1. David Yaron, Donovan Lange, Michael Karabinos, Rea Freeland and D. Jeff Milton Department of Chemistry, Carnegie Mellon University Gaea Leinhardt and Karen Evans Learning Research and Development Center, University of Pittsburgh Funded by the National Science Foundation (CCLI and NSDL Programs) and the Howard Hughes Medical Institute Virtual Chemistry Laboratory

2. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Goals Shift current chemistry courses so that they –Promote inquiry –Promote chemical literacy Methods –Homework for college classes and capstone activities for high school classes –Community building and support activities

3. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Learning challenges and interventions Promoting flexibility and applicability –From mathematical procedures to chemical phenomena ( use in chemistry ) Virtual laboratory – From chemical phenomena to real world ( transfer to real world ) Scenario based learning Promoting coherence –Scenarios that touch down at various points in the course

4. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Use in chemistry: Virtual laboratory Flexible simulation of aqueous chemistry New mode of interaction with chemical concepts Ability to “see” inside a solution removes one level of indirection in chemical problem solving

5. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Authoring a virtual lab activity Add chemical species and reactions (if desired) –Can create “fictional” proteins, drugs etc. Create Stockroom Solutions Specify available functionality –Viewers For example, turn off “Solution Contents” for exercises involving unknowns –Transfer mode Precise: student enters exact amount to transfer –Facilitates comparison with paper and pencil problems Realistic: simulates accuracy attainable in real lab –Forces student to use correct apparatus (buret for titration) HTML problem description can be included Of 35 current problems, 15 are by community of users

6. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Virtual lab problem types Check paper-and-pencil work –Encourages students to see connection between calculations and an experimental procedure –Provides intermediate results Virtual experiments “Measure the free-energy of binding between the DNA base pairs” –Requires experimental design –Involves use of concepts to achieve goals  G o related to equilibrium constant K –Relates material to an interesting system

7. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Goal oriented problems Typical textbook problem “What is the pH of a buffer made by mixing 10ml of 0.10M HAc with 12ml of 0.10M NaAc?” –Textbook author makes the interesting choices –Student just does the calculation; leads to a narrow focus on skills Virtual lab (goal oriented) problem “The stockroom contains a cyanine dye that binds to DNA when it is protonated. Create a solution in which 50% of the dye will be bound to DNA.” –Involves mix of experimental design and paper-and-pencil calculations –Realistic feedback (Did you achieve your goal?)

8. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Puzzle problems “The stockroom contains 1M solutions of A, B, C and D. What is the reaction between these chemical species?” Possible because the virtual lab allows students to rapidly do experiments and to see “inside” solutions. Represents a new mode of interaction with the material.

9. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Observational studies with virtual lab Situation –30-35 students working alone or in pairs –2-3 instructors Outcomes –Can give insight into student difficulties and nature of their qualitative understanding –Unanticipated error types provide opportunities for instructional design

10. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Observational studies: Student difficulties Students who could perform the textbook procedure on exams, had difficulty designing the experiment, and needed help from human tutor. The procedure is apparently not triggered in response to relevant prompt. Virtual lab activities may reinforce activation. Typical textbook problem “When 10ml of 1M A was mixed with 10ml of 1M B, the temperature went up by 10 degrees. What is the heat of the reaction between A and B?” Virtual lab problem “Construct an experiment to measure the heat of reaction between A and B?”

11. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Observational studies: Unanticipated errors “The virtual lab contains 1M solutions of A, B, C, and D. Construct experiments to determine the reaction between these reagents” Intent was to give practice with determining reaction coefficients A + 2B  3C + D Observation When A is mixed with B, some A remains, so the reaction must be: A + B  C + D + A Reveals misunderstanding of limiting reagent concept (even though they could easily perform textbook limiting reagent problems) This type of mistake is good opportunity for Elicit-Confront-Resolve instructional strategy

12. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Observational studies: Qualitative reasoning Challenge problem involving multiple interacting chemical equilibria (“a weak acid dye binding to DNA”) Student strategies were much more sophisticated than instructors anticipated based on students ability to manipulate these concepts algebraically Suggests the new manipulatives of the virtual lab have potential for supporting and assessing qualitative understanding

13. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Transfer to real world: Scenarios Scenario based learning –Embed the procedural knowledge of the course in a scenario that highlights its utility –Scenarios that touch down at various points in the course may promote coherence –Examples: forensics, biological and medicinal chemistry, environmental chemistry, space exploration/colonization Outcome of design process –Attempt to organize scenario development lead to a “concept map” of the domain

14. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Scenarios: Examples Mixed reception –Murder mystery activity for first few weeks of high-school/college chemistry –Illustrates that contextualization can be done early in course Arsenic poisoning of wells in Bangladesh –Scenario that touches down at various points in the course (stoichiometry, titration, spectroscopy)

15. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Scenarios: Design process Concept map of chemistry benchmarked against –CA state content standards - Nobel prizes –Best selling textbooks - NY Times/Scientific American Reveals flaw in current course structure –Top level: Three subdomains in which chemists work Scientific literature spread equally between these three subdomains –Lower levels: Toolbox Textbooks and standards found only in Toolbox and Analyze subdomain

16. EXPLAINANALYZE SYNTHESIZE Hypothesis Generation Hypothesis Testing Goal (What do you want to know?) Process (How to determine What you have) Functional Motifs Structural Motifs Assembly Motifs TOOLBOX Representational Systems Quantification Systems Chemistry Concept Map

17. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Challenges for active content in digital libraries Simulation and visualization tools often require a flexible development environment such as JAVA Evidence of a problem –Thousands of applets are available on the web –Indicates nascent developer community –But most applets are used only by team that developed them Root of the problem –Current development approach puts too much of the process in the hands of programmers rather than educators

18. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Student interface as a dividing line Programmers develop components –Produce materials for use by instructors and curriculum developers –Takes advantage of their ability to produce interactive, domain- specific learning objects Curriculum designers provide student interface –Provide student interface, guidance and scaffolding –Takes advantage of classroom and pedagogical expertise

19. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Configuration as authoring A component that save its state to a file serves as a domain-specific “authoring tool”. Configuration file specifies –Available chemicals, instruments, and visualization tools –User interface features Other examples: Physletts (Davidson), Interactive Physics 2000, VGEE earth science project (UCAR) Matlab, Mathematica, and SPSS can also be viewed as using “configuration as authoring”

20. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Linked active content Allow assembly of multiple components each of which is configurable and saves its state to a file Image maps with hotspots that send messages to the simulation Simulation

21. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon CREATE digital library architecture simulation Image map 1 Image map 2 Software Components Linked Active Content start reset File that specifies number of planets to show, scale etc. Planets.xml File containing image and links to start and reset simulation. Control_panel1.xml File containing image and links to set fuel parameters. Control_panel2.xml Content files and software are stored separately in the digital library collection

22. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Mixed reception

23. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Promoting reuse and maintenance Reuse –CreateStudio allows you to step through an activity in preview mode to where you want to make a change –Switch to edit mode to make a change and the –Save modified content back to library Maintenance benefits of separating content from software –Can update viewers without needing to change content –Can tag and search content files Supports iterative approach to development

24. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon CreateStudio as an authoring portal for the NSDL Component (viewer) discovery –Search NSDL for viewers –Search for activities that use a particular viewer (for examples) Content discovery –Search for movies when using a movie player –Search for virtual lab configurations Publish into the library Extract from library, edit, and republish into the library

25. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Current dissemination strategies Web site (http://ir.chem.cmu.edu/)http://ir.chem.cmu.edu/ –1000 page requests per day, 80 instructors on mailing list, 36 requests to become test sites in past year –>7000 students have performed one or more activity in the virtual lab Booths at conferences –Demonstrate materials for about 75 instructors per day of 3 to 4 day conference

26. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Community partners Most active sites –University of British Columbia (4200 students over 3 semesters) –Florida Atlantic University (1500 students over 2 semesters) –University of West Virginia (150 students over 2 semesters) All three sites recruited at conferences, with no prior affiliation Sites attracted by ability to author/customize –UBC, FAU: ability to customize to existing physical lab program –UWV: enabling technology for own curriculum development Of 35 activities on web site, 15 were developed by user community

27. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Using DLs to build educational communities Community must share a specific educational goal DLs can combine expertise through remote and asynchronous collaboration –Learning technology: As with the CREATE architecture –Learning science: Design of components (virtual lab), DL organization (concept map), and assessment tools (instruments and tracing technologies) –Domain/classroom experience: By having teachers author material, can shortcut (develop-assess-disseminate) cycle DLs can support an iterative development process –Carnegie Mellon’s OLI is developing technology to collect data online and use this to inform the development process

28. Chemistry Virtual Laboratory http://ir.chem.cmu.edu/ Carnegie Mellon Credits: Programmers / Writers/ Educational Research Carnegie Mellon Donovan Lange D. Jeff Milton Michael Karabinos Rea Freeland Giancarlo Dossi Katie Chang Emma Rehm Erin Fried Jason Chalecki Greg Hamlin Brendt Thomas Stephen Ulrich Jason McKesson Aaron Rockoff Jon Sung Jean Vettel Rohith Ashok Joshua Horan LRDC, University of Pittsburgh Gaea Leinhardt Karen Evans