Design, development, and implementation of take-home laboratory kits for measurement of water vapor transmission rate of polymeric films

Introduction

Laboratory-based courses are a vital part of science and engineering curricula because they help students connect theories to applications and build both practical skills and conceptual knowledge. Studies show that laboratory activities enhance students’ conceptual understanding and improve cognitive growth (Trifan, 2011; Kolil et al., 2020). Several studies have reported that students place a higher value on learning through practical hands-on experiences than virtual experimentation, and laboratory experiments help them understand these concepts better (Accettone, 2022). However, many students struggle to gain access to laboratory facilities and physical experiments for a variety of reasons, including shrinking financial resources and the scarcity of expensive laboratory equipment at many universities. Further, during the COVID-19 pandemic, many institutions were forced to switch to virtual classes and halt in-person laboratory activities.

The development of remote laboratories offers a promising solution to support the teaching of science and engineering courses, and it provides equal access to all, including students in online degree programs and others with physical disabilities that prevent them from accessing laboratory buildings. Remote laboratories (RL) can be designed to employ low-cost, take-home kits that replicate laboratory equipment using inexpensive scale models. Several science and engineering disciplines have considered virtual laboratories to replace or intensify physical laboratory experiences, but the development of RL is a relatively new concept. The development and utilization of virtual and remote laboratories have been quite limited in the teaching of food science and engineering courses. This paper describes the design and development of a low-cost experimental take-home kit for use in RL.

In a Kelley (2020) study, three different high school chemistry RL laboratory formats implemented during the COVID-19 pandemic were examined and their effectiveness measured. These formats were at-home experiments, virtual simulations, and instructor-provided data given to the students. The study concluded that kits used during the at-home experiments ensured a more effective and equitable laboratory experience for all students when compared with the other two formats. Miles and Wells (2020) also found that a “Lab-in-a-Box” model for delivering an instrumental analysis laboratory course in a remote learning environment enabled students to perform experiments at home while achieving key learning outcomes related to data collection, instrument operation, and analysis. The portable laboratory kits used in that study contained essential instruments, materials, and detailed instructions. The authors reported that this approach helped in maintaining the students’ interest in the discussion topic and that their experimental skill development met expectations. At the same time, the use of this remote instruction method was logistical, safe, and cost-effective. Additionally, Amsen (2021) described the development of take-home laboratory kits, virtual tools, and online platforms to engage students with hands-on and interactive activities in a distance learning environment. The authors discussed logistical hurdles such as equipment distribution and remote guidance, as well as instructional adaptations like artificial intelligence-driven reaction prediction tools and virtual instrument control, showing that remote lab modalities can effectively support student learning and may influence future teaching.

Traum and Hadi (2019) developed, implemented, and evaluated a miniaturized circular hydraulic jump laboratory kit intended for remote online fluid mechanics instruction. The researchers developed a small, inexpensive hands-on kit that was mailed to students, who then assembled the apparatus, conducted experiments, gathered data, analyzed the results, collaborated with their peers, and wrote their laboratory reports while at their home. This exercise allowed the students to learn key features of traditional fluid mechanics while in a distance learning context. An assessment of the mechanical engineering students under simulated remote learning conditions showed that they achieved the intended learning outcomes, indicating that this remote laboratory approach can effectively substitute for conventional in-person fluid mechanics instruction.

Motivation is a fundamental component of science and engineering education, shaping students’ learning experiences across a wide range of instructional settings (Pintrich, 2003; Rosenzweig et al., 2019). It is essential for promoting active participation, goal orientation, and persistence, particularly when learners are exposed to complex or unfamiliar instructional information and methods (Abumalloh et al., 2021; Diwakar et al., 2023). An important question is whether students demonstrate comparable levels of motivation in remote laboratory environments, such as those involving food packaging kits, relative to traditional, campus laboratories, and how prior knowledge may influence these motivational responses. Gaining insight into how remote laboratories, as an emerging pedagogical approach, affect student motivation is especially important in food science and engineering education. The overall project probes student perceptions of motivation by examining key motivational factors associated with the use of take-home food packaging kits as a remote laboratory tool (Fabiyi et al., 2026, manuscript under review). Food packaging is essential for maintaining food safety and extending shelf life, and food packaging courses require students to integrate knowledge from multiple disciplines to address challenges related to food security, safety, quality, and environmental sustainability. However, take-home experimental kits tailored to food packaging courses are currently lacking, and the proposed modules aim to address this important educational gap.

