Research to Practice, Practice to Research
Connected Science Learning November-December 2021 (Volume 3, Issue 6)
By Lynn D. Dierking, John H. Falk, Neta Shaby, and Nancy L. Staus
Many researchers use “ecological” perspectives to frame their learning studies, which is the idea that a STEM learning ecosystem contains varied resources—both in and out of school (Falk et. al 2015; Staus et al. 2020; Traphagen and Traill 2014)—and youth construct unique STEM interest and participation pathways (SIPPs) as they traverse the ecosystem. Research suggests that rather than typical influences (i.e., grades and courses taken in school), factors such as interest, identity, and participation in out-of-school, informal/free-choice learning activities during the middle-school years—as well as social and cultural capital factors (income, education, and geographical access to resources)—are collectively the best predictors of future engagement and participation in STEM (McCreedy and Dierking 2013; Fortus 2014; Maltese, Meilki, and Wiebke 2014). As a result, many researchers, educators, and policymakers have begun to advocate for an ecosystem approach to STEM learning: an approach that supports STEM interest and participation across the day and settings, both in school and outside school (Dierking and Falk 2003; NRC 2015; Traphagen and Traill 2014).
A six-year US–National Science Foundation (NSF) project, SYNERGIES: Customizing Interventions to Sustain Youth STEM Interest and Participation Pathways investigated the STEM Interest and Participation Pathways (SIPPs) of multiple youth (11–14 years old) within the same low-income, urban community over time (days, months, and years), both in school and during out-of-school (OOS) informal/free-choice learning time. Quantitative findings from the analysis of the STEM Interest and Participation Pathway survey questionnaire administered once a year include a factor analysis study (Falk et al. 2015; Staus et al. 2020a; Staus et al. 2020b) and latent profile and transition analyses. These studies show that, in aggregate, youth pathways were influenced by complex factors; in particular, youth persisted in and sustained STEM interest if they had family support for “their” interest and engaged in OOS activities.
In a qualitative component of the research, we tracked three participating youth, Charlie, Steve, and Stella (pseudonyms that students created), who at the start of the study were interested in STEM (Shaby et al. 2021). We interviewed each of them over two to three years as they moved through the learning ecosystem to “see” in an in-depth manner whether their interests persisted and were sustained. These data demonstrate how three individual youth living in the same community were influenced by
This study highlighted the opportunities and obstacles each youth faced in pursuing their STEM interests, as well as factors influencing the further development of and persistence in the interest. Even though the three youth live in the same learning ecosystem and attended the same middle school and afterschool program, the mere presence of STEM learning resources did not guarantee that each youth was aware of the resources or if they were aware of the learning resources that they felt the resources were available and accessible to them. Finances (economic capital), geography, available transportation options, and other social or cultural capital resources all were factors. Each of the pathways were exceedingly unique and from the perspective of each youth the ecosystem was different.
This article attempts to deepen and extend the qualitative findings of the three youths’ SIPPs by moving beyond their three unique pathways, which are difficult to generalize to the overall STEM learning ecosystem. We also strive to define the characteristics essential in addressing fundamental questions such as whether the ecosystem is thriving; if it is, who does it work for, and if it is not thriving, what is the reason? We argue that the ecological sciences have applied systematic approaches and empirical methods to study ecosystems, while educational research and applications have primarily used the ecosystem concept as a descriptive metaphor, which is only helpful to a degree (Hecht and Crowley 2020; Falk, Dierking, and Staus 2020). We make the case that researchers and educators using the ecosystem model for learning contexts could significantly benefit from adapting the analytical and application approaches pioneered within the ecological sciences that have enabled the development of adaptive management strategies such as those used in ecosystem restoration and recovery efforts.
Like ecologists, we defined a STEM learning ecosystem in a very specific way, focusing on three qualities of thriving ecosystems: (1) productivity; (2) durability; and (3) resilience (see Falk, Dierking, and Staus 2020). Also, as with living things in natural ecosystems, we assumed that learners perceived the learning opportunities in their ecosystem in varied ways for different reasons. Thus, although all youth ostensibly live within the same learning ecosystem (one that contains comparable resources and opportunities), reality is likely not this straightforward. The mere presence of STEM learning resources in a community does not guarantee that learners are aware of them; if they are aware, they feel resources are available and accessible to them because of either finances, geography, transportation, or other social/cultural capital reasons. In addition, not all learning ecosystems contain the same number or variety of STEM offerings, depending upon systemic factors such as the ecosystem’s size, affluence, or location (e.g., rural/urban). It is our hope that this approach will help inform the management and adaptability of entire learning ecosystems.
