The Learning Assistant model for Teacher Education in Science and Technology
The Learning Assistant model [1, 2], at the University of Colorado at Boulder uses course transformation as a mechanism to achieve three related goals: (1) to recruit and improve the preparation of future mathematics and science teachers, (2) to improve the education of all students enrolled in our mathematics and science courses, and (3) to engage science faculty more thoroughly in the preparation of future teachers.
The Learning Assistant model was initially developed as a part of the STEM Colorado project headed by Richard McCray in response to studies that demonstrate that a majority of our nation's youth are not performing proficiently in mathematics and science [3, 4] that many of our teachers, especially in the physical sciences, are under prepared--having neither a major nor minor in their field,  and that large research universities are not producing adequate numbers of mathematics and science teachers.
At large research universities, few mathematics and science majors pursue careers in K-12 teaching. Those who do, typically learn about pedagogy only after they have completed most of their content courses. There is generally little or no interaction between disciplinary faculty and education faculty and between undergraduate programs designed to teach content to students and programs designed to help future teachers learn to teach that content. We regard this disconnect between disciplinary and education programs as a missed opportunity to both improve the effectiveness of undergraduate mathematics and science education and to recruit and prepare mathematics and science K-12 teachers.
The Learning Assistant Model
The Learning Assistant model is based on the premise that teacher preparation begins in the College of Arts and Sciences, where students begin their content preparation. In order to explicitly help undergraduate students integrate their content learning with their understandings of how content is learned and to encourage talented students to become teachers, we needed to establish a close collaboration between faculty members from the School of Education and faculty members from content-based departments. This collaboration was achieved through the LA model which is designed to couple mathematics and science departments',efforts to transform large-enrollment undergraduate courses with efforts to recruit and prepare talented mathematics and science majors to become K-12 teachers.
The transformation of large-enrollment courses involves creating environments in which students can interact with one another, engage in collaborative problem solving, and articulate and defend their ideas. To accomplish this, faculty members teaching in large-enrollment courses need several assistants to help facilitate small group interaction. Learning Assistants (LAs) fill this role. LAs are talented undergraduate students who are hired to facilitate small group interaction in our large-enrollment courses, and at the same time, they make up the pool from which we recruit new K-12 teachers.
Since the program began in 2003, we have recruited 18 LAs to teacher certification programs, most of whom have reported that they did not consider teaching as a career until participating as LAs. The most common reasons reported for making the decision to become a teacher were recognizing the complexities of teaching and encouragement and support from mathematics and science faculty.
The differences between the LA model at the University of Colorado at Boulder and other standard models for undergraduate teaching assistants are (a) our focus on teacher recruitment and preparation, (b) a special seminar targeted at helping LAs integrate content, pedagogy, and practice, (c) a collaborative educational research program designed to evaluate the effects of the LA model, and (d) the involvement of mathematics and science research faculty in the recruitment and preparation of future teachers.
Course Transformation and the Role of LAs
LAs are paid a modest stipend to work approximately 10 hours per week in various aspects of course transformation. Approximately 60 LAs are hired each semester to work in six mathematics and science departments: Physics; Astrophysical and Planetary Sciences; Molecular, Cellular and Developmental Biology; Geological Sciences; Chemistry; and Applied Mathematics. Specific courses that have been supported by LAs are listed in Table 1.
