You are here
Leveraging the LibreTexts to Facilitate a Community-built Chemistry Textbook Library: From Start to Finish
The LibreTexts Project is a multi-institutional collaborative venture developing the next generation of open-access textual educational materials to improve postsecondary education at all levels of higher learning by developing an Open Education Resource (OER) platform. The project currently consists of 15 independently operating and interconnected libraries that are constantly being optimized by students, faculty, and outside experts to supplant conventional paper-based books. A primary goal of the project is to complete the OER textbooks to enable a zero textbook cost option for a American Chemical Society ACS certified curriculum for a Bachelor's degree. This will be the first comprehensive set of ZTC OER textbooks for a STEM Bachelor's degree curriculum. Given that chemistry is a “central science” and plays a supporting role to many other degrees, the complete Chemistry LibreTexts will have broad impact and will be a powerful example for other fields to follow. The project organization entails effort in five teams: (1) The Construction Team, (2) The Harvesting Team, (3) The Technology Team, (4) The Dissemination Team, and (5) the Assessment and Analysis team. The development approach is highly collaborative and expands through the efforts of a growing community, which has been increasingly active in recent years. All are encouraged to participate.
Narrative
It is becoming ever clearer that new and innovative educational efforts are required to facilitate the greater creativity, flexibility, and increased learning capability needed for post-secondary education in the future. The increasingly diverse American educational landscape requires more personalized approaches. Unfortunately, rapidly rising undergraduate fees and textbook costs impede access to higher education for many students. In particular, high textbook costs are a serious barrier for under-served, at-risk students. The amount students are asked to pay for textbooks overburdens governmental and familial financial resources. A 2005 US Government Accountability Office study estimated that the average annual cost of textbooks for a student in 2003-04 was $898 at four-year colleges with an average 3.8-year lifespan per textbook (US Government Accountability Office, 2005). Textbook costs continue to rise at a rate twice that of inflation; according to the National Center of Education Statistics, the average budget for textbooks and supplies during 2016-17 was $1,263 at four-year schools and $1,458 for those attending community colleges (College Board, 2013). The high cost leads many students to sell their textbooks onto the secondary market to raise money for the next semester’s book budget.
The LibreTexts Project is a multi-institutional collaborative venture developing the next generation of open-access textual educational materials to improve postsecondary education at all levels of higher learning by developing an Open Education Resource (OER) platform. The project currently consists of 15 independently operating and interconnected libraries that are constantly being optimized by students, faculty, and outside experts to supplant conventional paper-based books. The functional linking of the LibreTexts Libraries makes them ideal for adaptive learning to support individual students as they seek to master subjects. These free textbook alternatives are organized within a central platform that is both vertically (from basic to advanced) and horizontally (across different fields) integrated.
The ChemWiki is the predecessor project to the LibreTexts with a focus on building and disseminating Chemistry content. The initial construction, dissemination, evaluation, and adoption of the ChemWiki was supported with National Science Foundation grants (CCLI, IUSE and CHE programs - 1246120, 1525057, and 1413739) and is currently supported by the US Department of Education (FIPSE OER Pilot - PT12701438) and the State of California (California Education Learning Lab).
The impact of OER projects is often formulated within a financial perspective of the number of dollars saved of student textbook expenditures. While the LibreTexts has exceeded $40M to date and represent an excellent return on investment within this metric, we prefer to quantify the impact of LibreTexts in terms of educational metrics, e.g., how much students use the resource and for how long. Since its inception 13 years ago as the ChemWiki, the LibreTexts project has grown dramatically and delivered over half a billion (541,068,770 as of 12/15/2020) pageviews with an accumulated 4.2 millennia of confirmed student reading (https://tinyurl.com/LibreStats). The overall project serves over 25 million pageviews per month with chemistry as the most popular library as is responsible for 80% of the traffic (Figure 1).
