CONTEXT Please describe the type and the location of school in which you are working (small liberal arts college in eastern USA, large university in Brazil, etc.). San Francisco State University is a large, urban comprehensive university enrolling (Fall 2007) 24,000 undergraduates, 4200 graduate students and 1500 "post-bac" students. The student body is highly diverse: undergraduates are 58 percent female, 42 percent male; 34 percent white, 24 percent asian and 18 percent hispanic. Detailed demographic information about SF State is can be found at: http://www.sfsu.edu/~ubp/welcome.htm How many students are enrolled to take Statistics? How many sections are there? If some students take the course online and others do not, please describe how selection is made. As of third week spring semester 2008, 775 students are enrolled in 16 sections of statistics. (There is an additional section being taught to about 20 parolled convicts through a special program.) All sections but mine meet 2-3 times per week for a total of 2.5 hours, and all use the text Introduction to the Practice of Statistics by Moore and McCabe. My class of 50 students meets once a week on Tuesday evenings 7:00-9:45 and uses the OLI notes as their only text. A small number of my students may take the course "on-line" in the sense that I will excuse them from all class meetings except for two midterms and a final exam. The only homework in my class is completion of the OLI quizzes and Lab Exercises, and students who choose not to attend class are still obliged to complete these exercises on schedule. What is the composition of the class in terms of year, level, gender, ethnicity, etc? year and level freshman: 14 sophomore: 10 junior: 5 senior: 12 post-bac: 6 graduate: 3 28 women 19 men 3 names do not indicate gender I cannot determine ethnicity Does this course fulfill a requirement and if so what requirement? This course fulfills an undergraduate general education requirement in quantitative reasoning. Most students take the course to fulfill that requirement, although a few need statistics for their majors. The graduate students are all preparing for a graduate research project. Is there anything we should understand about the course context that we have not asked? The prerequisite for the class is intermediate algebra or algebra II, which is also a prerequisite for admission to the university (although many students fail an entrance test in math and must retake algebra after being admitted). On a survey I did the first day, the three major concerns expressed by students were: 1. I won't have enough time to do the homework 2. I won't be able to stay focussed through a three-hour class 3. I don't do well in math classes The first concern is not unusual with our older student body. One student wrote that she was a full-time teacher and single mother, and she was concerned about having enough time to study. As for the second concern, I've tried to structure class time to alternate every 10 minutes or so between lecture and group activities. I try to meet the third concern by focussing on the practical meaning of statistical concepts and calculations and by using software and calculators where possible to reduce the amount of mathematical symbolism that the students have to master.

PROCEDURE How do you teach the course and use the OLI material (e.g. One introductory explanatory lecture and then I do not see them until the mid-term; one recitation per week plus email advising hours)? The class meets 17 times on Tuesday evenings 7:00-9:45. Students are assigned sections of the OLI notes to complete between class sessions. Each week I review the past week's reading and introduce concepts for the week to come. In addition to the class hours, I hold an office hour before class each week in my office and an on-line office hour (through a chat session on the University's Moodle learning management system) the day before class. The online office hour is not much used by students except just before an exam. I use the learning management system as the principal means of communication between myself and the students. I post the syllabus and all the weekly assignments there, and I post my weekly lecture notes at least 24 hours in advance of the class and some spreadsheets to help the students with the Learn by Doing exercises. The system includes a forum that I check at least twice a day for student questions. Who is responsible for contact with the students: you, a TA, a combination? I am the sole teacher. I do not have a grader or TA. Which units and/or modules of the course are you using? I use the entire Statistics text with nothing added. How are you grading students in the course (quizzes, attendance, midterms, etc) In-class groupwork: 20 percent (THE KEEP SYSTEM DOES NOT PERMIT THE USER TO ENTER A PERCENT SIGN) Participation (on-line forum, My Response) 5 percent OLI quizzes and lab exercises: 25 percent Midterms (2): 25 percent Final: 25 percent I've told the students that their grade will be the maximum of their overall grade and their grade on the final. I will determine letter grades by a class curve which I expect to be around A=85 C=60 If you have used the OLI statistics course before, is there anything different in the proceedure that you are planning now than the first time you used it. If so, please describe the difference and explain why you are making the change. I used the OLI notes in fall semester with a night class much like the one I am teaching now. This semester I have fewer graded assignments than last time. No more weekly in-class quizzes. The students hated them and felt that they didn't have time to prepare for them. Last semester I tended to talk at the beginning and end of class for 30-45 minutes with some kind of student activity in the middle. This semester my lectures are interrupted 7-10 times by a group activity, so students seldom listen to me for more than 10 minutes without a break to try something for themselves. Is there anything we should understand about how you are running the course that we have not asked about?