Table 1. Learning Assistant supported Courses
||APPM 1350: Calculus I for Engineers
APPM 1360: Calculus II for Engineers
APPM 3310: Matrix Methods
APPM 3570: Applied Probability
GEEN 1340: Calculs I (semester 1 of a 2 semester sequence)
GEEN 1345: Calculs I (semester 1 of a 2 semester sequence)
|Astrophysical and Planetary Sciences
||ASTR 1010: Introductory Astronomy
ASTR 1120: General Astronomy Stars and Galaxies
ASTR 2000: Ancient Astronomies
||CHEM 1021: Introductory Chemistry
CHEM 1111: General Chemistry
CHEM 4411: Physical Chemistry/Biochemistry Applications 1
||GEOL1010/1030: Introduction to Geology
|Molecular, Cellular, and Developmental Biology
||MCDB 1111: Biofundamentals
MCDB 1041: Fundamentals of Human Genetics
MCDB 2150: Principles of Genetics
MCDB 4650: Developmental Biology
||PHYS 1010/1020: Physics of Everyday Life I and II
PHYS 1110/1120: General Physics with calculus
PHYS 2130: Modern Physics for Engineers
There is no dictated design of what course transformation should look like. Instead, faculty members who request LAs must (1) use LAs to promote interaction and collaboration among students enrolled in the course, (2) meet in weekly planning sessions with the LAs who support their courses, (3) attend biweekly meetings with other faculty participating in the program, (4) attend a summer session targeted at building a community of university faculty, high school teachers, and future teachers, and (5) actively evaluate transformations and assess learning in their own courses. Because there is little dictation as to exactly what a transformed course should look like, there exist several models of course transformation among our participating departments. For example, one of two models of transformation in the physics department utilizes the University of Washington's Tutorials in Introductory Physics  in recitation sections each headed by one graduate TA and one undergraduate LA. The Tutorials involve conceptually-based group problem-solving activities which are based on research in physics education. LAs who work in Tutorial sessions formatively assess student understanding, ask guiding questions, and facilitate collaboration within groups. These tutorial sessions are supplemented by weekly lectures which are made interactive through infrared response systems and collaborative peer instruction (Mazur, 1996). Average normalized learning gains in these courses, as measured by conceptual instruments such as the Force and Motion Conceptual Evaluation  range from 40% to over 60% , far above the learning gains measured for traditional courses (23%) .
A different model for course transformation is used in the Applied Mathematics department. In weekly LA-led problem-solving sessions, each small group of students uses a 2-ft × 3-ft dry-erase board to collaboratively construct problem solutions. A more radical form of transformation is found in the Astrophysical and Planetary Sciences department where one lecture per week is replaced by Learning Team sessions headed by the LAs. In this model, enrolled students are assigned to one of several learning teams each headed by an LA who facilitates collaboration among groups as they analyze real astronomical data and generate and compare models to fit these data.
Although the LA experience is somewhat different for each course, the experience for all LAs involves three related activities: (1) LAs facilitate collaboration among learning teams by formatively assessing student understanding and asking guiding questions; (2) they meet weekly with their faculty instructor to plan for the upcoming week, reflect on the previous week, and analyze assessment data; and (3) LAs from all departments attend a special Mathematics and Science Education seminar where they reflect on their own teaching and learning and make connections to relevant education literature.
The Mathematics and Science Education Seminar
The Mathematics and Science Education seminar is jointly conducted by a faculty member from the School of Education and a K12 teacher. In this course, new LAs reflect on their own teaching practice, reflect on the transformations of the course in which they are working, investigate relevant educational literature, and engage in in-depth discussions about their own teaching and learning. Seminar readings and discussions include topics such as discussion techniques, learning theory, cooperative learning, student epistemologies, metacognition and argumentation, self-explanations and tutoring, multiple intelligences and differentiated instruction, the nature of science and mathematics, national standards, teaching with technology, and qualities of an effective teacher. Students in this course try out new ideas each week in their learning teams and report their results in seminar. In many cases, LAs provide guidance to one another regarding managing issues that typically arise in their learning teams. Each week, LAs complete online reflections on their teaching and the learning of the students in their learning teams. In addition, throughout the semester LAs turn in two reflective essays that integrate the education literature with their own teaching and learning experiences. LAs often report that by studying and reflecting on student learning, they have become better learners themselves. At the end of each semester, LAs in the seminar present a poster session attended by their lead instructors, School of Education faculty, University of Colorado administrators, graduate students, and their peers. Each LA or small group of LAs present a poster that focuses on aspects of the LA experience that influenced their thinking both as a learner and as a teacher.
