https://orcid.org/0000-0002-8809-2776
Figures
Abstract
This study examined the effectiveness of lectures and inquiry-based instruction in supporting learning for language minority (LM) students in science gateway courses at a large public research university. Utilizing institutional data from 6,911 students across nine years, we fitted cross–lagged panel designs to model associations between lecture courses and inquiry-based laboratory courses for both LM and non-LM students in two-course sequences of introductory college Physics and Chemistry. We found that initial performance in lectures and laboratory sessions can be a predictor of subsequent course performance across disciplines and independent of LM status. Notably, while LM students performed worse in the initial lecture course, LM status resulted in neither worse performance in inquiry-based laboratory courses nor in worse performance in subsequent courses in the science gateway course sequence. Thus, this study suggests that interventions intended to support LM students in college science should target the initial courses in the corresponding science gateway course sequences.
Citation: Fischer C, Nguyen H, Estrella G, Collins P (2022) Examining lecture and inquiry-based laboratory performance for language minority students in science gateway courses. PLoS ONE 17(4): e0267188. https://doi.org/10.1371/journal.pone.0267188
Editor: Mingming Zhou, University of Macau, MACAO
Received: April 22, 2021; Accepted: April 4, 2022; Published: April 28, 2022
Copyright: © 2022 Fischer et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files. The Supporting Information file contains the minimal data set. Additional data cannot be shared publicly because of FERPA restrictions. Raw data may be requested for eligible researchers through the University of California, Irvine Education Research Initiative (ERI): http://dtei.uci.edu/tlrc-requesting-data-from-the-tlrc/.
Funding: This study was supported by the Open Access Publishing Fund of University of Tübingen through an awarded given to CF.
Competing interests: The authors have declared that no competing interests exist.
Introduction
As careers in science, technology, engineering, and mathematics (STEM) take greater precedence in the 21st-century job market, educational researchers and practitioners have begun to argue for greater integration of hands-on, authentic learning activities in science education. This is particularly true for the natural sciences, where inquiry-based learning through laboratory instruction has become crucial to successful STEM education. Inquiry-based instruction may be thought of as the educational practice whereby students construct knowledge by following the methods and practices used by professional scientists [1, 2]. When compared to traditional forms of lecture, inquiry-based learning environments have been found to significantly improve student learning, performance, and motivation in STEM [3–5]. However, despite these potential benefits, relatively few studies in higher education research have attempted to evaluate the effectiveness of inquiry-based laboratory instruction for language minority (LM) students [6–8]. As such, it remains unclear how inquiry-based instruction may support the academic achievement of LM students in postsecondary science courses [9].
Few would argue with the fact that science learning occurs not only within the lecture hall but also in laboratories in higher education settings. Though most current research focuses on modifying lecture-based instruction to be more aligned with more active learning and inquiry-based models of learning [5, 10], laboratory courses are particularly well suited for supporting inquiry-based learning environments [11–13]. Interestingly, previous studies indicated that student participation in laboratory experiments had positive effects on student learning and academic achievement [14, 15]. Indeed, the experiments conducted by students in laboratory courses are intended to foster the self-directed learning processes that characterize inquiry-based instruction [2]. Therefore, we intend to better understand the importance of laboratories as an inquiry-based environment to student learning in postsecondary science education. The primary objective of this study is to investigate associations of inquiry-based instruction and how they compare to traditional lecture-based instruction on student achievement in subsequent college science courses with a focus on LM students.
Theoretical framework
There are compelling reasons for choosing inquiry over more traditional teaching methods in higher education settings. Inquiry instruction is thought to produce more meaningful learning outcomes than direct instruction due to the deep levels of processing that occurs when students attempt to make sense of, explain, and communicate their scientific findings with others [16]. This advantage is often mirrored with findings that information is better remembered when it is actively self-generated (e.g., explaining the process of photosynthesis to oneself or a fellow peer) rather than passively consumed (e.g., solely reading about photosynthesis in a textbook) [17]. To this degree, recent textbooks even incorporate active learning elements though direct links to computer-based simulations alongside experimentation and reflection tasks (for instance, “Wave Motion as Inquiry” for introductory college level Physics) [18].
Indeed, active learning has been found to increase undergraduate student achievement in STEM classes by almost half a standard deviation when compared to traditional lectures [5]. Similarly, a recent study found that students in courses taught with textbooks focused on inquiry learning showed better performance compared to courses using more traditional textbooks [19]. Inquiry-based instruction is also believed to be exceptionally well-suited for building students’ problem solving and critical thinking skills, as it requires students to construct knowledge for themselves by engaging in sophisticated forms of scientific reasoning in the context of authentic problem solving, as opposed to merely relying on rote memorization [20]. In addition to these academic and cognitive benefits, research on academic motivation suggests that the hands-on, collaborative nature of inquiry instruction increases students’ positive attitudes and intrinsic motivation towards learning science [21–24].
