As part of the Open Learning Initiative (OLI) project, Carnegie Mellon University was funded to develop an online introductory statistics course designed to effectively support learners without an instructor. The OLI Statistics course is perhaps the most systematically studied of all online university courses and has been shown to be effective in a variety of higher education settings both in the online-only and hybrid instructional models. This paper discusses how the OLI Statistics course and its design features support both students and teachers, presents the course assessment results, and describes how these results have placed statistics education at the forefront of the discourse on how technology-enabled instruction can be used to improve learning outcomes while taking on the so called, “Cost Disease” in higher education. . INTRODUCTION In the mid 1960s the two economists Baumol and Bowen described the “Cost Disease” in sectors that relay heavily on human interaction such as education, where there are fewer opportunities for technological advances that lead to increase in labor productivity (Baumol and Bowen, 1966). In the context of higher education, the cost disease refers to the push to higher salaries (and hence cost in general) in response to rising salaries in goods producing sectors (such as car manufacturing) that have been able to significantly increase labor productivity through technological innovations and thus keep cost increases within the rate of inflation. Baumol and Bowen’s theory seem to present higher education with a dilemma. On the one hand, increasing productivity by reducing the cost of instruction per student (i.e., large lecture classes) will lower the quality of education provided. On the other hand, if we do not increase productivity, the Baumol/Bowen cost disease guarantees that more and more people will be priced out of quality education (Thille & Smith, 2012). Carnegie Mellon’s Open Learning Initiative (OLI), an open educational resources project that began in 2002 with a grant from the William and Flora Hewlett Foundation, had hoped to challenge Baumol and Bowen’s dilemma by providing access to high quality education while at the same time increase productivity. As part of OLI project, Carnegie Mellon University was funded to develop an online introductory statistics course. The goal of the funder was to provide open access to high-quality post-secondary educational materials to individual learners who otherwise would be excluded or not encouraged to pursue higher education (Smith and Thille, 2004). The course was developed by a team of statistics faculty members who specialize in statistics education, learning scientists, human-computer interaction experts, and software engineers in order to make best use of multidisciplinary knowledge for designing effective instruction.

As part of the Open Learning Initiative (OLI) project, Carnegie Mellon University was funded to develop a web-based introductory statistics course, openly and freely available to individual learners online and designed so that students can learn effectively without an instructor. In addition, the course is often used by instructors in the hybrid form, to support and complement face-to-face classroom instruction. This paper documents two studies where we investigated the OLI-statistics courses’ effectiveness in the hybrid instructional model. We describe the design, results and limitations of the studies and discuss the implication of the results for finding the “perfect” blend between an instructor and an online course for teaching introductory statistics.

The Open Learning Initiative (OLI) is an open educational resources project at Carnegie Mellon University that began in 2002 with a grant from The William and Flora Hewlett Foundation. OLI creates web-based courses that are designed so that students can learn effectively without an instructor. In addition, the courses are often used by instructors to support and complement face-to-face classroom instruction. Our evaluation efforts have investigated OLI courses’ effectiveness in both of these instructional modes – stand-alone and hybrid. This report documents several learning effectiveness studies that were focused on the OLI-Statistics course and conducted during Fall 2005, Spring 2006, and Spring 2007. During the Fall 2005 and Spring 2006 studies, we collected empirical data about the instructional effectiveness of the OLI-Statistics course in stand-alone mode, as compared to traditional instruction. In both of these studies, in-class exam scores showed no significant difference between students in the stand-alone OLI-Statistics course and students in the traditional instructor-led course. In contrast, during the Spring 2007 study, we explored an accelerated learning hypothesis, namely, that learners using the OLI course in hybrid mode will learn the same amount of material in a significantly shorter period of time with equal learning gains, as compared to students in traditional instruction. In this study, results showed that OLI-Statistics students learned a full semester’s worth of material in half as much time and performed as well or better than students learning from traditional instruction over a full semester. Editor: Stephen Godwin (Open University, UK). Reviewers: Tim de Jong (Open University, NL), Elia Tomadaki (Open University, UK), and Stephen Godwin (Open University, UK). Interactive elements: A demonstration of the StatTutor statistics tutorial is available for playback from http://jime.open.ac.uk/2008/14/stattutor_tour/ . The demonstration is in Flash format. http://jime.open.ac.uk/2008/14/stattutor_tour/

Carnegie Mellon University was funded to develop a introductory statistics course, openly and freely available to individual learners online. The goal of this project is to develop statistical literacy among people who do not have access to academic institutions because of remote locations, financial difficulties or social barriers. In order to achieve this goal, the design of the course has been a collaboration among statistics faculty, cognitive scientists and experts in human computer interaction. This paper discusses the challenges in developing such a learning environment and ways in which the course tries to address them. We also describe the design and results of a pilot study where the degree to which the course is successful in developing statistical literacy has been examined. INTRODUCTION As part of the Open Learning Initiative (OLI) project, Carnegie Mellon University has been funded by The William and Flora Hewlett Foundation to develop an online introductory statistics course. The Foundation's interest is in providing open access to high-quality post- secondary education and educational materials to those who otherwise would be excluded due to geographic, economic or time constraints (Smith and Thille, 2004), as well as for those who due to social barriers are not encouraged to pursue higher education. In other words, we were asked to develop a introductory statistics course that will be openly and freely available to individual learners online. The use of instruction can take many forms. According to Utts et al. (2005), the options range from using applications in a traditional course to full-blown online courses in which there is no face-to-face contact with an instructor. The latter, as we understand it, refers to distance learning where an instructor exists and his/her interaction with the students is mediated electronically, An online course that meets the Foundation's goal adds a new end-point to this continuum; a complete stand-alone, or self-sufficient online course which does not require an instructor in the background. Such a course can, of course, be used in any form on the web-based instruction continuum, and in fact, using the course in hybrid forms is aligned with one of The Hewlett Foundation's other priorities - California Reform - supporting California's community colleges in providing high-quality education to all students even as the state experiences a vast increase in enrollment known as Tidal Wave II (CPEC, 1999). In addition, in their research on the market for online statistics courses, Griffiths and Rascoff (2005) reported that two-year colleges in general, due to their limited ability to innovate, are good candidates for a new online statistics course which will bring their statistics curriculum up to date in substance and teaching methodology. Even though the course can support introductory statistics instruction in a variety of ways, its design and development processes was guided by The Hewlett Foundation's primary interest - the needs of the individual learner using the course to develop statistical literacy as a complete stand-alone course. THE COURSE DESIGN Our general approach to the task of developing a stand-alone course was to create a course that would be as close to a fully online enactment of instruction as possible. The course design used the wealth of experience and knowledge of statistics faculty members and is informed by general cognitive theory and by learning principles that are specific to statistics. The course design was also informed by earlier research conducted at Carnegie Mellon into how students learn statistical reasoning. In this section we'll present some of the challenges we confronted when making the transition from the classroom to an online format, and demonstrate how teaching experience and learning principles helped us address them.

This paper describes a computerized learningtool that we have developed to help overcome this obstacle. This tool is a cognitive tutor in whichstudents solve data-analysis problems and receive individually tailored feedback. We discuss ourcognitive tutor's use in the course and its measured effectiveness in a controlled experiment