Open Learning Initiative

Introduces the basic concepts, logic, and issues involved in statistical reasoning. Topics include Exploratory Data Analysis, Producing Data and Study Design, Probability and Statistical Inference.

Select one of the two new courses below. Both courses include all expository text, simulations, case studies, comprehension tests, interactive learning exercises, and the StatTutor labs. Both courses contain all of the instructions for the four statistics packages options we support. To do the activities, you will need your own copy of Microsoft Excel, Minitab, the open source R software, or a TI calculator. The only difference between the two new courses is the path through probability.

The course called Probability and Statistics includes the classical treatment of probability as it is in the earlier versions of the OLI Statistics course.

The course called Statistical Reasoning includes a new streamlined probability option that forgoes the classical treatment of probability in favor of an empirical approach using relative frequency. This path includes only those concepts that are necessary to support a conceptual understanding of the role of probability in inference.

You can also choose to work through only the StatTutor labs. You will need your own copy of Microsoft Excel, Minitab, R, or a TI Calculator to do the StatTutor activities.

• StatTutor Labs

The Open and Free Full Courses

The Probability and Statistics course is comparable to the full semester course on Statistics taught at Carnegie Mellon University. Your access includes the complete online course including all expository text, simulations, case studies, comprehension tests, interactive learning exercises, and StatTutor. The course covers the topics of Exploratory Data Analysis, Producing Data and Study Design, Probability and Statistical Inference.

The Statistical Reasoning course is the same as the Probability and Statistics course except that it includes a new streamlined probability option that forgoes the classical treatment of probability in favor of an empirical approach using relative frequency. There is an early focus on probability distributions as a way to describe patterns arising in a long series of repetitions of random phenomenon. This approach includes only those concepts that are necessary to support a conceptual understanding of the role of probability in inference. Examples use contexts from real world problems and simulations set in real world contexts.

Both probability paths culminate in a newly enhanced discussion of sampling distributions that is grounded in simulation and includes many new activities to support student understanding of the use of normal distributions to model sampling variability.

The Open & Free Statistics courses do NOT include access to the graded exams or to the course instructor. No credit will be awarded from Carnegie Mellon for completing either of the Open & Free Statistics courses.

Academic versions are offered through educational institutions that award credit. Students taking an Academic Course have access to the same course materials as the students taking an Open & Free Course PLUS access to graded exams. The Academic courses track student's learning of key concepts and give the student and the instructor formative feedback to improve learning outcomes.

Course Description

This course introduces students to the basic concepts, logic, and issues involved in statistical reasoning. Major topics include exploratory data analysis, an introduction to research methods, probability, and statistical inference. The objectives of this course are to give students confidence in manipulating and drawing conclusions from data and provide them with a critical framework for evaluating study designs and results.

An important feature of the course is the use of an intelligent tutoring system developed at Carnegie Mellon called "StatTutor." StatTutor aims to facilitate understanding of statistical ideas and analytical techniques by helping students construct useful knowledge representations and thereby develop effective problem-solving skills. It uses a specified outline of steps to follow in solving problems, or "scaffolding". StatTutor will use scaffolding and immediate feedback flexibly, tracking and responding to individual students as they navigate the learning environment.

New Release Available Now

In 2010, faculty from multiple institutions across the US joined the OLI development team to adapt and improve the course. The team also used the data we have collected over multiple semesters of student use to drive the redesign. The 2010 release represent one of the largest improvements in the course since its original design. Major changes to the new course include:

Learning Objectives: We have revised the Learning objectives throughout the course to be more specific, clear, and measurable and to align with objectives developed by the Carnegie Foundation (CFAT) "Statway" project. As part of their "Statway" project, CFAT convened statistics faculty from around the country to build agreement on the learning outcomes for an introductory level college statistics course. We also now show the relevant learning objective(s) at the top of every page.

Graded Assessments: We added detailed feedback to all of the quizzes throughout the course, increased the number of items for each quiz to create item pools and broke the quizzes into more frequent "checkpoint" activities to serve several needs: (1) provide additional graded activities that instructors can assign as homework, (2) provide instructors with more frequent feedback on student performance, so that misconceptions can be addressed sooner, and (3) better enable instructors who are not covering all topics in the course to assign graded activities that cover the specific parts of the course.

Learning Activities: Throughout the course, we added new interactive activities to fill gaps that we have identified by analyzing student log data and by gathering input from other college professors on where students typically have difficulties. We have increased the number of interactive activities by over 30%. We added new simulation-style activities to the course in Inference and Probability. We also moved many of the activities that were previously located behind links into the main content flow. We moved the activities because the student log data gave compelling evidence that students are more likely to complete the activities if they are placed in the flow of the page rather than behind a link.

Course Structure: We refined the statistics “big picture” (the coherent structure of the course) and now feature the Big Picture more prominently throughout the course. We also changed the structure of the unit on Inference. We broke the inference modules into smaller sections that isolate different tests and yet are structured to reflect the overarching logic of inference. We made this change to make it easier for instructors to choose different tests to include or exclude from their courses. We renamed the substructure of the variable role type classification from "Case I", "Case II", and "Case III" to Case C->Q (Categorical -> Quantitative), Case C->C, and case Q->Q respectively, based on feedback from instructors.

Learning Strategies: We added a new section to the introductory content of the course on Learning Strategies for taking an online course.

Probability: Student log data have shown that the unit on probability is the part of the course that is most difficult for students. We rewrote the content of the unit and now offer two paths through probability as described at the beginning of this email.

Supporting Statistics Packages: We changed the course architecture to create a new "unified course" where instructions for all statistics packages are available for all students. We also added instructions for using the TI calculator for almost all exercises.

Instructor Support: Based on general feedback from instructors and on more focused research and formal user testing with specific instructors, we redesigned the instructor reports (the learning dashboard) to better provide instructors with information about student performance that is meaningful and useful and to better support instructors to interpret the information provided by the learning dashboard. We added a new instructor resources section that will host instructor-contributed problem sets that instructors can use for additional practice on concepts where students are struggling.