Figure 1: The Chemistry library Traffic Analytics. A more comprehensive set of statistics can be viewed online at https://datastudio.google.com/u/0/reporting/818412e1-0914-47df-b35b-ae8028667681/page/EFWjB
Building the Chemistry Library
The current development efforts of the LibreTexts are multifaceted with activity in five general thrusts: (1) Construction new materials, (2) Harvesting for inclusion of materials developed by others under open license, (3) developing leading edge technology to guide student learning, (4) dissemination of LibreTexts OER and recruiting new collaborators, and (5) formative and summative assessment and analysis. A cursory review of our Youtube channel shows that these efforts are spread broadly over all libraries (https://www.youtube.com/LibreTexts), but this report focuses on primarily on the “state of affairs” and future efforts in the construction of the Chemistry library.
While we pursue a "no gap left behind" policy of broad construction across all LibreTexts libraries, a special priority in the construction effort is dedicated on the Chemistry library. We intend to complete a suite of OER textbooks to enable a zero textbook cost (ZTC) option for an American Chemical Society ACS certified curriculum for a Bachelor's degree. This will be the first comprehensive set of ZTC OER textbooks for any STEM Bachelor's degree curriculum. Given that chemistry is a “central science” and plays a supporting role to many other degrees, this library will have broad impact and will be a powerful example for other fields to follow.
Figure 2: Organization of the Chemistry LibreTexts into the Bookshelves and the Course Hubs. https://Chem.LibreTexts.org
The organization of content on the Chemistry library (and all other LibreTexts libraries) involves storing of texts either within the Bookshelves or the Course (Figure 2). The Bookshelves “hold” the canonical and centrally curated texts, while the Course provides institutional hubs for faculty to build their own text or customize existing texts from the Bookshelves. We focus here on the building of canonical texts.
The Chemistry construction effort is broken into five working groups that are independently operating on filling existing gaps in the library to build a comprehensive set of OER textbooks for to support a complete ACS undergraduate chemistry degree program. Each team is led by a faculty expert in the specific chemistry subdiscipline:
- Layne Morsch (University of Illinois, Springfield) focuses on the Organic content.
- Kathryn Haas (Saint Mary’s College) focuses on the Inorganic Chemistry content.
- Delmar Larsen (UC Davis) is leading the effort to develop the Physical Chemistry content
- David Harvey (DePauw University) focuses on development of the Analytical Chemistry.
- Henry Jakubowski (College of Saint Benedict/ Saint John's University) focuses on the Biological Chemistry.
- Development of first year General chemistry content is extends over all development teams.
Construction efforts in each team involve range diverse activities including the integrating of available openly licensed content (“Harvesting”) into the platform and the construction of new content to fill in gaps. It is through the efforts of numerous contributors and developers that enable these teams move forward; unfortunately, it difficult to properly acknowledge everyone here - attribution is given on each page on the LibreTexts.
General Chemistry: As with the greater LibreTexts project, it is our belief that one-size does not fit all when it comes to textbooks and that it is important to have a range of options available for faculty to customize their own textbooks subject to the nature of the instructor, students, and campus (Figure 2). Hence, the general chemistry focus involves building and augmenting a three-level suite of general Chemistry texts: basic-Introductory, University-level, and University-honors-level.
The basic-Introductory level is reasonably handled by existing OpenStax textbooks although with some editing. As part of our effort, we harvested the OpenStax chemistry textbook (https://chem.libretexts.org/Bookshelves/General_Chemistry/Book%3A_Chemistry_(OpenSTAX)) and augmented it by adding easily editable latex equations (with other editing activities). We have also harvested their exercises into our homework system (see below) for use as formative or summative assessments.
The University-level texts have focus on the developed of building three Textmaps (i.e., openly licensed surrogates that follow commercial textbook pedagogy and organization. We have a working versions of Brown, LeMay et al. Textmap and both of Tro’s texts (Atoms first and traditional organization): https://chem.libretexts.org/Bookshelves/General_Chemistry
The University Honors-level text is less developed at this stage and focuses on building the Textmap for Oxtoby et al. General Chemistry text (editions circa 2000). This effort has focused primarily on the honing the middle third of the text that focuses on thermodynamics, equilibria, and intermolecular interactions. This portion of the text has been used in UCD five times:
https://chem.libretexts.org/Courses/University_of_California_Davis/UCD_Chem_002BH. Development of the other parts of the text will be completed next year.
