem-solving, and communication skills vital for effective inquiry (Meyer et al. 2014). The blend of technological and practical elements in web-based inquiry provides an interactive and authentic platform, deepening students’ understanding of scientific concepts.

Moreover, web-based learning environments are characterized by their interactivity, visualization, openness, autonomy, and virtuality, providing scientifically authentic learning experiences (Donnelly et al. 2012; Meyer et al. 2014; Omarchevska et al. 2021; Pyatt and Sims 2011; Winkelmann et al. 2014). These attributes not only boost students’ interest in science and inquiry but also increase their participation, immersion, and confidence in conducting independent inquiries, leading to an engaging and immersive learning experience (Reilly et al. 2021). Engaging in practical science activities builds students’ confidence in their inquiry skills and fosters their self-perception as budding scientists. Consequently, this enhances the “recognition”, “competence” and “performance” dimensions of their science identity.

The current study posits that web-based inquiry, analogous to traditional inquiry, plays a pivotal role in fostering students’ science identity. This hypothesis is underpinned by the active and experiential characteristics shared by both forms of inquiry, which are crucial for acquiring scientific knowledge, comprehending scientific reasoning, and mastering scientific methodologies. By engaging in authentic scientific practices, students parallel the investigative processes of scientists. The theoretical framework and research hypothesis that assert the effectiveness of web-based inquiry in fostering students’ science identity are depicted in Fig. 1.

Fig. 1
figure 1

Theoretical framework and research hypothesis for fostering students’ science identity through web-based inquiry.

Full size image

Methodology and materials

Measuring tools

This study employs a comprehensive measurement tool consisting of three distinct sections. The first section delves into primary school students’ demographic information, encompassing age, gender, place of origin, and family economic situation (see Appendix 1). The second section, rooted in prior research highlighting the impact of parental and peer influences, as well as students’ informal science learning experiences on science identity, investigates these variables as control measures before the commencement of the experiment (see Appendix 1). The third section employs the Chinese version of the science identity measurement tool adapted from Schon (2015). English science identity measurement is used for elementary school students based on Carlone and Johnson’s (2007) “recognition-competence-performance” three-dimensional framework. Comprising 12 items rated on a 5-point Likert scale, ranging from “strongly agree” (5) to “strongly disagree” (1), the science identity questionnaire is included in Appendix 2. Rigorous validity checks were conducted through questionnaire reliability analysis and confirmatory factor analysis using AMOS 18.0. The outcomes revealed that IFI > 0.9 and CFI > 0.9, with NFI and GFI, though slightly below 0.9, remaining close, indicating satisfactory empirical values. Correlation analysis demonstrated a high positive correlation between factors and overall concepts, attesting to good consistency. The Cronbach’s α coefficient for the total questionnaire was 0.918, split-half reliability stood at 0.876, and coefficients for the three factors ranged between 0.822–0.871, with split-half reliability between 0.822–0.876, affirming the questionnaire’s robust main factor structure.

Participants

The study involved 106 fifth-grade students from an elementary school in Shenzhen, China, distributed across two classes. Before the experiment, all students in six fifth-grade classes completed a pre-test on their science identity scores through a questionnaire, and their science scores from the second grade of the fourth year were collected. Two classes with similar levels were then selected for the study, one class, consisting of 52 students, was designated as the experimental group, while another class, with 54 students, served as the control group. The average age of the participants was 10.76 years, with 47 boys (44.3%) and 59 girls (55.7%). To minimize the impact of teacher differences on the experimental results, the two classes were taught by the same teacher.