The objectives of this study were to design, develop, and deploy low-cost, user-friendly water vapor transmission rate (WVTR) laboratory kits for food packaging-related courses and to evaluate their feasibility, student learning outcomes, and potential educational value.

Materials and methods

Materials

In this study, the method used to measure WVTR of polymeric films was based on ASTM E96/E96M-21 (ASTM International, 2021), which is widely applied for packaging films, coatings, and sheets. In this method, a film specimen is sealed over the mouth of an impermeable test cup containing either a desiccant (desiccant method) or water (water method) to create a defined water vapor pressure gradient across the film. The assembly is then placed in a controlled relative humidity (RH) chamber. The change in mass of the cup over time is due to water vapor transfer through the film, and it is measured gravimetrically. The materials that comprised the WVTR kits were: 1) a desiccator designed to maintain a constant RH; 2) impermeable test cups; 3) an analytical balance capable of measuring the change in mass of the water in the test cups; 4) RH and temperature sensors; 5) vacuum grease/sealant; 6) a desiccant; and 7) the test films. The test cups used in this study had an open, exposed mouth of 5 cm in diameter (Figure 1). They were made from acrylic sheets and were fabricated at the Engineering Shop (Washington State University, Pullman, WA). The process began by carving the test cup from 7.5 cm-thick acrylic sheets using a CO₂ laser cutting machine (Speedy 400 Flexx, Trotec Laser GmbH, Marchtrenk, Austria) at 100% power and 0.2% speed. The cup was subsequently refined on a milling machine (VM10i CNC vertical machining center, Hurco Companies, Inc., Indianapolis, IN, USA) to produce an internal pocket, O-ring groove, chamfers, and threaded holes to accommodate the lid. The lids were cut from 6.5 mm thick acrylic sheets using the same CO₂ laser at 100% power and 0.5% speed. The lids were designed to be sealed to the cups using four 3 mm stainless steel screws. Following cutting, both the lid and base cup were laser-engraved on the Trotec Speedy 400 Flexx using raster engraving settings of 100% power and 20% speed for easy matching of cups and lids.

The test film samples were prepared from polyethylene terephthalate and polypropylene-based multilayer materials obtained from Toppan Inc. (Tokyo, Japan). They both had a thickness of 110 microns. All other items were purchased from Amazon.com, Inc. Table 1 provides a cost breakdown of the components, including in-house fabricated parts, maintenance expenses, expected reuses, and commercial water vapor permeation equipment.

TABLE 1. Components and costs of the take-home WVTR test kit.

Laboratory instructions for students and course teachers

The students given the take-home laboratory kits were expected to perform the assigned experiments independently and without assistance from teaching assistants or course instructors. Therefore, the laboratory instructions were written in a simple, step-by-step manner and were supported by schematics, photographs, and videos to guide the users throughout the experimental procedure, data collection, and analysis. The instruction manual (Supplementary Appendix B) also included a checklist, material safety data sheets, student learning outcomes, a brief background on the experiment, and the practical significance and applications of water vapor permeation through polymeric films used for food packaging. Additionally, instructional guidelines for the course instructors (Supplementary Appendix A) were provided to help them with the maintenance and long-term use of the laboratory kits. The kits were developed with the understanding that they would be reusable and would be deployed at multiple institutions across universities in the United States.