While not part of the ACS curriculum, secondary efforts have also been focused on building and improving General, Organic, and Biological (GOB) chemistry and pre-general chemistry Introductory and Contextual chemistry courses: https://chem.libretexts.org/Bookshelves/Introductory_Chemistry.
Organic Chemistry: Building viable organic chemistry texts has been a difficult task in part due to the richness of the field and the varied pedagogical approaches used in the community. The focus of this team has been in building a Textmap around McMurry’s Organic Chemistry text: https://chem.libretexts.org/Bookshelves/Organic_Chemistry/Map%3A_Organic_Chemistry_(McMurry)
To date, a working bare-bones framework has been created and is currently in use in several campuses including Athabasca University (Canada). As part of the development effort, we will revamp all molecules into new ChemDraw 2D images with optional pull ups of the 3D structure (E.g., via JSMol or GLMol), modern cheminformatics (via PubChem) and semantic accessibility for the visually impaired (via ProgressiveAccess- https://progressiveaccess.com/chemistry/).
Inorganic Chemistry: The focus of this team has been on building the Textmap for Miessler, Fischer, Tarr’s Inorganic Chemistry text: https://chem.libretexts.org/Bookshelves/Inorganic_Chemistry/Map%3A_Inorganic_Chemistry_Miessler_Fischer_Tarr). A bare-bones skeleton has been constructed and has been used in several campuses including Saint Mary’s College and UCD.
Physical Chemistry: The text that is used as templates for physical chemistry content is McQuarrie and Simon's “Physical Chemistry: A Molecular Approach” (https://chem.libretexts.org/Bookshelves/Physical_and_Theoretical_Chemistry_Textbook_Maps/Map%3A_Physical_Chemistry_(McQuarrie_and_Simon)). The efforts in building a textmap alternative has focused primarily on the first 10 chapters (Quantum mechanics) and has been used in seven introductory physical chemistry courses at UC Davis to date. Much of this content was based on the openly licensed Introduction to Quantum Mechanics text by David Hanson, Theresa Julia Zielinski, Erica Harvey, and Robert Sweeney (J. Chem. Educ. 2005, 82, 12, 1880).
Analytical Chemistry: This team has focused on building/augmenting three texts: (1) Harvey’s Analytical Chemistry text, (2) a new Chemometrics and (3) an Instrumental Analysis Textmap.
Harvey’s Analytical Chem text has been harvested and updated since last year: https://chem.libretexts.org/Bookshelves/Analytical_Chemistry/Book%3A_Analytical_Chemistry_2.1_(Harvey). A goal is to augment exercises with embeddable R code for statistical analysis (see below). The Chemometrics book is still in development and will be released soon. The Instrumental Analysis text will be a Textmap along Skoog’s ubiquitous text and has only the just started: https://chem.libretexts.org/Bookshelves/Analytical_Chemistry/Map%3A_Principles_of_Instrumental_Analysis_(Skoog_et_al.). To facilitate its construction, the Analytical Sciences Digital Library has recently been harvested, which will provide valuable resources in this building effort.
Biological Chemistry: The focus of this team has been on building the Textmap for Lehninger’s Biochemistry textbook: https://bio.libretexts.org/Under_Construction/Map%3A_Principles_of_Biochemistry_(Lehninger). The magnitude of scope and complexity in building a biochemistry book at this level has made development of this text a slow effort. Fortunately, much of the content from Jakubowski’s Biochemistry Online! text can be repurposed for this construction effort. A key aspect in building this text is to introduce NCBI's (National Center for Biotechnology Information) iCn3D, a biomacromolecule structure visualization program into the textbook for enabling interactive structures. In addition, the online text contains interactive mathematical graphs for binding and kinetic equations that allow the user to move sliders to change values of constants and immediately see the effects on the graph. We are also working on revamping all images into SVG format with embedded alt-texts for accessibility requirements and hyperlinks (e.g., to Pubchem).
Advanced Technology
Learning complex 3-Dimensional concepts from paper-based textbooks can be difficult for students. Seeing concepts “come to life” on the computer helps students to better form their understanding (Clements, 2000) and gives them a concrete visual framework to which they can connect the more abstract properties and concepts. LibreTexts currently exposes students to text, mathematical equations, videos, figures, code snippets, and charts, all of which are fundamentally non-interactive in nature (e.g., https://chem.libretexts.org/Bookshelves/Ancillary_Materials/Interactive_Applications).