Learning materials

This study utilized the “Solar Oven” STEM learning unit from WISE as the learning material, focusing on engineering, technology, and applications of science. This learning unit aims to familiarize students with the conversion of solar radiation to heat through hands-on projects and interactive models. The “Solar Oven” STEM learning unit involves designing, building, testing, and evaluating solar ovens, fostering critical thinking and problem-solving skills (McBride et al. 2018). While generally suitable, the study identified areas where the learning unit required adaptation to better align with the context of Chinese students. Consequently, a secondary development process was undertaken, involving adjustments in learning steps selection, sequence, problem descriptions, contextual introductions, grading criteria, characters, and images in the original “Solar Oven” STEM learning unit on WISE. Two adapted versions, the WISE-adapted STEM project, and the traditional adapted STEM project, were developed, with the latter designed based on the “Solar Oven” STEM learning unit.

Considering the content of the “Solar Oven” STEM learning unit and the specifics of experimental schools and samples, secondary development was undertaken with three objectives: firstly, to appropriately reduce the difficulty by adjusting content complexity, teaching sequence, and evaluation criteria; secondly, to localize the content by providing Chinese names and reviewing relevant fifth-grade science curriculum knowledge on “light and heat” as well as “solar water heaters”. thirdly, reorganize learning materials by supplementing WISE tips on tool operation and offering task sheets for learning tasks. The adapted versions underwent rigorous scrutiny and approval from a science education expert along with three science teachers.

Both the WISE-adapted and the traditional adapted versions represent innovative STEM project-based learning grounded in constructivist teaching, student-centered approaches, and interdisciplinary methods. The learning units address real-world problems, tasking students with constructing a “warm house” for tomatoes in winter. Students integrate scientific knowledge, mathematical calculations, engineering design, language expression, and writing to propose hypotheses, design solutions, make solar ovens, collect data, analyze and compare results, and reflect on the learning process. However, the distinction lies in that the WISE-adapted STEM project’s learning activities are grounded in the KI-oriented inquiry learning model, whereas the traditional adapted STEM project’s learning activities are founded upon the general inquiry learning model. The KI-oriented inquiry learning model pays more attention to the integration of knowledge, facilitating students to integrate new knowledge with existing knowledge to achieve deep understanding, while the general inquiry learning model focuses more on developing inquiry skills and solving specific problems.

The experimental group experienced the WISE-adapted “Solar Oven” featuring seven learning activities and 33 learning steps in the web-based inquiry setting. Subsequently, the group collaboratively designed and produced their “Solar Oven” The traditional “Solar Oven” STEM learning unit was adapted to the WISE original version, guided by the inquiry model, resulting in five learning task sheets. control group engaged in traditional inquiry with nine activities and 35 learning steps before designing their “Solar Oven”.

Experimental procedure

The experimental procedure encompasses three phases: pre-experimental, experimental, and post-experimental, as illustrated in Fig. 2.

Fig. 2
figure 2

Experimental procedure.

Full size image

Before the experiment, a pre-test assessed the science identity of the two groups, revealing comparable statistical levels. Subsequently, the experimental group and control group underwent a 30-min training session covering inquiry topics, group assignments, inquiry learning guidance, and learning requirements. Additionally, the experimental group received a 20-min training on computer and WISE platform operation to familiarize students with the online environment and mitigate the novelty effect.

Throughout the experiment, the experimental group engaged in a 6-lesson (240 min) Web-based inquiry of the traditional adapted “Solar Oven” on WISE, followed by 2 lessons (80 min) of work presentation and exchange, totaling 8 lessons (320 min). The control group, in a traditional classroom setting, studied the traditional adapted “Solar Oven” with 2 lessons (80 min) of scientific concept learning, group discussion, and classroom data analysis, 4 lessons (160 min) of two rounds of solar oven inquiry experiments in the laboratory, and 2 lessons (80 min) of working presentation and exchange, also totaling 8 lessons (320 min), all groups received identical instructional time.

Following the experiment, all students in the two groups underwent a post-test to evaluate their science identity.