Over the three years we tracked Charlie, Steve, and Stella, we wanted to understand what the STEM learning ecosystem “looked like” for each of them and what supported or hindered their strong interest in STEM, with the broader goal of understanding their individual journeys and better understanding critical elements of a thriving STEM learning ecosystem. Here we provide highlights of the findings from this perspective; for a detailed description of the methods and each youth’s pathway, refer to Shaby et al. (2021).
Fortunately, when Charlie entered high school, he became interested in video and took two elective courses, Advanced Video and Newsroom. Charlie’s teacher noticed his piqued interest and offered him a job using the video class equipment to record school board meetings. This permitted Charlie to continue his interest in storytelling—in this case, visual storytelling. He now had an effective guide in his teacher, who helped him economically by offering him a job. It is possible that Charlie’s interests all along were more in storytelling than technology, computers, and coding—an emerging line of practice (Azevedo 2011) that he and his family (and even the supportive teacher) may not have entirely understood. Despite his strong desire to learn about programming generally, and coding in particular, Charlie may have had difficulty identifying and locating relevant resources and opportunities in the ecosystem because he lacked the adult support to help him determine exactly what his interests were. These constraints in the structure of the ecosystem itself, and Charlie's inability to access additional resources related to computers and coding (and perhaps storytelling), decreased his participation in such activities, making it challenging for his interest to persist. Charlie had consistently taken advantage of opportunities in the ecosystem readily available to him, but his choices were limited both in and outside of school due to his and his family’s limited financial and social capital. The ability for youth to fully identify their interests and find resources within the ecosystem is critical to their pathways and persistence in a specific STEM area. Fortunately, he had a high school teacher who noticed his interest and was an effective guide in helping him pursue opportunities in video production.
Steve takes a path that at times is difficult to recognize, until guides connect with his interest and assist him in realizing that the ecosystem includes more visible and accessible learning resources. When we first met Steve, he was totally infatuated with ants; however, he quickly exhausted his ability to pursue this interest, in part because of the lack of social capital that his guides (in this case his extended family) had related to his topic of interest. Although family members tried to capture ants for him, collectively they were unable to help him find additional resources, hindering his interest pathway.
Beyond his and his family’s lack of social capital, there was another critical reason that Steve’s initial path seemed difficult to recognize. Both Steve’s specific interest in ants—and his more general interest in entomology—are exceedingly specific interests with very few visible resources or guides available in the Parkrose STEM learning ecosystem, either in school or outside school. School provided little support, so he primarily pursued this interest at home, via the internet, and on his own by observing captured ants and trying to build a colony. If his parents had additional social capital, they might have taken Steve to the local public library or to observe an ant swarming after a nuptial flight in Central Oregon; they might have also perused the Portland State University faculty directory to see if there was a local expert on ants or entomology, but seemingly they did not understand that these options were available.
However, Steve—still a resilient STEM-interested youth—began pursuing an alternate interest he had in computer science, perhaps related to his extended family (two of his uncles work in Silicon Valley). This was a STEM area in which his extended family had social capital and one in which both in-school and out-of-school learning resources existed; collectively these resources allowed Steve to successfully pursue this interest.
There was also one other critical reason why Steve was able to progress in this interest. When Steve first identified this “new” interest, his parents gave him a great deal of support, another form of social capital. This was not because they necessarily shared his interest, but because of the lived experience of their extended family, they could see that their son might be able to make a good living in this area; an important issue for them. Thus, his parents were motivated to support his interest. Even though Steve still struggled to find opportunities to push his coding knowledge and skills to ever-higher levels, he felt like he got more support for this area of interest than he did for his interest in ants. The support offered by his extended family—on a topic in which they have social and cultural capital—allowed them to effectively guide Steve through the ecosystem. This influenced his ability to participate in activities related to his interest, which facilitated his persistence.
Stella takes a well-lit path, with visible signposts and skilled guides, as well as connections between her interests and available learning resources. Early on Stella expressed a strong interest in astronomy and made broad use of the Parkrose and extended Portland STEM learning ecosystem to pursue this interest. Initially her expressed interest was in astrophysics, until she appreciated how much math was required; she switched to astrobiology after taking a biology course in high school. During her middle school years, she participated in the SUN afterschool program, an amateur astronomy club, and an astronomy-focused Girl Scout troop. Her parents, who also were interested in astronomy, not only helped initiate her interest but also had the social capital to connect her to relevant resources. This alignment between her and her parents’ interests served as a mutually reinforcing motivation.