Focus on Teacher Recruitment
Although the LA experience (represented in Figure 1) is valuable for undergraduates who continue to any career, our program is specifically designed to actively recruit talented undergraduate students to careers in teaching. Therefore, a student can continue to be an LA for a second semester only if he or she shows commitment to finding out more about teaching. This may be evidenced by taking an education course or participating in an early K-12 field experience. LAs can be hired for a third semester only if they have been accepted to a teacher certification program at which time they are eligible for NSF funded Noyce Teaching Fellowships of up to $10,000 per year.  As Noyce Teaching Fellows, students can become Lead LAs who mentor novice LAs, participate in the development course educational technology, or work with mathematics, science, and education faculty conducting educational research.
Faculty members who use LAs evaluate their own course transformations by systematically investigating student learning in their courses. In some cases this involves the design or modification of assessment instruments to measure students's levels of conceptual understanding of the content of the course. This type of research and development has been conducted by individual faculty members since the beginning of the STEM Colorado program in 2003. However, a coordinated research program to test the effectiveness of the LA model on multiple levels will officially begin in Fall 2006.
The LA-TEST research project
The NSF-funded Learning Assistant model for Teacher Education in Science and Technology (LA-TEST) research project  was designed to test the effectiveness of the LA model specifically in terms of LAs' development of content knowledge, pedagogical knowledge, and their practice in K-12 schools. Faculty members from education, mathematics, and science, K-12 teachers, graduate students, and Noyce Fellows comprise three interacting research teams: the Discipline-Based Educational Research (DBER) team, the Conceptions of Teaching and Learning (CTL) team, and the K-12 team. These interacting research teams investigate teacher recruitment rates as well as the research questions shown in table 2 and synthesize results on an ongoing basis.
Table 2. Research Questions for the LA-TEST project
|DBER: Content Knowledge
||CTL: Pedagogical Knowledge
||K-12: Teaching Practice
|(a) What effects can be observed on student achievement in courses that are supported by LAs?
(b) How do LAs compare to other mathematics and science majors in terms of their content understanding, beliefs about the discipline, and beliefs about learning in the discipline?
|(a) What is the effect of the LA model on the sophistication of LA pedagogical understanding?
(b) Does sophistication of pedagogical understanding vary by length of exposure to the LA model?
(c) How is the pedagogical sophistication of LAs different from the sophistication of non-LAs who become teachers?
|How do teachers and teacher candidates who participated as LAs compare to those who did not in terms of:
(a) Practicum-based coursework
(b) Their teaching practices
(c) Their own studentsҠ(K-12) attitudes and beliefs about mathematics and science (d) Retention and attrition rates
Since the program's inception in 2003, 28 faculty members from 6 mathematics and science departments have used LAs to transform 23 courses; 125 mathematics and science majors have participated as LAs (18 LAs have enrolled in teacher certification programs); and 4 education faculty have been involved in this process. LAs recruited to teacher certification programs have an average cumulative GPA of 3.4, well above the average GPAs for mathematics and science majors at our university. Table 3 compares enrollments in certification programs in the state of Colorado and enrollments at the University of Colorado at Boulder (not including LAs) to the numbers of students that have been recruited to certification programs through the LA program. In table 3, 6 of the 7 mathematics majors recruited through the LA program are from Applied Mathematics, a discipline traditionally underrepresented in our certification programs but specifically targeted through the LA program. The departments of Chemistry and Geological Sciences joined the program in mid-2006, so it is not surprising that the LA program has not yet recruited any students from these majors.
Table 3. Undergraduate students enrolled in teacher certification programs
||State of Colorado* (2004-2005)
||University of Colorado Boulder Not including LAs (2004-2005)
||Recruited from LA Program (2004-2006)
*Data from 18 colleges and universities with 10,869 candidates, 385 science majors
These data are evidence that the LA program has some effect on recruiting mathematics and science majors to teaching careers.
