This item is only available as the following downloads:
1 ANALYSIS OF SELF DIRECTED MASTERY LEARNING OF HONORS PHYSICS By WENDY ATHENS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCT OR OF EDUCATION UNIVERSITY OF FLORIDA 2011
2 2011 Wendy Athens
3 To Alexis, Andrew, and Katie, who have sacrificed to earn this degree with me
4 ACKNOWLEDGMENTS This dissertation would not be possible without the unwavering support of my com mittee including Kara Dawson (chair), Alyson Adams, Ana Puig, and Chris Sessums. Their insight guided this research project. I respect their ability to discern nuggets of gold in a pile of data. I also want to thank my honors physics students who tolerated filling out daily Time Tickets and surveys to extract greater meaning out of routine school and my school administration for allowing me to conduct this classroom research.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 7 LIST OF FIGURES .......................................................................................................... 8 LIST OF ABBREVIATIONS ............................................................................................. 9 ABSTRACT ................................................................................................................... 11 CHAPTER 1 INTRODUCTION .................................................................................................... 13 2 LITERATURE REVIEW .......................................................................................... 17 Physics Education Research .................................................................................. 18 Mastery Learning .................................................................................................... 19 Self Directed Learning ............................................................................................ 23 Technology ............................................................................................................. 26 Animations and Simulations in Science ............................................................ 26 Blended Learning ............................................................................................. 27 Computer Tutors .............................................................................................. 30 Google Apps/Web 2.0 .................................................................................... 33 Summary of Literature Review ................................................................................ 36 3 RESEARCH DESIGN AND METHODOLOGY ....................................................... 42 Context ................................................................................................................... 43 Research Design .................................................................................................... 44 Pilot Study ........................................................................................................ 45 Infrastructure .................................................................................................... 46 Goog le Apps platform .............................................................................. 46 VOLOs ....................................................................................................... 47 Instructional Strategy ........................................................................................ 48 Data Col lection ....................................................................................................... 53 Unit Tests ......................................................................................................... 53 SDL Activities ................................................................................................... 54 Post survey ....................................................................................................... 55 Data Analysis .......................................................................................................... 56 Summary ................................................................................................................ 57
6 4 RESULTS ............................................................................................................... 63 SDL Activities Chosen and Completed (RQ #1) ..................................................... 63 Choices and Time Allocation of Successful Students (RQ #2) ............................... 64 Perception Differences of Successful versus Unsuccessful Students (RQ #3) ....... 65 Summary ................................................................................................................ 69 5 FINDINGS, DISCUSSION, CONCLUSIONS, RECOMMENDATIONS, AND FUTURE WORK ..................................................................................................... 78 Overview of the Study ............................................................................................. 78 Limitations ............................................................................................................... 82 Findings .................................................................................................................. 83 RQ #1 Findings SDL Activities Chosen ......................................................... 83 RQ #2 Findings Relationship between SDL Activi ties and Concept Mastery ......................................................................................................... 84 RQ #3 Findings Differences in Perception between Successful and Unsuccessful Students .................................................................................. 85 Discussio n .............................................................................................................. 89 Conclusions ............................................................................................................ 96 Recommendations and Future Work ...................................................................... 98 APPENDIX A TIME TICKET ....................................................................................................... 102 B POST STUDENT SURVEY .................................................................................. 103 C SAMPLE TEST ..................................................................................................... 107 D IRB ........................................................................................................................ 109 LIST OF REFERENCES ............................................................................................. 114 BIOGRAPHICAL SKETCH .......................................................................................... 128
7 LIST OF TABLES Table page 2 1 Collection of noteworthy instructional interventions within physics education research (PER) literature .................................................................................... 38 2 2 Learning gains cited in literature for various educational interventions ............... 40 2 3 Shifts in thinking between teacher directed and student directed learning (Gibbons, 2002) .................................................................................................. 40 3 1 Data sources and analyses that answer the research questions ........................ 58 4 1 VOLO activities selected by percentage, and average educational value and enjoyment scores. .............................................................................................. 71 4 2 Comparison of post survey average Likert responses of successful versus unsuccessful students (N=24). ........................................................................... 72 4 3 MannWhitney U test indicated post survey Likert responses were statistically different between successful and unsuccessful groups ( = 0.05) ..................... 74 4 4 Likert ratings by successful and unsuccessful students of accessibility and quality of three tutor supports ............................................................................. 74 4 5 Comparison of successful vs. unsuccessful group responses to the post survey question, If you had a choice of two tools or resources to use to learn physics, what would you pick and why? ............................................................ 75
8 LIST OF FIGURES Figure page 2 1 The literature review of Chapter 2 explores these four conceptual areas: Physics Education Research (PER), Learning for Mastery (LFM), Self Directed Learning (SDL), and Technology. ........................................................ 41 2 2 Mastery learning is a feedback corrective res ponse instructional strategy that uses formative assessments to guide learning. .................................................. 41 3 1 Overview of the capstone design of a supported and assessment based self directed mastery learning environment. .............................................................. 59 3 2 Snapshot of a personal st udent web page in the class website. ........................ 60 3 3 The VOLO for the conservation of energy unit shows three required activities PhET, masteringphysics, and PBB Lab. ........................................................ 61 3 4 Instructional sequence during the four week unit. Note the time allocation of approximately 25% wholeclass versus 75% SDL. ............................................. 62 4 1 The tally of SDL activity choices indicates all students compl eted the required activities and all used expert and peer tutors. ..................................................... 76 4 2 Average minutes invested per SDL activity per student. Time data based on Time Tickets and masteringphysics time logs. ................................................. 76 4 3 The number of students who cleared the unit test in four tries is shown. Seven students cleared the unit test within four tries and 17 students did not. .. 77 4 4 SDL activities evaluated based on minutes invested by successful and unsuccessful groups. Overall, successful students spent 25% more time on their activities than unsuccessful students did. ................................................... 77 5 1 Epilogue: Gains in average unit test scores over five tests. Gain was calculated as [(Recent Score Previous Score) / (100 Previous Score)]. ..... 101
9 LIST OF ABBREVIATION S FCI Force Concept Inventory (FCI) is a v alidated assessment tool that measures conceptual understanding of physics without requiring formal education in physics. This tool was developed by Halloun and Hestenes (1985) and its widespread adoption has enabled secondary and university physics instr uctors to gauge the impact of pedagogical changes on students conceptual understanding. LFM Learning for Mastery (LFM) is an instructional practice attributed to Benjamin Bloom (1968) that is based on the foundational belief that every student can attain concept mastery given enough time and adjustment of instructional methods. MOPS Minds on Physics (MOPS) is a curriculum developed by the Physics Education Group at the University of Mas sachusetts ( http:/ /www.physicsclassroom.com/mop/ ). MOPS is built on constructivist principles and is interactive and increases in complexity, requiring effort on the learners part. PBB Lab The Pendulum Box Bash Lab was adopted from Zou (2000) of Ohio State University, and demonstrated conditions under which conservation of energy laws DO NOT apply and conservation of momentum laws DO apply. The central idea of the lab was to account for energy losses in inelastic collisions PhET Physical Educational Technology PhET from University of Colorado at Boulder provides fun, interactive, research based simulations of physical phenomena ( http://phet.colorado.edu) PhET enables students to make connections between real life pheno mena and the underlying science, deepening their understanding and appreciation of the physical world. PIVOT Physics Interactive Video Tutor (PIVOT) was developed and implemented at MIT (Lipson, 2001 and encompassed video lectures are interspersed with i nteractive questions to create a virtual mentoring environment. PSI Personalized System of Instruction (PSI) is a form of mastery learning attributed to Fred Keller (1968). PSI involves self paced, largely text based learning with retesting on identical t ests until mastery is reached. RQ Research Question TEAL Technology Enabled Active Learning (TEAL) has evolved over 10 years as a joint research partnership between the MIT Teaching and Learning Laboratory and the MIT Physics Department. The TEAL project has transformed freshman physics from a
10 lecture/recitation format to a studio physics model that combines short presentations with inclass desktop experiments and conceptual problem solving teams (Breslow, 2004). VOLO A VOLO an acronym for Vocabular y Objectives Learning Options is a learning contract Typically, the VOLO appears as a 9 squared tic tac toe box on the students web page, and the student contracts with the instructor for which five activities he or she will complete for the learning unit. A student must earn a score of 80% to receive a completion mark for the VOLO task. The grading scale for VOLOs is typically: 5 of the 9 tasks completed = 100%, 4 tasks = 85%, 3 tasks = 60%, 2 tasks = 30%, 1 task = 0%.
11 Abstract of Dissertatio n Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Education ANALYSIS OF SELF DIRECTED MASTERY LEARNING OF HONORS PHYSICS By Wendy Athens August 2011 Ch air: Kara Dawson Major: Curriculum and Inst ruction Self directed learning (SDL) is an important life skill in a knowledgebased society and prepares students to persist, manage their time and resources, use logic to construct their knowledge, argue their views, and collaborate. The purpose of this study was to facilitate mastery of physics concepts through self directedness in formative testing with feedback, a choice of learning activities, and multiple forms of support. This study was conducted within two sections of honors physics at a privat e high school (N=24). Students learning activity choices, time investments, and perceptions (assessed through a post survey) were tracked and analyzed. SDL readiness was linked to success in mastering physics c oncepts The three research questions pursued in this study were: What SDL acti vities did honors physics students choose in their self directed mastery learning environment ? How many students achieved concept mastery and how did they spend their time? Did successful and unsuccessful students perceive the self directed mastery learning environment differently? Only seven of 24 students were successful in passing the similar concept based unit test s within four tries, and these seven students were separat ed into a successful
12 group and the other 17 into an unsuccessful group. Differences between the two groups were analyzed. A profile of a self directed secondary honors physics student emerged. A successful self directed student invested more time learning from activities rather than simply completing them, focused on learning concepts more than rote operations, intentionally selected activities to fill in gaps of knowledge and practice concepts, actively constructed knowledge into a cognitive framework, engaged in academic discourse with instructor and peers as they made repeated attempts to master content and pass the test given constructive feedback, used a wide variety of learning resources, and managed their workload to meet deadlines. This capstone study found that parallel instruction in content and SDL skills could be important for improving learning outcomes and better equipping secondary honors physics students for college and life in general. Mastery learning principles coupled with modeling in self direction appear mutually reinforcing and, when more explicitly approached, should yield dual benefits in concept mastery as well as self efficacy.