Unlike Steve’s interest in ants, Stella’s interest in astronomy also was more readily fulfilled as there were numerous out-of-school astronomy resources in Parkrose and the greater Portland community. These resources were effectively signposted in a way that Stella and her parental guides could readily access; if they were outside Parkrose, Stella’s parents drove her to them. It is unlikely that these resources were all free and there was no indication that the cost of her interest was an issue for the family financially. In fact, at her last interview, she was pleased to share that her parents had bought her a new telescope for her birthday (that she asked for), in addition to the two other telescopes the family already owned and the one she made at Girl Scouts.
There also was a serendipitous event during the study, a complete solar eclipse, which further reinforced Stella’s interests and her parents’ involvement in supporting the interest. It is critical to note that while high school offered Stella more choice and opportunities to explore her astronomy interest, she began to push back a bit about being identified only as a STEM-interested youth. She talked about her “new” interests in performance (she joined the drama class, band, and choir). Although Stella was proud of "branching out," she still strongly claimed at her last interview that she planned to do something STEM-related in the future, although she did not know in what area. Research findings indicated that Stella's pathway—including her persistence in STEM overall and in astronomy specifically, as well as a clearly developing STEM identity—was quite unique among most Parkrose youth at this age.
These cases offer useful insights into how three specific youth perceived the resources available to them in a STEM learning ecosystem, highlighting the affordances and constraints each faced in pursuit of their interests, as well as the curious role of serendipity. Importantly, each youth’s interest in STEM persisted over the three years we followed them, although not always in the same specific areas of focus. Collectively, the STEM Interest and Participation Pathways (SIPPs) of these youth uniquely demonstrate the critical importance of understanding the nature of the ecosystem itself and its characteristics, including its structure and the availability and access to learning resources (with access defined primarily by a youth's family social, cultural, and financial capital). Also important was whether resources to support specific interests, activities, or practices were numerous and signposted in ways that made them visible and relatively easy to access and use.
However, to apply these ideas more generally to learning ecosystems writ large, it is critically important to transcend the individual pathways so that we can actively and intentionally create and tweak ecosystems in ways that increase the number of children who have access to quality STEM learning early on; find it interesting; and then begin a supported, connected, and valued journey toward a life that includes STEM. As outlined by Falk, Dierking and Staus (2021), it is helpful to frame these findings using the three qualities of healthy, thriving biological ecosystems: (1) productivity, (2) durability, and (3) resilience. It is important to point out that although we will discuss these three dimensions as separate entities, this is not the case at all; it is difficult to categorize the characteristics and processes we observed in the three youth SIPPs (as well as in our entire sample), into just one of the three dimensions (see Figure 1). Collectively, these three qualities contribute to a healthy and thriving ecosystem.
Understanding what, and how much an ecosystem “produces” is critical to understanding a system. The biological study of ecosystems was revolutionized by the consistent use and quantification of productivity measures such as energy (McIntosh 1985). Within a learning ecosystem, productivity is a function of its structure and the visibility, availability, and access learners have to its resources; the three youth SIPPs presented here demonstrate the varying interaction with these resources that can occur within the same ecosystem. Specifically, youth pathways were significantly influenced by the quantity and characteristics of in-school and out-of-school resources and activities related to their interests that were available within a variety of learning settings and contexts during middle school and high school years.
There were additional structural issues in the ecosystem that strongly influenced youth STEM interest and participation pathways. In terms of formal schooling, no youth reported that science and math classes at the middle school (ages 12–14) were important in either triggering or sustaining specific STEM interest. By contrast, high school (ages 15–16) classes were consistently identified as important resources for supporting STEM interest. It is fair to generalize that there are many more ecosystem opportunities at the high school level since youth have more choice and control over courses that align with their interests and aspirations. Like Bricker and Bell (2014), we found that offerings in middle school seemed to potentially constrain the maintenance of STEM interest.