Because of its cognitive and motivational benefits, inquiry-based instruction may be particularly beneficial for students in need of greater support. A long-standing body of research in postsecondary settings provides strong evidence of persistent disparities in retention and graduation rates for students from underrepresented minority groups, including language-minority students from non-English speaking households and those who are continuing-generation, nonminority students [25, 26]. Numerous studies show an achievement gap for LM students in STEM courses and that they are more likely to leave STEM majors [6, 27–29]. On the one hand, the greater attrition rate for LM students may reflect the perceived chilly climate of STEM, or the student perception that faculty, and other instructors are unapproachable, intimidating, cold, and indifferent [30, 31]. On the other hand, given the growing proficiency in English for LM students, the achievement-gap may reflect the complexity of scientific language [32], such as its informational density, heavy use of infrequent words, and technical vocabulary [33, 34]. Inquiry-based instruction, with its emphasis on experimentation and student-constructed knowledge, may mitigate the language-demands of science and decrease the achievement gap for LM students. Further, inquiry-based instruction may provide a more inclusive learning environment for LM students, as it fosters an environment of collaboration and group work, encourages discussion among students and increases the perceived availability and approachability of instructors [35]. A recent meta-analysis reported that active learning in STEM courses reduced the achievement gap for minority and low-income by 33% in their examination grades and reduced the gap in pass rates by 45% [36]. Further, moderator analyses revealed increasing the intensity or class-time devoted to active-learning activities yielded narrower achievement gaps for minoritized groups of students. Similarly, studies that examined mediation effects found that increases in student self-efficacy mediated the increase in student performance due to active learning pedagogy in particular for URM students [37]. Despite these promising findings, it is noteworthy that laboratory-based classes were just one of several types of active learning activities examined in the aforementioned meta-analysis, including activities such as problem-based learning, collaborative groups, and the use of clickers. Because the goal was not to examine each type of activity’s pedagogical benefits, it is unclear the degree to which inquiry-based laboratories may affect the achievement gap for URM students. That said, a recent study that examined the pairing of lecture courses with accompanying laboratory sessions did not find significant effects on student performance by different subpopulations of students, while overall indicating that laboratory participation increased students’ lecture performance [14].
While many studies find that active learning may enhance equitable STEM learning opportunities for minoritized groups of students [24, 36–38], it is less clear whether it would have differential benefits for LM students. Although much of the higher education literature has addressed the underrepresentation of women, low-SES and black, indigenous, and other people of color (BIPOC) in STEM, relatively little higher education research has focused on serving LM students [29, 39]. Further, the LM student population is heterogeneous, varying widely in the proficiency in English, academic preparedness, and SES. Although LM students in general are more likely to enroll in community colleges than their non-LM peers and less likely to enroll in non-selective four-year institutions. However, at selective four-year institutions the proportion of LM students is comparable to the proportion of non-LM students [7]. This bifurcation may reflect important differences in the LM populations, as LM students attending selective four-year institutions may be more English-proficient than those who attend community colleges and less selective institutions. Despite their greater English-proficiency, little is known about how best to serve LM students at selective four-year institutions. To this end, we ask the following research questions (RQs):
- RQ1: What differences in academic preparedness, language ability, and demographic background exist between LM and non-LM students?
- RQ2: How are initial student performance in inquiry-based labs and lectures associated with subsequent course performance in inquiry-based labs and lectures?
- RQ3: How are associations of initial student performance in inquiry-based labs and lectures with subsequent performance in inquiry-based labs and lectures different between LM and non-LM students?
Methodology
Data sources and study setting
The current study is situated at a large public research university in California. The university has been federally designated as a Hispanic-Serving Institution (HSI) and an Asian American and Native American Pacific Islander-Serving Institution (AANAPISI). Data for this study has been provided from multiple institutions on campus including the Registrar’s Office, the Office of Institutional Research, the Office of Information Technology, and Admission, among others. This study utilizes seven cohorts of students (2009–2017) who enrolled in two large introductory science gateway course series (i.e., General Chemistry and Classical Physics). This study was approved by the UC Irvine Institutional Review Board (HS#2018–4801). Informed consent was not required.
Each course series required students to simultaneously enroll in a lecture course with an associated inquiry-based laboratory session in the same term for two subsequent terms. The laboratory experiments were closely tied to the corresponding lecture content knowledge in each term. Notably, these two-course series represent the only such science course series at this university. For Chemistry, this two-term course series (with the unique lecture-laboratory pairing in the same term) was only offered in 2009–2011, whereas it was offered in 2009–2017 for Physics.