Testing of the kits and implementation of the laboratory module

To evaluate the viability and robustness of the WVTR kit, it was beta-tested at five collaborating institutions. These included Washington State, California State Polytechnic Pomona, Cornell, The Ohio State, and Rutgers Universities. At these institutions, course instructors, undergraduate and graduate students evaluated them for evidence of defects and/or deviations from the intended design. Based on feedback received from the students and course instructors, both the test kits and the instructional manual were refined. These were subsequently redistributed to the collaborating and cooperating institutions (eight in total) for implementation into undergraduate- and graduate-level food packaging courses throughout seven semesters (Table 2). The students were expected to complete the assigned experiments using these kits at their homes, in their dormitories, or in the laboratory, without help from the course instructors or teaching assistants. If the kits were insufficient for distribution to each student in a given class, they were allowed to work together in small groups. However, each student was responsible for their own work, and those elected to participate in the research study were asked to complete the pre-and post-laboratory quizzes and surveys. Students recorded and analyzed their experimental measurements using the Excel worksheet supplied with the laboratory instructions. This worksheet served as a data-recording and calculation tool rather than as an assessed research outcome. The formal 8–10 slide PowerPoint laboratory-report requirement (as seen in the laboratory instructions) was waived during the pilot implementation, and data from either format were not included in the educational-outcome analyses reported here. Students were offered extra credit for completing the laboratory experiment as an incentive for participation in the activity, while participation in the associated pre-and post-laboratory quizzes and surveys used for the research study was voluntary.

TABLE 2. Participating Institutions and student/class levels throughout seven semesters.

Assessment of the impact of take-home kits on student learning

The pre-and post-laboratory quizzes provided to students along with the kits were included in the instruction manual for the class exercise. The laboratory quizzes were available via web links (Qualtrics) and QR codes and were accessible by smartphones or computers. Supplementary Appendix C presents the quiz questions via the Qualtrics interface for pre-and post-laboratory quizzes. The student demographics and motivation survey questions are summarized in Supplementary Appendix D. The instruction manual also contained links to the survey answers provided by the students. These answers included input from the students about their learning experience and their demographics (e.g., graduate versus undergraduate, food science vs. agricultural engineering major, upper division vs. lower division course being taken). In addition to the demographic survey, the pre-laboratory exercise also contained a consent form. Immediately after completion of the WVTR experiment, students participating in the research study were asked to complete the post-laboratory quiz and a motivation/learning experience survey. Motivational constructs were assessed across four dimensions—interest/enjoyment, perceived competence, effort/importance, and perceived value/usefulness—using 5-point Likert-type items described in Supplementary Appendix D. The broader analysis and application of these motivational constructs to remote food science laboratory experiences are reported separately by Fabiyi et al. (2026), manuscript under review). This information was useful in assessing the effectiveness of the laboratory kits on student learning experience.

The pre-and post-laboratory quizzes (Supplementary Appendix C) contained three identical multiple-choice questions that assessed the students’ ability to conceptualize the relationships between the gravimetric cup measurement of WVTR and the following: (i) the effect of film thickness, (ii) the effect of temperature, and (iii) the effect of film surface area on the data collected. The students were not required to submit their names; instead, an access code was used to associate them with their respective universities and with the semester in which they used the kits. The access coding was designed as a combined letter and number system: a 3-letter abbreviation (such as WSU for Washington State University), followed by a 2-digit number for the laboratory kit to be used in the exercise (such as 01 for this WVTR test method), followed by a 2-letter abbreviation (representing the initials of the first and last names of the instructor who helped assess the chosen kit from a cooperating institution), and lastly, a 3-digit randomly generated code.

In summary, the WVTR laboratory test kits were deployed at eight cooperating institutions, involving a total of ten instructors, over a period of seven semesters from Fall 2022 to Fall 2025. Before the beginning of this exercise, approval from the Institutional Review Board was obtained from Washington State University (WSU IRB # 19616-001) and other participating universities.