Data analysis

The data analysis is structured into three key phases. Initially, an independent sample t-test was applied to control variables including gender, science final exam scores, family economic status, extracurricular science training, peer influence, and parental influence. This first step was designed to establish baseline levels for the two groups and to mitigate potential confounding effects from these variables. The second phase involved conducting an ANCOVA on pre-experimental science identity data for both the experimental and control groups, using science final exam scores as covariates to adjust for the influence of academic performance on science identity. Subsequently, a post-test on science identity was administered to both groups. A covariate analysis was then performed, with the inquiry method as the independent variable, and pre-test and post-test science identity scores as covariates and dependent variables, respectively. This analysis sought to compare the effects of web-based inquiry and traditional inquiry on elementary students’ science identity. In the final phase, detailed analyses were conducted separately for the experimental and control groups, focusing on the subdimensions and indicators of science identity. Paired t-tests were used to compare pre-test and post-test scores for recognition (SI-R), competence (SI-C), and performance (SI-P) within each group. Additionally, effect size analyses were performed to evaluate the benefits of each inquiry type on students’ science identity, clarifying which inquiry method had a more significant impact.

Analysis of control variables

Literature posits that students’ science identity is shaped by social, pedagogical, and individual factors derived from various sources such as schools, classes, families, teachers, peers, science lessons, and personal attributes (Hazari et al. 2010; Carlone and Johnson 2007; James 1998; Kane 2012; Kelly 2019; Schon 2015). It is worth noting that this study was conducted using a sample of students from the same school who were taught by the same science teacher. Therefore, this study only accounted for differences in gender, final science exam scores, family economic status, number of science training outside of school, peers’ influence on science, and parents’ influence on science as control variables, and did not consider variations in influences from science teachers and schools.

The “Questionnaire on science learning for elementary school students” (see Appendix 1) was administered to collect data on these control variables across two groups. independent sample t-test detailed in Table 1, gaged whether significant differences in the control variables existed among the experimental group and control group.

Table 1 The independent sample t-test results of the control variables for the two groups.
Full size table

The results of the independent samples t-test on the control variables, as shown in Table 1, indicate that there were no significant differences among the experimental group and control group in terms of gender (t = −0.37, p > 0.05), final science exam scores (t = −0.97, p > 0.05), family economic status (t = −1.05, p > 0.05), number of science training outside of school (t = 0.33, p > 0.05), peers’ influence on science (t = 0.14, p > 0.05), and parents’ influence on science (t = −0.73, p > 0.05). It should be noted that the study was conducted in the same school and taught by the same full-time science teacher with over 20 years of teaching experience. Therefore, it is reasonable to assume that there were no significant differences between the school and the science teacher.

Before experimenting, the researchers investigated and analyzed various factors that could influence the results, including demographic information, final science exam scores, family economic status, peers’ and parents’ influence on science, and the number of science training outside of school. Through the analysis of these factors, it was determined that there were no significant differences between the experimental group and control group in terms of important influencing factors related to personal, teacher, class, school, family, and out-of-school experiences. Therefore, it can be concluded that the two groups were at equivalent levels before the experiment, providing a solid foundation for the subsequent analysis of the impact of different inquiry methods on science identity.

Comparative effects of web-based inquiry and traditional inquiry on elementary students’ science identity

To explore the impacts of web-based inquiry and traditional inquiry on elementary students’ science identity, this study analyzed pre- and post-test data from experimental and control groups using paired sample t-tests. Science identity scores for the experimental group significantly increased from the pre-test (M = 38.27, SD = 9.48) to the post-test (M = 43.40, SD = 9.28), p < 0.001. Likewise, the control group’s scores improved from the pre-test (M = 38.17, SD = 10.74) to the post-test (M = 41.92, SD = 11.77), p < 0.05. These findings suggest that both inquiry methods significantly bolstered science identity among elementary students, as illustrated in Fig. 3.

Fig. 3
figure 3

Box plot comparing pre-test and post-test results for the experimental and control groups.