13 CHAPTER 1 INTRODUCTION Although science education is considered critical to national security and economic vitality, much evidence points to unremarkable performance of American students in science (Gates, 2007; Hart & Rudman, 2001). Our success as a nation depends on strengthening Americas role as the worlds engine of discovery and innovation" ( Wh itehouse Press 2010). The United States received an average rating for science in the 2009 Program of Inter national Assessment (PISA, 2009). The 2005 National Assessment of Educational Progress (NAEP) science section showed 46% of American 12th grader s do not meet the basic threshold for science, 18% are "proficient," and only 2% are advanced (NAEP, 2005). The National Science Foundation ranked the United States 73 out of 91 countries surveyed in the fraction of its college students obtaining bach elors degrees in science and engineering (NSF, 2006). Only 17% of American students chose such careers, compared to 52% of Chinese, 41% of Korean, and 38% of Taiwanese college graduates (NSF, 2006). In 2003, China graduated 5.8 times as many engineering students as the U.S. (NSF, 2006). American universities bemoan the preparation of incoming freshmen in science and other subjects (Report of the Academic Competitiveness Council, 2007). To foster learning of physics in particular, this c apstone projec t revived a triedand proven instructional strategy mastery learning (Bloom, 1968; Guskey, 2010) in a self directed and technology enabled secondary honors physics course. Self directed learning (SDL) was incorporated because it is r ecommended as an i mportant life skill and can contribute to more meaningful learning as learners follow their interests, learning styles, and skill sets (Abdullah, 2001; Partnership for 21stcentury skills, 2009).
14 The focus of the project was to determine the types of learning activities and support structures learners preferred in their self directed learning of a secondary physics unit on conservation of energy and analyze how learning activities and support structures impacted concept mastery. The corresponding research questions (RQs) are listed and then detailed below: RQ #1: What SDL acti vities did honors physics students choose in their self directed mastery learning environment ? RQ #2: How many students achieved concept mastery when SDL activities were infused into the honors physics classroom for a given time period, and how did they spend their time? RQ #3: Did successful and unsuccessful students perceive the self directed mastery learning environment differently? RQ1: What SDL activities did honors physics students choose in their self directed mastery learning environment? The SDL aspect was designed and integrated in the form of learning contracts called VOLOs ( Vocabulary Objectives Learning Options ) which provided freedom to choose among a limited number of learning pathways. Learning contracts provide a way to develop individualized skills and knowledge while respecting differences in learning styles, readiness, and interests (Guglielmino, 2000). The three types of support investigated were oneto one t utoring with the instructor (also called expert tutor ) peer instruction, and computer tutorial support. The focus on tutor support whether from an expert, a peer, or a computer was an outgrowth of mastery learning emphasis on tutors (Bloom, 1968). Bl oom felt that the ultimate learning environment was oneto one expert tutor ing, and in such an environment he believed every student could reach concept mastery given sufficient time, feedback, and support (Bloom, 1968). More recent work in mastery learni ng gives evidence that computer tutor support improved learning outcomes as effectively as an expert tutor (Morote &
15 Pritchard, 2009; Ogilivie, 2001). To address this question, data collection included VOLO choices and also daily Time Tickets (Appendix A) that tracked how the student allocated their time among the SDL activities meaning learning activities and three support types RQ 2 : How many students achieved concept mastery when SDL activities were infused into the honors physics classroom for a given time period, and how did they spend their time? The most important form of feedback in mastery learning is formative testing that serves as a guiding beacon for the efforts of lear ners and tutors (Bloom, 1984). At the end of the four week learning unit, the students who had passed the unit test with at least an 80% given four tries were separated into the successful group. Choices and time investment in SDL activities of successful versus unsuccessful students were compared. RQ 3 : Did successful and unsuccessful students perceive the self directed mastery learning environment differently? A post survey (Appendix B) probed student perceptions of SDL, attitudes and approaches to learning physics, and views on the capstones self directed mastery learni ng environment The survey instrument was developed from three validated instruments Guglielminos (1977) Self Directed Learning Readiness Scale (SDLRS), Hallouns (2001) Views About Sciences Survey (VASS), and the Colorado Learning Attitudes About Science Survey (CLASS, 2006) as well as custom questions about the capstones course design. Responses were compared between successful and unsuccessful groups to determine differences in SDL readiness, problem solving strategies and development of expert thinking about physics, and preferences in learning environment. Statistical significance of the
16 comparison was determined using the MannWhitney U Test due to the small sample size (N=24) and nonnormal distribution of responses. To summarize, this capston e project sought to improve l earning outcomes in a secondary physics course by incorporating self directed mastery learning elements in a technology rich environment. This was a year long effort and the capstone focused on one four week unit, the conserv ation of energy unit. The SDL activities of students that were/were not successful in mastering concepts were compared. The determinant of concept mastery was passing the unit test with an 80% within four tries, given constructive feedback and correctives. A post survey was administered to determine perceptions of SDL, attitudes and approaches to learning physics, and opinions about the capstones self directed mastery learning environment. To develop a profile of a self directed secondary honors physic s student, SDL activity choices, time allocations, and survey responses between successful and unsuccessful students were compared.
17 CHAPTER 2 LITERATURE REVIEW Figure 21 presents an image of the capstone niche that merges four conceptual domains and preludes this literature review. These four conceptual domains are Physics Education Research (PER), Learning for Mastery (LF M), Self Directed Learning and Technology. PER is a well developed branch of science education driven by the availability of pedagogical assessment tools that provide essential feedback on instructional innovation (Hake, 2007). The review of rich PER literature highly influenced this work to blend traditional instruction with peer interaction, hands on inquiry, and technologies suc h as animations. Figure 21 illustrates the influence of LFM and SDL on the capstone design. The LFM aspect was incorporated in many forms of feedback and support, a flexible timeline, and multiple tries to reach at least an 80% on the unit test. Student s were encouraged to work oneto one with the instructor to review tests and glean feedback to overcome their misconceptions. As mentioned, SDL was fostered by the use of VOLOs, plus the allocation of approximately 75% of class time to independent work. Technology was integral to the capstone design both from infrastructure and educational technology perspectives. The literature review of technology encompasses Web 2.0, Google Apps, computer tut ors, animations and simulations and blended learning. A Go ogle Apps infrastructure was essential to smooth operation of the capstones learning environment. Through Google Apps, students were editors of the class website (a Web 2.0 feature), which enabled them to manage their personal web pages containing thei r VOLOs and electronic file cabinets. Google Apps also provided a content management system and enabled easy communication between instructor -
18 student and student student. Google Forms was used to capture daily Time Ticket data Computer tutors were inc orporated because they served as a primary means of self directed learning and provided another feedback source for students Animations and simulations were used to focus the learners attention on the science phenomenon of interest and allow manipulation of variables for causeeffect analysis (Wieman, 2008). Finally, this capstones learning environment can be classified as a blended learning environment because it employed approximately 25% face to face (FTF) a nd 75% SDL, which was primarily online work The literature review thus considers the emerging understanding of blended learning models and considers literature recommendations in the creation of FTF and SDL activities Physics Education Research Over three decades of substantive and extensive res earch correlating pre/post validated test scores with various instructional strategies in mostly college physics courses provided great support for this capstone. Physics education reform has outpaced other scientific disciplines largely because Halloun and Hestenes (1985) developed the Force Concept Inventory (F CI) assessment tool which was designed to determine conceptual understanding without formal training in mechanics (Hestenes, 1995). The widespread adoption of the FCI as a comparative tool has allowed physics instructors everywhere to assess their classroom performance relative to an average and receive feedback on their adaptations in pedagogy (Hake, 2007). A bedrock PER belief is that "incorrect science concepts [are] tenacious and resistant t o extinction," according to University of North Carolina's Joel Mintzes (2007, p. 366). Carrying these misconceptions forward results in rote equationgrabbing approaches rather than in reflective problem solving. It is difficult to foster conceptual
19 cha nge through traditional instruction, and one important role of the tutor, whether human or computer, is to highlight the discrepancy and create conceptual conflict (Ozmen, 2008). Table 21 summarizes noteworthy PER contributions that relate to this capst one project, including computer tutorials, modeling, interactive engagement, active learning, and online lectures. The highlights include the role of interactive engagement and breaking away from traditional lecture ( deemed largely ineff ective by Crouch, 2001; Hake, 1998; Mazur, 1997; Meltzer, 2002; Reddish, 1999). As a result of such research some major universit ies have transformed large lecture halls int o engaging, small group problem solving sessions through interactive quiz and response systems (Crouch, 2001; Mazur, 1997). Given the small class sizes in this capston e and the predominance of smallgroup work, interactive engagement is integrated into the capstone. Another highlight is the role of active learning and hands on experiments Elaborate designs have emerged over the years at MIT, University of Washington, North Carolina State University, and other universities (Beichner, 2000; Breslow, 2007; McDermott, 2002). While t he capstone does not employ such elaborate learning spaces it does incorporate animations collaborative learning, hands on experiments, and a commitment to SDL Mastery Learning Mastery learning defined as a feedback corrective response instructional strategy based on form ative testing gained prominence over four decades, peaking in the 1970s. Few strategies have been implemented more broadly or evaluated more thoroughly (Guskey, 2010). For instance, Whiting (1995) compiled 18 years of data gathered from more than 7000 high school students and showed mastery learning