LAs' content knowledge is beyond that of their peers and LAs learn content more deeply trough the LA experience. For example, results from the Brief Electricity and Magnetism Assessment  given to students enrolled in second semester introductory physics show average pre-test score of 27% for enrolled students and an average post-test score of 59%, with an average normalized gain of 0.44. The LAs who had taken the course the prior semester had pre-test score (the beginning of the semester of working as an LA in this course) of 75%, higher than their peers' post-test scores. More interesting is the fact that LAs' average post-test score (at the end of one semester of being an LA) was 90%, with an average normalized gain of 0.56.  Thus, LAs developed their content knowledge as a result of teaching as an LA. Similar results are being found in other LA-supported courses and this is the subject of ongoing research.
LAs tend to view their students in terms of their students' learning processes rather than in terms of whether a student is good or bad or whether they do or do not understand the material. Our studies show that while faculty and teacher candidates tended to view students in terms of the learning process they also viewed their students in terms of whether the students do or do not understand, and in terms of whether they are good, bad, lazy, smart or dumb. Results are summarized in the table 4 below.
Table 4. Views of students by faculty, teacher candidates, and LAs
|View of Students
||Specific statements indicating view
||Faculty (% of codes)
||Candidates (% of codes)
||LAs (% of codes)
||Students are trying to learn, are constructing understandings, must articulate/defend ideas
|Condition of Student
||Students want to learn/do not want learn, get it or they donҴ, have misconceptions
|Property of Student
||Students are smart/dumb, good/medium/bad, have/do not have ability, lazy/do least they can
The results shown in table 4 may indicate that LAs have a greater sensitivity to the struggles of learning mathematics and science among their peers. This may be an indication that there is great value in beginning the teacher preparation process early in students' undergraduate careers rather than waiting until they have decided to become teachers of students who are much younger and more inexperienced than themselves.
The LA model integrates the development of content knowledge, pedagogical knowledge, and practice for all participants by beginning the teacher preparation process early in students' undergraduate careers and by involving mathematics and science faculty in this process. Although we recruit approximately 15% of the LAs who participate in this program to teacher certification programs, the experience is valuable for students who move on to any career. The participation of mathematics and science research faculty in the active recruitment of teachers has led to departmental cultures that encourage rather than discourage teaching as a legitimate and valuable career option for our most talented mathematics and science students.
1. Supported in part by National Science Foundation Grant DUE-0302134.
2. Supported in part by the PhysTEC program of the American Physical Society, AIP, and AAPT.
3. National Center for Education Statistics (2003). The Nation's Report Card: Mathematics Highlights 2003. NCES. 2004-451 Washington,.DC: U.S. Department of Education.
4. National Center for Education Statistics (2003). The Nation's Report Card: Science 2000. NCES 2003-453. Washington DC: U.S. Department of Education.
5. Neuschatz, M. and McFarling, M. (2003). Broadening the Base: High School Physics Education at the Turn of the New Century. American Institute of Physics, College Park, MD.
6. National Science Board (2006). Science and Engineering Indicators 2006. National Science Foundation, Arlington, VA: 2006, volume 1, NSB 06-01; volume 2, NSB 06-01A.
7. McDermott, L., Shaffer, P., and the Physics Education Group (2002). Tutorials in Introductory Physics. Saddle River, NJ: Prentice Hall.
8. Thornton, R.K. & Sokoloff, D.R. (1998). Am. J. Phys. 66 (228), 338-352.
9. Finkelstein, N.D. and Pollock, S.J. (2005). Phys. Rev. ST Phys. Educ. Res. 1, 010101.
10. Hake, R. (1998). Interactive-engagement vs. traditional methods: A six-thousand-student survey of mechanics test data for introductory physics courses. Am. J. Phy. 66 (1), 64-74.
11. Supported by the National Science Foundation Grant DUE-0434144.
12. Supported by the National Science Foundation Grant ESI-0554616.
13. Ding, L. Chabay, R. Sherwood, B. and Beichner, R. (2006) Phys. Rev. ST Phys. Educ. Res. 2, 010105 (2006).
14. Pollock, S. (2006). Replicating Physics Education Reforms: How and why to keep a good thing going. Presented at the annual meeting of the American Physical Society, Dallas, TX., April 2006.
Valerie Otero is an Assistant Professor of Science Education at University of Colorado at Boulder