Full size image

Additionally, to evaluate differences in science identity impact between the two methods, an independent sample t-test was performed on the baseline scores of both groups before the experiment, which showed no significant difference (t = 0.05, p > 0.05), indicating comparable science identity scores at the outset. To account for the potential impact of academic performance, science final exam scores were used as covariates in a covariance analysis. The homogeneity of variance test (F = 0.03, p > 0.05) verified that the data satisfied the analysis conditions. A subsequent covariance analysis, treating the inquiry method as the independent variable and the pre-test and post-test science identity scores as covariates and dependent variables, respectively, found no significant differences in the enhancement of science identity between the groups, as detailed in Fig. 3.

The comparison of the effects of web-based inquiry and traditional inquiry on elementary students’ science identity shows that while both methods effectively foster science identity, they do not differ significantly in their effectiveness. This result supports the study’s hypothesis that both inquiry methods positively influence students’ science identity.

Differences and effect sizes in science identity subdimensions and indicators between web-based inquiry and traditional inquiry

Both web-based inquiry and traditional inquiry were effective in fostering science identity development in elementary school students. However, it remains unclear whether there are any significant differences between the two groups and what these differences might be. To address this issue, further analysis is required, focusing on the changes in science identity scores across the three dimensions and 12 indicators in the experimental group and control group. Such analysis would provide valuable insights into the effectiveness of each intervention and help identify any unique features or advantages of web-based inquiry and traditional inquiry in fostering science identity.

To compare the pre-test changes in the experimental group and control group, the scores of three dimensions were analyzed separately for each group. Paired t-tests were conducted on the pre-tests and post-tests of recognition (SI-R), competence (SI-C), and performance (SI-P) in the experimental group. The results of the paired t-tests for the pre-test and post-test of science identity and its three dimensions are presented in Table 2.

Table 2 Results of pre and post-test paired t-test of three dimensions in the experimental group (n = 52).
Full size table

As shown in Table 2, presents the results of the paired t-tests conducted on the pre-test and post-test scores of three dimensions (SI-R, SI-C, SI-P) for the experimental group. The results for the three dimensions reveal a strong correlation between the pre-test and post-test scores, with correlation coefficients above 0.6. The mean value of the “recognition” dimension after the experiment (M = 3.00, SD = 1.11) was significantly higher (p < 0.001) than before the experiment (M = 2.21, SD = 0.97). The mean value of the “competence” dimension after the experiment (M = 3.88, SD = 0.74) was also significantly higher (p < 0.05) than before the experiment (M = 3.65, SD = 0.80). Similarly, the mean value of the “performance” dimension after the experiment (M = 3.60, SD = 0.93) was significantly higher (p < 0.001) than before the experiment (M = 3.11, SD = 0.97).

The evaluation of the actual significance of the experimental effect can be inferred from the effect size d, In Cohen’s definition, “d = 0.2” indicates a “small” effect size; “d = 0.5” represents a “medium” effect size, and “d = 0.8” means “large” effect size (Cohen 1988). The results of the paired t-tests and the calculation of the effect size d suggest that the differential effect of the pre- and post-tests of science identity is medium (t = 5.71, p < 0.001, d = 0.6). Among the three dimensions of science identity, the “recognition” dimension has the largest differential effect, indicating a “large” effect (t = 6.33, p < 0.001, d = 0.8), followed by the “performance” dimension with a “medium” effect (t = 4.85, p < 0.001, d = 0.5), and then the “competence” dimension with a “small” effect (t = 2.67, p < 0.05, d = 0.36). Therefore, it can be concluded that web-based inquiry and traditional inquiry have a positive impact on students’ science identity development, and the “recognition” dimension had the most significant impact, followed by the “performance” dimension.

The findings of the data analysis indicate that web-based inquiry has a positive impact on elementary school students’ science identity development, specifically on their recognition, competence, and performance dimensions. The effect size analysis suggests that the “recognition” dimension had the strongest effect, followed by the “performance” dimension, while the “competence” dimension had a relatively smaller effect. These results suggest that web-based inquiry can be an effective approach to fostering science identity of elementary school students.