Statistical analysis of assessed data

All responses from the quizzes associated with the kits were reviewed for completeness and conformity to the guidelines. Each multiple-choice question was scored as correct/incorrect using the following answer key: Q1 (thickness) = “WVTR decreases with increasing film thickness”; Q2 (temperature) = “WVTR increases with increasing temperature”; Q3 (surface area) = “WVTR does not change with increasing surface area”. For Q3, WVTR rate (g/(m²·day) is normalized by the exposed film area; therefore, under otherwise identical test conditions, the calculated WVTR is not expected to change solely because of a change in exposed surface area. For each respondent, a total knowledge score (0–3) was computed as the sum of correct responses across Q1–Q3; for example, if an individual student answered all three questions correctly, the respondent’s score would be recorded as 3/3. If a student misses all three questions, the respondent would receive a score of 0/3. These individual scores were aggregated to summarize overall pre-and post-laboratory performance across the three conceptual questions.

In performing the primary comparisons of the learning outcomes, the pre- and post-laboratory responses were treated as independent samples because the dataset did not contain a unique student identifier that could reliably link all individual pre/post responses. For each question, the pre-vs. post-laboratory percent correctness was compared using a two-sided chi-square test method (equivalently, a two-proportion z test), and the effect size (i.e., the difference between the two variable groups) was summarized as an odds ratio (OR), with a 95% confidence interval (CI). Because the total knowledge score was a discrete and bounded variable ranging from 0 to 3, the pre-and post-laboratory score distributions were compared primarily using the Mann–Whitney U test. A Welch two-sample t-test was additionally performed as a sensitivity analysis to quantify differences in mean scores.

As a secondary sensitivity analysis, a “matchable” subset was created for the access codes that appeared exactly once in both pre-and post-laboratory datasets (i.e., one-to-one match by access code). Within this subset, changes in item-level correctness were assessed using McNemar’s exact test. All tests were two-sided in design, with α = 0.05.

To examine potential heterogeneity associated with the multi-institutional implementation, quiz outcomes were also summarized descriptively by participating institutions. For each institution, the proportions of correct responses to Q1–Q3 and the mean total knowledge score were calculated separately for the pre-and post-laboratory samples. Because institution-specific sample sizes were relatively small and unequal, and students were further clustered within instructors, courses, and semesters, institution-specific hypothesis tests were not performed. These stratified results were used to evaluate the consistency and variability of response patterns across sites. Responses with access codes that could not be reliably assigned to an institution were retained in the overall pooled analyses but excluded from the institution-stratified summaries.

The item-level comparisons of Q1–Q3 were exploratory; therefore, the reported p-values were unadjusted for multiple comparisons and were interpreted together with effect estimates and 95% confidence intervals.

ChatGPT (GPT-5.2 Pro; institutional license through the California State University system) was used to support data cleaning, sorting (e.g., grouping by institution/instructor/year), and statistical analysis of the survey dataset. Survey instruments were administered and raw response data were collected using Qualtrics (institutional license through California State University). All analyses and outputs were independently reviewed and verified by the authors.

Results and discussion

Students from eight different institutions, enrolled in both undergraduate and graduate-level courses, evaluated the WVTR take-home kits. All students, whether they worked independently or in teams, completed the experiment without assistance from their instructors or teaching assistants. Students recorded and analyzed their WVTR experimental measurements using the Excel worksheet supplied with the laboratory instructions. The worksheet served as a data-recording and calculation tool and was not used as an educational outcome measure in this study. Additionally, the students completed a set of three questions as part of the pre- and post-laboratory quizzes.