Topics of the Chemistry course series included properties of gases, liquids, solids, and solutions; changes of states; and energetics of chemical reactivity (course 1); as well as equilibria, aqueous acid-base equilibria, solubility equilibria, electrochemistry, and nuclear reactions (course 2). Topics of the Physics course series included force, energy, momentum, rotation, and gravity (course 1) and electricity and magnetism (course 2). Our study only investigates students who enrolled in all four courses of the corresponding course series in the same academic year. We excluded international students as their language experiences may differ from U.S. born students, which may confound results of the analysis. Also, we excluded transfer students as these courses are introductory gateway courses that are typically taken early in students’ college career. Transfer students who enroll in these courses may have already enrolled in similar courses, which may introduce underlying biases. Robustness checks that included both international and transfer students indicated that the results remained similar to the more conservative analytical sample of the study (see Tables A1, A2 and Figs A1-A4 in S1 File).
The analytical sample of this study includes 6,911 students (2,865 Chemistry students; 4,046 Physics students). About 61% of students enrolled in the Chemistry course series are female (39% male), whereas 26% of students in the Physics course series are female (74% male). Most students have an Asian, Asian American, or Pacific Islander racial/ethnic background (60% Chemistry, 51% Physics), followed by White (15% Chemistry, 20% Physics), and Latino and Hispanic (13% Chemistry, 22% Physics). Notably, 19% of students are language minority students in the Chemistry course series, whereas 25% of students are language minority students in the Physics course series. Table 1 provides full demographic information of the study sample.
Measures
The dependent variable is a continuous variable indicating student course performance on a 4.0 scale (A+ = 4.0, A = 4.0, A- = 3.7, B+ = 3.3, …, D = 1.0, D- = 0.7, F = 0). The letter to numerical grade conversion mirrors the university grading policy.
The core independent variable of interest represents whether students come from a language minority background. The categorization is based on student self-report data of growing up in monolingual-English, bilingual-English, or non-English homes. As academic and background differences between monolingual- and bilingual English students proved negligible, these categories were collapsed to a non-LM comparison group. To adjust for potential confounding with underlying differences in students’ demographic and academic backgrounds [5, 7, 40], two blocks of covariates were included in analyses, academic preparedness and family background. Academic preparedness controls included the number of pre-college enrollment units, as well as students’ mathematics, reading, and writing scores of college admission tests (i.e., ACT/SAT examinations). Family background characteristics included family household size, first-generation college student status (i.e., neither parent received a bachelor’s degree), low-income status (as classified based on household income and size at the 185% U.S. poverty line cutoff), and fathers’ education level. Notably, fathers’ educational level was used as model fit indices were better compared with mothers’ educational level (see Table A3 in S1 File). That said, robustness checks indicated that the findings are similar in both magnitude and direction (see Fig A5 in S1 File).
Missing data on the LM-variable was list-wise deleted as it represents the grouping variable. For all other variables, the analyses applied full information maximum likelihood (FIML) approaches [41].
Analytical methods
Research question 1 uses descriptive analysis on students’ academic preparedness, language ability, and socio-demographic background variables and estimated standardized mean differences between LM and non-LM students, as well as the corresponding effect sizes (Cohen’s d).
Research question 2 utilizes structural equation modeling to examine whether inquiry-based lab instruction was associated with greater student performance compared to traditional lecture-based instruction. A cross-lagged panel design modeled the reciprocal effects of both inquiry and lecture-based instruction on subsequent course performance [42]. Latent constructs measuring students’ academic preparedness, language ability, and family background were included as covariates to control for the potentially confounding influences [43]. Model parameters used maximum likelihood estimation with Satorra-Bentler scaled chi-squared values and robust standard errors to adjust for biases associated with non-normal data. All models exhibited acceptable model fit; CFI > .90, TLI > .90, RMSEA < .08, SRMR < .08 [44].
Research question 3 fits multi-group structural equation models to examine whether associations of lecture and inquiry-based lab instruction with subsequent course achievement differed by LM status. This analysis used χ2 difference test to provide evidence of moderation [45], comparing the model fit of constrained single-sample models (path estimates fixed to equality based on LM status) to nested multi-group models (path estimates freed to vary between groups).
Results
Academic preparedness and demographics by language background (RQ1)
To compare students’ academic preparedness and family background characteristics, we calculated the effect sizes of the standardized mean differences by LM status (Table 2). Regarding students’ academic preparedness, LM students entered the university with substantially lower reading scores (Chemistry: d = .31, p < .001; Physics: d = .32, p < .001) on the college admission exam for both course series when compared to non-LM students. Similarly, LM students’ writing scores on the college admission exam were significantly lower for both course series (Chemistry: d = .19, p < .001; Physics: d = .18, p < .001), when compared to non-LM students, although the effect sizes are considerably smaller. In Physics, LM students had significantly lower mathematics scores on the college admission exam (d = .15, p < .01) compared to non-LM students, whereas in Chemistry students did not significantly differ on their mathematics scores by LM status (d = 0.04, p = .94). Interestingly, LM students entered the college with more pre-college enrollment units in Physics (d = .13, p < .001), whereas this difference was negligible for Chemistry (d = .03, p = .06).