Paired t-tests were performed in the control group to examine the pre-test and post-test differences in recognition (SI-R), competence (SI-C), and performance (SI-P). Table 3 displays the results of these t-tests for the three dimensions of science identity as a whole in the control group.

Table 3 Results of pre-test and post-test paired t-test of three dimensions in the control group (n = 54).
Full size table

As shown in Table 3, it can be observed that there is a substantial correlation between the pre-test and post-test for the recognition, competence, and performance dimensions of science identity endorsement in the control group. Furthermore, there are significant changes in these dimensions before and after the experiment. After the experiment, the mean for the “ recognition” dimension (M = 2.86, SD = 1.34) is significantly higher (p < 0.05) than the pre-experiment mean (M = 2.45, SD = 1.10). The mean for the “competence” dimension after the experiment (M = 3.74, SD = 0.98) does not show a significant increase (p > 0.05) compared to the pre-experiment mean (M = 3.56, SD = 0.97). However, the mean for the “performance” dimension after the experiment (M = 3.50, SD = 1.22) is significantly higher (p < 0.01) than the pre-experiment mean (M = 3.09, SD = 1.03).

The actual significance of the experimental effect was assessed through the calculation of the differential effect between the pre-test and post-test science identity and its three dimensions. The results indicate a “small” differential effect for science identity (t = 2.39, p < 0.05, d = 0.3), a “small” differential effect for the “recognition” dimension (SI-R) (t = 2.25, p < 0.05, d = 0.33), a “small” differential effect for the “performance” dimension (SI-P) (t = 2.78, p < 0.01, d = 0.36), and a “small” differential effect for the “competence” dimension (SI-C) (t = 1.17, p > 0.05, d = 0.18).

After analyzing the data, it can be concluded that traditional inquiry has a positive impact on elementary school students’ science identity, recognition, competence, and performance dimensions. However, the effect sizes of this intervention on science identity, performance, and recognition dimensions were small. Conversely, the effect on the competence dimension was negligible, as evidenced by the minimal change in mean values. Additionally, there was almost no effect on the “competence” dimension.

To investigate the differences between pre-test changes in the experimental group and control group, the changes in 12 science identity indicators under the three dimensions of science identity were analyzed separately. Specifically, paired t-tests were performed on the pre-test and post-test data of the 12 indicators in the experimental group. The mean values and paired t-test results for the science identity indicators in the experimental group are presented in Table 4.

Table 4 Means and paired t-test results for science identity indicators in the experimental group (n = 52).
Full size table

Table 4 displays the results of paired t tests conducted on the pre-test and post-test data of the 12 indicators of science identity for the experimental group. The data reveals that the correlation coefficients for both the pre-test and post-test data are greater than 0.4, indicating a moderate to large correlation. Out of the 12 indicators, four (SI1, SI2, SI3, and SI11) showed a significant increase in post-test means compared to pre-test means (p < 0.001), there (SI4, SI9, and SI10) showed a significant increase in post-test means compared to pre-test means (p < 0.01). However, five indicators (SI5, SI6, SI7, SI8, and SI12) exhibited non-significant differences (p > 0.05) between pre-test and post-test means. Paired t-tests were also performed on the pre-test and post-test data of the 12 indicators of science identity in the control group. The mean values and paired t-test results for the science identity indicators of the control group are reported in Table 5.

Table 5 Means and paired t-test results for science identity indicators in the control group (n = 54).
Full size table

Table 5 displays the means and paired t test results for the pre-test and post-test of the 12 indicators of science identity in the control group. The analysis revealed that out of the 12 indicator data (SI2, SI4, and SI9) had significantly higher post-test means compared to the pretest means (p < 0.05), SI10 had significantly higher post-test means compared to the pretest means (p < 0.01). Conversely, the pre-test and post-test mean for eight indicators (SI1, SI3, SI5, SI6, SI7, SI8, SI11, SI12) did not show significant differences (p > 0.05).