Assessment of learning outcomes

The responses from the pre- and post-laboratory quizzes and the student motivation surveys were analyzed, and the results were used to gain an understanding of the respondents’ learning and interest in take-home laboratory activities. Across all implementations, 162 and 118 students submitted valid, complete responses to the pre- and post-laboratory quizzes, respectively. As shown in Table 3, the proportion of correct responses to Q1, which assessed the effect of film thickness on WVTR, was higher in the post-laboratory sample group than in the pre-laboratory sample. Correct responses increased from 63.0% in the pre-laboratory group to 76.3% in the post-laboratory group, corresponding to a difference of +13.3 percentage points. The odds of a correct response were higher in the post-laboratory group (OR = 1.89, 95% CI: 1.11–3.21; p = 0.018). For Q2, which assessed the effect of temperature on WVTR, the proportion of correct responses was similar between the pre- and post-laboratory groups (74.1% vs. 75.4%, respectively; p = 0.798), indicating little difference in group-level performance. For Q3, which assessed the effect of film surface area on WVTR, the proportion of correct responses increased from 17.9% in the pre-laboratory group to 25.4% in the post-laboratory group; however, this difference was not statistically significant in the pooled unpaired comparison (p = 0.127). The relatively low proportion of correct responses to Q3 in both groups suggests that the concept of surface-area normalization remained challenging for many students.

TABLE 3. Pre-and post-laboratory quiz performance by item (Q1–Q3) based on pooled responses.

In a subset of the student responses (N = 52) with access codes that could be uniquely matched (i.e., the same access code was used in both pre-and post-laboratory quizzes), the percentage answering Q3 correctly (surface area effect on WVTR) increased from 13.5% (7/52) to 30.8% (16/52), with McNemar’s exact p = 0.022, meaning the same group of students demonstrated better understanding of this concept after the laboratory exercises (Table 4). Among the 52 uniquely matched respondents, the mean total knowledge score increased from 1.62 pre-laboratory to 1.88 post-laboratory; this within-individual change did not reach conventional statistical significance (Wilcoxon signed-rank p = 0.065). In the full unpaired samples, the mean total score increased from 1.55 ± 0.95 before the laboratory activity to 1.77 ± 0.88 afterward (Mann–Whitney p = 0.045; Welch sensitivity analysis p = 0.045), corresponding to a small standardized mean difference (Hedges’ g = 0.24).

TABLE 4. Pre-and post-laboratory quiz performance by item (Q1–Q3) based on matched responses.

Institution-stratified learning outcomes

Because the WVTR laboratory kit was implemented across eight institutions, the pre- and post-laboratory quiz results were additionally examined by institution to assess the consistency of the observed response patterns across instructional settings (Table 5). The proportion of correct responses to Q1 was numerically higher in the post-laboratory sample at each of the eight institutions, although the magnitude of the difference varied among sites and some institution-specific post-laboratory sample sizes were small. Performance on Q2 was generally high and showed comparatively little pre/post difference across institutions. Greater between-institution variability was observed for Q3, with improvements at several institutions but little change or modest decreases at others. Mean total knowledge scores likewise varied across institutions.

TABLE 5. Institution-stratified pre-and post-laboratory quiz performance.

Institution-stratified results are descriptive because pre-and post-laboratory responses could not be completely matched at the individual level and institution-specific sample sizes were unequal. Twenty-four responses with nonstandard or ambiguous access codes could not be reliably assigned to an institution; these responses were retained in the pooled analysis but excluded from Table 5. No institution-specific inferential tests were conducted because of small sample sizes and potential clustering of students within institutions, instructors, courses, and semesters.

These institution-stratified findings should be interpreted descriptively rather than as independent estimates of institutional effectiveness. The pre-and post-laboratory samples were not fully matched within institutions, sample sizes were unequal across sites, and responses from students within the same institution, course, or instructor may be correlated. Consequently, potential intraclass correlation may influence the precision of the pooled statistical estimates. Nevertheless, the stratified results demonstrate that implementation and student participation occurred across diverse institutional settings and provide useful context for interpreting the overall pooled and matched analyses.