Regarding student family characteristics, LM students came from less-formally educated households (Chemistry: d = .45, p < .001; Physics: d = .74, p < .001). Household sizes did not substantially vary by LM-status. Also, more LM students were classified as low-income students compared to non-LM students (Chemistry: 44% LM vs. 23% non-LM; Physics: 54% LM vs. 23% non-LM). Similarly, more LM students were first-generation college students in chemistry (47% LM vs. 32% non-LM), although the percentage is similar for Physics (49% LM vs. 47% non-LM). Overall, these findings suggest that LM students may enter the university relatively less prepared to perform in introductory science gateway courses than non-LM students.
Subsequent student performance in inquiry-based labs and lectures (RQ2)
Cross-lagged panel models examined whether inquiry-based labs were stronger predictors of student achievement in subsequent courses than lecture-based instruction. For Chemistry (Fig 1), initial laboratory performance was a statistically significant predictor of subsequent laboratory achievement, β = .06, p <. 05, but not of subsequent lecture performance, β = .01, p > .05. Initial lecture performance did neither significantly predict subsequent lecture performance, β = -.03, p > .05., nor subsequent laboratory performance, β = .03, p > .05. Interestingly, students’ academic preparedness was only significantly predictive of students’ initial lecture performance, β =. 68, p < .001, but not for students’ initial laboratory performance, β = -.02, p > .05.
CFI = .97, TLI = .96, RMSEA = .04 [.03, .04], SRMR = .03, N = 2,865; p < .05., **p < .01., ***p < .001.
For Physics (Fig 2), students’ academic preparedness was similarly only significantly predictive of students’ initial lecture performance, β = 1.05, p < .001, but not for students’ initial laboratory performance, β = .01, p > .05. In contrast to Chemistry, students’ initial lecture performance was a statistically significant predictor of subsequent lecture performance, β = .11, p <. 05, but it did not significantly predict subsequent laboratory performance, β = .01, p > .05. Similarly, initial inquiry-based lab performance did neither significantly predict subsequent lecture performance, β = .00, p > .05., nor subsequent laboratory performance, β = -.03, p > .05.
CFI = .99, TLI = .99, RMSEA = .02 [.02, .03], SRMR = .02; N = 4,046; p < .05., **p < .01., ***p < .001.
Achievement in inquiry-based labs and lectures across language groups (RQ3)
Multi-group structural equation models examined whether the impact of lecture and inquiry-based lab instruction on subsequent course achievement differed between LM and non-LM students. For chemistry (Fig 3), only LM-students had a significant association between initial and subsequent performance, namely the initial laboratory performance on subsequent laboratory performance (β = .15, p < .001). However, a chi-square test of equality indicated that this association did not differ between language groups (χ2 = 2.63, p = .10). Interestingly, academic preparedness had a significantly higher influence on initial lecture performance for LM students (β = 1.04, p < .001) compared to non-LM students (β = .64, p < .001), χ2 = 6.33, p < 0.05. All other paths were not significantly different across LM groups.
left: LM students, right: non-LM students; latent covariates for academic preparedness and SES are included but not shown; dashed lines describe non-significant path estimate; *p < .05., **p < .01., ***p < .001.
For physics (Fig 4), initial lecture-performance was a statistically significant predictor of subsequent lecture achievement for both LM students (β = .14, p < 0.01) and non-LM students (β = .10, p < .01) students. However, a chi-square test of equality indicated that this association did not differ between language groups (χ2 = 169.53, p = .27). Also, social economic status had significantly higher influence on students’ initial lab performance for LM students (β = .03, p > .05) compared to non-LM students (β = -.01, p < .05), χ2 = 7.75, p < 0.05. All other paths were not significantly different across LM groups.
left: LM students, right: non-LM students; latent covariates for academic preparedness and SES are included but not shown; dashed lines describe non-significant path estimate; *p < .05., **p < .01., ***p < .001.
