Smartphone-Assisted Experimentation as a Medium of Understanding Human Biology Through Inquiry-Based Learning

Abstract

The integration of Inquiry-Based Learning (IBL) and mobile technologies can transform science education, offering experimentation opportunities to students from budget-constrained schools. This study investigates the efficacy of smartphone-assisted experimentation (SAE) within IBL to enhance pre-service science teachers’ understanding of human physiology and presents a newly developed and validated rubric for assessing their scientific skills. Students (N = 286) from a Science and Mathematics Teacher Education Program participated in a summative IBL activity (“Investigating the Human Physiology”—iHPhys) where they designed experimental projects using smartphone applications to collect body sign data. The scoring rubric, assessing seven criteria including hypothesis formulation, methodological design, data presentation, and conclusion writing, was validated as substantial to almost perfect inter-rater reliability. Results reveal that students exhibited strong skills in hypothesis clarity, theoretical grounding, and experimental design, with a high degree of methodological innovation observed. However, challenges persisted in predictive reasoning and evidence-based conclusion writing. The students were strongly interested in inquiring about the cardiovascular and nervous systems. Correlational analyses suggest a positive relationship between project originality and overall academic performance. Thus, integrating SAE and IBL fosters critical scientific competencies, creativity, and epistemic cognition while democratizing access to scientific experimentation and engaging students in tech-savvy pedagogical practices.

1. Introduction

Active learning represents a pedagogical approach that places students at the center of the educational process, promoting their engagement in activities such as discussions, problem-solving, and critical reflection. According to Prince (2004), although challenges remain in measuring its effectiveness due to the diversity of outcomes and variables involved, there is consistent evidence that active methodologies enhance academic performance, develop higher-order cognitive skills, and foster positive attitudes toward learning. Moreover, studies have shown that the intensive use of active learning strategies can help reduce educational inequalities by significantly narrowing the achievement gaps between underrepresented students and their peers in science, technology, engineering, and mathematics (STEM) courses (Theobald et al., 2020).

Within this context, Inquiry-Based Learning (IBL) stands out as an active learning methodology that shifts the focus from the mere transmission of content to the active construction of knowledge. The IBL cycle includes question formulation, hypothesis generation, the design of data collection methods, data analysis and interpretation, the explanation of phenomena based on gathered evidence, and the communication of results—stages that align with the 5E instructional model (Engage, Explore, Explain, Elaborate, Evaluate), widely recognized as effective in promoting inquiry-based learning grounded in constructivist principles (Bybee & Landes, 1990; Jumaa & Ismail, 2023). Thus, IBL emphasizes student agency and supports the development of scientific competencies such as investigative autonomy, critical reasoning, data analysis, and evidence-based argumentation. Furthermore, IBL projects may vary in the degree of autonomy granted to students, ranging from structured approaches—where the problem and method are defined by the instructor—to open-ended formats, in which students determine the problem, formulate hypotheses, design experiments, and interpret data, thereby expanding their agency in the learning process (Přinosilová et al., 2013).

By adopting IBL as a guiding framework, this study also draws upon Project-Based Learning (PjBL), which organizes educational experiences around complex, extended projects. Despite conceptual differences, IBL and PjBL can be complementary. While IBL emphasizes scientific inquiry and problem-solving through systematic experimentation, PjBL fosters engagement with broad, authentic issues that require students to integrate interdisciplinary knowledge and devise creative solutions. According to Panasan and Nuangchalerm (2010), integrating PjBL and IBL can enhance learning by allowing the application of scientific concepts in meaningful, collaborative contexts.

In IBL, activities such as designing data collection methods demand that students operationalize variables, select measurement instruments, and anticipate potential sources of error—essential aspects of scientific inquiry. This phase fosters not only technical skill development but also metacognitive reflection, as students must justify their methodological choices based on criteria such as validity, reliability, and ethics. Manz (2015) argues that enabling students to formulate and test their own experimental procedures empowers them to understand how knowledge is constructed and strengthens their ability to critically evaluate empirical evidence as long as the students comprehend the social context of evidence construction. Similarly, Nahar and Machado (2025) demonstrated that students engaged in IBL projects focused on methodological design showed substantial gains in creativity and critical thinking. These findings are supported by Wulandari et al. (2025), whose bibliometric analysis highlights the role of IBL in developing procedural competencies and fostering more refined epistemic cognition. The presence of this approach at various educational levels underscores its importance in the training of future teachers, particularly those enrolled in undergraduate science education programs. Pre-service teachers who engage in authentic investigative practices become better equipped to understand the challenges of science teaching and to design instructional interventions aligned with the nature of science.