20 yielde d not only higher test scores and GPAs but also better attitudes towards school (Guskey, 2007). Exam scores, retention rates, and student attitude evaluations were superior in mastery learning to those in traditional classrooms, with learning gains measur ing up to the 1 sigma range (Guskey, 2007). On a typical bell curve distribution, a 1 sigma learning gain corresponds to the midpoint moving from 50% to 84%. Bloom believed time not aptitude, is the determinant of performance (Bloom, 1968). Aptitude t ests simply indicate how much time will be required for the student to reach mastery. Mastery learning theorists suggest that rather than holding instructional time constant and allowing achievement to vary (as in traditional instruction), achievement level should be held constant and time allowed to vary (Slavin, 1987 A p. 197). Figure 22 illustrates the formative testing and feedback loops of LFM as well as the branching that occurs as a dvanced students enter enrichment activities while lagging stud ents lay firmer foundational knowledge. Interestingly, LFM research indicates classrooms that start off staggered due to readiness differences, but emphasize correctives, yield powerful benefits as lagging students catch up (Guskey, 2007). Bloom (1968) promoted the personal expert tutor as the ideal learning environment. An expert tutor enabled the average student to outperform 98% of nontutored students, a 2sigma improvement, yet was expensive to execute (Bloom, 1984). Bloom sought which critical el ements in oneto one tutoring could be transferred to wholeclass settings. Bloom focused on assessments being prescriptive and on implementing feedback corrective cycles (Bloom, 1984). In later work (Table 22), Bloom sought additive effects by combining mastery techniques with other reasonable
21 adjustments a classroom educator could make, such as reviewing prior material one week before starting new material (a 1.2sigma gain), increased classroom participation (1 sigma), cooperative learning (0.8sigma) graded homework (0.8sigma), study skill instructi on (1sigma), and improved time on task (1 sigma) (Bloom, 1984). Several of these interventions are incorporated into the capstone design including oneto one tutor, computer tutor, review, improved time on task, mastery learning, classroom participation, cooperative learning, and graded homework. Bloom believed the bell curve was the greatest disservice of the American educational system and that 90% of students could master material if the educator ada pted methods and materials and gave them enough time, just as a personal tutor would do (Bloom, 1968). An unexpected outcome of mastery learning was the cooperative learning environment it fostered through its peer support system, which was enhanced by the fact that high grades were no longer a scarce commodity and that all could earn an A (Bloom, 1978). Another interesting outcome was the similarity of learning rates that occurred over time in the mastery learning groups versus the control groups. All c lassrooms began with an estimated 5:1 ratio in rate of learning, meaning top students finished material five times faster than slower students, which practically disappeared in the mastery classroom but became more pronounced in control classrooms (Bloom, 1978). Bloom attributed this to increased self confidence resulting from academic success and the fact that constant feedback enabled students to develop self efficacy sooner (Bloom, 1978). Fred Keller of Arizona State Universit y was a contemporary of Bloom's. He designed the Personalized System of Instruction (PSI) and saw excellent results (Keller,
22 1968). Because PSI involved feedback with cor rectives (from peer, not expert tutors) and frequent assessments, it was also a form of mastery learning, and i n fact during the mastery learning heyday, twothirds of all published mastery learning case studies were PSI initiatives (Kulik, Kulik & Bangert Drowns, 1990). Modern PSI involves computer tutors that support multiple attempts to master the same test (Cracolice & Roth, 1996; Eyre, 2007). In contrast, Bloom's approach, based on wholeclass instruction followed by the expert tutor crafting an individual corrective (or enrichment) pathway for each learner, was called Learning for Mastery or LFM, and is mor e closely associated in modern terms with differentiated instruction, Response to Intervention, and Understanding by Design ( Guskey, 2010). Commonalities across these instructional strategies include bedrock highquality group instruction, regular formati ve assessments, individualized corrective instruction, and enrichment activities for advanced learners while slower learners catch up (Guskey, 2010; Tomlinson & McTighe, 2006). For all of the successes of mastery learning, today the term rarely appears in the literature, common only in pharmacy and law schools (Eyre, 2007; Lockman, 2008). There may be many reasons for this including the repackaging just mentioned. Eyre (2007) attributes inertia and cost efficiencies of the lecture system, which discourag e any change from traditional instruction. Another key factor cited is the daunting workload of the first year of mastery implementation, including the generation of worksheets to accompany texts, multiple forms of quizzes and tests, increased tutoring hours, and classroom management complexities of asynchronous work (Eyre, 2007). Guskey (2007) states that students are more actively engaged in mastery learning
23 classrooms and that a firmer foundation is laid early such that more rapid progress is possible t han for the majority of students in traditional classrooms. Self Directed Learning Self direction is r ecommended as an important life skill to be fostered by K 12 science education (Partnership for 21stcentury skills, 2009). A self directed learner has a desire to learn and the motivation to persist in the learning process ( Guglielmino 1989 ; Taylor, 1995). Self discipline is coupled with basic study skills and timemanagement skills to support the learners motivation (Taylor, 1995). Granting the self di rected learner greater autonomy can contribute to more meaningful learning as the learner chooses learning strategies that interest them and complement their skill set (Abdullah, 2001). Some studies indicate greater retention for active and self directed learning (Dori, 2007). The first SDL literature appeared in 1967 when Alan Tough discussed self teaching as adults decided what, when, and how to learn (Tough, 1967, p.3 ). Knowles (1975) extended these ideas with rationale, stating SDL was beneficial because proactive learners retained more than reactive learners, which was a mark of maturity, and this cultivated the initiative essential to educational success. The goals of SDL are realized when learners think for themselves in a systematic way and construct their own perspective (Candy, 1991). SDL helps students learn how to learn and challenges them to challenge themselves (Gibbons, 2002). Relating SDL to science education, engaging learners in scientific inquiry helps them experience the scientifi c process (not just facts) with its strengths and limitations (van Joolingen, 2007). Inquiry learning is a favored instructional strategy because it simulates real world challenges and develops necessary life skills such as finding creative alternatives, persisting, using logic, and
24 negotiating ones view ( van Joolingen, 2007). Research shows that scienti researchers experience research as a complex, cyclic and iterative process that often involves argumentation to produce scientific knowledge (van Rens, 2010). One of the most important tasks of the SDL instructor is raising student awareness of their responsibility to manage their own learning (Idros, 2010; Taylor, 1995). Other contrasts between SDL and traditional, teacher directed learning are highlighted in Table 23 (Gibbons, 2002). Table 23 shows a gradual release of control from the instructor to the student between teacher directed and student directed learning. In SDL, the student increasingly manages setting personal learning goals, processes, and even assessments (Abdullah, 2001; Gibbons, 2002). As learner s as su me greater responsibility to diagnose t heir learning needs, strategize on how to acquire t he necessary knowledge, and set personal learning goals including timelines, they are developing critical career survival skills ( Guglielmino, 2000). Bolhuis (2003) describes how a teacher must strike a balance between content oriented and process oriented instruction (process oriented instruction relates to SDL skill development). The balance involves providing content instruction in a positive emotional climate wi thin a social context in parallel with pushing students a little outside their comfort zone for self direction (Bolhuis, 2003). The capstones learning unit provided an SDL environment but did not incorporate explicit instruction in SDL strategies; however, future iterations should, as will be discussed in Chapter 5, Future Work. Interestingly, although SDL is inherently individualistic, it requires increased collaboration between the student, the teacher and peers (Abdullah, 2001; Hogg, 2008). SDL has roots in social constructivist theory (Gibbons, 2002; Idros, 2010). Instructors
25 support SDL through constructive discourse (Her bal Eisenman & Breyfogle, 2005) and by modeling problem solving methods and Learning to Learn strategies (Pape, 2000). Teachers can create envi ronments in which students do not fear making a mistake, and collaboration is frequent so socio cognitive construction of knowledge is possible (Buschman, 2003). By incorporating more real world aspects into the learning environment, the instr uctor can lend significance and meaning to the learning process ( Bransford, 2000). Another important environmental factor that the instructor can use to support SDL is offering a choice of learning activities. Learning contracts allow learner s to choose t heir pathway to acquire needed skills and knowledge while respecting individual differences in learning styles, readiness, and interests. The pertinent questions that the learner should ask (initially under the guidance of the instructor) when developing a learning contract include ( Guglielmino, 2000 ) : What do I need to be able to do? What resources will I use to learn this? What is the targeted completion date? How will I measure that I have attained the goal? The VOLO structure of the capstone and m ultiple forms of support provided capstone students with choice. A validated SDL assessment tool developed in 1997 by Guglielmino is called the Self Directed Learning Readiness Scale (SDLRS). Over 150 studies have employed the SDLRS, mostly in adult set tings (Chou, 2 008). The SDLRS uses a 58item Likert scale analysis along with eight factors to assess readiness (Guglielmino, 1989): o penness to learning opportunities self perception as an effective learner d riving initiative as an effective learner a cceptance of responsibility for ones own learning l ove of learning c reative spirit f uture orientation, and ability to use basic study and