Based on the analysis of the pre- and post-measurement changes of the 12 indicators of science identity, it was observed that the number of significant indicators in the experimental group was higher than that in the control group. Notably, three indicators, namely, SI1, SI3, and SI11, showed significant differences between the experimental group and the control group. These differences were related to “SI1 self-recognition” “SI3 scientific knowledge” and “SI11 conclusion of experiments through reasoning and argumentation”. These findings suggest that the web-based inquiry was more effective in promoting these three indicators of science identity compared to traditional inquiry.

Discussions

This study aims to evaluate the effectiveness of web-based inquiry in fostering elementary school students’ science identity by comparing changes in science identity before and after experiments conducted under two inquiry methods: web-based inquiry and traditional inquiry. The goal is to expand the pathways for cultivating students’ science identity and enhance the sense of science identity among “digital natives”.

The investigation reveals that both web-based inquiry and traditional inquiry effectively foster primary school students’ science identity, aligning with prior research emphasizing the positive impact of engaging in scientific practices. Such participation, akin to scientists’ methods, aids students in acquiring essential skills, knowledge, and the scientific exploration process crucial for shaping their science identity (McGee 2016; Rivera Maulucci et al. 2014). This study contributes by scrutinizing the differential effects of web-based inquiry and traditional inquiry on science identity, filling a gap in existing research. Additionally, our theoretical hypothesis posits that web-based inquiry synergistically combines “technology” and “practice” to influence students’ science identity positively. The interactive, visual, open, autonomous, and virtual aspects of web-based inquiry promote interest, active engagement, and confidence in scientific exploration, enriching the “competence” and “performance” dimensions of science identity. This finding accentuates web-based inquiry’s suitability for “digital natives” and its potential to enhance their involvement in scientific activities.

What are the differences between web-based inquiry and traditional inquiry in fostering science identity among elementary school students? The web-based inquiry has a superior capacity to enhance the dimensions of “recognition” and “performance”. In contrast, both web-based inquiry and traditional inquiry exhibit minimal impact on the “competence” dimension. Participation in scientific practices is identified as a crucial factor in fostering students’ science identity, wherein heightened “recognition” results from students perceiving themselves to think and work “like a scientist” during the inquiry process, subsequently boosting self-confidence. The elevation of the “performance” dimension aligns with Carlone and Johnson’s perspective, explicitly stating that active engagement in scientific practices serves as the tangible manifestation of science identity (Carlone and Johnson 2007).

The limited impact of both forms of inquiry on the “competence” dimension can be attributed to the challenges outlined in Carlone and Johnson’s science identity framework, where the “competence” dimension is deemed more resistant to rapid improvement, especially concerning indicators such as “scientific experiments” “scientific tasks” and “scientific exam performance”. Investigation into the science classes at experiment school reveals a predominant emphasis on the impartation of knowledge from science textbooks, neglecting the development of students’ practice skills and scientific thinking abilities. Additionally, communication with science teachers highlights the prevalence of challenging pen-and-paper exams at the end of the term, contributing to students’ perceived difficulty and lack of confidence in the “competence” dimension.

Web-based inquiry has been shown to significantly enhance seven key indicators of students’ science identity, whereas traditional inquiry only improves four of these indicators. This discrepancy highlights the considerably greater effectiveness of web-based inquiry in nurturing the multifaceted development of science identity in elementary school students. Two primary factors are proposed to account for this notable contrast: firstly, the instructional theoretical model of the WISE, based on the KI-oriented inquiry learning model, proves more effective in encouraging students to think and work like scientists compared to traditional models (Linn and Eylon 2011). Secondly, the characteristics of the STEM learning unit within WISE, emphasizing the use of powerful learning technologies to promote cognitive development and encourage students to think and work like scientists, facilitate a more comprehensive improvement in students’ science identity, especially in the acquisition of knowledge related to solar radiation (Rivera Maulucci et al. 2014). In conclusion, the study highlights the distinctive effectiveness of web-based inquiry, particularly within the WISE, in promoting the nuanced development of elementary school students’ science identity.