Paul Seeburger (Monroe Community College) is enhancing LibreTexts pages by integrating the CalcPlot3D JavaScript app developed as part of NSF project (DUE-IUSE #1524968). This visualization tool was designed to allow students and faculty to visually explore and “play” with concepts from multivariable calculus and differential equations and will extended to visually explore concepts in other Libraries (e.g., engineering, physics, chemistry). For example, it was recently used for aiding students to visualize 3D wavefunctions for quantum chemistry courses (Figure 3), CalcPlot3D also allows the user to create STL files for surface plots that can be used to print these surfaces on a 3D printer, providing an even more hands-on way for students to study and master these concepts. Generating these 3D objects offers an extremely important aid for students who are blind or have low sight.
Figure 3: Probability Wave Function using CalcPlot3D: https://chem.libretexts.org/Bookshelves/Ancillary_Materials/Interactive_Applications/CalcPlot3D_Interactive_Figures/Chemistry_Wave_Functions_on_a_2D_Box/Probability_Wave_Function_-_Linked.
While the CalcPlot3D application provides simple-to-create, specialized 3D visualizations of mathematical functions, many LibreTexts authors also desire arbitrary generated text and visualizations illustrating concepts in any scientific domain. It is known that quality figures enhance learning when adjunct to text regardless if the figures are static, dynamic, or interactive (Mayer 2002, Carney and Levin 2002), but research demonstrating whether interactivity improves learning is nascent. In computer science education, there is evidence that interactive visualizations of computer algorithms improve learning when they are self-paced and of high quality (Saraiya et al. 2004, Amerishi et al. 2008, Guo 2013) but may hamper learning if poorly designed (Schnotz 1999). This enables LibreTexts to become a platform for research into the benefits of interactive figures in addition to providing authors with infinite interactive visualization possibilities to convey concepts (Figure 4).
Jason Moore (UC Davis and University of Delft) has implemented the general functionality for authors to easily include any type of interactive source code in a LibreTexts page which can then be executed by the reader to display both static and interactive in-browser visualizations using open source Project Jupyter (jupyter.org) technologies: JupyterLab, Binder, ThebeLab, ipywidgets, etc.
Figure 4: An example of this capability is an interactive Particle in an Infinite Potential Box page written by Valaria Kleiman and co-workers (JCE 2019, 96, 8, 1663–1670).
Jupyter is a popular open source web application that allows users to create and share interactive documents that contain equations, visualizations, narrative text, and the execution of code (Ragan-Kelley, et. al 2014). LibreTexts authors are able to write high level code in a variety of programing languages (e.g. Python, R, Octave, Sage, etc) in the LibreTexts editing interface. The visualizations will be generated on the fly with the requested cloud services and instantly embedded in pages This system will support any visualization that can be generated using thousands of available scientific libraries. Developers can substitute existing code and figures in the Libraries with new interactive versions.
Integrated Homework System
Over the past decade, several online homework systems have been developed, often as commercial extensions of, or supplements to, existing paper-based textbooks (e.g., Mastering Chemistry, OWL, WebAssign and QBank, ALEKS, and ARIS). To address this important educational aspect within the extended LibreTexts Project, we started to build our own homework system titled: ADAPT. Principle programming of ADAPT was spearheaded by Eric Kean (Western Washington University) and Henry Agnew (University of California Davis).
The system is under active development and has been used with success in a Quantum Mechanics course at UCD this past Fall (Figure 5). The approach follows a similar centralization approach used in the greater LibreTexts project with a general repurposing of existing content or technologies in this case. The ADAPT system currently integrated three existing technologies: H5P, IMathAs (used in MyOpenMath and Lumens Ohm), and Webwork within a centralized system. This development scheme hides the differences in how each technology operates from the faculty and students and provides an integrated approach (versus a piece-wise system) (Figure 6).
Figure 5: ADAPT assignment screen for Chem 110A Introductory Quantum mechanics at UCD.