26 problem solving skills The SDLRS is thought to be most appropriate for adult audiences with a higher educational status (Chou, 2008; Hendry, 2009). Questions from the SDLRS were selected and incorporated into the capstones post survey. Technology Jonassen (2003) identifies several ways technology can be thoughtfully integrated into the instructional desi gn, which is ideally anchored in a real world problem T hese technology tools include multimedia presentation of ideas information access simulations and animations to represent difficult concepts collaboration and communication tools The technology literature review focuses on four aspects relevant to the capstones self directed mastery learning environment: animations and simulations, blended learning, computer tutors, and Google Apps/Web 2.0. Animations and simulations were embedded in the VOLOs and funct ioned as learning tools. Blended learning is reviewed because the capstone learning environment was fundamentally a blended learning environment with 25% wholeclass face to face instruction and 75% SDL. Computer tutors functioned as the third major supp ort for students in combination with expert and peer support. Finally, Google Apps/Web 2.0 is reviewed because the technology infrastructure for the capstone learning environment was based on a Google Apps for Education platform. Animations and Simulati ons in Science Researchbased computer animations such as PhET ( Physical Educational Technology ) allow the learner to interact with parameters and help to create conceptual conflict (Ozmen, 2008; Wieman, 2008). Bransford (2000) states that i dentifying a misconception is the beginning of learning and animations are powerful tools to surprise the user with an unexpected outcome. L earners must make a conscious effort
27 to recompose their thinking as they manipulate parameters and engage in causeeffect analysis (Wieman, 2008). Wieman (2008) perceives animations as more focused and filtered than real world demonstrations, which eas es the learner's cognitive load. PhETs multiple representations are reinforcing (Wieman, 2008) and appeal to different intelligences (Bransford, 2000) The value of PhET is enhanced by the collaborative contribution of the PER community in providing supporting worksheets to guide explorations of the animations. Minds on Physics (MOPS ; http://www.physicsclassroom.com/mop/ ) is a tutorial system developed by the Physics Education Group at the University of Mas sachusetts. MOPS is built on constructivist principles and is interactive and increases in complexity, requiring effort on the learners part. The learner climbs ten (usually) levels per learning unit and gains a sense of achievement as well as understanding in the climb. Each learning unit is focused on real world scenarios and asks conceptual and quantitative questions. I f a user misses more than two questions, he or she drops down to the beginning and must restart the climb (Gerace, 1999). Blake & Scanlon (2007) consider features for effective use of simulations in distance science educat ion. Specifically, effective simu lations use multiple representations, such as graphs and images, and are based on real data. Help is integrated and tiered according to the ability level of the user. This latter recommendation concurs with cognitive load considerations described by Clar k (2005) and van Merrienboer ( 2003). Blended Learning Educationally useful research on blended learning needs to focus on the relationships between the different modes of learning (for example, faceto face and online) and especially on the nature of thei r integration. In
28 particular, such research needs to generate usable evidence about the quality of the students' learning experiences and learning outcomes. In turn, this demands appropriately powerful methodologies, rooted in a firm theoretical foundati on. (Bliuc, 2007) This capstone project seeks to address Bliuc's recommendation above to optimize the design of the blend with respect to learning outcomes based on a firm conceptual foundation. The capstones self directed and supported mastery learning environment is a blended learning environment, with intentional integration between FTF and online components. The U.S. Department of Educations (2009) metaanalysis of 51 studies comparing online, FTF, and blended learning found classes with online lear ning (either completely online or blended) outperformed classes based solely on FTF instruction. The mean effect size for all 51 contrasts was +0.24, p<0.001. If only blended courses were compared to FTF courses, the effect size was even more significant +0.35, p<0.0001. Blends of online and FTF instruction, on average, had stronger learning outcomes than did FTF instruction alone (US DOE, 2009). Koohang (2009) has developed a learner centered model for higher education and considers the instructional d esign of the blend. Under FTF activities, he groups lectures, individual/group discussions, labs, presentations, and assess ments. Under online activities he groups individual learning activities, collaborative learning activities, and assessments. He det ails the form and function of each of these activities and explains how each contributes to the elements of constructivism. Online learning provides an ideal space for self paced, constructivist learning (Koohang, 2009). Combining the work of Koohang and B liuc with the latest US DOE report (2009), Handbook of Blended Learning (2006), and Bates book (2003), it is apparent that
29 tremendous opportunity exists for research and development in the blended learning arena. As Bates (2003) states, it is relatively easy to find the technology skill set, but to couple this with instructional design talent is a rare find. It was hoped that the capstone project could link learning outcomes with blended learning design, in particular relating VOLO activities and support choices with conceptual development Rapid growth in blended learning challenges instructors and students to view education differently. Bates and Poole (2003) state the design of blended courses challenges instructors to reevaluate the use of class time. Students must assume more responsibility for their learning, and instructors must evolve their role to becoming a facilitator of learning rather than a transmitter of knowledge. At the University of Central Florida, 87% of instructors indicated that th ey have changed their approach to teaching as a result of their online teaching experience (Bonk, 2006). They became more responsive to student needs, changed their course development and delivery methods, incorporated more technology into teaching, and used more resources (Bonk, 2006). This viewpoint is pertinent to the mastery learning environment, in which instructors serve as expert tutors and instructional designers, providing feedback and correctives to individualize instruction. In her open access document "Lessons learned: Findings from MIT initiatives in educational technology (20002005)," Teaching and Learning Lab Director Lori Breslow discusses the shift of responsibility to the learner in an active learning classroom: The shift from passiv e recipient of information to active user requires a major re conceptualization of who is responsible for what, under what circumstances, and to what end. This kind of substantial redefinition of roles takes both time and effort. The ability of education al technology to deliver more information, in more forms, anytime and anywhere is going to
30 be one of the prime movers in redefining that relationship whether students and their teachers are ready for it or not. (Breslow, 2007, p. 293) Technology learning curves are an additional factor to consider when transitioning to an active and self directed learning environment. Instructors must offer more structure and guidance, especially in the beginning. Skill & Young (2002) present a hybrid model that captures the best of the physical and virtual worlds P owerful FTF teaching is coupled with online content richness and interactivity. Online work empowers the individual and emphasizes time on task activities such as virtual teamwork and interactive simulations Applying Skill & Youngs ideas to this capstone project, an optimal blend was sought between FTF /online interactions (student with expert/peer/computer), wholeclass, and self paced VOLO activities. Time allocation was tracked for a better sense of individual time investment in SDL activities. Computer Tutors Some scholars assert that simulations and computer based models are the most powerful resources for the advancement and application of mathematics and science since the origins of mathematical modeling during the Renaissance. ( Bransford 2000, p.215) Computer tutors offer sever al advantages over human tutors. Besides 24x7 access, computer tutors provide immediate feedback a huge advantage in the learning process (Chickering & Gamson, 1991). Feedb ack is essential, in fact, "meaningful feedback is likely the most powerful tool that can be used to improve performance" (Anglin, 2008). A lthough the feedback is only as valuabl e as the background programming, i ncreasingly more sophisticated programming will one day allow computer tutors to better reflect expert reasoning and pedagogy, becoming a virtual extension of the expert tutor. Computer tutors are "amazingly patient" (Littlejohn, 2002)
31 and can increase time on task. Time on task is closely related to learning (Cavanaugh, 2009; Morote, n.d.). Computer tutors are comparatively inexpensive. For instance, the annual registration for masteringphysics wa s $20 for the capstones physics student s. Computer tutors offer animations and simulations that a h uman cannot equal. The computer allows scaling down of inquiry tasks to reduce cognitive load (van Joolingen, 2007). Wong (2001) found frequent computerized formative assessments stimulated students to work harder. Keller's Personalized System of Instruc tion (1968) had a self paced structure that adapts well to computer tutorial environments (Cracolice, 1996; Eyre, 2007; Gagne, 1998). Some disadvant ages of computer tutors include the following: f eedback is never original (granted, artificial intelligenc e is advancing); IT skills and cognitive load considerations are important, but difficult to program (Breslow, 2007; Littlejohn, 2002; Park, 2009); and c omputer tutors can only simulate the warmth and connection of another human. As far as design of computer tutors, Mayer's multimedia strategies (2005) speak of the need for an uncluttered learni ng space with graphics and text or graphics and narration in proximity. Navigation bars are doubly effective in clearly setting expectations and organizing c ontent (Chickering & Gamson, 1991) and preventing less technically savvy students from getting lost in descending web pages (Clark ,2005; van Merrienboer, 2003). Van Joolingen (2007) and Chang (2008) investigated different aspects of computer support of self di rected inquiry learning in science. Van Joolingen (2007) sought to optimize the balance between preserving inquiry as the process of learning science with the conflicting need to provide learners with some framework.