PEDAGOGICAL APPROACH If you have a recitation or formal contact with the students how are you structuring it (e.g. discuss quiz problems and their questions from the instructor feedback reports? Conduct pre-planned lab activities. Review examples and concepts covered in the module)? I meet my students once a week in a three-hour night class. That's pretty long, so I try to alternate every 10 minutes or so between lecture and group activities. I begin by answering any questions that they have posed during the week through My Response or the class on-line forum. We take a 15 minute break in the middle of the class. My lectures follow the OLI notes closely. Sometimes I use examples from the notes, and sometimes I bring in my own examples, but in either case the topics are presented in the same order as they are in the notes. What if anything will you share with students about effective ways to study or use the online materials? I've told them that they are unlikely to pass unless they study a minumum of six hours per week, and that the most effective studying will be one hour every day. I emphasize the importance of doing all the interactive modules in the OLI notes. I repeat all this at least once every class, but I know that many of my working students try to cram all their studying into the weekend. When you are discussing a new or difficult idea do you start with a clean and abstract definition and then instantiate it with interesting examples or do you start with an example and try to move to the abstract? Or do you do something entirely different? I usually try to give an general description of the concept and the need for it, then the precise definition, then some examples. When I do correlation next week, I'll say: 1. We want to know if scatterplot data approximately follows a line. The answer isn't "yes" or "no" but "how much". 2. We have a numerical measure...definition...properties 3. Examples Are you giving any additional assignments besides the questions in the modules and the quizzes and the StatTutor labs and if so what sort? Two midterms and a final exam. Also group work in class, but that isn't carefully graded. If you have used the OLI statistics course before, is there anything different in the approach you are taking now than the first time you used it. If so, please describe the difference and why you chose to make the change. I used the OLI notes in fall semester with a night class much like the one I am teaching now. This semester I have fewer graded assignments than last time. No more weekly in-class quizzes. The students hated them and felt that they didn't have time to prepare for them. Last semester I tended to talk at the beginning and end of class for 30-45 minutes with some kind of student activity in the middle. This semester my lectures are interrupted 7-10 times by a group activity, so students seldom listen to me for more than 10 minutes without a break to try something for themselves. Is there anything we should understand about your pedagogical approach that we have not asked about?

LEARNING Please describe how you think learning takes place? Do you, for example, imagine that a student develops a vague sense of the material and over time and with practice fills in the details? Do you believe that learning takes place as a series of fairly discrete building blocks each of which needs to be mastered in turn, and, when mastered results in understanding and competence of the content? Or, do you have another view altogether? I've been teaching for 40 years, and I've finally learned that I know little about how people learn. That's why I'm so excited to work with the researchers at CMU, who may actually have knowledge about learning. I do think the following principles are valid: 1. People have different levels of natural ability in mathematics, and all people can improve their ability to do mathematics. 2. Successful mathematicians (at all levels) apply a small number of powerful ideas in a flexible way to solve problems. 3. Most people fail to succeed in mathematics because they try to memorize specific methods for solving each problem that they might encounter, and they usually are unable to remember or successfully apply the memorized methods when needed. 4. People don't all learn a subject in the same linear order. Key ideas must be presented repeatedly. On the other hand there is a general order to how a subject must be learned. A statistics student who hasn't mastered means and standard deviations will probably not understand hypothesis testing. 5. People have to use ideas and methods, not just read or hear about them, to learn them. Please describe, in as detailed a way as you can, how you imagine a good student might go through the on line material in the most effective way possible. They will read until they come to an online activity, then they will try the activity. They will go back and re-read as necessary to complete the activity. Then they will go on. They will not permit themselves to continue without getting the on-line activities right or fully at least fully understanding why the right answers are correct. Students who start the quizzes early (at least a day before they are due) are able to go back and find the right answers in the text. Those who start late just put in what they remember, which is often wrong. Some sort of index to the notes would be very helpful to the student who wants to look something up while doing a quiz or other interactive exercise.

Motivation and Results Please describe why you have used or are using the OLI statistics course. Since I would see my students only once a week, I knew that they would have to learn a lot on their own. I didn't believe that they could learn from Moore and McCabe, and the OLI course seemed well-written and equipped with a lot of support in terms of interactive activities. My expectations were borne out during the first semester my students used the course. What pedagogical, teaching or learning challenge(s) are you trying to address in using this course? For example, you may want to discuss a particular learning problem that your students faced, or difficulty in teaching a particular concept. My main challenge is getting students to put in enough time with the notes. The students say that the sections on probability and parts of hypothesis testing are difficult to understand, but I cannot disentangle real difficulties with the text from ineffective or inadequate study methods. I know that if I look at the time they start each week's assignment, most of the students begin just a day or two before class. Last semester--for one week--I tried assigning an on-line midweek quiz between class sessions to get the students working all week. The class rebelled, saying that I was not giving them enough freedom to study as they wanted or needed to. I decide that they were right, and that there was a limit to how much negative motivation I could usefully provide. Motivating students to study outside of class is always a challenge, whether we are talking about immature freshmen or overburdened working seniors. The latest fad on our campus aimed at this problem is educational games. I think that the OLI couse is well-written, as short as possible while still covering the material, and reasonably interesting. The motivation problem is not due to poor quality materials but lack of dedication on the part of students. It's up to me to find ways to make the class more interesting and more compelling. If you have used the OLI statistics course before, please describe the results for you and your students in using it. Last semester my grades were: A: 7 B: 10 C or CR: 23 D: 2 F, NC or W: 15 Of the students who failed, only two attempted the final exam. You could say that almost every student who completed the class passed it. However I don't see the class as very successful. Too many student quit, and too many earned C's. At the end of the semester I calculated some correlations. I only considered students who took the final. The correlation between class activities (essentially class attendance) and the final exam grade was 0.35; the correlation between OLI activities (quizzes and lab exercises) and the final exam grade was 0.67. Students who completed the OLI activities learned statistics.