The theoretical foundation of this study lies at the intersection of inquiry-based learning (IBL), Project-Based Learning (PjBL), and smartphone-assisted experimentation (SAE)—which is further explained—in science education. While IBL and PjBL share common constructivist underpinnings, they differ subtly in their focuses and structure. The IBL framework focuses on fostering scientific thinking by engaging students in the formulation of questions, designing investigations, and analyzing empirical data, thereby simulating the authentic practices of scientists (Bybee & Landes, 1990; Furtak et al., 2012). The PjBL framework, in contrast, centers around complex, often interdisciplinary problems that extend beyond the classroom and may not always require empirical investigation (Panasan & Nuangchalerm, 2010), often centered on producing a final product or solution that serves as the culmination of the learning process (Sahin, 2013). Both inquiry-based learning (IBL) and Project-Based Learning (PjBL) informed the pedagogical design of the didactic activity proposed in this study, although neither framework was used to guide the methodological procedures of the research per se. While these approaches share a student-centered orientation that promotes learner autonomy and active engagement, they diverge in important ways. A common feature of both frameworks is their emphasis on empowering students to take ownership of their learning by posing questions, exploring phenomena, and constructing knowledge through experience (Pedaste et al., 2015; Kokotsaki et al., 2016). However, one of the defining characteristics of PjBL—the strong emphasis on collaboration and peer interaction as mechanisms for social learning (Lee & Lee, 2025)—was not emphasized in our implementation. This decision was motivated by the need to design an activity that could also be conducted asynchronously and independently, particularly in remote or non-school settings. Moreover, both frameworks highlight the importance of real-world relevance, encouraging students to address problems grounded in their everyday lives (Kurt & Akoglu, 2023). In our implementation, this principle was integrated into the instruction phase, where students were guided to formulate inquiries using physiological variables observable in their own bodies through smartphone-based data collection. A notable distinction between the two approaches lies in their structural demands. While IBL allows for more flexible inquiry paths, PjBL typically involves the production of a predefined final product, such as a report or presentation, with clearly articulated evaluation criteria (Hasni et al., 2016). Our study incorporated this structured element by requiring students to submit written reports based on predefined sections, which were assessed using a validated rubric—thus blending the openness of inquiry with the formality of academic reporting. The integration of IBL and PBL has been recommended in several studies, especially in contexts that require both disciplinary depth and authentic engagement with real-world phenomena (Wulandari et al., 2025; Rodríguez et al., 2019). In our study, IBL provided the primary pedagogical framework, while PBL informed the structure and scope of the student projects. This hybrid model encouraged students to define scientific questions grounded in human physiology, design empirical investigations using smartphone technologies, and communicate their findings through structured reports.

With the advancement of digital technologies, new opportunities have emerged to expand the scope of IBL. The constant presence of smart whiteboards, tablets, 3D printers, and smartphones materialize Papert’s (1990) pioneering view that the presence of computers in education should be seen as an opportunity to rethink the role of teaching and learning in a technologically mediated society. For Papert, the meaningful use of technology goes beyond merely introducing devices into the classroom; it should transform teaching and learning processes by connecting them to students’ digital language and culture. From a constructionist standpoint, technological tools should not be treated as external to the educational process but as integral components of learning, capable of materializing ideas, enabling experimentation, and mediating significant conceptual reflections (Harel & Papert, 1991). However, the impact of educational technologies depends on how they are used. Studies such as Gaudreau et al. (2014) show that positive behaviors related to computer use—such as taking digital notes, conducting complementary research, and using pedagogical resources—are associated with better academic performance. Conversely, the use of social media and unrelated applications during class can lead to distraction and lower performance. Moreover, Wacks and Weinstein (2021) warn of the negative effects of excessive mobile device use among adolescents, including decreased intelligence quotient, learning difficulties, and issues with sleep and socialization.

Regarding this panorama of possibilities, there is an urgent need to reconsider the role of technology in education—not as an obstacle, but as a tool that can enhance learning. Smartphones have surpassed computers as the primary means of internet access. Additionally, mobile devices can support culturally situated pedagogical practices and promote educational equity in contexts marked by economic disparities and limited access to digital resources (Martins, 2023). Nevertheless, institutional and sociocultural resistance continues to hinder the consolidation of such practices. As Anshari et al. (2017) discuss, many educators still grapple with the tension between the pedagogical potential of smartphones and the risks of distraction or misuse in the classroom. These authors emphasize that effectively integrating such devices into classrooms requires clear rules to mitigate distractions and maximize their pedagogical benefits. In this context, school smartphone bans, although often driven by legitimate concerns about focus and discipline, may prove counterproductive. By prohibiting these devices, we risk excluding from the educational environment one of the most accessible and versatile tools for conducting investigative, collaborative, and experimental activities—especially in the context of inquiry-based learning.

Emerging from this reality are proposals such as mobile learning (m-learning) strategies, which leverage mobile devices to provide temporal and spatial flexibility in education (Holzinger et al., 2005). However, implementing experimental activities—particularly in subjects such as human physiology—remains a challenge, especially in budget-constrained schools and remote learning contexts. Specific pedagogical tools for such content are scarce, requiring creative and low-cost solutions. In order to overcome those financial barriers, we developed didactic strategies named Mobile Learning Laboratories, which are mobile device applications that transform smartphones into portable laboratories. These tools use built-in or external sensors to measure actual body signs and variables such as heart rate, temperature, and motion, enabling authentic experimental activities both inside and outside the classroom (Nadal & Lellis-Santos, 2015; Lellis-Santos & Abdulkader, 2020). The MobLeLabs stem from earlier approaches such as computer-assisted experimentation and computer-assisted learning, which evolved with the advent of the internet into e-learning platforms and, subsequently, into m-learning environments.

Given the potential of mobile-device-mediated scientific experimentation to support investigative practices in human biology education, this study is justified by the need to better understand the impacts of this approach on the learning of pre-service science teachers. Considering the challenges educators face when implementing active methodologies with technological support, especially in early teacher training, it becomes crucial to develop assessment tools that evaluate not only the content learned but also the scientific competencies mobilized throughout the process. Therefore, the objective of this study is to develop an assessment rubric aimed at evaluating the learning of first-year students in a Science and Mathematics Teacher Education Program who participated in inquiry projects involving smartphone-mediated experimentation. Through the application of this rubric, we aim to identify which skills and competencies were developed by the students, and to reflect on the formative potential of this didactic intervention in the teaching of human physiology. In light of the gaps identified in the literature, our study poses the following research questions:

R1:

How can an analytic rubric be developed and validated to assess students’ scientific reasoning in smartphone-assisted inquiry-based learning activities?

R2:

What critical challenges in student outcomes can be observed when using smartphone-based IBL strategies, particularly in scientific method reasoning and presentation?

R3:

In terms of the diversity and variability in selecting vital signs and body variables, to what extent does smartphone-assisted experimentation foster creativity, innovation, and methodological design in pre-service science teachers’ projects?