32 When supported computer environments f ailed to yield cognitive gains, van Joolingen (2007) attributed this to either insufficient time for knowledge building or a mis match of support structure to the knowledgebuilding process Interaction is a desirable feature for tutorial systems because un derstanding is more likely to come if learners are actively engaged (Littlejohn, 2002). The term "interactive" requires some clarification and should involve students manipulating parameters to be truly effective (Littlejohn, 2002; Wieman 2007). Park's (2009) work with presenting physics concepts to 5thgrade students showed that students with limited prior knowl edge had difficulty with highly interactive simulations, but high prior knowledge students did not. Park (2009) recommends adaptive simulations that stage according to cognitive load. "Novices without guidance quickly become disoriented" (Morote, n.d., p 14). MIT's Pritchard created masteringphysics based on mastery learning principles (hence the name) (Morote & Pritchard, 2009). He wanted to capture the large amounts of time students invested doing homework and use that time to build accurate conceptual models. Pritchard states that only 60% of students answer a physics problem right the first time, but with the masteringphysics tutorial sys tem, 94% of students achieve mastery (Pritchard, 2010). The computer tutorial system incorporates scaffolding and Socratic questions, and on average 15% learning gains are realized with each successive hint. Pritchard's work demonstrates 2sigma improvem ent in final exam performance for those students who choose to use masteringphysics over those who do not, which equals learning gains associated with a personal tutor (Morote &
33 Pritchard, 2009). M asteringphysics served as the primary computer tutor in the capstone study. In one study Ogilvie (2001) integrated masteringphysics tutorial system, written homework, small group tutoring, collaborative complex problem solving, and instructional videos and correlated their usage with FCI performance gains Ma steringphysics generated double the FCI perfor mance gains of written homework. Collaborative complex problem solving contributed to more modest learning gains, and the latter two were unremarkable in their effects. Ogilvie (2001) concluded that masteringphysics was more than twice as effective as written homework in improving students' conceptual understanding. An interesting study at Carnegie Mellon's HumanCompu ter Interaction Institute shows that the ACT Programming Tutor generated 1.76sigma gains over traditional instruction. The ACT tutor incorporated three aspects: modeling (0.75sigma), individualized feedback (0.89sigma), and scaffolding (0.42 sigma). Arguably, this is on par with an expert human tutor (Corbett, 2001). Another commercial system is ALEKS which adjusts questions according to student readiness and periodically checks for retention (Eyre, 2007). Google Apps /Web 2.0 Google Apps i s a cloud computing environment hosted on Google server farms using Python programming language and Bigtable distributed storage systems, and in these early stages is offered free to schools and universities (Hayes, 2008). As a cloud application, Google Apps provides schools and universities with virtually unlimited remote server capacity with outsourced maintenance and constant software development (Hill, 2009). Getting more functionality for less money is irresistible for not
34 only funding strapped educational institutions, but American corporations as well. Gartner Group's Vice President Darr yl Plummer estimates that $8 out of every $10 spent on technologies in corporations goes to maintenance rather than innovation (King, 2009). Merrill Lynch estimates that within the next five years the annual global market for cloud computing will surge to $95 billion (King, 2009). In a May 2008 report, Merrill Lynch estimated that 12% of the worldwide software market would go to cloud computing in that period (King, 2009). Microsoft has made cloud computing one of five priorities for fiscal 2009, accordi ng to a memo from CEO Steve Ballmer (King, 2009). "The focus of innovation indeed seems to be ascending into the clouds" (Hayes, 2008). Google Apps is uniquely positioned for educational sectors in relation to alternative cloud vendors such as IBM, Dell Amazon, and Hewlett Packard (Hill 2009). While Google Apps is viewed as an evolutionary step towards efficiency and scalability, some question whether privacy and security concerns are sufficiently addressed (Hayes, 2008). From the student perspective, the free Education Edition of Google Apps provides learning tools such as Gmail, word processing/spreadsheet/presentation software, calendar, chat, and video/photo processing (Oishi, 2007). From an instructor perspective, the Google Apps ecology reduc es "transactional distance" (Moore, 1992) via its improved communication and collaboration tools. In addition, links to computer test generators and grading systems recoup potential time expended in mundane grading tasks and potentially free the instructor to spend more time with students In one study at Minnesota State University (N=73), computer assisted grading yielded a 200% efficiency gain over hand grading, a 300% efficiency gain over hand grading with
35 rubrics, and 350% efficiency gain over typing feedback into the Learning Management System, with no attenuation of student satisfaction with respect to feedback quality (Anglin, 2008). Integrating homework tutorial systems such as masteringphysics into the learning environment extends the potential tutor list to include the computer tutor, and extends the workday to after school hours. From a management perspective, schools, universities, and classrooms register their private domain with Google to install the Education Edition of Apps. This provi des a secure environment with administrator functions for user management, firewall, email filtering functions, and more (Oishi, 2007). Google Apps provides the collaborative framework necessary for mastery learning of secondary science. This technology framework facilitates the essential mastery learning components of frequent feedback, correctives, and assessments. Google Apps enables the instructor to have increased personal contact with students and increased efficiency by several means. Google Apps provides Gmail, chat, video chat, c ollaborative calendaring and documents (word processing, spreadsheet, and presentation) c ontent management e asy linkage to web tutorials, video lectures, and assessment tools (i.e., test generators with automatic grading) and domain management and control tools. Web 2.0 is a polymorphic term coined by publisher and champion Tim O'Reilly in 2006, and its range of applications includes collaboration and communication tools such as blogs and wikis. The core idea behind Web 2.0 is that "the web and its structure should support people in their work; it should be a medium for communication between people" (Rollett 2007, p. 95). Web 2.0 technologies, including Google Apps,
36 empower the instructor to meld web resources, inc luding RSS feeds, to class websites (Rollett, 2007). "Many members of the Web 2.0 community regard it as their moral right to remix content, which of course flies in the face of traditional business models based on restricted access to intellectual proper ty" (Rollett, 2007, p. 97). The Open Access movement, such as MIT's Open Courseware (http://mitworld.mit.edu), Creative Commons (http://creativecommons.org), and Science Commons (http://sciencecommons.org), empower instructors to combine materials from different sources (Rollett, 2007). The capstone environment embeds many open resources in the customized framework of the class website. Within the connected Web 2.0 learning environment, instructional strategies give way to learning environments, beca use just as Friedman (2008) expressed so influentially, the world is flat, and that includes a flattening of education. Traditional, vertical, transmission modes of learning are being supplanted by horizontal, networked, learning ecologies (Downes, 2005; Siemens, 2008). Summary of Literature Review The capstone focus on self directed and supported mastery learning evolved through extensive literature review and was anchored around Blooms landmark article on mastery learning (Bloom, 1968). Decades o f research lend support for mastery instructional practices to shift the bell curve and enable more students to master concepts given sufficient time and feedback with correctives (Bloom, 1968; Gagne, 1988; Guskey, 2007; Kulik, Kulik & Bangert Drowns, 1990). The prevailing and resonating themes from the literature that drove the capstone project were 1) traditional teacher led instruction is largely ineffective (Crouch, 2001; McDermott, 2001; Meltzer & Manivannan, 2002); 2) blended learning is more effec tive than either faceto face or
37 wholly online instruction (US DOE, 2009); 3) self directed learning, like mastery learning, is interactive and involves the gradual release of control from the instructor to the student (Abdullah, 2001; Gibbons, 2002); 4) i nteractive engagement and peer instruction have strong PER literature support (Crouch & Mazur, 2001; Hake, 1998, 2007; Meltzer & Manivannan, 2002); 5) research based computer tutors such as masteringphysics produce learning gains equivalent to an expert t utor (Morote & Prit chard, 2010; Ogilvie, 2001); and 6) decades of educational technology research by MITs Teaching and Learning Lab have produced impressive learning gains in physics and other branches of science (Breslow, 2007; Dori, 2007; Lipson, 2001). The result of these extensive and sometimes disparate readings was a coalescence of self directed learning, mastery learning, guided inquiry, educational technology, and blended learning ideas into the form of a self directed learning contract called the VOLO in conjunction with multiple formative tests to serve as the core of the capstone learning environment, more fully described in Chapter 3. Chapter 4 will present results and Chapter 5 will discuss the findings and conclusions.
38 Table 21 Collectio n of noteworthy instructional interventions within physics education research (PER) literature Category Instructional Strategy Synopsis and Capstone Integration Reference Mastery learning using computer tutors masteringphysics Computer tutors provide feedback and Socratic questioning necessary to build problem solving skills in physics students and are available 24x7. Learning gains measured in 2 sigma range. In addition, computer tutors provide feedback to instructors, highlighting areas in which to focus instruction. This capstone integrates masteringphysics as the primary computer tutor. Morote & Pritchard, 2009; Ogilvie, 2001; Warnakulasooriy a, 2005 University of Colorado at Boulder's PhET project, MIT's PIVoT and TEAL projects Well designed interactive simulations can be an engaging and effective tool for learning physics, especially when scaffolded by semi structured activity plus feedback and assessment. This capstone integrates PhET. Wieman, 2008; Breslow, 2007 Interactive simulations Minds on Physics (MOPS) curriculum developed by the Physics Education Research Group at the University of Massachusetts is built on constructivist principles and is interactive and increases in complexity, requiring effort on the learner's part. Gerace, 1999 Modeling Teaching methodology that rests heavily on pedagogical expertise As few as seven models, for instance conservation of energy, can unify physics instruction and organize developing schema. Students are familiarized with the models, then eval uate new problems and apply the appropriate model to the problem. High school physics teacher Malcolm Wells developed the modeling method of instruction and his students achieved FCI performance equivalent to Harvard physics students. Modeling is highl y regarded within the Physics Education community, but is not emphasized within this capstone project. It is mentioned here because the same Arizona State PER group (headed by Hestenes) that created modeling also created the FCI assessment and because it is a future emphasis for the capstone site. Wells, Hestenes & Swackhamer, 1995
39 Table 21. Continued Category Instructional Strategy Synopsis and Capstone Integration Reference Interactive engagement ConcepTests, peer instruction, and response systems Small group tutorials and problem solving Active learning is sought even in large lectures. Techniques include short 15 minute lectures interspersed with short ConcepTests, which challenge students to apply concepts collaboratively, cast votes for best answer, and then receive immediate feedback. Such interaction guides the content of the ensuing lecture, as the instructor focuses on common misconceptions. Students learn very little from traditional lectures.... Students develop complex reasoning skill s most effectively when actively engaged with the material they are studying (Crouch & Mazur 2001, p. 970). Given the small class sizes in this capstone and the predominance of small group work, interactive engagement is integrated. Noteworthy investig ations include the impact of peer instruction (Mazur, 1997; Crouch & Mazur 2001), interactive engagement (Hake, 1998; Meltzer, 2002; Reddish, 1999), small tutorial groups (Hake, 1998), and cooperative problem solving (Heller, 1992). Active learning Labor atory explorations may be digital simulations, or simple to elaborate hands on exercises. The TEAL project at MIT doubled test performance in an electromagnetism course, but required years to perfect. Specially designed learning spaces allow nine students to sit at a round table with a PC on which two and three dimensional visualizations and simulated laboratory data are generated. The capstone does not employ such carefully designed and orchestrated learning spaces as TEAL, but approaches something simi lar using open resources and hands on experiments. Breslow, 2007
40 Table 22. Learning gains cited in literature for various educational interventions Intervention * Interventions addressed in this capstone. Sigma Change Reference One to one tutor* 2 Bloom, 1984 Computer tutor* 1.76, 2 Corbett, 2007; Morote & Pritchard, 2009 ; Ogilvie, 2001 Reviewing prior material* 1.2 Bloom, 1984 Study skill instruction 1 Bloom, 1984 Improved time on task* 1 Bloom, 1984 Mastery learning* 0.5 1 Bloom, 1984 ; Kulik, Kulik Bangert Drowns, 1990 Classroom participation* 1 Bloom, 1984 Cooperative learning* 0.5 0.8 Bloom, 1984; Johnson, Johnson & Stanne, 2000 Graded homework* 0.8 Bloom, 1984 Table 23. Shifts in thinking between teacher directed and student directed learning (Gibbons, 2002) In teacher directed learning, the teacher In student directed learning, the teacher Decides the course goals and the content to be studied Teaches students to set their own goals and eventually choose what they will st udy Presents course content to students in lessons Teaches students the goals and processes involved in setting goals, making plans, and initiating action Sets exercises and assignments for study Negotiates student proposals for learning and acting Moni tors completion and assesses accuracy of student work Guides students through self directed challenge activities Tests and grades student performance Review s students assessment of their work
41 Figure 21. The literature review of Chapter 2 explores these four conceptual areas : P hysics E ducation R esearch (PER), L earning for M astery (LFM), S elf D irected L earning (SDL), and T echnology. Figure 22. Mastery learning is a feedback corrective response instructional strategy that uses formative assessm ents to guide learning.