Several limitations should be considered when interpreting these educational outcomes. First, complete individual-level linkage between the pre-and post-laboratory assessments was unavailable; therefore, the pooled comparison of 162 pre-laboratory and 118 post-laboratory responses represents a group-level association rather than a within-individual or causal estimate of learning. The difference in response counts may also reflect differential participation or nonresponse. Second, although 52 respondents could be uniquely matched, the smaller matched sample limited statistical power. Third, no concurrent control group receiving conventional laboratory instruction or another instructional intervention was included. Fourth, implementation varied across institutions, instructors, courses, academic levels, and semesters, creating potential clustering of responses within instructional settings. Fifth, extra-credit incentives and repeated administration of the same three quiz items may have contributed to motivation, selection, or testing effects. Finally, item-level statistical tests were exploratory and were not adjusted for multiple comparisons. These considerations support interpreting the findings primarily as evidence of feasibility and potential educational value, rather than as definitive evidence of comparative effectiveness.

In addition to the quantitative assessment of conceptual learning, two open-ended free-response questions in the post-laboratory survey (Supplementary Appendix C) were analyzed to identify challenges encountered by students and perceived limitations or opportunities for improving the laboratory kit and instructional materials. Analyses of the responses to these questions provided a basis for understanding the “challenges” that the students encountered (Q4), and the “limitations” of the laboratory kit and the experiment (Q5). These responses provided an opportunity for “improvements” to the overall exercise. A total of 115 out of 118 (97.5%) students responded to the “challenges” question, and 105 out of 118 (89.0%) of the students responded to the “limitations/improvements” question. For Q4 (challenges), several students described the slow rate of water vapor through the test film as being a problem. This was so because the test kit required a relatively long storage time before differences in the daily weights of the test cup could be recognized. The students also reported problems with leaks in the seal of the test cup, and fluctuating environmental factors such as temperature and humidity in their homes or dorms. For Q5 (limitations and suggested improvements), 17.1% (18/105) of the students reported “no issues/no suggestions”; 43.8% (46/105) reported mixed responses (generally positive but still suggesting improvements); and 39.0% (41/105) responded with primarily limitations/issues (little or no positive response). From the data and responses received, practical and respondent-driven improvements that were needed were determined to be:

• More explicit guidance for controlling/recording of the RH and temperature in the home settings (and how deviations may affect the WVTR).
• The inclusion of an “acceptable measurement schedule” option (e.g., fewer weighing time-points and longer experiment duration) in the guideline.
• Improving the quality of the cup seal by providing a better gasket, a troubleshooting checklist, and the use of a larger quantity of sealing grease.
• A calibration check step and extra hygrometer batteries.
• Revision of the instructional manual, to include information about the exposed surface area of the test film.

Based on participant feedback, the laboratory kits and instructions require further refinement. For instance, untracked temperature and relative humidity (RH) fluctuations across decentralized student locations represent a significant experimental limitation and a primary source of data variability. Gathering these environmental metrics in future iterations will help clarify how environmental conditions affect the water vapor transmission rate, allowing for more comprehensive laboratory instructions.

Hurdles encountered during the development and implementation of the take-home kits

This required significant planning in sourcing affordable components, assembling the kits, and shipping them safely to instructors/students at different locations. In some cases, delivery restrictions and delays complicated the timely implementation of the exercise. Another challenge was ensuring the safe use of the desiccant used in the desiccator, and the necessary Institutional Review Board (IRB) compliance approvals. Without direct supervision at the time when the students would be doing the WVTR testing, the exercises needed to be designed to minimize injury from hazardous chemicals and sharp objects. This required careful selection of materials, clear safety documentation, and detailed instructions to ensure that the experiments could be conducted safely in non-laboratory environments. Kelley (2020) also reported a major challenge with student safety and supervision during at-home laboratory experiments.