Discussion
This large quantitative study extends the research base two-fold. This study contributes to discussions on the antecedents and benefits of inquiry-based and traditional lecture-based instruction for student learning in higher education. Also, this study extends these discussions to students with linguistically diverse backgrounds, which are currently underrepresented in the literature. The two main findings of this study are as follows:
First, this study suggests that initial performance in lectures or laboratory sessions can be a predictor of subsequent course performance in corresponding lecture and laboratory courses, depending on the disciplinary content area. These findings support the somewhat divergent literature base. One the one hand, researchers argue that lectures can be more effective mode of delivering complex information without detriment to student learning and performance [9, 46, 47]. This correspond to our findings regarding the Physics course series indicating that initial lecture performance was only associated with subsequent lecture performance but not with subsequent laboratory performance (while initial Physics laboratory performance predicted neither subsequent lecture or laboratory performance). On the other hand, other studies argue that inquiry-based learning is more effective than traditional lecture-based instruction [5, 10, 48]. This corresponds to our findings for the Chemistry course series indicating that initial laboratory performance predicted subsequent laboratory performance (but not subsequent lecture performance). This suggests that both disciplinary content and institutional context matter. Therefore, college instructors and educational policy-makers are encouraged to carefully consider their particular disciplinary content area when suggesting subsequent benefits of inquiry-based or lecture-based instruction.
Second, although LM students entered college with less beneficial academic preparation, compared to non-LM students–which mirrors current literature [49]–we did not find support that LM status influenced how initial lecture and inquiry-based instruction impacted subsequent performance in the course series. That said, we still found that LM students had overall lower courses grades in the initial lecture course but not the initial inquiry-based laboratory course, which aligns with research that suggests that active learning may be a means of reducing the achievement disparities often experience by LM and URM students [16, 26, 35, 50]. Further, our study indicates that LM students are not additionally penalized in subsequent lecture or laboratory courses in science gateway course series.
These findings have important implications for STEM education. Given that academic preparedness, rather than LM status, contributed to student performance in the initial course, we encourage educational administrators and university instructors to aim interventions and support systems at the first course in college gateway science courses sequences for LM students and others who may be less well prepared, such as first generation or low SES students. These may include short preparatory online courses with foci in the intersection of the science content and the challenges posed by scientific language prior to the first course [51, 52] or nudging intervention to increase motivation, self-efficacy and self-regulatory abilities [37, 53, 54].
Future work
With the increased availability of large institutional data sets at many universities [55], we encourage researchers to replicate this study at their own institutions to mirror the nationwide diversity in student demographics and institutional contexts to assess the generalizability of our findings. We particularly encourage these replications to take place at community colleges, as most LM students attend community colleges rather than four-year institutions [7, 56]. There are important differences between LM students who attend selective four-year universities, such as those in our study, and LM students attending less selective institutions and community colleges, particularly in terms of their English proficiency and academic preparedness [7]. Also, we encourage researchers to utilize additional data sources that are available at scale to supplement such analysis. Potential data sources may include clickstream data from the learning management systems that capture student behavior during the science courses [57]; or data from college entrance surveys, which may include measures of stable psychological constructs such as student personalities characteristics [58]. From a methodological perspective, this study utilizes a correlational research design which is common for such institutional research. That said, future research may attempt to get closer to generate causal inference, for instance through a randomized controlled field design. However, the university setting may posit some practical limitation hindering random sampling, for instance, regarding student course selection. In addition, future work may uncover factors that contribute to the performance gap of LM students in initial lecture classes. Such factors may include student-level factors such as self-regulatory skills, student motivation, and study skills [59–61]; instructor-level factors such as teaching quality, teaching experience, instructional styles, and choice of the textbook [5, 19, 62]; and context-level factors such as scheduling of the classes, department funding, and the perceived chilly climate often reported in STEM learning environments [27, 63, 64]. Another strand of research may develop and evaluate interventions attempting to narrow this initial gap in lecture course performance for LM students such as short preparatory courses [51, 52]; or nudging interventions [53, 54].
Conclusions
Our study adds to and extends the growing body of literature examining the more equitable approaches to STEM education for LM students [23, 28, 29]. Even among students enrolled at a selective research university, LM students, who were more likely to be first generation and come from lower SES backgrounds, had weaker academic preparation for higher education than their non-LM peers. For all students, weaker academic preparation contributed to worse performance in the initial lecture classes, which in turn contributed to weaker performance in subsequent lecture courses in the science gateway course sequences across all disciplines. Notably, while LM students performed worse in the initial lecture course, LM status resulted in neither worse performance in inquiry-based laboratory courses nor in worse performance in subsequent courses in the science gateway course sequence. The comparable performance of LM and their non-LM peers in the laboratory classes suggests that inquiry-based instruction and the active learning it promotes supports LM students and can be leveraged to enhance equity in STEM education. Further, this study suggests that colleges might adopt interventions to support LM students in college science that target the initial courses in the corresponding science gateway course sequences.