42 CHAPTER 3 RESEARCH DESIGN AND METHODOLOGY The first chapter presented the challenge of improving American science education in the context of an increasingly competitive, technology driven world (Friedman, 2008). This is an over whelming problem, but an achievable first step in its solution is to improve science education one classroom at a time. An instructor could pursue myriad angles in the pursuit of better learning outcomes, and a literaturebased rigor with careful reflecti on drove this author towards the self directed mastery learning approach. The broad problem of improving science education then focused on an instructional design challenge to create a supported and assessment based self directed mastery lear ning environment. This caps tone study was conducted over four week s in two honors physics classrooms in a private high school in Fort Myers, FL (N=24). The course design entailed approximately 25% FTF/ wholeclass format of traditional lecture, problem solving practic e, and laboratory experiments, and 75% SDL rendered through a Google Apps infrastructure. The overarching goal was to build mastery of conservation of energy concepts by incorporating SDL such that the instructor could spend more time in oneto one tutor ing while also providing supplemental peer and computer tutors During the self directed portion of the study, students were given three required activities and told to choose two more learning activities from a limited array presented as the VOLO The V OLO activities were managed by the student within the class website. The SDL choices of students, i.e. their time investment in learning activities and support structures, were tracked through daily Time Tickets, which were
43 rendered using Google Forms (A ppendix A) and fed directly into a spreadsheet. Computer tutor time was logged by the masteringphysics system. Figure 31 provides an overall view of the capstones self directed and supported mastery learning environment. The overall technology infrastructure was Google Apps, which provided improved communication between instructor, learner, and peers. Within Google Apps, each learner maintained a personal web page (box on the left), which contained his or her VOLOs, blogs, and e portfolio. The VOLO appears as the tic tac toe box in the lower left corner of Figure 31 The three tutor types are represented in the figure (note computer tutors are embedded in the VOLO box). Weekly assessments were part of the mastery learning process, which was a cycle of formative assessments with associated feedback and correctives. Context The two sections of honors physics (N = 24) were 50/50 male/fem ale and 20% minority. The minority students were of Asian descent. Less than 10% of the students were on economic scholarship. Student learning styles were determined before the capstone study using the Visual Auditory Read/Write Kinesthetic (VARK) Survey ( http://www.vark learn.com/english/page.asp?p=younger ). VARK results indicate 33% of the students were kinesthetic learners, 25% were auditory, 24% were visual, and 17% were read/writeoriented. Classroom resources included a tech nology lab equipped with 16 networked PCs and a science lab. On a typical day students chose between the classroom or these other two locations to work. Ever y student also had a PC or a personal laptop at home. Every student owned a scientific calculator and had purchased a masteringphysics registration at the start of the school year, which functioned as the homework tutorial
44 system and served as the primary "computer tutor." The class textbook was Physics: Principles and Problems (2009). Columbus, OH: McGraw Hill. The class website was http://tinyurl.com/ecsphysics and can be viewed through username capstone and password capstone Research Design The central problem for this study was to investigate w hether self directed mastery learning improved conceptual understanding of physics. The central research questions of the capstone study were: RQ #1 What SDL acti vities did honors physics students choose in their self directed mastery learning environm ent ? RQ #2 How many students achieved concept mastery when SDL activities were infused into the honors physics classroom for a given time period, and how did they spend their time? RQ #3 Did successful and unsuccessful students perceive the self di rected mastery learning environment differently? These three research questions are aligned with corresponding data sources and analyses that answer the questions, as shown in Table 31. RQ # 1 data collection and analysis: The first question about SDL activities that students chose was addressed using VOLO selection and Time Ticket data. Bar charts of these data revealed the most popular and time intensive activities/supports. RQ # 2 data collection and analysis: The second question separated the students into two groups, successful and unsuccessful, according to whether they had cleared the unit test with at least an 80% given up to four tries during the timeframe of
45 the learning unit The SDL activities of successful versus unsuccessful students were compared on a time basis and this data was presented in a bar chart and data table. RQ # 3 data collection and analysis: The final research question about student perceptions was addressed using post survey data that was Likert style as well as freeresponse. Three categories of questions were evaluated: SDL readiness, basic problem solving and development of expert thinking about physics, and perceptions of the self directed mastery learning environment. The average Likert scores were reviewed and the Mann Whitney U test was run to determine statistical significance of the differences in responses between successful and unsuccessful students. These results were summarized in Tables 42 and 4 3. Pilot Study Several implementation decisions were made as a result of a pilot study conducted during the first semester of the year long project. In the pilot study, students average test scores were higher on the second attempt of the test (typically a 20% gain) but then went down on the third and fourth tries. Students were strongly encouraged to retake tests until they reached an 80%, but it was not mandated because the instructor felt compelled to move forward with the majority Students could be content with their grade and stop trying. The challeng e remained to help the struggling students achieve mastery in the capstone study, which occurred early in Semester 2. The capstone project incorporated several new elements based on pilot study feedback and the need to collect more data: The percentage of whole class faceto face time was increased. Based on student feedback from Semester 1, students desired more exposure to example problem solving in a wholeclass setting. In Semester 2, the percentage of wholeclass
46 instruction was increased to approxim ately 25%. More classes began with brief (1015 minutes) wholeclass instruction, which provided an opportunity to help with challenging problems from the homework set and allowed a second and third presentation of key concepts. Students filled out Time Tickets each day. Time tracking by students occurred through Time Tickets that were rendered using Google Forms and automatically fed into a Google Spreadsheet ( Appendix A ). Students specified how many minutes they engaged in each VOLO activity each day and how many minutes they used the three tutor supports Knowing minutes of each SDL activity informed the instructor about which activities were most/least time intensive and popular. It was possible the successful students needed little support and the unsuccessful students needed much support, such that an inverse relationship would emerge or perhaps a positive relationship or perhaps none at all. Also, since the purpose of the VOLO design was to free the instructor for more oneto one help with students, it was important to determine if that goal was achieved. A student post survey was added. Students were surveyed at the conclusion of the learning unit (Appendix B ). The goal of the survey was to appraise student attitudes about SDL, LFM, and lea rning physics. Infrastructure Google Apps platform The design of the self directed and supported mastery learning environment began in summer 2010 with the creation of a dedicated Google Apps for Education domain for the capstone classroom ( http://tinyurl.com/ecsphysics ). This freed the instructor from internal IT restrictions and provided the essential Web 2.0
47 communication (Gmail, chat, Groups) and collaboration (Groups, calendar) tools. It also provided site management, security, word processing, presentation, spreadsheet, drawing, imaging, video, and work flow and content m anagement functions. The instructor created a class roster, gave each student a Gmail account, and gave the students editor status on the class website. The instructor then created a personal page for each student and embedded a file cabinet on their page for their eportfolio items. One of the students first class assignments wa s to personalize their web page and create a physics learning blog and embed it on their page. Students were also required to take the VARK (visual auditory read/write kinesthetic) learning style survey and embed their results on their personal page f or future reference. Figure 32 is a snapshot of an indivi dual student page. Notice the prominent VOLO boxes, which are the student's learning contracts, and which are detailed below. The instructor ported over the previous course website and established a navigation bar to link to each learning unit. VOLOs VOL Os provided a differentiated pathway to learning by allowing students to choose their learning pathway according to their individual learning styles and interests. The VOLO contained eight learning activities and one wildcard activity that permitted the s tudent to propose how he/she woul d learn the concepts (Figure 33 ). Some VOLO tasks were required and are hig hlighted in Figure 33 specifically the Pendulum Box Bash Lab, masteringphysics and PhET Energy Skatepark. A great deal of thought and care was taken in the selection of VOLO activities. The instructor was cognizant of appealing to different l earning styles and also applied years of experience with PER resources. Helpful tools for finding qualified resources included such excellent websites
48 as http://merlot.org and http://learningscience.org/ The VOLO grade was derived from this simple rubric: five tasks completed was a 100%, four units completed was an 85%, three units completed was a 60%, and two units completed was a 30%. Instructional Strategy The capstone crafted a four week instructional unit that blended wholeclass FTF and self paced online learning as shown in the unit schedule (Figure 34) Instruction began in a wholeclass setting and involved three days of conceptual/problem solving lecture, an inquiry activity about pendulums followed by the PBB Lab. The VOLO for the unit was then revealed (Figure 33 ), and students contracted with the instructor for the five tasks they would undertake over the next two weeks to master the concepts The contract consisted of students pasting the VOLO box onto their personal web page and indicating their choices. Wholeclass meetings interspersed the next two we eks of SDL work. The team based Rube Goldberg project consumed the third week of the unit (Figure 3 4). Bates & Poole's (2003) book on blended learning facilitated the instructional design process through a series of very practical questions, which provided a framework for complete development of the instructional strategy: What are the intended learning outcomes for the course? How will they be assessed? The o verall goal of the course was to prepare learners for college physics The l earning goals for t he conservation of energy unit were to use a model to relate work and energy ; c alculate kinetic, potential, and work energy ; i dentify how potential energy is stored in a pendulum and spring ; s how that conservation of energy laws hold in pendulums and springs ; s olve problems using conservation of energy ; d esign and build a Rube Goldberg machine that folds a piece of paper twice, incorporating at least
49 five energy conversions in the process ; d esign a virtual roller coaster that meets amusement park requirements ; i dentify cases in which conservation of energy laws do NOT apply. The s tandards addressed in the learning unit included S tandard Number 1.0 Mechanics: The student will investigate the laws and properties of mechanics ; i nvestigate the definitions of force, work, power, kinetic energy, and potential energy ; and analyze conservation of energy including friction Unit 11 assessments included m ultiple unit tests including a pre test the Rube Goldberg project and VOLO completion How much time each week do students have to devote to this course? How can the teaching best be organized over the weeks? Students had typically 52 minutes of class time each day. Because the materials and videos were ava ilable online, nothing precluded students from working outside of class. Figure 34 illustrates the four week cycle, which kicked off with three days FTF instruction that created a big picture view of the unit and introduced important concepts VOLO choices were revealed The three days also included the who le class pendulum inquiry activity followed by the Pendulum Box Bash Lab. Common activi ties during the SDL time were required activities such as masteringphysics and test review with the instructor. Whole class summaries and problem solving were common at the beginning of many class periods. Students were free to work individually or in groups. The Rube Goldberg project consumed the third week. Could technology enable achievement of learning outcomes? Technology enabled achievement of learning outcomes. Computers and a science lab were always available. The class website was the central hub of all class activity and provided a
50 24x7 repository and central communication portal Google Apps streamlined communication and collaboration and provided content management functions and office suite tools. The seamless integration of web resources was essential to computer tutor function Students completed daily Time Tickets. Student choice was preserved with the VOLO scheme (Figure 33). VOLO choices were blended with respect to technology or traditional and with respect to learning styles. Rather than the instructor making the technology choice for the student, the student chose his or her preferred learning pathway, with or without technology. What are th e main presentation requirements for the content? What media and technologies could best meet these presentation requirements? A class field trip to an amusement park preceded the learning unit and provided the real world framework for the conservation of energy investigation of roller coaster design. The Pre Test was administered on Day 1. The instructor gave a Powerpoint lecture that reviewed work and emphasized that work was a form of energy. To introduce the ideas of conservation of energy, Roller Coaster Tycoon was showcased as the unifying theme and as an illustration of various energy forms. Actual data collected during the field trip was used to compute kinetic and potential energies The instructor demonstrated how to determine the amount of potential energy at the top of the ramp and the amount of kinetic energy at the bottom, then suggested inbetween values could be determined using the Work Energy Theorem. This was reinforced by use of http://ph ysicsclassroom.com animations of roller coasters, which included bar charts indicating the kinetic, potential, and total energies at each moment.