2. Material and Methods

2.1. Description of “Investigating the Human Physiology” (iHPhys)

The activity “Investigating the Human Physiology” (iHPhys) is a summative assessment implemented as part of the overall performance evaluation of students enrolled in the course Human Body: Structure and Function offered by the Institute of Environmental, Chemical, and Pharmaceutical Sciences at the Federal University of São Paulo (Unifesp), Diadema campus. This course is part of the curriculum for first-year undergraduate students enrolled in a Science and Mathematics Teacher Education Program, which prepares future educators in biology, chemistry, physics, and mathematics. Unifesp is a public university in Brazil, often ranked in the top 10 best universities of the country. Our teacher training program accepts students from a broad spectrum of socioeconomic status, with the prevalent participation of first-generation students, minority groups, workers, and low-income citizens. The iHPhys activity was developed and implemented in a single institutional setting, chosen for its consolidated experience with mobile learning initiatives, commitment to active learning pedagogies, and convenience regarding the authors’ affiliation to the institution. The course Human Body: Structure and Function takes place during the second semester of the academic year, and accounts for 4 credits in a total workload of 72 h. The syllabus covers the fundamental principles of human anatomy, histology, and physiology. Active learning methodologies are employed throughout the semester, including the iHPhys project, which constitutes 20% of the final grade. In addition to iHPhys, two traditional summative exams and one weekly formative activity also contribute to the final grade. Instructions for conducting the iHPhys activity are introduced during the first class following the midterm exam. Students are guided by the course instructor to consider all stages of a scientific investigation project, creating a hypothesis that can be tested and either supported or refuted through data on human body variables. Students are individually asked to think about which human body variables could be investigated using smartphone applications. Subsequently, students receive a handout with detailed instructions and use a draft sheet to sketch their research plans. The objectives of the iHPhys activity are (i) to engage students in the authentic practice of scientific investigation by designing and executing their own experiments in human physiology; (ii) to promote the development of scientific reasoning competencies, including hypothesis formulation, methodological design, data interpretation, and conclusion writing; and (iii) to encourage creativity and innovation through the use of smartphone-based tools for measuring physiological variables. By working on inquiry-driven projects that involve real-time data collection, students are expected to deepen their understanding of human biology, gain familiarity with experimental procedures, and reflect on the role of digital technologies in contemporary science education. The instructor presents six evaluation criteria from the assessment rubric and explains that students must find one or more apps capable of collecting variables through internal smartphone sensors or external sensors attached to mobile devices. During the instructions, emphasis is placed on the “methodological innovation” criterion, which encourages students to explore variables other than heart rate, since this variable was already used during the cardiovascular physiology module. After classroom orientation, students had 4 to 5 weeks to complete all stages of the project and submit their research reports. Reports were limited to six pages, divided into four sections: hypothesis, methodology, results, and conclusion. Reports were evaluated according to the rubric, and results were analyzed based on the scores assigned to each criterion. As the data were non-parametric, the Kruskal–Wallis test, performed using GraphPad Prism 5.0a, was applied to compare students’ performance across the different levels. This study involved multiple artificial intelligence programs (Gemini 2.5 Flash, DeepSeek-V3, ChatGPT–4o) for writing and translating the text, not for writing the rubric or data analysis.

2.2. Development of the Rubric

The rubric aimed to evaluate students’ performance in designing and conducting a research project involving smartphone-mediated experimentation. Since the assessment instrument targets scientific reasoning competencies developed through inquiry-based learning (IBL), it was designed to evaluate the final scientific investigation product—namely, the research report. Although developed specifically for the iHPhys activity, the rubric can also be used in any inquiry-based learning project involving the use of smartphones for data collection related to human physiology. Thus, it serves both instructors and teaching assistants as an assessment tool and as a guide for students aiming to achieve the highest grades. Given that the activity involved short-term project completion with the results presented in written reports, the rubric was intentionally compact. An analytic rubric was chosen because it allows independent evaluation of multiple criteria with different weightings. The criteria assessed students’ competencies in formulating a scientific hypothesis, designing an appropriate methodology using smartphone sensors, collecting and organizing data into graphs, and writing a conclusion based on the obtained results. Since iHPhys contributed to the students’ final course grade, the rubric assigned numerical values to four performance levels: unsatisfactory (2.5 points), fair (5.0 points), good (7.5 points), and excellent (10 points).

2.3. Rubric Validation

Four former teaching assistants (TAs) of the course were invited to serve as evaluators to validate the rubric. These TAs were familiar with smartphone-assisted experimentation methodologies, having assisted students with similar experiments during their own service. Evaluations were conducted through individual online interviews, where each evaluator received four student project reports. The first report was used for acclimatization, allowing evaluators to familiarize themselves with the rubric’s criteria, the structure of student portfolios, and the online evaluation questionnaire. The remaining three reports—selected to represent low-, medium-, and high-quality projects—were used for statistical analyses. Cohen’s kappa statistic, calculated using GraphPad QuickCalcs, was applied to measure inter-rater agreement for each criterion. This measure was chosen because the rubric initially used qualitative ratings converted into a nominal scale. Interpretation of kappa values followed Landis and Koch (1977): 0.41–0.60 as moderate agreement, 0.61–0.80 as substantial agreement, and 0.81–1.00 as almost perfect agreement. Weighted kappa with linear weights was used, considering that scores were ordinal and proximity between scores mattered (e.g., a score of 7.5 compared to 10 points was treated as closer than 5 to 10 points). Additionally, intraclass correlation coefficients (ICCs) were calculated using IBM SPSS version 20 for Windows to complement the kappa analysis, offering a variance-based view of agreement, particularly suitable for ordinal data (Fleiss & Cohen, 1973). ICC interpretation followed the same categories as for kappa. Moreover, exact and adjacent agreement percentages were calculated according to Eppich et al. (2015). Exact agreement was defined as the percentage of identical ratings assigned by two evaluators for the same item in the same project. Adjacent agreement included identical ratings plus those differing by one level.