Limitations
This study has a variety of limitations that should be considered when interpreting findings and implications. Most notably, data for this study was provided from students at a selective research university, which can be mirrored in the high academic preparedness scores (i.e., SAT/ACT performance) for both LM and non-LM students compared to a national average. Descriptive analyses for monolingual- and bilingual English students indicated a relatively small variance by LM status. Therefore, we decided to collapse these two student groups to a single non-LM students comparison group. This limits the generalizability of our findings, as well as the potential to detect finer-grained differences among LM students. That said, we would expect a larger magnitude in effect sizes in replication studies with more diverse student populations (e.g., community colleges, private universities, open enrollment universities). Additionally, we only examined two course sequences for two introductory science courses sequences (i.e., General Chemistry and Classical Physics) suggesting replication studies in other science disciplines (e.g., Biology) and for more advanced course series (e.g., organic chemistry) that were not available in these lecture-laboratory combinations at this university.
Methodologically, the models have an omitted variable bias as this study only included constructs that were already available within the institutional data sets. Other potentially important variables for explaining variance in student performance such as self-regulation, self-efficacy, interest in STEM, among others, as well as variables important for LM students (e.g., years of exposure to English, participation in scientific English courses or dual language programs) were not measured. Similarly, variables describing the instructional context in the analyzed courses beyond the lecture/laboratory classification were not measured at the institutional level and available for this study. Also, the study design is correlational in its nature; therefore, we cannot infer causal claims about the presented relationships.
Supporting information
S1 File. Additional information, data tables, and robustness checks.
https://doi.org/10.1371/journal.pone.0267188.s001
(DOCX)
References
- 1. Keselman A. Supporting inquiry learning by promoting normative understanding of multivariable causality. J Res Sci Teach. 2003 Nov;40(9):898–921.
- 2. Pedaste M, Mäeots M, Siiman LA, de Jong T, van Riesen SAN, Kamp ET, et al. Phases of inquiry-based learning: Definitions and the inquiry cycle. Educ Res Rev. 2015;14:47–61.
- 3. Armbruster P, Patel M, Johnson E, Weiss M. Active Learning and Student-centered Pedagogy Improve Student Attitudes and Performance in Introductory Biology. CBE—Life Sci Educ. 2009;8(3):203–13. pmid:19723815
- 4. Deslauriers L, Schelew E, Wieman C. Improved learning in a large-enrollment Physics class. Science. 2011;332(6031):862–4. pmid:21566198
- 5. Freeman S, Eddy SL, McDonough M, Smith MK, Okoroafor N, Jordt H, et al. Active learning increases student performance in science, engineering, and mathematics. Proc Natl Acad Sci. 2014 Jun 10;111(23):8410–5. pmid:24821756
- 6. Kanno Y. High-Performing English Learners’ Limited Access to Four-Year College. Teach Coll Rec. 2018;120:1–46.
- 7.
Nunez A-M, Cecilia Rios-Aguilar, Yasuko Kanno, Stella M. Flores. English Learners and Their Transition to Postsecondary Education. In: Higher Education: Handbook of Theory and Research. Cham, Switzerland: Springer; 2016. p. 41–90.
- 8. Zhang L. Is Inquiry-Based Science Teaching Worth the Effort?: Some Thoughts Worth Considering. Sci Educ. 2016;25(7–8):897–915.
- 9. Dunlosky J, Rawson KA, Marsh EJ, Nathan MJ, Willingham DT. Improving Students’ Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology. Psychol Sci Public Interest. 2013 Jan;14(1):4–58. pmid:26173288
- 10. Arthurs LA, Kreager BZ. An integrative review of in-class activities that enable active learning in college science classroom settings. Int J Sci Educ. 2017;39(15):2073–91.
- 11. Carmel JH, Herrington DG, Posey LA, Ward JS, Pollock AM, Cooper MM. Helping Students to “Do Science”: Characterizing Scientific Practices in General Chemistry Laboratory Curricula. J Chem Educ. 2019;96(3):423–34.
- 12. Hofstein A, Lunetta VN. The laboratory in science education: Foundations for the twenty-first century. Sci Educ. 2004 Jan;88(1):28–54.
- 13. Kurdziel JP, Turner JA, Luft JA, Roehrig GH. Graduate Teaching Assistants and Inquiry-Based Instruction: Implications for Graduate Teaching Assistant Training. J Chem Educ. 2003;80(10):1206.
- 14. DeFeo DJ, Bibler A, Gerken S. The Effect of a Paired Lab on Course Completion and Grades in Nonmajors Introductory Biology. Schussler E, editor. CBE—Life Sci Educ. 2020 Sep;19(3):1–12. pmid:32822278
- 15. Freedman MP. Relationship among laboratory instruction, attitude toward science, and achievement in science knowledge. J Res Sci Teach. 1997;34(4):343–57.
- 16. Lee HS, Anderson JR. Student Learning: What Has Instruction Got to Do With It? Annu Rev Psychol. 2013;64(1):445–69.
- 17. Bertsch S, Pesta BJ, Wiscott R, McDaniel MA. The generation effect: A meta-analytic review. Mem Cognit. 2007;35(2):201–10. pmid:17645161
- 18.