Currently over 100,000 problems in all three technologies have been integrated into a central gallery (https://query.libretexts.org/). While we are still building the search capability, simple search of chemistry topic (and especially math) will result in multiple questions being found. Any of these problems can be then imported into the homework system which has its own gradebook and authentication system: https://ADAPT.LibreTexts.org. The ADAPT system has basic LTI capability for integration into Learning Management System (LMS).
The ADAPT system is built to support three types of questions:
- Open Ended (requiring human grading)
- Online graded (using one of our existing technologies)
- Learning Trees (adaptive learning infrastructure
There is overlap between these types though. For example, the Learning Trees work best with real-time Online Grading and Mol file or excel files of work can be uploaded for real time grading (e.g., using cheminformatics to grade molecular modeling exercises). The flexibility of the system allows for the construction of Frankenstein problems that easily couple two dissimilar technologies like combining PhET, ChemCollective labs, Concord Consortium or JSMol molecules questions with assessments. This provides an easy mechanism for building the questions for organic and inorganic chemistry texts.
Figure 6: Adapt delivering an open-ended question. Students upload their answer, although this question also exists a real-time online graded version. Student statistics are shown for the problem.
We are at the infancy of the ADAPT project and are excited about the prospects. More information about this project can be gleaned from these YouTube videos: https://youtu.be/LU0lxFE4hpU and https://youtu.be/HIl2WAt9aRQ.
Concluding Comments and a Community Call
The 13-year effort in building the LibreTexts has resulted in the most popular OER textbook platform on the internet today. This is driven by the success of the Chemistry library, which is the successor to the ChemWiki project. A major thrust of our current construction effort is to build the textbooks to support an ACS approved BS curriculum, which entails the construction of books for 15+ courses. Furthermore, these “texts of the future” are being augmented with advanced visualization and computational features including CalcPlot3D, juytper code, ADAPT homework system and much more.
The utility for the project is the direct result of thousands of contributors and developers (both faculty and students) that have work with content created by thousands of other content authors. Because our efforts are community based, we welcome the donation from anyone of existing content for harvesting, or new volunteers to aid in construction efforts. Moreover, we welcome feedback on our efforts and the content hosted on our pages. This is the power of hosting living content enabled by our technology.
References:
- US Government Accountability Office. (2005). College Textbooks: Enhanced Offerings Appear to Drive Recent Price Increases (No. GAO-05-806). Retrieved from http://www.gao.gov/new.items/d05806.pdf
- College Board. (2013). Trends in Student Aid. (College Board, Ed.).
- Mayer, R. E. (2002). Multimedia learning. Psychology of Learning and Motivation, 41, 85–139.
- Saraiya, P., Shaffer, C. A., McCrickard, D. S., & North, C. (2004). Effective Features of Algorithm Visualizations. In Proceedings of the 35th SIGCSE Technical Symposium on Computer Science Education (pp. 382–386). New York, NY, USA: ACM. https://doi.org/10.1145/971300.971432
- Schnotz, W., Böckheler, J., & Grzondziel, H. (1999). Individual and co-operative learning with interactive animated pictures. European Journal of Psychology of Education, 14(2), 245–265.
- Ragan-Kelley, M., Perez, F., Granger, B., Kluyver, T., Ivanov, P., Frederic, J., & Bussonnier, M. (2014). The Jupyter/IPython architecture: a unified view of computational research, from interactive exploration to communication and publication. AGU Fall Meeting Abstracts, 44. Retrieved from http://adsabs.harvard.edu/abs/2014AGUFM.H44D..07R
Comments
Instructional Design
Dr. Larsen,
On a text like the LibreText for University-level Gen Chem, how many authors would be involved in writing a chapter? How many in the text overall?
In terms of the instructional design of that text, are writers told to assume that students solve end-of-chapter-type problems by reasoning from concepts presented in the text, or by recall of algorithms taught in the worked examples?
Are there any agreed-to rules on how the explanations in the worked example algorithms will be structured?
Pedagogy
I will be a bit less specific than you may want. The answer is as many authors as needed to get the chapter or book up to speed. I never consider a text to be complete, concepts are constantly developed, new technologies and new pedagogies implemented. Hence, the content in the LibreTexts is designed as living content to be curated by the community.