51 On Day 2, after revi ew of the previous days energy conversion ideas, a pendulum demonstration with accompany ing virtual pendulum animation ( http://www.elmer.unibas.ch/pendulum/index.html ) illustrated how the pendulum alternated between 100% potential and 100% kinetic energy. The various factors that might influence the period of the pendulum were discussed. Giant pendulums were attached to the classroom ceiling and students were told to isolate those factors that influenced the period of the pendulum. At the conclusion of the class period, the ins tructor wrote students suggestions on the board, then revealed the pendulum equations. The instructor discussed the setup and purpose of the next days lab, the PBB Lab. VOLO contract choices were presented and clarified. On Day 3 students conducted the PBB Lab, developed at Ohio State University (Zou, 2000). The basic setup of the lab consisted of a sandbag pendulum striking an ordinary cardboard box. Students measured how high the sandbag was raised and how far the box traveled. The purpose of the lab was to consider energy losses and conditions in which conservation of energy laws do not apply A wholeclass discussion followed the lab. The ideas presented in the PBB Lab were embedded in the unit tests (a sample test is shown in Appendix C) Appr oximately 10 days of independent and small group work followed as students engaged in their VOLO work. The VOLOs provided a variety of technology rich, traditional, and hands on activities for students. Everyone had already completed one VOLO requirement namely the PBB Lab. The PhET Energy Skate Park animation was the second required activity and continued the roller coaster theme and challenged the students to manipulate variables to see the effects on energy types. The third required
52 activity was mast eringphysics which required approximately 106 minutes to complete and consisted of 19 problems involving work, kinetic energy, potential energy, springs, pendulums, skate boards, bullets, slides, amusement parks, and more. As for VOLO choices the learn er selected from an AP level lab which tested Hookes Law (a visual and kinesthetic choice), another pendulum lab which found the acceleration of gravity (a visual and kinesthetic choice), Roller Coaster Tycoon software design of a roller coaster with specified parameters (visual, auditory, and kinesthetic choice), a rather simple online KE PE quiz (read/write choice), and reading Chapter 11 and taking the associated standardized test (read/write choice). Approximately every week s tudents took unit tests until they achieved at least an 80%. Brief wholeclass meetings interspersed the SDL phase to address troublesome areas in the masteringphysics probl em set and to guide activities. For instance, a spring was suspended from a cross bar with a mass attache d. The instructor illustrated the periodic m otion of the spring mass system and how it alternated between 100% potential and 100% kinetic energy. These ideas were part of the masteringphysics homework assignment. Parameters were manipulated physically and virtually ( http://phet.colorado.edu/en/simulation/mass spring lab ) to demonstrate how changing the spring constant, mass and amplitude impacted the period. Students learned how to e xperimentally determine the spring constant. Students computed kinetic and potential energies for the spring. The Rube Goldberg project was a grand team effort that challenged the students to think about energy conversions and apply the principles learned with a hands on activity The Rube Goldberg challenge was to fold a piece of paper twice by the most
53 inef ficient means possible. The project was graded according to simple criteria: Did the team incorporate at least five energy transformations in the pr ocess of folding a piece of paper into quarters? Points were deducted if human intervention was necessary to complete the process. Points were also deducted for noncontribution, if necessary. Because the Rube Goldberg project did not differentiate betw een students (because all team members shared the same grade), the activity was not included in the data analysis. What kind of interaction will help students acquire the content and develop the skills for this course? What technologies will best facilita te these interactions? Students received support from instructor, peer, and computer tutors. Feedback was to be timely and individualized. Through the use of a Web 2.0 class website, which made students not only passive consumers but also authors and communicators, higher levels of engagement were expected (National Survey of Student Engagement, 2009). How will the student be assessed? The balance of the students course grade was 45% test, 45% VOLO completion, and 10% homework (submitting daily Time Tickets and post survey ). This overall assessment balanced concept mastery (test scores) with diligence (VOLOs and homework) and collaboration (Rube Goldberg project, PBB Lab). Data Collection Unit T ests Frequent formative assessments are central to mas tery learning philosophy, proved troublesome in Semester 1, and were streamlined in Semester 2. Unit tests were multiple paper versions that were hand graded (a sample test is shown in Appendix C) Unit tests were similar, not the same, and shortened so they could be
54 completed in 30 minutes, which ensured that the instructor could quickly grade and review them with the student. The limit on the number of test retakes was four. Students retained their best test score; it was not averaged with previous at tempts. The SDL structure provided the flexibility for students to retake a test during the regularly scheduled class period. SDL Activities Students were assessed on completion of VOLO components. A VOLO activity was complete when the student scored at least an 80% on the activity. The student needed to complete five VOLO activities to receive a 100% for the VOLO grade. Masteringphysics was a required activity and was automatical ly graded. One lab, the Pendulum Box Bash Lab, was required and was com pleted in small teams E ach member received the same grade when the report was submitted. The third required VOLO activity was the PhET Energy Skate Park simulation, completed on an individual basis using a worksheet. The remaining VOLO activities were according to student choice, and students communicated the completion of their work through their personal web pages. As indicated in Figure 33 the VOLO options for this unit were: Roller Coaster Tycoon, KE PE quiz, f ind g Lab, vi rtual spring lab, read C hapter 11 and answer standardized test questions, or a wildcard activity of their choosing (with permission). All VOLO activities were counted and also tracked according to minutes invested in the activity. In this way, the instructor could note t he most time intensive and popular activities. To identify which learning activities were most effective in building concept mastery, the SDL choices and time investment of successful and unsuccessful students were compared.
55 A student could choose to use expert, peer, or computer tutor support. The Time Ticket specified which types of support, if any, students used and for how many minutes. Computer tutor times (minutes spent using masteringphysics ) were logged by the system and accurately known. Peer a nd expert times were gathered through Time Tickets and were less accurate than computer tutor times. Post survey Student perceptions of the self directed mastery learning environment were captured through a post survey and analyzed (Appendix B ). The first part of the survey focused on SDL, atti tudes about learning physics, and learning environment preferences. The post survey was derived from three sources in addition to custom questions about the capstones learning environment Several questions were d erived from Guglielminos (1977) Self Directed Learning Readiness Scale (SDLRS), Hallouns (2001) Views About Sciences Survey (VASS), and the Colorado Learning Attitudes About Science Survey (CLASS, 2006) Example Likert scaled items included I cannot l earn physics if the teacher does not explain things well in class, I enjoy solving physics problems, and I do not invest time understanding the derivations of equations. The second part of the survey used the following four open response questions to explore preferences of lear ning activities and support structures particular to the capstones learning environment: 1. Three VOLO activities were required masteringphysics PhET Energy Park, and the Pendulum Box Bash experiment. Identify the other two V OLO activities that you chose, explain why you chose them, rate how much you enjoyed them, and rate their usefulness to your learning. Use the scale of 1 (low) to 5 (high). 2. Describe how accessible your teacher was in meeting your learning needs, and rate the quality of that support on a scale of 1 (low) to 5 (high).