2.4. Students’ Outcomes and Product Analysis

To identify the most critical evaluation criteria in students’ project reports produced under inquiry-based learning, data from 286 student reports submitted between 2016 and 2021 were analyzed. Participants were pre-service teachers enrolled in science teacher education programs specializing in biology, chemistry, physics, or mathematics. Most students were female (64.3%), in their expected academic term (78.1%), and enrolled in the daytime program (58.1%). Only students who provided informed consent for participation and authorized the use of their project reports were included. This study was approved by the Unifesp Research Ethics Committee (protocol number 64195922.2.0000.5505).

Reports were scored according to the validated rubric criteria. In addition, the types of variables investigated and the corresponding human physiology topics were categorized. As a single project could involve multiple variables and topics, the results are presented as percentages relative to the total number of reports. To investigate creativity and innovation competencies, projects were categorized based on app usage: imitation, utilization of the same type of app demonstrated in class (heart rate measurement); improvement, utilization of a heart rate app plus an additional app measuring a different variable; innovation, utilization of apps measuring variables other than heart rate. Project originality frequencies are presented using boxplots with median and interquartile range (IQR) values. Kruskal–Wallis tests were conducted using GraphPad Prism for statistical analysis.

2.5. Correlational Study

To explore potential relationships between student performance in iHPhys and other variables, Spearman’s rank correlation (rs) analyses were conducted using Past 3 software. Variables included midterm exam scores, final exam scores, formative assessment scores, final grades, and sociodemographic variables (gender, term, and study period). The results are expressed using correlation diagrams where gray-highlighted squares indicated significant correlations (p < 0.05). Stronger correlations were reflected by more compact ellipses and more intense coloration: positive correlations in blue and negative in red. Correlation strength was classified as follows: up to 0.333 as weak, up to 0.666 as moderate, and above 0.667 as strong. For qualitative data analysis, sociodemographic characteristics were coded as binary variables: gender (1 = male, 2 = female), academic term (1 = freshman, 2 = senior), study period (1 = daytime, 2 = evening), and originality (1 = imitation, 2 = improvement, 3 = innovation).

3. Results

3.1. The Rubric Is Concise and Practical

Considering the importance of each criterion for the development of scientific reasoning, the rubric (

Table 1. Investigating the Human Physiology (iHPhys) Rubric.

Criteria	Unsatisfactory (2.5)	Regular (5.0)	Good (7.5)	Excellent (10.0)
(1) Hypothesis (Clarity)	There is no mention of which conditions/estate the hypothesis will be applied.	The cited conditions/estate is incorrect or naive, AND did not specify which conditions/estate the hypothesis will be applied.	Correctly cited the conditions/estate but did not specify which conditions/estates the hypothesis will be applied.	Made it clear in which conditions/estate the hypothesis will be applied.
(2) Hypothesis (Rationale)	The hypothesis has no logic/foundation based on physiological principles.	The hypothesis was elaborated based on common sense (misconception/urban legend) of physiology OR the hypothesis is well-known, has been extensively tested, and the answer is obvious and premeditated.	The hypothesis was elaborated based on a logical question, BUT the theoretical basis was not centered on physiology.	The hypothesis was elaborated based on a question with logic AND a theoretical basis in physiology.
(3) Hypothesis (Prediction)	No predictions of outcomes or results arising from the proposed experiment were made.	Explicit or implicit in the question/text. It was wrongly or superficially cited/mentioned consequences and possibilities for predicting results.	Implicit. It was mentioned in some way the consequences and possibilities of predicting results, with or without examples/descriptions.	Explicit. It was objectively cited/mentioned the consequences and possibilities for predicting results, with examples/descriptions.
(4) Methodology (Design)	There was unorganized data collection, no use of methodological principles and no correlation between the app and the variable.	It presented reasoning in the collection of the variable with flaws in the choice of the detection app and did not develop the experiment based on scientific rigor.	Presented a rationale for collecting the variable with the chosen app, BUT presented “simplistic” errors of scientific rigor (e.g., lack of comparison groups/situations).	It complied with scientific rigor, including controls/basal conditions or mentioned the necessity of calculating variation (delta) AND presented a rationale for collecting the variable with the chosen app.
(5) Methodology (Innovation)	Only apps previously demonstrated in class to measure heart rate were used.	Used apps previously demonstrated in class BUT added new apps that also measure heart rate OR Used apps demonstrated in class for heart rate measurement plus a non-app collection tool.	Apps demonstrated in class were used, AND new apps that collect data from other variables were added.	Only unprecedented apps were used.
(6) Results	None of the essential elements of a graph/table were presented OR the graph only contains images taken directly from apps (screenshot) without any student-created data visualization.	The graphs present 1 to 2 of the items presented at the excellent level.	The graphs present 3 to 4 of the items presented at the excellent level.	The graphs present ALL of the following items: (a) all axes of the graph were identified; (b) experimental groups and conditions were correctly identified; (c) units of measurement were indicated, especially on the axes of the graph; (d) the type of graph is appropriate for presenting the collected data; (e) the graph was correctly described so that the data can be understood with the elements presented.
(7) Conclusion	No conclusions were made, only a report of the data OR the conclusion is not supported by the collected data.	The collected data does not necessarily contribute to the conclusion, despite the presence of an explanation of the relationship between the data and the concluding text in a manner that was not just a writing up of the results AND the conclusion is not related to physiology’s principles.	The collected data partially supported the conclusion, OR there was an explanation of the relationship between the data and the concluding text so that it was not just a description of the results AND the conclusion was presented based on physiology.	The collected data supported the conclusion entirely, AND there was an explanation of the relationship between the data and the concluding text, AND the conclusion was presented based on physiology.