We don't force a specific pedagogy or organization design onto the library as whole. We favor to allow different "flavors" of texts to be available and let faculty select/customize their textbook based on the goals and realities of the faculty/student/department/campus and of stakeholders. We have been working on a framework to help centralize different approach, but we are at the infancy of that thrust.
As for ground-rules of example algorithms: no. As long as content is correct, it meets our bar. I have been conflicted over the years in not pushing my personal preferences onto other faculty's approaches. What works well for my students may not work well for others (e.g., I am in strong support of teaching activities to my students). My hopes are that we can flesh out a community forum to guide faculty away from less-supported outdated approaches and provide pages that faculty can easily adopt/edit to pursue this. That is the power of a centralized, consistently formatted warehouse of content like the LibreTexts. Building this community forum infrastructure is a goal this year.
Web-based pedagogy
I am not directly involved with LibreTexts, but I do use it as one of my reference materials. However, your question about how problem-solving is presented caused me to reflect, because BestChoice (http://www.bestchoice.net.nz) is all about systematic development of content through integrating information pages and practice questions.
From the start I was keen on emphasising strategies for problem-solving so for multistep stuff like stoichiometry. Thus I put a big emphasis on planning - that is getting them to construct a plan for solving the problem before doing the calculations. At first this was done with minimal or very general information on the page introducing that activity, and the students were not very happy about that approach - it might have been them - it might have been the way I presented it as this was early days. So then I put general plans on the preceding information, not worked examples. Then through answering the BestChoice question pages, students construct their own worked examples including planning and calculations as in the concentration activity below (sorry about the complicated link - I seem to keep losing ones that I modify.)
https://www.bestchoice.net.nz/?i=0&s=1&c=25&cs=17070&p=30192
Looking through my stoichiometry sections, I mostly have a general method (algorithm) on the preceding information pages. Integrated with this is often an example of what those words mean, but the following problems use the principles in that example and are not exactly the same type. My tendency is to use a minimalist approach and then go by the data as to what is needed. If the preliminaries are not sufficient support for the following questions, they are altered, inserting a small example of what the words mean.
For non-quantitative areas, the principles precede the questions with no worked examples. So I guess I do a mixture of the two, again driven by the data.
I have recently started doing something different which I think has merit and encourages students to read the principles pages, namely, inserting interactions in these where students choose answers to build the small examples that I often used on these pages. The answers to these are all pretty obvious. That seems to work quite well. The key is developing and modifying pedagogy is to have a system which lets your mind go free when constructing interactive question pages - one size does not fit all. It is lots of fun - I have learned so much.
accessibility
Hi Delmar,
I hope all is well. Are all of the LibreTexts accessible for the visually impaired? Do all of the images have alternate text that that describes them, including all of the equations and unit analysis setups? Is the reading order clear for the electronic readers?
Mark Bishop
These are very important
These are very important questions. The answer is not to the level that we want, but we have an active effort to revamp much of these things. Part of the slowdown is that content comes from multiple sources, each with unique code and accessibility standards (if at all). We have a new bot system that goes through to resync headings, add aria landmarks and other things that a non-human can do. Alt-texts take more time and we are slowly moving forward on that.
I mentioned in the paper that we are revamping the molecular images to be standardized (with chemdraw) that we want to couple to JSMol (of GLmol) for 3D viewing and coupled to ProgressiveAccess for semantic screen reading.
We have just updated the color schemes to handle all the contrast requirements for WCASG2.1. There many other aspects that we are either actively pursuing or the bot handles already.
Learning Trees and Adaptive Learning
Hi Delmar,
Thank you for providing us with an update on LibreText, which I use extensively, and I am always amazed at how your are moving forward and taking on new challenges. Are there any existing Learning Trees that we can look at to get a feel for how you have used them? I realize this is under development, but I am really excited about this. And could you possibly give us a little bit on your vision on where these will be going?
I mean for example, you describe general chemistry as sort of three tracks, prep, gen chem and honor's chem. Some schools are small and do not have the option for an honor's class, would that be possible through a learning tree approach, that forked into honor's content for students who could handle it?
