WWW.DISSERTATION.XLIBX.INFO
FREE ELECTRONIC LIBRARY - Dissertations, online materials
 
<< HOME
CONTACTS



Pages:     | 1 || 3 | 4 |   ...   | 20 |

«University of California Los Angeles Bridging the Gap Between Tools for Learning and for Doing Statistics A dissertation submitted in partial ...»

-- [ Page 2 ] --

At the primary school level, the necessity of a computational tool is less obvious, and students’ reasoning is typically not to the level of abstract thinking required to grasp concepts like variation at a deep level. The Guidelines for Assessment and Instruction in Statistics Education (GAISE) outline three levels of statistical reasoning, A, B, and C, which build on one another (Franklin et al., 2005). The authors of the guidelines suggest pre-secondary school students should be exposed to levels A and B so they are prepared for level C (including randomization, comparing groups using measures of variability, and considering model fitting, including variability in model fitting) when they arrive in high school. The final reason for excluding younger students from this work is existing tools for teaching statistics seem to work well at this level (Biehler et al., 2013).

Going forward, the term ‘novice’ is used to refer to someone at the secondary school level or above, who does not have statistical or programming skills.

1.3 Technology as a component in the educational ecosystem The problems and opportunities in statistics and statistics education are wide reaching. Certainly, access to statistics and data analysis will continue to broaden as material trickles down from innovations in academia and industry. However, for novices to learn statistics, there is a need for much more than technology. As has been the case in the evolution of all technology (e.g., moving pictures, radio, television, computers, and now hand-held devices like smartphones and tablets), just having the technology does not improve education.

The most crucial educational need is for good instructors. At the secondary school level, this appears to be particularly hard (the politics of education in the United States still have not resolved to a state where teachers are paid like other professionals) but even at the post-secondary level practitioners are valued more highly than educators.

However, simply having qualified teachers is still not enough. Teachers need training and curriculum in order to scaffold the material they are teaching.

While modern statistical methods have begun to permeate courses at the college level, there has been limited trickledown to the high school level. The Introduction to Data Science course discussed in Section 5.1.3.3 is the only curriculum we are aware of focusing on computational statistics and data analysis at the high school level.

It is possible new methods of educational publishing will help remedy some of these problems. For example, Massive Open Online Courses (MOOCs) have been gaining traction, and Johns Hopkins University has developed a data science sequence on the Coursera platform, which introduces students to exploratory data analysis, R, reproducible research, statistical inference, and modeling, from simple linear modeling to machine learning (Caffo et al., 2015). Perhaps the availability of good computational statistics instruction on the web will reduce the need for teachers in other contexts, but it is likely to be an incremental shift rather than a large one.

The problems of good teachers, curriculum, and teacher training are all difficult. This work only touches on these problems briefly, although I acknowledge they are equally if not more important than the technological tool used. However, in the context of statistics and data analysis, a computational tool is necessary. There is simply no way to do data analysis without a computer, and the current tools leave a lot to be desired.

–  –  –

One question I am often asked when discussing this work is why I use the term ‘tool’ to mean computer software or a programming language. Essentially, I am looking forward to a future where computers do more than just amplify human abilities – they enhance them (Pea, 1985). In the same way things we traditionally think of as tools (like hammers, levers, and sewing machines) allow us to do much more than we can do on our own, I believe computers are going to give us superpowers. Particularly in the context of data, where so much of it is highdimensional, humans need some assistance ‘seeing’ the existing patterns.

1.5 The difference between users and creators Throughout this work, I often reference the difference between being a ‘user’ of a tool and being a ‘creator’ of statistics. My use of these terms comes from the literature on equity in computer science education.

In her 2001 book “Unlocking the Clubhouse,” Jane Margolis describes the difference between being a user of technology, which might include learning word processing software and typing skills, and being a creator of technology (Margolis and Fisher, 2001). In her book, she found women were less likely to be creators of technology than their male counterparts. Margolis followed this book by another, called “Stuck in the Shallow End,” which presented similar results for minority students – minorities were less likely to have access to high school courses focused on creation (e.g., AP Computer Science) and more likely to take classes labeled as computer science but focused on using software (Margolis et al., 2008).

This distinction has continued to be forefronted in discussions of computer science education. In a 2014 article in Mother Jones, a National Science Foundation representative was quoted lamenting, “We teach our kids how to be consumers of technology, not creators of technology” (Raja, 2014).