These structural issues were also a factor with the out-of-school learning resources in the ecosystem. Programs such as afterschool, weekend, and summer opportunities are now appreciated as important in supporting STEM engagement, learning, interest, and motivation, particularly among low-income youth and youth of color (McCreedy and Dierking 2013; Bevan et al. 2010; Clark 1990; NRC 2009 2015; Stocklmayer, Rennie, and Gilbert 2010). Although all youth in this study participated in afterschool programs during middle school, the programs provided varying and mostly incomplete levels of support for interests. A further constraint was that these programs were only available for one month in the summer and most did not continue into high school. Summertime was consistently a STEM learning “desert” in this ecosystem, despite our efforts to support family STEM engagement during these months. Thus, although out-of-school resources were important, they suffered from similar structural issues to those identified in the middle school. As a result, collectively, the ecosystem was not sufficiently productive to consistently sustain specific STEM interests over time at this critical age.
Healthy natural ecosystems are characterized by considerable durability that arises from long-term intersecting relationships between and among resources. Therefore, measuring ecosystem durability requires that the entire system be studied as a whole, rather than merely as individual or isolated parts of the system. Learning ecosystems are durable and persistent only to the degree that they support access and participation across settings and social arrangements over long periods of time. One important aspect of durability is redundancy; in learning ecosystems, complexity develops as the number, richness, and diversity of learning resources increase; for example, individuals can learn about geology, both in school and through out-of-school opportunities such as afterschool programs, special interest clubs, websites, online resources, etc.
As highlighted in this study, social and cultural capital (as well as effective guides) play important roles in helping youth pursue long-term interests. This study supports the idea that in a healthy learning ecosystem youth have opportunities to identify and use appropriate learning resources across settings and over time (durability) that extend their interest. The mere availability of learning resources, productivity, is insufficient by itself to establish a healthy and thriving ecosystem. As suggested above, the capacities afforded by durability also need to be present so youth can locate and use the "next" resource that will help them continue to pursue their interest.
The ability of a youth in this ecosystem to continue to pursue their interests was greatly influenced by both youth and their families’ social and cultural capital, which influenced their ability to find and leverage potential resources. The role of familial social, cultural, and financial capital was observed in all three cases most often when it was missing. We found that Steve exhausted his ability to pursue his interest in ants in part because of his parents and extended family’s lack of social or cultural capital related to his specific interest but also their understanding of the overall STEM learning ecosystem, its learning resources, and how to access them. Financial constraints also emerged as a limitation for some youth’s ability to pursue their interests.
The reverse was also true. Stella’s parents possessed ample quantities of social, cultural, and financial capital (including understanding how to access the resources), so there were seemingly endless possibilities to pursue her STEM interests with no apparent geographical or financial constraints. Stella’s perceived ecosystem was more extensive than that of the other two youth, in part because her parents could locate and access a wide variety of often geographically dispersed resources.
One other family-related issue of critical importance often not considered in the literature—likely because it can be sensitive—relates to whether the family values support a child’s interest. Important to youth at this point in their development, particularly youth from underrepresented communities, is that they feel their family supports their interest (Wang 2020). Value placed on making a good living—combined with the social capital of knowing people who work and succeed in a variety of STEM occupations, including as technologists and other support roles within the scientific enterprise—can make a tremendous difference.
To be resilient, the species and communities within an ecosystem (in our case, in-school and out-of-school learning resources, the educators offering them, and the learners and families engaging with them) must be able to buffer disturbance, reorganize and renew, and learn to adapt and transform in response to change (Lavorel et al. 2015). We all observed this starkly with the advent of COVID-19, as both in-school and OOS educators scrambled to learn from and adapt to these new realities. Also, as is true of biological ecosystems (e.g., Mahonge 2010), the more complex and highly integrated a learning ecosystem, the healthier and more resilient it can be.
New research into climate adaptation in biological systems has identified three primary mechanisms and traits that support the resilience of ecosystems and facilitate their capacity to adapt: structural diversity, the role of keystone species, and connectivity. In a biological system, this may be maintaining perennial vegetation to reduce the risk of future desertification; preserving intact, diverse, connected forest stands; or focusing on greater management of fire-sensitive species. We can view the resilience of a learning ecosystem similarly. For example, resilience increases when educational organizations and resources foster substantial collaboration and connections (i.e., synergies within and between themselves). It is also critical that the ecosystem includes key learning resources to support specific interests, activities, or practices that are perceived as abundant, visible, and accessible. In some cases, specific youth interests were not supported by the ecosystem under study, either because resources were not diverse enough or were not effectively signposted in ways that made them relatively easy to identify, access, and use.