Conclusions

This study addressed an important research issue regarding web-based inquiry used to foster students’ science identity. Comparative experiments provide empirical evidence confirming that both web-based inquiry and traditional inquiry are effective in fostering students’ science identity. Previous studies have only discussed how inquiry can promote the development of students’ science identity. In the information technology environment, the inquiry also contributes to the development of students’ science identity (Asbell-Clarke et al. 2012; Clarke and Dede 2009; Johnson-Glenberg et al. 2023). However, the information technology environments studied previously were games or virtual communities, rather than web-based inquiry environments. Thus, the findings drawn in this study are generally consistent with those of previous research. In comparison, the effectiveness of web-based inquiry in enhancing students’ science identity is higher than that of traditional inquiry. We hypothesize that the inquiry in the STEM learning units within WISE is more in line with the learning characteristics and needs of digital native students, which can better attract them to participate in the inquiry. Moreover, we propose that the KI-oriented inquiry learning model plays a pivotal role in shaping students’ science identity. Engaging with this model, students immerse themselves in authentic scientific practices, cultivating an identity that aligns with the roles of scientists, which in turn bolsters their self-confidence and enhances their recognition as emerging scientists. These findings contribute substantive insights to the extant body of literature concerning science identity.

These findings provide credence to the idea that science education and instruction have benefited immensely from the information technology environment’s innovation, which has the added benefit of piquing students’ interest in the subject and increasing their enthusiasm to learn. Students’ interest in and engagement with inquiry can be raised by web-based inquiry, which can also strengthen their science identity. On the other hand, traditional science education that emphasizes teacher-centered training lowers students’ enthusiasm for science and makes it challenging for them to develop a stronger sense of themselves as scientists. Our study endeavors to illuminate the contemporary landscape of science education in China and to prompt science educators and researchers to recognize the importance of science identity.

This study, which overcomes earlier constraints in science identity research, identifies the changing nature of education in the digital era and the unique learning traits of students who are digital natives. Examining the connection between web-based inquiry and science identity from an IT standpoint not only introduces new ways to develop science identity but also fits in with the requirements of creative teaching. In addition, this study offers a new perspective on scientific literacy development in the context of ongoing reforms in Chinese science education, with a particular focus on Chinese primary school pupils. It goes beyond the realm of cognition to highlight the significance of non-cognitive elements, particularly science identity.

Research limitations

The present study investigated the effects of web-based inquiry on students’ science identity. within a short 6-week period. The selection of experimental content was based on several criteria, including the appropriateness of the content, feasibility of the experimental operation, and moderate difficulty. The “Solar Oven” STEM learning unit within the WISE was chosen as the experimental content and was adapted to the actual teaching situation. The results indicated that the adapted “Solar Oven” was more suitable for the study subjects, the experimental process proceeded smoothly, and the teaching effect was positive. However, during the student interviews, two situations were observed: some students expressed disinterest in the learning topic and preferred to explore other topics, while others had a strong knowledge base and were already familiar with the content of the STEM learning unit and therefore were not interested in exploring it in depth. Thus, it is evident that the selection of experimental topics can have an impact on the development of students’ scientific interests, scientific participation, and science identity. Future research should consider offering students a variety of alternative learning content to choose from to enhance their engagement and interest. Additionally, what factors contribute to the greater appeal of web-based inquiry, compared to traditional inquiry, in enhancing students’ science identity with a larger effect size? This represents another limitation of the current study, as it has not been explored within this research. Future work will involve a thorough analysis of students’ learning behavior data and interview responses to address this question.