In the realm of statistics education, there has been less explicit discussion of the difference between using statistics and statistical tools, and being a creator of statistical products. However, this distinction is crucial. Students are often taught statistics as a static set of formulas to be applied in particular situations.

They may have a formal definition of a standard deviation memorized, and be able to apply a hypothesis test to some data, but they do not have deeper conceptual knowledge. With the recent focus on p-hacking (Head et al., 2015) and at least one journal banning the use of p-values (Trafimow and Marks, 2015), it is clear that even within the ‘client disciplines’ there is need for richer statistical understanding.

Being a ‘creator’ of statistics and statistical products requires a person to think creatively with data and to move through a cycle of exploration and conrmation with their data. In the context of a statistical tool, a creator of statistics should be able to build new methods and data visualizations based on existing functionality within the system.

1.6 Overview of the rest of the dissertation This dissertation is a mash-up of case study, opinion piece, and provocative technology experiment.

Chapter 2 outlines the history of statistical programming, both for practitioners of statistics as well as for learners of statistics. Then it calls attention to the gap between tools for doing statistics and tools for learning statistics. The chapter ends with a call to action to close this gap by developing tools that can support a learning-to-doing trajectory.

Chapter 3 discusses existing tools and their success at spanning the gap, including Excel, R, TinkerPlots and Fathom, SAS, SPSS and STATA, as well as a number of bespoke data visualization tools.

Chapter 4 outlines requirements for future statistical programming tools, including the need for easy entry for novice users, interactivity at every level, and simple support for narrative, reproducibility, and publishing.

Chapter 5 discusses work I have done to close the gap between tools for teaching and learning statistics, and those for doing statistics. This chapter includes my work on the Mobilize project (the initial inspiration for this work), my joint work with Aran Lunzer of the Communications Design Group, called LivelyR, and two illustrative Shiny widgets I created to augment my teaching.

Chapter 6 brings these thoughts together, and outlines my vision for the future of this work. However, understanding this vision will take years to come to fruition, I also provide recommendations for best practices using currentlyavailable tools.

We will be remiss in our duty to our students if we do not see that they learn to use the computer more easily, flexibly, and thoroughly than we ever have;

we will be remiss in our duties to ourselves if we do not try to improve and broaden our own uses.

–  –  –

This chapter sets up the gap between tools for learning and tools for doing statistics. The current qualities of both types of tools were highly influenced by the time of their development and the specific goals the developers were considering. We begin by considering the history of tools for doing statistics, and follow with the history of tools for learning statistics. Looking at the diverging history makes it clear there is a gap between the two types of tools, so we consider the reasons (both educational and political) for the gap. Finally, I argue the gap should be eliminated in future tools.

2.1 History of tools for doing statistics Many tools have been used for statistical computing over the years. It is not within the scope of this work to capture every one of them, but some of the largest stepping stones can be noted.

According to Jan De Leeuw, the history of statistical software packages began at UCLA in 1957 with the development of BMDP (De Leeuw, 2009). BDMP was part of the first generation of statistical software packages, the others being SPSS and SAS, both introduced in 1968. The second generation includes Data Desk, JMP and STATA, all of which were developed in the 1980s (De Leeuw, 2009).

The next generation were the Lisp-inspired languages, S, LISP-STAT, and R (De Leeuw, 2009). Both S and LISP-STAT could be extended using Lisp, and R was based on Scheme, a dialect of Lisp. S was written by John Chambers in the 1970s when he grew frustrated with Fortran and developed S to provide more statistical power and flexibility (Becker, 1994).

The versions of S are often spoken of in terms of ‘books’ – the ‘brown book’ ‘blue book’ and ‘white book’ correspond with versions of S. The goal of S was initially to create an interactive functional programming language with a strong sense of data format (Becker, 1994). Initially, S was free to academics, but it was eventually sold to the Insightful corporation (De Leeuw, 2009) and now exists in the form of S-PLUS, sold by TIBCO Software, Inc.

However, R has essentially captured the former market of S and XLISPSTAT, at least in academia (De Leeuw, 2004). R was developed by Ross Ihaka and Robert Gentleman at the University of Auckland to provide an open-source alternative to S, incorporating features from both S and Scheme (De Leeuw, 2009; Ihaka and Gentleman, 1996). R is discussed in more detail in Section 3.8.

2.2 History of tools for learning statistics

Running parallel to the history of statistical programming tools is a trajectory of tools for making statistics easier for novices to understand. The history of tools for learning statistics is shorter than that of tools for doing statistics, in part because learning tools require more developed computer graphics, and also because there has been a long debate about whether computer tools are necessary for statistics education at all.