I look forward to trying these features and thank you again for your time and contribution to the Newsletter.
Bob
The learning trees are still
The learning trees are still new and while we have many constructed, they are not formally presentable. Many of these were taken over from SmartSparrow's adaptive learning platform that we are essentially recreating (minus the fancy graphics). The gist is to provide a curatable set of tree (a forest) that faculty can pick from like the the QUERY gallery, but with a root assessment and then multiple branches addressing specific learning objectives for the project and then multiple remedaition nodes with secondary assessments. We want the utility of cross-linking these trees for a comphrehensive overview of content and the utility of customization at the tree level. We want to avoid the sort of infrastructure that ALEKS has built and similarly without the financial profit component.
We are working on ussing the ACCM as one of several frameworks, including one based on learning objectives, to tie the content of the trees together and for easy searching. Included in this alignmnet efforts will be tagging for academic level. We can use machine learning to check if thes alignments are correct (second year outcome). So, we could build a progressive system that delivered harder questions to students that require/desire it (like hte modern SAT), subject to faculty and student consent.
Choice of trees to climb
Thank you for sharing your description of your group's work. The scope is amazing. There are is a huge family of students who stand to benefit from it. The flexibility of the model sounds like it can accommodate virtually anyone at any skill level. Clearly the learning tree inventory is a work in progress. it may evolve perpetually. The overall project description lists the classical instructional domains for chemistry. Are all of them at the same level of learning tree development? Which ones are more mature? For the more complete forests do users have much choice while exploring the forest that fits their needs?
The learning tree
The learning tree infrastructure is available for anyone to build, as we grow the assessment nodes and remediation nodes, making new trees to target specific audiences will be easier. Our California Education Learning Lab grant is focusing on general chemistry, albeit at the community college, four-year and University levels. I intend to focus on learning trees for my pchem class this summer. My primary concern in this endeavour is not building the trees (e.g., https://youtu.be/LU0lxFE4hpU?list=PL83Q_gTbFatRwgbrfjbAx6Ge2pWjziLxs&t=3416) or the nodes as part of the trees. I am most concerned about which topology(ies) is(are) effective and how to guide students along the tree.
Naturally, if anyone has a desire to get involved in this effort, either in building, adopting or evaluating the trees, please contact us. This is a big project and the more people involve in it, the faster it will evolve beyong the testing phase (although I am using ADAPT already in my classes).
How can people get involved?
Delmar,
Having an entire undergraduate major supported by OER would be an impressive milestone. There is a significant need for OER that works for upper-level courses.
I learned how to remix texts on Libretexts through a workshop this summer ("Librefest 2020"), and subsequently attended the OpenEd 2020 conference to showcase some of the work. Inspired by one of the OpenEd 2020 talks, our institution is hosting a local OER faculty cohort next semester to support our faculty in adopting, remixing or creating OER. Some of the applicants (from biology and music) have fairly mature plans of using multiple open textbooks, some of which are on Libretexts and others which aren't. They are looking for a platform where they can tie the different resources together into a coherent course resource.
My question is whether training session for the spring semester are already scheduled. Or is there a possibility to get a sandbox to start experimenting without the formal training (I'd be happy to assist on a local level as far as I can)?
Karsten
Westfield State University
Oops, I should have mentioned
Oops, I should have mentioned that. Accounts are freely available via this form:
https://tinyurl.com/Register4Libre
We just need a little verification to be more streamlined in our activities. This account gives a sandbox that faculty can play with constructing content or customizing existing content. The account also provide a mechanism to submit page specific feedback (at bottom of the page); although the unverified community accounts do that too. This aspect enables what I refer to as a "passive curation" or "community curation" where issues/errors can be submitted and are often addressed same day. I use this as extra credit for my classes as it encurouages students to careful read the material.
We have a LibreFest planned for the middle of January. It is a campus specific workshop for St. Mary's College in Notre Dame (we provide these of our LibreNet member campuses), but I bet there is space for people to jump in. I think it will be Jan. 13-15 for two hours a day. There will be general LibreFest planned later Winter/early Spring and another in July as part of a California center OER meeting that will be announced soon. Naturally, if a critical mass is formed, we can do an impromptu workshop.