Youth used digital resources (e.g., YouTube videos) to pursue interests, but even this resource appeared to be insufficient for sustaining engagement in the absence of other supports in the ecosystem. The topic of the interest also made a difference; there were more resources and opportunities aligned with some specific areas than others, which in turn influenced outcomes. For example, as discussed earlier, there were far fewer clearly visible resources about ants in this ecosystem than there were about astronomy.
Although youth could pursue an existing interest or an interest could be triggered, it was clear that if youth had less social, cultural, or financial capital, it often influenced their ability to effectively navigate the ecosystem, and thus they found it challenging to locate learning resources that could sustain their interest. As has been discussed in the out-of-school, informal, and free-choice STEM learning arenas, a healthy, resilient learning ecosystem contains many diverse resources and opportunities in school and out of school that have the potential to excite youth—and possibly even adults and families—about a topic. We found that the STEM learning ecosystem under study was less rich and diverse in terms of resources and effective signposting, making it difficult for some youth to be inspired and sustain an interest (Staus, Falk, and Dierking, forthcoming).
Based on the findings and their interpretation, here are some recommendations for building a productive, durable, and resilient learning ecosystem healthy enough to support and sustain youth’s interest and participation in STEM, both in adolescence and beyond:
We have argued throughout this article that researchers and educators using an ecosystem model in learning contexts could benefit from adapting ecological analytical and application approaches to develop adaptive management strategies, like those used in biological ecosystem restoration efforts. Complex ecosystems like the Parkrose STEM learning ecosystem are dynamic, both at the individual level of learners and at the overall ecosystem level. Ideally, the goal is to create and foster a community-wide effort to build a productive, durable, and resilient STEM learning ecosystem healthy enough to support and sustain youth’s interest and participation in STEM, both in adolescence and beyond.
This work was supported in part by a grant from the U.S. National Science Foundation (DRL-1516718). We also want to acknowledge Dr. Yoon Ha Choi for her assistance in data collection and interpretation; Tanya Kindrachuk for coordinating all SYNERGIES data collection; and Kiyauna Williams, SUN Afterschool Coordinator, for her support. Finally, we thank Charlie, Steve, and Stella, and their parents, for participating in this study and so graciously sharing their STEM Interest and Participation Pathways with us.
Lynn D. Dierking (Lynn.Dierking@freechoicelearning.org) is Principal Researcher at the Institute for Learning Innovation and Professor Emeritus at Oregon State University in Corvallis, Oregon. John H. Falk is Executive Director at the Institute for Learning Innovation and Professor Emeritus at Oregon State University in Corvallis, Oregon. Neta Shaby is a Lecturer in Science Education in the Southampton Education School and a member of the MSHE (Mathematics, Science and Health Education) research group at University of Southampton in the U.K.; she served as a post-doctoral scholar with the SYNERGIES project from January 2019 through June 2020. Nancy L. Staus is a Senior Researcher at the Center for Research on Lifelong STEM Learning at Oregon State University in Corvallis, Oregon.
citation: Dierking, L.D., J.H. Falk, N. Shaby, and N.L. Staus. 2021. Thriving STEM learning ecosystems—for all? Connected Science Learning 3 (6). https://www.nsta.org/connected-science-learning/connected-science-learning-november-december-2021/thriving-stem-learning
Azevedo, F.S. 2011. Lines of practice: A practice-centered theory of interest relationships. Cognition and Instruction 29 (2): 147–184.
Bevan, B., J. Dillon, G.E. Hein, M. Macdonald, V. Michalchik, D. Miller, D. Root, L. Rudder, M. Xanthoudaki, and S. Yoon. 2010. Making science matter: Collaborations between informal science education organizations and schools. Technical Report. Washington, DC: Center for Advancement of Informal Science Education.
Bricker, L.A. and P. Bell. 2014. “What comes to mind when you think of science? The Perfumery!”: Documenting science-related cultural learning pathways across contexts and timescales. Journal of Research in Science Teaching 51 (3): 260–285.
Cresswell I.D., and H. Murphy. 2016. Biodiversity: Factors affecting resilience capacity. In: Australia state of the environment 2016, Australian Government Department of the Environment and Energy, Canberra. https://soe.environment.gov.au/theme/biodiversity/topic/2016/factors-affecting-resilience-capacity, DOI 10.4226/94/58b65ac828812
Clark, R.M. 1990. Why disadvantaged students succeed: What happens outside school is critical. Public Welfare (Spring): 17–23.