) was designed to encompass all necessary qualitative aspects with appropriate weighting: three criteria (clarity, theoretical foundation, and prediction) were assigned to hypothesis formulation (weight 3); two criteria (scientific rigor and creativity/innovation) to experimental design (weight 2); one criterion to the results, focusing on the quality of graphs and data (weight 1); and one criterion to the conclusion, primarily assessing synthesis and coherence (weight 1).

The clarity of the hypothesis criterion aimed to assess students’ ability to present all necessary elements for understanding the main research question. The theoretical foundation criterion evaluated whether students were able to provide a sound theoretical basis for their experimental approach. The prediction criterion was designed to encourage students to propose experimental research projects rather than merely reporting trivial observations. It is important to note that students had previously participated in a smartphone-mediated experimentation class on heart rate measurement; therefore, the innovation criterion specifically encouraged students to explore different smartphone sensors and applications. For grading purposes, each of the seven criteria could receive one of four scores: unsatisfactory (2.5 points), fair (5.0 points), good (7.5 points), or excellent (10.0 points). The final activity score was calculated as the sum of the points obtained across all criteria, with a maximum possible score of 70 points. A score of zero (0) was assigned when an item was missing. Terms in bold in the rubric highlight critical elements for evaluation, as identified during the rubric validation stage.

3.2. The Rubric Is Realiable

Validation results among evaluators were summarized in a matrix (

Table 2. Inter-rater weighted Kappa (wKappa) and Intraclass Correlation Coefficient (ICC).

	Rater 1 wKappa/ICC	Rater 2 wKappa/ICC	Rater 3 wKappa/ICC	Rater 4 wKappa/ICC
Rater 1	1
Rater 2	0.641 b/0.859 c	1
Rater 3	0.760 b/0.906 c	0.682 b/0.895 c	1
Rater 4	0.464 a/0.769 b	0.553 a/0.798 b	0.501 a/0.821 c	1

^a Moderate agreement (Kappa and ICC between 0.41 and 0.60); ^b substantial agreement (Kappa and ICC between 0.61 and 0.80); ^c almost perfect agreement (Kappa and ICC between 0.81 and 1.00) according to Landis and Koch (1977).

) containing the weighted kappa (wKappa) and intraclass correlation coefficient (ICC) values. For the wKappa analysis, three evaluator pairs showed moderate agreement, and three pairs showed substantial agreement, with an overall mean agreement value of 0.6, classified as moderate. Although mean kappa values indicated substantial agreement, the weighted kappa test is considered more appropriate for this rubric type. Changes to the rubric did not significantly improve wKappa values. The ICC analysis yielded highly satisfactory results, confirming that the point attribution system helped evaluators in their judgment tasks. The ICC values showed two substantial agreements and four almost perfect agreements (Table 2), with a mean ICC value of 0.841, classified as almost perfect agreement.

To further strengthen the validation, exact and adjacent agreement percentages were calculated, as recommended by Jonsson and Svingby (2007). The mean percentages were 66.7% for exact agreement and 93.7% for adjacent agreement (

Table 3. Overall percentage of exact and adjacent agreement.

Pairs of Raters	Exact Agreement (%)	Adjacent Agreement (%)
1 & 2	71.43	95.24
1 & 3	80.95	95.24
1 & 4	57.14	85.71
2 & 3	71.43	100.00
2 & 4	66.67	90.48
3 & 4	52.38	95.24

). Variations ranged from 52.38% to 80.95% for exact agreement, and from 85.71% to 100% for adjacent agreement. Notably, the exact agreement fell within the expected 55–75% range for most rubric validation studies, and the adjacent agreement exceeded 90%, representing an excellent level of consistency.

3.3. Students Master Methodological Reasoning but Struggle in Predicting and Concluding

Analysis of the first criterion (hypothesis clarity) revealed that most students performed at good (45.84%) or excellent (38.57%) levels. Performances categorized as good (REGU 1.58 ± 1.01 vs. GOOD 45.84 ± 2.90, p < 0.001) and excellent (REGU 1.58 ± 1.01 vs. EXCE 38.57 ± 3.21, p < 0.001) were significantly higher compared to fair performances (Figure 1A), indicating students’ competence in formulating clear hypotheses. For the “hypothesis foundation” criterion, results were even more satisfactory, with 87.38% of students achieving good or excellent scores (Figure 1B). Excellent performances significantly prevailed over fair and unsatisfactory performances (EXCE 60.76 ± 4.53 vs. REGU 7.74 ± 1.02, and vs. UNSA 4.88 ± 1.82, p < 0.001), showing students’ ability to support scientific hypotheses with theoretical knowledge rather than common-sense arguments. However, the third criterion, concerning the predictive nature of the hypothesis, showed slightly unsatisfactory results: 46.53% of students scored unsatisfactory or fair (Figure 1C). Although there was a higher occurrence of good scores compared to fair (GOOD 33.21 ± 4.35 vs. REGU 7.96 ± 2.14, p = 0.0013), a significantly greater number of students obtained unsatisfactory scores (UNSA 38.57 ± 5.55 vs. REGU 7.96 ± 2.14, p = 0.0013). This suggests a difficulty among first-year students in formulating predictive hypotheses, likely due to limited prior academic exposure to scientific methodology and writing.