Espinoza F. Wave Motion as Inquiry: The Physics and Applications of Light and Sound. Cham, Switzerland: Springer; 2017.
- 19.
Espinoza F. The impact on cognitive development of a self-contained exploratory and technology-rich course on the physics of light and sound. In: Sinha G, Suri J, editors. Cognitive Informatics, Computer Modelling, and Cognitive Science. Academic Press; 2020. p. 55–70.
- 20. Lazonder AW, Harmsen R. Meta-Analysis of Inquiry-Based Learning: Effects of Guidance. Rev Educ Res. 2016;86(3):681–718.
- 21. Cairns D, Areepattamannil S. Exploring the Relations of Inquiry-Based Teaching to Science Achievement and Dispositions in 54 Countries. Res Sci Educ. 2019;49(1):1–23.
- 22. Jiang F, McComas WF. The Effects of Inquiry Teaching on Student Science Achievement and Attitudes: Evidence from Propensity Score Analysis of PISA Data. Int J Sci Educ. 2015;37(3):554–76.
- 23.
Renninger KA, Hidi S, Krapp A. The role of interest in learning and development. New York, NY: Psychology Press; 2014.
- 24. Snyder JJ, Sloane JD, Dunk RDP, Wiles JR. Peer-Led Team Learning Helps Minority Students Succeed. PLOS Biol. 2016 Mar 9;14(3):1–7. pmid:26959826
- 25.
Seidman A. Minority Student Retention: The Best of the Journal of College Student Retention: Research, Theory & Practice. New York, NY: Routledge; 2019.
- 26.
Olson S, Riordan DG. Engage to Excel: Producing One Million Additional College Graduates with Degrees in Science, Technology, Engineering, and Mathematics. Washington, DC: Executive Office of the President; 2012.
- 27. Flynn DT. STEM Field Persistence: The Impact of Engagement on Postsecondary STEM Persistence for Underrepresented Minority Students. J Educ Issues. 2016;2(1):185–214.
- 28. Hurtado S, Cabrera NL, Lin MH, Arellano L, Espinosa LL. Diversifying Science: Underrepresented Student Experiences in Structured Research Programs. Res High Educ. 2009;50(2):189–214. pmid:23503690
- 29. LaCosse J, Canning EA, Bowman NA, Murphy MC, Logel C. A social-belonging intervention improves STEM outcomes for students who speak English as a second language. Sci Adv. 2020;6(40):1–10. pmid:33008912
- 30. Daempfle PA. An Analysis of the High Attrition Rates among First Year College Science, Math, and Engineering Majors. J Coll Stud Retent Res Theory Pract. 2003 May;5(1):37–52.
- 31. Doolen TL, Long M. Identification of retention levers using a survey of engineering freshman attitudes at Oregon State University. Eur J Eng Educ. 2007 Dec;32(6):721–34.
- 32. Shaw JM, Bunch GC, Geaney ER. Analyzing language demands facing english learners on science performance assessments: The sald framework. J Res Sci Teach. 2010;47(8):909–28.
- 33. Fang Z. Scientific literacy: A systemic functional linguistics perspective. Sci Educ. 2005;89(2):335–47.
- 34.
Schleppegrell MJ. The language of schooling: A functional linguistics perspective. Mahwah, NJ: Lawrence Erlbaum; 2004.
- 35. Dewsbury B, Brame CJ. Inclusive Teaching. CBE—Life Sci Educ. 2019 Jun;18(2):fe2. pmid:31025917
- 36. Theobald EJ, Hill MJ, Tran E, Agrawal S, Arroyo EN, Behling S, et al. Active learning narrows achievement gaps for underrepresented students in undergraduate science, technology, engineering, and math. Proc Natl Acad Sci. 2020;117(12):6476–83. pmid:32152114
- 37. Ballen CJ, Wieman C, Salehi S, Searle JB, Zamudio KR. Enhancing Diversity in Undergraduate Science: Self-Efficacy Drives Performance Gains with Active Learning. Dolan EL, editor. CBE—Life Sci Educ. 2017 Dec;16(4):ar56. pmid:29054921
- 38. Burke C, Luu R, Lai A, Hsiao V, Cheung E, Tamashiro D, et al. Making STEM Equitable: An Active Learning Approach to Closing the Achievement Gap. Int J Act Learn. 2020;5(2):71–85.
- 39. Estrada M, Burnett M, Campbell AG, Campbell PB, Denetclaw WF, Gutiérrez CG, et al. Improving underrepresented minority student persistence in STEM. CBE—Life Sci Educ. 2016;15(3):1–10. pmid:27543633
- 40. Callahan RM, Shifrer D. Equitable Access for Secondary English Learner Students: Course Taking as Evidence of EL Program Effectiveness. Educ Adm Q. 2016;52(3):463–96. pmid:27429476
- 41. Graham JW. Missing data analysis: Making it work in the real world. Annu Rev Psychol. 2009;60(1):549–76. pmid:18652544
- 42. Zyphur MJ, Allison PD, Tay L, Voelkle MC, Preacher KJ, Zhang Z, et al. From Data to Causes I: Building A General Cross-Lagged Panel Model (GCLM). Organ Res Methods. 2020;23(4):651–87.