Dierking, L.D., and J.H. Falk. 2003. Optimizing out-of-school time: The role of free-choice learning. New Directions for Youth Development 97 (Spring): 75–88.
Dierking, L.D., and J.H. Falk. 2016. 2020 Vision: Envisioning a new generation of STEM learning research in (Eds.) L.D. Dierking and J.H. Falk. Cultural Studies in Science Education. 11(1): 1–10. DOI: 10.1007/s11422-015-9713-5.
Falk, J.H., and L.D. Dierking. 2018. Learning from museums. Lanham, MD: Rowman and Littlefield.
Falk, J.H., and L.D. Dierking. Forthcoming. On-ramps to nowhere: Are free-choice science learning institutions/organizations fulfilling their goals? In Amplifying Informal Science Learning eds. J. Diamond and S. Rosenfeld. New York: Routledge.
Falk, J.H., L.D. Dierking, and N.L. Staus. 2020. The use of ecological concepts in the social sciences: Measuring the productivity, durability and resilience of learning ecosystems. Ecology and Conservation Science 1 (3): 555563. DOI: 10.19080/ECOA.2020.01.555563
Falk, J.H., L.D. Dierking, N. Staus, W. Penuel, J. Wyld, and D. Bailey. 2015. Understanding youth STEM interest and participation pathways within a community: The SYNERGIES project. International Journal of Science Education, Part B 6 (4).
Falk, J.H., L.D. Dierking, N.L. Staus, J.N. Wyld, D.L. Bailey, and W.R. Penuel. 2016. The Synergies research–practice partnership project: a 2020 Vision case study. Cultural Studies in Science Education 11 (1): 195–212. DOI: 10.1007/s11422-015-9716-2.
Fortus, D. and D. Vedder-Weiss. 2014. Measuring students' continuing motivation for science learning. Journal of Research in Science Teaching 51: 497–522. https://doi.org/10.1002/tea.21136.
Hecht, M., and K. Crowley. 2020. Unpacking the learning ecosystems framework: Lessons from the adaptive management of biological ecosystems. Journal of the Learning Sciences 29 (2): 264–284. https://doi.org/10.1080/10508406.2019.1693381.
Lavorel, S., M.J. Colloff, S. McIntyre, M.D. Doherty, H.T. Murphy, D.J. Metcalfe, M. Dunlop, R.J. Williams, R.M. Wise, and K.J. Williams. 2015. Ecological mechanisms underpinning climate adaptation services. Global Change Biology 21 (1): 12–31. https://doi.org/10.1111/gcb.12689
Maltese, A.V., C.S. Melki, and H.L. Wiebke. 2014. The nature of experiences responsible for the generation and maintenance of interest in STEM. Science Education 98 (6): 937–962.
McCreedy, D., and L.D. Dierking. 2013. Cascading influences: Long-term impacts of informal STEM programs for girls. Philadelphia, PA: Franklin Institute Science Museum Press.
National Research Council. 2009. People, places and design: Learning science in informal environments. Washington, DC: National Academies Press.
National Research Council. 2015. Identifying and supporting productive STEM programs in out-of-school settings. Washington, DC: National Academies Press.
Shaby, N., N.L. Staus, L.D. Dierking, and J.H. Falk. 2021. Pathways of interest and participation: How STEM-interested youth navigate a learning ecosystem. Science Education. 105 (4): 628–652.
Staus, N.L., K. Lesseig, R. Lamb, J.H. Falkand, and L. Dierking. 2020a. Validation of a measure of STEM interest for adolescents. International Journal of Science and Mathematics Education 18 (2): 279–293.
Staus, N.L., J.H. Falk, W. Penuel, L. Dierking, J. Wyld, and D. Bailey. 2020b. Interested, disinterested, or undecided: Exploring STEM interest pathways in a low-income urban community. EURASIA Journal of Mathematics, Science and Technology Education 16 (6): em1853.
Stocklmayer, S.M., L.J. Rennie, and J.K. Gilbert. 2010. The roles of the formal and informal sectors in the provision of effective science education. Studies in Science Education 46 (1): 1–44.
Traphagen, K., and S. Traill. 2014. How cross-sector collaborations are advancing STEM learning. Technical Report. Los Altos, CA: Noyce Foundation.
Wang, X. 2020. On my own: The challenge and promise of building equitable STEM transfer pathways. Boston: Harvard Education Press.
Equity Research STEM Middle School Informal Education