Regarding the ability to design scientific experiments, students demonstrated adequate competence (Figure 1D), with 75.22% scoring good or excellent. There was a significant prevalence of good performances over fair and unsatisfactory (GOOD 53.57 ± 3.14 vs. REGU 16.78 ± 3.33, and vs. UNSA 7.99 ± 3.20, p = 0.0013). The most common error was the improper selection or omission of a control or comparison group. The “innovation” criterion yielded the best qualitative performance (Figure 1E), with 65.16% of students achieving excellent scores. A significant difference was observed only between excellent and fair categories (EXCE 65.16 ± 3.03 vs. REGU 5.71 ± 1.82, p = 0.0011), demonstrating students’ ability to explore new physiological variables and applications. In terms of presenting results, most students (70.28%) achieved good or excellent ratings (Figure 1F), confirming expectations given their interdisciplinary training. There was a significantly lower frequency of unsatisfactory performances compared to good and excellent (UNSA 10.12 ± 2.85 vs. GOOD 34.30 ± 4.78, and vs. EXCE 35.98 ± 6.19, p = 0.0024). Finally, regarding the ability to write conclusions, students faced difficulties (Figure 1G). Although no significant differences were observed among categories, a higher frequency of unsatisfactory performances (32.31%) was noted. Common mistakes included merely restating results without drawing conclusions or making unsupported inferences.

3.4. Inquiry Leads to Innovative and Creative Thinking

One way to assess students’ creativity in developing new experiments was by analyzing the diversity of selected variables. Students selected variables using smartphone apps, either relying on internal sensors or externally connected devices.

Heart rate remained the most selected variable (32%), but the overall repertoire was notably diverse (Figure 2A): sleep signs (15.37%), respiratory rate (12.97%), blood pressure (6.59%), step counting (6.39%), calories expended (4.79%), reaction time (3.99%), hearing detection (3.59%), SpO₂ index (2.4%), vocal frequency (2.2%), temperature (2.2%), muscular vibration (2.0%), urine color (1.8%), visual acuity (1.0%), pupil diameter (1.0%), blood glucose (0.6%), stool color (0.2%), nail size (0.2%), and sclera color (0.2%). Over the years, a clear trend toward selecting novel variables beyond heart rate was observed. Categorizing the physiological topics addressed revealed the following descending frequency (Figure 2B): cardiovascular (40.35%), nervous system (35.55%), respiratory (15.4%), locomotor (6.25%), urinary (1.5%), digestive/nutrition (0.5%), integumentary (0.32%), and reproductive (0.13%) systems. Some projects addressed multiple topics. Notably, the proportion of topics other than cardiovascular increased over time. Finally, project originality was classified into three categories: imitation, improvement, and innovation. A significantly higher prevalence of innovation projects was observed compared to imitation (60.75% IQR: 52.3–67.2 vs. 19.15% IQR: 9.5–35.0, p = 0.003) and improvement (60.75% IQR: 52.3–67.2 vs. 18.1% IQR: 10.6–27.8, p = 0.003) (Figure 2C). These results highlight students’ ability to creatively explore technological resources in smartphone sensors and applications.

3.5. The iHPhys Activity Can Substitute a Summative Assessment

Figure 3 presents the correlation matrix among student variables (1. Gender, 2. Academic term, 3. Study period, 4. Midterm exam score, 5. Final exam score, 6. Formative assessment score, 7. Final grade) and performance on the iHPhys project (8. iHPhys score). Student performance on iHPhys was not influenced by gender, academic term, or study period. However, a trend of negative correlation was noted between study period (day vs. evening) and performance in all course assessments, especially iHPhys (rs = −0.092, p = 0.07), suggesting that evening students—typically working students—may have less time to dedicate to project development. Conversely, significant positive moderate correlations were observed between iHPhys scores and other summative assessments—midterm exam (rs = 0.347, p < 0.001), final exam (rs = 0.373, p < 0.001), and final grade (rs = 0.485, p < 0.001)—suggesting that students who excel in project-based assessments also perform well in traditional assessments. Additionally, greater originality in projects correlated with better iHPhys performance (rs = 0.277, p < 0.001). Although unrelated to iHPhys performance, formative assessment scores were negatively correlated with academic term (rs = −0.194, p < 0.001), indicating that students who took the course later in their academic trajectory performed better in formative activities.

4. Discussion and Considerations

The integration of smartphone-assisted experimentation into inquiry-based learning (IBL) represents a paradigm shift in science education, particularly for learning human biology, by bridging the gap between budget-constrained schools for acquiring physiological data-collecting equipment and tangible, student-driven investigations. The results of this study demonstrate that the “Investigating the Human Physiology” (iHPhys) activity, mediated by smartphone-assisted experimentation, enhances the development of essential scientific competencies among pre-service Sciences and Mathematics teachers. The analysis of the educational products, triangulated with the rigorous validation of the evaluation rubric, enables a critical discussion of the effectiveness of the adopted methodology, its contributions to initial teacher training, and its limitations and implications for future pedagogical practices. Our findings provide compelling evidence supporting constructionist learning theories (Harel & Papert, 1991), demonstrating how mobile devices serve as what Jonassen (1995) terms “cognitive partners” rather than mere tools. Although a tool in essence, our results evidenced how cognitive processes can materialize through research design thinking, which enabled learners to externalize and refine their scientific reasoning through real-time data collection. This challenges traditional lab-centric models, supporting the argument that mobile technologies democratize experimentation by placing laboratories in students’ pockets (Koydemir & Ozcan, 2018). The high levels of methodological innovation observed (65.16% of students achieving “excellent” ratings and 60.75% of projects categorized as innovative) suggest that smartphones facilitate what Hatano and Inagaki (1986) describe as “adaptive expertise”—the ability to apply knowledge flexibly to novel situations. This aligns with the work by Baya’a and Daher (2009) on mobile learning as a catalyst for situated cognition, where learning is deeply embedded in authentic contexts.