After obtaining feedback from the students who were sent the first set of kits, the data collected were analyzed, then used to make improvements to kits that were fabricated subsequently. One major modification that was made was the production of a pre-recorded video with additional instructions on the execution of the WVTR testing. Similar logistical challenges have been reported by Amsen (2021) when they developed and implemented remote chemistry laboratory kits. In similar publications, Traum and Hadi (2019) also mentioned that some students initially struggled with the correct assembly of take-home kits. This highlighted the need for clear instructions, diagrams, and possibly tutorial videos to guide the setup of the kits reported in this paper. One limitation of take-home kits for the performance of experiments is the absence of immediate access to an instructor or teaching assistant. In situations where a student may need help, responses could be delayed, or the use of virtual communication tools may have to be employed. Even if such tools are available, a delay in the problem resolution could add to the frustration of the students. Overcoming such logistical challenges requires creativity, adaptability, and substantial institutional and instructional support. Despite these challenges, several publications reported that take-home kits have successfully supported remote learning objectives (Traum and Hadi, 2019; Amsen, 2021; Kelley, 2020; Miles and Wells, 2020).

Conclusion

Remote laboratories offer a promising approach in supporting distance education. Asynchronous distance learning offers flexibility by allowing students to choose when, where, and how they learn. This paper reports on the successful development and implementation of an experimental take-home kit for the measurement of water vapor transmission rate through a polymeric film. This was done using low-cost devices, sensors, and in-house–fabricated components to assemble the device. These kits provided students majoring in food science and engineering with an essential hands-on learning experience. Although remote and virtual laboratory approaches have been increasingly reported across science and engineering disciplines, comparatively limited literature has focused on take-home experimental laboratories designed specifically for food science and food packaging education. The present study contributes to this emerging area by evaluating the feasibility of delivering a hands-on polymer packaging experiment to remote learners. Results from both indirect and direct assessments of student learning indicated positive student performance. Indirect assessment was conducted through student surveys, mainly as self-reported post-experiment experience connecting to four motivational factors such as “interest”, “competence”, “effort”, and “perceived value” (Fabiyi et al., 2026, manuscript under review), while direct assessment involved pre-and post-laboratory quizzes as reported in this study. The findings presented in this paper support the feasibility and potential educational value of using take-home WVTR laboratory kits in food science and engineering education. However, because the study did not include a concurrent control group and complete individual-level pre/post matching was unavailable, the results should not be interpreted as demonstrating superiority or equivalence to conventional in-person laboratory instruction. Participation and motivation effects should also be considered. The use of extra credit to encourage participation may have preferentially engaged students who were more motivated to complete the activity and associated assessments. Furthermore, the same three conceptual items were administered in both the pre-and post-laboratory quizzes; therefore, some improvement may reflect familiarity with the assessment instrument rather than the laboratory experience alone.

Overall, the findings indicate generally positive student experiences and support the feasibility and potential educational value of take-home laboratory kits for food science and engineering instruction. The kits also enabled students to conduct hands-on experimental activities without the continuous physical presence of an instructor. Further controlled studies with complete individual-level pre/post matching and objective measures of experimental performance are needed to establish their effectiveness relative to conventional laboratory instruction.

Practical application

This study demonstrated the feasibility of implementing low-cost take-home laboratory kits to provide hands-on experimental experiences in distance-learning and online instructional settings. The approach may serve as a model for developing laboratory activities for additional topics in food packaging courses as well as other food science and engineering programs, such as those focused on the physical properties of foods.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. ChatGPT (GPT-5.2 Pro; institutional license through the California State University system) was used to support data cleaning, sorting (e.g., grouping by institution/instructor/year), and statistical analysis of the survey dataset. Survey instruments were administered and raw response data were collected using Qualtrics (institutional license through California State University system). The authors reviewed and verified all analyses and interpretations and take full responsibility for the content of the manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

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

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/frfst.2026.1964059/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Seydi Yıkmış, Namik Kemal University, Türkiye

Reviewed by: Andres Ciolino, National University of the South, Argentina

Arlan Silva Freitas, Science and Technology of Maranhão, Brazil