In 1997, Rolf Biehler described what he saw as the difference between tools for doing statistics and those for learning statistics (Biehler, 1997). Biehler distinguishes between tools (used for real statistical analysis), microworlds (interactive systems encouraging play), resources (including data and data documentation) and tutorial shells (he mentions Hypercard being used to create tutorial shells to interface with other software) (Biehler, 1997).

In the late 1990s, Minitab and Splus could provide functionality for creating microworlds, and were therefore commonly used in teaching. The other interfaces Bielher mentions are tutorial shells, including SUCROS, StatView, and Data Desk (Biehler, 1997). Interestingly, Data Desk is the one tool covered by both De Leeuw as a tool for statistical computing and by Biehler as a tool for teaching and learning statistics. It was developed in 1985, and has been maintained by its software publisher to this day (currently, the software is on version 7). The functionality is very inspiring in the context of the bridging I am considering, but the user interface looks antiquated compared to current tools.

In the context of the tools that existed in 1997, Biehler was arguing for a new type of software. He wanted a tool which would solve three problems: the complexity of tools for doing statistics, the challenge of closed microworlds, and the necessity of using many tools in order to cover a particular curricular trajectory (Biehler, 1997).

Biehler’s paper expanded on the capabilities he believed were necessary for such a system, and his guidelines were used to develop TinkerPlots and Fathom in the early 2000s (Konold and Miller, 2005; Finzer, 2002a). Both tools are still in use today, and are discussed at more length in Section 3.4.

Since their development, TinkerPlots and Fathom have become extremely popular for teaching statistics in the K-12 context (Lehrer, 2007; Garfield and Ben-Zvi, 2008; Konold and Kazak, 2008; Watson and Fitzallen, 2010; Biehler et al., 2013; Finzer, 2013; Fitzallen, 2013; Mathews et al., 2013), at the introductory college level (Ben-Zvi, 2000; Garfield et al., 2002; Everson et al., 2008) and in training for teachers (Rubin, 2002; Biehler, 2003; Gould and Peck, 2004;

Hammerman and Rubin, 2004; Rubin et al., 2006; Hall, 2008; Pfannkuch and Ben-Zvi, 2011).

The main alternatives to TinkerPlots and Fathom are browser applets (what Biehler might call microwords) like those by the five Locks (Morgan et al., 2014) or by Rossman and Chance (Chance and Rossman, 2006). Applets are described more fully in Section 3.5. In general, tools for learning statistics have not moved forward since the early 2000s, and have not kept up with modern statistical practice.

2.3 The gap between tools for learning and doing



Pages:     | 1 || 3 | 4 |   ...   | 20 |


Similar works:

«Secular Conversions: Politics, Institutions, and Religious Education in the United States and Australia, 1800-2000 By Damon Walter Mayrl A dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy in Sociology in the Graduate Division of the University of California, Berkeley Committee in charge: Professor Margaret Weir, Chair Professor Marion Fourcade Professor Philip Gorski Professor David Hollinger Fall 2011 Abstract Secular Conversions:...»

«THE NEURAL CORRELATES OF EMOTIONAL NUMBING AND NICOTINE USE IN VETERANS DURING WAKE AND REM: AN [18F]-FDG PET IMAGING STUDY by Marissa H. Swanson Submitted to the Faculty of The Dietrich School of Arts and Sciences in partial fulfillment of the requirements for the degree of Bachelor of Philosophy University of Pittsburgh 2012 UNIVERSITY OF PITTSBURGH THE DIETRICH SCHOOL OF ARTS AND SCIENCES This thesis was presented by Marissa H. Swanson It was defended on April 6, 2012 and approved by Erika...»

«Reply to David Buller by Martin Daly & Margo Wilson The substantial excess risk of abuse and homicide incurred by stepchildren has been abundantly documented in dozens of studies using diverse methodologies (see companion document “The Cinderella effect”). Nevertheless, philosopher David Buller (2005a,b) has recently attempted to call the existence of this phenomenon into question by proposing “that all of the evidence cited in support” of it (Buller 2005b: 282; emphasis in original)...»

«DYNAMIC RESOURCE MANAGEMENT IN RESOURCE-OVERBOOKED CLOUD DATA CENTERS By Faruk Caglar Dissertation Submitted to the Faculty of the Graduate School of Vanderbilt University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY in Computer Science August, 2015 Nashville, Tennessee Approved: Dr. Aniruddha S. Gokhale Dr. Douglas C. Schmidt Dr. Gautam Biswas Dr. Christopher J. White Dr. Akos Ledeczi To my beloved wife Fatma for her patience, encouragement, and support...»