Considering the journal’s audience, one of our goals is to disseminate the methodology we have created, anticipating the needs of professors and teachers in assessing students’ learning through active learning methodologies. The development of practical assessment tools is crucial for evaluating student learning in inquiry-based science education. The rubric created in this study provides a structured framework for assessing key scientific skills, including hypothesis formulation, methodological design, data representation, and conclusion development. The validation process, which involved calculating Cohen’s kappa and the intraclass correlation coefficient (ICC), demonstrated moderate to substantial agreement among raters, indicating that the rubric can be applied consistently by different evaluators. Although the mean value of the weighted kappa did not reach the 0.8 threshold, studies have emphasized the stricter standards imposed by the weighted kappa compared to the original kappa coefficient. This discrepancy often arises when the evaluation process involves analyzing nominal data by individuals with varying levels of expertise, or when the material being assessed is subject to high variability. Konstantinidis et al. (2022) highlight the divergences observed when different analytical methods are employed, emphasizing that researchers should carefully select the method that best fits the specific context of their study. In this case, due to the small and heterogeneous sample, internal consistency calculations were deemed inappropriate. Instead, inter-rater agreement measures were prioritized, as achieving consensus among evaluators was crucial for accurately assessing students’ project work. Most IBL assessments focus on outcomes (e.g., exam scores) rather than processes (e.g., hypothesis refinement) (Furtak et al., 2012). Our rubric’s granular criteria (e.g., “Prediction” vs. “Rationale”) address this by operationalizing the E5 learning framework. The high inter-rater reliability (ICC = 0.841) surpasses benchmarks for analytic rubrics (Jonsson & Svingby, 2007), offering a replicable model for assessing open-ended inquiry. The levels of agreement observed in this study suggest that the rubric provides clear guidelines for evaluating student performance, reducing subjectivity in the assessment process. This is particularly important in the context of IBL, where assessment can be complex due to the open-ended nature of investigations and the diversity of student approaches. The high percentages of exact and adjacent agreement further support the rubric’s reliability. The exact agreement values fall within the acceptable range reported in other rubric validation studies, and the adjacent agreement values indicate a strong consensus among raters on the overall quality of student work. Scoring rubrics are dynamic, qualitative assessment instruments, and while they establish coherent evaluative standards for the proposed activity, they can be adjusted to meet the needs of each specific case. This also takes into account the student’s ability to argue and the teacher’s or evaluator’s capacity for judgment (Jonsson & Svingby, 2007). The adaptability of the present rubric reflects its need to meet criterion M2. Innovation is applicable only if the instructor offers an opportunity to practice smartphone-assisted experimentation for studying heart rate. Otherwise, this criterion must be excluded from the rubric or adapted for a chosen context. Its exclusion would reduce the number of criteria and consequently improve inter-rater reliability. The successful validation of the current rubric has significant implications for educators seeking to implement and assess IBL projects involving smartphone-assisted experimentation. The rubric can serve as a valuable tool for (i) providing clear expectations to students—the detailed criteria and performance levels outlined in the rubric can help students understand the expectations for their projects, promoting self-regulation and improved learning outcomes (Güneş et al., 2015); (ii) guiding instruction—the rubric can inform instructional decisions by highlighting areas where students may need additional support or guidance; (iii) ensuring fair and consistent assessment—the rubric provides a standardized framework for evaluating student work, reducing bias, and promoting consistency in grading; and (iv) facilitating feedback—the rubric can be used to provide specific and actionable feedback to students, helping them improve their scientific skills.

The analysis of student projects revealed a general strength in formulating clear and well-supported hypotheses, indicating an ability to connect existing knowledge of human physiology to inquiry thinking. Hypothesizing is a cornerstone of scientific reasoning principles, which science teachers must invest in their teaching practices (Osborne, 2014). However, a significant proportion of students encountered difficulties in proposing predictive hypotheses. This finding is consistent with research highlighting challenges in hypothetico-deductive reasoning among students (Zimmerman, 2007; Lavoie, 1999) and underscores a common gap in science education: differentiating between reporting observations and constructing testable predictions based on theories (Sandoval, 2005). Studies have reported that consolidating the predictive and explanatory nature of science requires explicit pedagogical interventions focused on the epistemology of science (Lavoie, 1999; Duschl, 2008). Factors contributing to these difficulties may include limited experience with experimental design, lack of explicit instruction in formulating predictive hypotheses, and the cognitive complexity of the task. The persistent challenge in predictive reasoning, with a notable percentage of students scoring “Unsatisfactory/Regular,” reveals a limitation: while smartphones facilitate data collection, they do not automatically cultivate epistemic cognition (Sandoval et al., 2016). Additionally, some students showed weaknesses in data analysis and drawing conclusions, struggling with data representation and the interpretation of results to formulate logically supported conclusions. Developing these skills is crucial for scientific literacy, as it requires students to critically evaluate data, identify patterns, and draw valid conclusions. The difficulties in drawing evidence-based conclusions suggest potential cognitive overload (Sweller, 2020), as the complexities of app selection and sensor calibration, despite simplifying data collection, can overwhelm novice learners. Nevertheless, minimally guided IBL, using rubrics, for example, might alleviate the overwhelming feeling (Lazonder & Harmsen, 2016). Moreover, the frequent omission of control groups suggests that students require more explicit instruction in aspects of experimental methodology. According to Brownell et al. (2014), students tend to oversimplify sample size and replicates, focusing on data collection without considering the logical structure of the experimental design. To address these challenges, incorporating meta-reflection cycles into projects, allowing students to review and refine their designs before data collection, might improve experimental design thinking. In two editions of iHPhys, we have implemented a peer-ethics analysis, where classmates evaluate each other’s projects.