- 43. Little TD, Preacher KJ, Selig JP, Card NA. New developments in latent variable panel analyses of longitudinal data. Int J Behav Dev. 2007;31(4):357–65.
- 44.
Kline RB. Principles and practice of structural equation modeling. 4th ed. New York, NY: Guilford Press; 2011.
- 45.
Hair JF, Black WC, Babin BJ, Anderson RE. Multivariate Data Analysis. Upper Saddle River, NJ: Pearson; 2014.
- 46. Kirschner PA, Sweller J, Clark RE. Why minimal guidance during instruction does not work: An analysis of the failure of constructivist, discovery, problem-based, experiential, and inquiry-based teaching. Educ Psychol. 2006;41(2):75–86.
- 47.
Tobias S, Duffy TM. Constructivist instruction: Success or failure? New York, NY: Routledge; 2009. https://doi.org/10.1177/1084713809356701 pmid:20150187
- 48. de Jong T. Moving towards engaged learning in STEM domains; there is no simple answer, but clearly a road ahead. J Comput Assist Learn. 2019;35(2):153–67.
- 49. Contreras F, Fujimoto MO. College Readiness for English Language Learners (ELLs) in California: Assessing Equity for ELLs under the Local Control Funding Formula. Peabody J Educ. 2019;94(2):209–25.
- 50. Teig N, Scherer R, Nilsen T. More isn’t always better: The curvilinear relationship between inquiry-based teaching and student achievement in science. Learn Instr. 2018;56:20–9.
- 51. Fischer C, Zhou N, Rodriguez F, Warschauer M, King S. Improving College Student Success in Organic Chemistry: Impact of an Online Preparatory Course. J Chem Educ. 2019 May 14;96(5):857–64.
- 52. Chaytor JL, Al Mughalaq M, Butler H. Development and Use of Online Prelaboratory Activities in Organic Chemistry To Improve Students’ Laboratory Experience. J Chem Educ. 2017 Jul 11;94(7):859–66.
- 53. Baker R, Evans B, Dee T. A Randomized Experiment Testing the Efficacy of a Scheduling Nudge in a Massive Open Online Course (MOOC). AERA Open. 2016;2(4):1–18.
- 54. Baker R, Evans B, Li Q, Cung B. Does Inducing Students to Schedule Lecture Watching in Online Classes Improve Their Academic Performance? An Experimental Analysis of a Time Management Intervention. Res High Educ. 2019;60(4):521–52.
- 55. Fischer C, Pardos ZA, Baker RS, Williams JJ, Smyth P, Yu R, et al. Mining Big Data in Education: Affordances and Challenges. Rev Res Educ. 2020 Mar;44(1):130–60.
- 56. Kibler AK, Bunch GC, Endris AK. Community College Practices for U.S.-Educated Language-Minority Students: A Resource-Oriented Framework. Biling Res J. 2011;34(2):201–22.
- 57. Baker R, Xu D, Park J, Yu R, Li Q, Cung B, et al. The benefits and caveats of using clickstream data to understand student self-regulatory behaviors: opening the black box of learning processes. Int J Educ Technol High Educ. 2020;17(1):13.
- 58. Cobb-Clark DA, Schurer S. The stability of big-five personality traits. Econ Lett. 2012;115(1):11–5.
- 59. Broadbent J, Poon WL. Self-regulated learning strategies & academic achievement in online higher education learning environments: A systematic review. Internet High Educ. 2015;27:1–13.
- 60. Eccles JS, Wigfield A. From expectancy-value theory to situated expectancy-value theory: A developmental, social cognitive, and sociocultural perspective on motivation. Contemp Educ Psychol. 2020;61.
- 61. Rodriguez F, Fischer C, Zhou N, Warschauer M, Massimelli Sewall J. Student spacing and self-testing strategies and their associations with learning in an upper division microbiology course. SN Soc Sci. 2021;1:1–24. pmid:34693299
- 62. Papay JP, Kraft MA. Productivity returns to experience in the teacher labor market: Methodological challenges and new evidence on long-term career improvement. J Public Econ. 2015;130:105–19.
- 63. Liefner I, Clark BR. Funding, resource allocation, and performance in higher education systems. High Educ. 2003;46:469–89.
- 64. Preckel F, Lipnevich AA, Schneider S, Roberts RD. Chronotype, cognitive abilities, and academic achievement: A meta-analytic investigation. Learn Individ Differ. 2011;21(5):483–92.