Learning trees
The concept of learning trees where students are provided with remedial material has been around for a while. It certainly has merit for practice problems, and I am pleased to see data is being collected. I presume that the first question in a chemistry learning tree would be one that may involve multiple steps and the remediation would then provide support for each of these steps. The places where I have seen this being used generates separate areas for each of the remediation bits, and loses the flow of the initial overall problem. Would the remediation stuff be presented in separate windows? Is there any possibility for direct feedback (for example, under the multichoice answers)?
Additionally, again from my experience - students are amazingly linear. The BestChoice tree menu is always there in the left hand panel, and labels and tables of contents are given for each activity. There is very little skipping around. They start at either the top of the menu or the first of a group of activities and work their way down. Even for activities which are multichoice quizzes (so no obvious order), they start with number 1 and proceed to number whatever which is the last one.
Learning Trees
<p>We are flexible at this stage and I don't want to commit to a specific formulation of the root assessment. I think there are merits in different approaches , which will vary depending at what level the tree is supposed to address. We intend to provide the option of feedback at different levels, but it is not in the existing system (we will be actively building for the next 18 months though).</p><p>As for navigation. We have a system in place that let's students choose their path. This is one scenerio to build agency in students, although we have plans to provide direction based on statistics of past students. Can you share the screenshot of your page organization that you described? I am unclear about it</p>
Remediation in trees
This was in a commerical package that I was reviewing years ago when I first started my own work. But basically it had a questions like "Hydrogen reacts with chlorine to give hydrogen chloride. What is the mass in grams of hydrogen chloride from 25 g chlorine?" They they typed in the answer. If they got it wrong, then they were asked something like "What is the amount in moles of HCl produced from 25 g Cl2?" This was a separate block of screen - I can't remember, but it might have even covered the first one. If they got it wrong, they kept going backwards and you ended up have well more than a screenful of stuff with no obvious connections between these.
In my view this working backwards from the correct answer approach is ok for assessment, but is not so good for teaching or introducing strategies for problem-solving - not alogorithms but stuff like identifying the unknown and its units, in cases where a formula is involving, giving the unknown and known symbols, identifying a pathway between the known and the unknown. Starting with scaffolding and then gradually taking it away. I think that is what you are worrying about with how to guide them through.
An example of a scaffolded problem that approaches the task from a problem-solving instead of an answer perspective.
Then after choosing the 15 things that you have to think about (with opportunities to remediate mistakes on the way), an annotated model answer results. The display of the blocks for the steps depends on the width of the screen.
Linearity or leap frogging
I have seen the same tendencies among my students. Most typically follow the content sequentially. It would be surprising if they didn't That is what they have done during years of education.
The Science of Learning Chemistry
Dr. JudithAnn Hartman and I just finished a draft of a paper on “Working Memory Limits and Chemistry Instruction” which may be of interest to those who write for textbooks or their own instructional materials. It is an 18 page review of how cognitive neuroscientists and cognitive psychologists say students learn physical sciences: what science says the student brain can and cannot do.
A preprint is posted at www.ChemReview.Net/TheScienceOfLearningChemistry.pdf
An email address is included at the top for critical comments (which the authors would appreciate).
What we found was, according to the consensus of cognitive experts, students in introductory courses for science majors can only solve the kind of problems assigned at the end of the chapter in most textbooks by applying memorized algorithms. The most efficient and effective algorithms at each step rely on applying well-memorized facts.
This is, I think, not what anyone wanted to hear, but science is like that.
Cognitive experts say conceptual understanding is the right goal, but it takes quite a bit of thorough memorization to achieve.
And, even if concepts are understood, because of the brain’s working memory limits, to solve problems of any complexity, students still must memorize the fundamental facts and algorithms that instructors suggest they “overlearn:” memorize to perfection, repeatedly.
Chem ed journals often deprecate memorization and algorithmic problem solving, but on questions of how the brain works, I think it might be wise to defer to those whose scientific expertise is the study of how the brain works.
For textbook writers and instructors, these findings mean a key question for chemistry instructors to research mightbe: Overall and for each topic, what are the best problem-solving algorithms?
- rick nelson