In contrast, students demonstrated strong performance in methodological design and innovation, effectively utilizing smartphone sensors and applications to design creative experiments. This capacity to propose novel questions and creatively use technology reflects a desirable investigative profile, particularly given the tendency of undergraduate students to reproduce traditional experiments (Kind & Kind, 2007). The use of mobile devices to measure diverse physiological variables exemplifies the “technological appropriation” described by Papert (1990), where technology extends thought and scientific investigation. Our findings corroborate previous studies, which show that IBL enhances higher-order cognitive skills, such as critical thinking and creativity (Rodríguez et al., 2019; Setiawan et al., 2018). The high performance of students in the criteria “Methodological Innovation” (65.16% “Excellent”) suggests that smartphone-assisted experimentation effectively bridges theoretical knowledge and practical application. The variety of physiological variables investigated by students demonstrates the versatility of smartphone-assisted experimentation as a tool for inquiry-based learning, underscoring the versatility of mobile technologies to democratize access to experimental science (Nadal & Lellis-Santos, 2015; Monteferrante et al., 2018; Lellis-Santos & Abdulkader, 2020). The increasing trend of students choosing novel variables over time suggests that they are becoming more proficient in utilizing the technological resources available to them and in designing more sophisticated experiments. This finding supports the idea that integrating technology into science education can foster innovation and creativity (Rodríguez et al., 2019). However, the persistent preference for heart rate measurements (32% of projects) highlights a need for scaffolding to encourage broader exploration of less conventional variables. The higher frequency of projects investigating heart rate is likely a consequence of the greater availability of applications in app stores and the increasing use of smartwatches for tracking fitness performance. Therefore, we strongly recommend that students be previously exposed to a practical SAE (smartphone-assisted experimentation) class on heart rate and that the innovation criterion be adopted when applying the iHPhys activity. Otherwise, the diversity of variables and research topics will be low. The cardiovascular and nervous systems were the most frequently investigated topics, likely due to the availability of user-friendly apps and sensors for measuring related variables. By observing this data, instructors can identify areas where additional resources or support may be needed to facilitate student inquiry in other physiological domains. The classification of projects based on their originality further highlights students’ capacity for innovation. The significant prevalence of “innovation” projects demonstrates that students are not simply replicating existing experiments but are actively exploring new research questions and developing novel methodologies. This finding underscores the potential of IBL and mobile technology to foster creativity and scientific thinking.

Although the iHPhys activity did not show significant differences by gender or academic term, a slight negative correlation was observed between student performance and enrollment in the evening program. This finding suggests that evening students, who may face time constraints due to work or other commitments, may require additional support or time to succeed in inquiry-based learning activities, such as smartphone-assisted experimentation. Briefly, our results confirm the advantages of active learning methodologies, especially inquiry-based learning (IBL), in reducing performance gaps in STEM. Importantly, student performance in the iHPhys activity was positively and significantly correlated with their performance in other summative assessments, including the midterm exam, final exam, and final grade. Therefore, iHPhys activity effectively assesses students’ overall understanding of the course content and their ability to apply scientific concepts. In some contexts, project-based assessments like the iHPhys activity could potentially serve as a valuable alternative or complement to traditional exams, providing a more authentic and holistic evaluation of student learning.

The collection of physiological data in educational settings, particularly involving human subjects, raises important ethical considerations. In the iHPhys activity, data were collected solely from consenting students, who were informed of the nature, objectives, and voluntary nature of the task. All procedures were conducted in accordance with the ethical standards of Unifesp’s Institutional Review Board, and at the beginning of the semester, students declared consent to participate in the studies developed throughout the course. In addition, during the instruction of the activity, the professor explained the ethical issues in experimentation involving human subjects. No sensitive or invasive data collection were allowed. This information was also discriminated in the instructional sheet used by the students. Nevertheless, the implementation of similar smartphone-assisted experimentation (SAE) activities in other educational contexts should adhere to a strict ethical protocol, especially in primary or secondary schools. We recommend that instructors or lecturers wishing to replicate this methodology in their own classrooms (i) obtain informed consent from participants, ensuring they understand that participation is voluntary and that personal data will be anonymized and not used for punitive or evaluative purposes; (ii) consult local ethics committees when implementing SAE projects—if the school board does not have regimentation for this purpose, the schools can establish a partnership with universities; (iii) avoid requiring students to share data with peers or anyone unless they opt to do so voluntarily—in some countries, there are laws to protect student’s data, such as the Brazilian General Data Protection Law (LGPD 13.709/2018); (iv) state the nature and risks of human physiology experimentation, and guide students to design experiments and opt for physiological measurements that are non-invasive, and unlikely to cause discomfort or embarrassment. Ethical awareness is a critical dimension of scientific literacy, and integrating discussions about research ethics into the instruction of experimental design may foster a deeper understanding of responsible conduct in science.

We acknowledge that the restricted sample and the focus on a specific discipline limit the generalization of the results. Multi-campus and interdisciplinary studies could broaden the understanding of the impact of the proposed methodology. Future investigations should also include qualitative analyses, such as group semi-structured interviews with open-ended questions, to capture the dynamics of scientific reasoning that do not fully manifest in the written products.

The current study demonstrates the potential of smartphone-assisted experimentation as a valuable tool for promoting inquiry-based learning for the initial training of science teachers. Those pre-service teachers might have their didactic repertoire amplified and might have increased probabilities of adopting the proposed methodology in their teaching practices. The validated rubric provides a reliable framework for assessing student work in this context, and the analysis of student projects reveals important insights into their development of scientific competencies. Thus, smartphone-assisted experimentation fosters creativity in IBL settings, boosting the methods of data collection and analysis and promoting a deeper understanding of scientific concepts. Students might experience how to effectively integrate mobile technology into science education and meet their demands for a contemporary tech-savvy science education.