«University of Iowa Iowa Research Online Theses and Dissertations 2011 T regulatory cells and the germinal center Carla-Maria Alana Alexander University of Iowa Copyright 2011 Carla-Maria Alexander This dissertation is available at Iowa Research Online: http://ir.uiowa.edu/etd/1117 Recommended Citation Alexander, Carla-Maria Alana. T regulatory cells and the germinal center. PhD (Doctor of Philosophy) thesis, University of Iowa, 2011. http://ir.uiowa.edu/etd/1117. Follow this and additional...»

«K O E FANĀ FOTU´: SUCCESS IN MOTION, TRANSFORMING PASIFIKA EDUCATION IN AOTEAROA NEW ZEALAND 1993-2009 A thesis submitted in fulfilment of the requirements for the Degree of Doctor of Philosophy in Pacific Studies and Education at the University of Canterbury Christchurch New Zealand Lesieli Pelesikoti Tongati‘o © ii MINISTRY OF EDUCATION STATEMENT 15 April 2010 Approval is given for Lesieli Pelesikoti Tongati‘o to use and analyse information and data, gathered during the course of her...»

«ABSTRACT Title of Dissertation: ON THE DEVELOPMENT OF POSTURAL STABILITY DURING INFANCY AS A PROCESS OF GROWTH AND ACTIVE, EXPLORATORY SENSORIMOTOR TUNING Jason S. Metcalfe, Doctor of Philosophy, 2007 Dissertation directed by: Professor Jane E. Clark, Ph.D. Department of Kinesiology The process by which humans stabilize bipedal stance represents a confluence of changes associated with musculoskeletal maturation and experience-based sensorimotor learning. While investigations have documented a...»

«An Appraisal of Naturalism in Contemporary Meta-Ethics David Christopher Lahti B.S., Gordon College Dissertation submitted for the degree of Ph.D. in philosophy (ethics) on 28 February 1998. The Whitefield Institute, Oxford.Co-supervisors: E. David Cook, Fellow, Green College, Oxford University R. J. Berry, Emeritus Professor of Genetics, FIBiol, FRSE, University College London Reformatted and repaginated for Letter sized paper, 2000. None of the following has been submitted previously for any...»

«Necessity in Kant; Subjective and Objective DAVID T. LARSON University of Kansas Kant suggests that his contribution to philosophy is analogous to the contribution of Copernicus to astronomy—each involves the revolutionary reorientation of a discipline. Kant describes his revolution as the move from taking objects to be determinative of knowledge to taking knowledge to be determinative of objects. This revolution should be seen in its proper context; it is Kant's response to such philosophers...»

«http://unllib.unl.edu/LPP/ Library Philosophy and Practice 2011 ISSN 1522-0222 Doctoral Research in Library and Information Science by Pakistani Professionals: An Analysis Dr. Rais Ahmed Samdani Associate Professor Department. of Library & Information Sciences University of Sargodha, Pakistan Dr. Rubina Bhatti Assistant Professor Department. of Library & Information Sciences The Islamia University of Bahawalpur, Pakistan Introduction Research is a process of enquiry that draws data from the...»

«DEVELOPMENT AND VALIDATION OF A MUSICAL BEHAVIOR MEASURE FOR PRESCHOOL CHILDREN By Gina Jisun Yi A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Music Education Doctor of Philosophy ABSTRACT DEVELOPMENT AND VALIDATION OF A MUSICAL BEHAVIOR MEASURE FOR PRESCHOOL CHILDREN By Gina Jisun Yi The purpose of this study was to develop a measure for use in assessing musical behaviors of preschool children in the context of regular music...»

«ABSTRACT Title of Dissertation: BECOMING A RECOGNIZED HISPANICSERVING INSTITUTION (HSI): INSTITUTIONAL CULTURE & TITLE V Rebecca C. Villarreal, Doctor of Philosophy, 2014 Directed By: Associate Professor Noah D. Drezner, Department of Counseling, Higher Education, and Special Education The purpose of this study was to understand the influence of institutional culture at an HSI on the Title V program. The findings highlight the forces that influence the process of securing, evaluating, and...»





 
<<  HOME   |    CONTACTS
2016 www.dissertation.xlibx.info - Dissertations, online materials

Materials of this site are available for review, all rights belong to their respective owners.
If you do not agree with the fact that your material is placed on this site, please, email us, we will within 1-2 business days delete him.