56 3. Describe how accessible your peers were in meeting your learning needs, and rate the quality of that support on a scale of 1 (low) to 5 (high). 4. If you had a choice of two tools or resources to use to learn physics, what would you pick and why? Data Analysis Quantitative data sources for this capstone are listed in Table 31 and included unit test grades, Time Ticket data, masteringphysics times and scores, VOLO activities chosen and complete d, plus the post survey data. Data analysis was broken down according to the three research questions as follows: RQ #1: What SDL activities did honors physics students choose in their self directed mastery learning environment? Completed VOLO activities were tallied from student web pages, and the minutes invested per activity were tracked using Time Tickets. These summary data provided perspective of the most/least popular or time intensive activities Minutes invested with expert and peer tutor were t racked using Time Tickets and minutes invested with masteringphysics were logged by the system RQ #2: How many students achieved concept mastery when SDL activities were infused into the honors physics classroom for a given time period, and how did they spend their time? To determine the impact of SDL activity choices and time on concept mastery, the number of students who cleared the unit test given four tries was determined. Then the students time allocation per SDL activity data was separated int o two groups, successful and unsuccessful; sideby side comparisons were made and similarities and differences were noted. RQ #3: Did successful and unsuccessful students perceive the self directed mastery learning environment differently? The post survey was designed to probe students perceptions of the self directed mastery learning environment using Likert
57 ratings and freeresponse questions. A side by side comparison was made between the average Likert ratings of successful versus unsuccessful students and MannWhitney U tests were conducted to determine statistical significance. In addition, free responses were evaluated in successful versus unsuccessful groupings. Summary T o develop mastery of conservation of energy concepts for the 24 honors physi cs students at a private high school in Fort Myers, FL, this c apstone project revived a triedandproven instructional strategy mastery learning (Bloom, 1968; Guskey, 2010) in a self directed, technology enabled learning environment Self direction was incorporated because it is recommended as an important life skill and can contribute to more meaningful learning as learners follow their interests, learning styles, and skill set s (Abdullah, 2001; Partnership for 21stcentury skills, 2009). The focus of the project was to determine the types of learning activities and support structures learners preferred and analyze how SDL activities impacted concept mastery. Data collected included Time Tickets, masteringphysics time logs, VOLO choices, and post surv ey data. Data analysis pursued the nature of the relationship between SDL activities and concept mastery. Attitudes about SDL, learning physics, and the self directed mastery learning environment were assessed using a survey and perception differences be tween successful and unsuccessful students were explored.
58 Table 31. Data sources and an alyses that answer the research q uestions Research question Data Sources Data Analysis What SDL acti vities and support structures did honors physics students choose i n their self directed mastery learning environment ? Time Tickets specified time spent in each SDL activity Completed SDL activities records on personal web pages Detailed time data from masteringphysics computer tutor Average number of SDL activit ies completed Minutes invested per SDL activity Average educational and enjoyment ratings of VOLO activities How many students achieved concept mastery when SDL activities were infused into the honors physics classroom for a given time period, and how did they spend their time ? Number of students who achieved mastery level (80%) on unit test within four tries. Time Tickets specified time spent in each SDL activity for successful students Completed SDL activities were recorded on personal web pages for successful students Detailed time data from masteringphysics computer tutor for successful students Number of students who cleared unit test SDL activities completed for successful students Minutes invested per SDL activity for successful students Did successful and unsuccessful students perceive the self directed mastery learning environment differently? Student post survey Spreadsheet of average Likert responses of successful versus unsuccessful students per post survey question MannWhitney U Test of statistical significance of responses Favorite resource list of successful versus unsuccessful students Table of Likert ratings from successful and unsuccessful students of three support structures
59 Figure 31. Overview of the capston e design of a supported and assessment based self directed mastery learning env ironment.
60 Figure 32 Snapshot of a personal student web page in the class website.
61 Conservationof Energy Learning Unit 2 Tests: Rube Goldberg Project & Ch 11 Test required Rubric: 5 boxes = 100 4 boxes = 85 3 boxes = 60 2 boxes = 30 Figure 33 The VOLO for the conservation of energy unit shows three required activities PhET, masteringphysics and PBB Lab.
62 Figure 34 Instructional sequence during the four week unit. Note the time allocation of approximately 25% whole class versus 75% SDL.
63 C HAPTER 4 RESULTS The capstone results focus on SDL activities, defined as time allocation choices made by the student and include both VOLO activities and support structures. Three VOLO activities were mandatory masteringphysics, PBB Lab, and PhET Skate Park and the rest were elective. To reiterate, the three support structures were the instructor (expert tutor), masteringphysics (computer tutor), and the peer tutor. T he minutes invested per activity were tracked using Time Tickets with the exception of the computer tutor times, which were collected from the more reliable source of the masteringphysics time logs. The results for this chapter are organized according to the three research questions. SDL A cti vities Chosen and Completed (RQ #1) Completed SDL activities were tallied from student web pages and are displayed as a bar chart in Figure 41. All students participated in the two mandatory SDL activities and all students used peer and expert tutor support. T he most popular elective VOLO activities were Roller Coaster Tycoon (16 students chose), the KE PE quiz (15 chose), read Chapter 11 and take standardized test (6 chose), the PhET springs lab (3 ch ose) and the Find g experiment ( 4 c hose) According to the post survey free responses, t he top reasons stated (in rank order) for these selections were (1) quick and easy, (2) enjoyment, and (3) educational VO LO completion rates were good 20 of 24 students completed all five activities and received a 100% VOLO grade, 3 students completed 4 activities for an 85% grade, and 1 student completed 2 activities for a 30% grade. The wholeclass PBB Lab counted for a VOLO activity for every student. Table 4 1 associates the VOLO activities with students perceptions of their
64 educational and enjoyment value. The most educational activities were two labs with low participation (one and two students completed) and the most enjoyable activity was Roller Coaste r Tycoon (score of 4.3). Figure 42 presents SDL choices on a time basis and it is evident that minutes invested with peer tutors (203 minutes ) dwarfed everything else. The second most time intensive SDL activity was masteringphysics (118 minutes ), foll owed by time spent with the instructor (84 minutes ). Preparing and retaking tests consumed an average of 45 minutes per student, and PhET Skate Park and Roller Coaster Tycoon both consumed an average of 37 minutes per student. The Find g lab consumed an average of 21 minutes and all the other activities required less than 10 minutes to complete. Choices and Time Allocation of Successful Students (RQ #2) Students could retake the unit test up to four times during the learning unit in an effort to achi eve at least an 80%. At the close of the learning unit, seven students cleared the unit test, and 17 did not (Figure 4 3). It is interesting to note that one of the successful students was a minority student of Asian descent and two were on economic scho larship. Henceforth, the seven students who cleared the unit test are grouped as the successful students, and the 17 students who did not are grouped as the unsuccessful students. These two groups are compared with respect to their SDL choices, time allocation, and perceptions of the self directed mastery learning environment. Figure 44 is a repeat of Figure 42 (Minutes invested per SDL activity), but is split into successful and unsuccessful groups. It is clear that successful students invested an average of 25% more time than unsuccessful students did per SDL activity, and
65 particularly more time was invested in doing the masteringphysics homework assignment, retaking tests, and tutoring with the instructor. In fact, the only SDL activity in whic h successful students invested less time than unsuccessful students was in Roller Coaster Tycoon. Perception Differences of Successful versus Unsuccessful Students (RQ #3) Averages of post survey Likert responses were determined for successful and unsucc essful groups and are shown in Table 42; many differences are apparent. Medians could have been used to equal effect. To test whether the differences in perception were statistically significant, and because of the small sample sizes of successful (N=7) and unsuccessful (N=16) groups, the categorical data type, and nonnormal distribution of responses, the decision was made to use the MannWhitney U test for this analysis (Huck, 2008). A summary of the statistical outcomes appears in Table 4.3. The MannWhitney U test was conducted on the raw data of the two groups and flagged differences to Prompt s #8, #14, #27, and #30 as statistically significant. The Mann Whitney U test outcome for Prompt #8 was (U = 91, p = 0.0209). Prompt #8 was, I cannot learn physics if the teacher does not explain things well in class. Prompt #14 was, Understanding what I read is a problem for me (U = 89.5, p = 0.0278). Prompt #27 was, I prefer traditional lecture with a weekly lab compared to the self paced VOLO work (U = 90.5, p = 0.0232). Prompt #30 was, I am satisfied with my grade for this unit (U = 15.5, p = 0.0076). The null hypothesis was that the two groups had identical frequency distributions across the five possible Likert ratings. The null hypothesis was rejected because the means were significantly different between the two groups ( = 0.05).
66 Notable differences were observed between successful and unsuccessful groups in the areas of SDL readiness, problem solving ability, and alignment with expert thin king about physics. Successful students, without exception, scored higher in questions pertaining to SDL readiness and unsuccessful students expressed greater need for teacher direction (consider Prompts # 7, 8, 1016, 27, 31, and 32). Successful students accepted responsibility for their own learning more than unsuccessful students (#12, 4.0 vs. 3.4) and expressed a preference for a self paced VOLO learning environment whereas unsuccessful students were neutral (#27, 1.7 vs. 3.0). Successful students expressed moderate success in developing a learning strategy for the unit but unsuccessful students were less sure (#32, 3.1 vs. 2.2). Unsuccessful students strongly believed that they could not learn physics if the teacher did not explain things well in cla ss whereas successful students felt that way a little more than half the time; this was a significant statistical difference (#8, 3.4 vs. 4.2; U = 91, p = 0.0209). Successful students expressed greater initiative and self efficacy in their statement, If I experience a difficulty in studying physics, I try to figure it out on my own (#16, 3.4 vs. 2.0). To their credit, unsuccessful students said it was helpful for them to work many problems when learning physics (#3, 3.3 vs. 3.7). The groups were alike in expressing an awareness of whether they were learning something well or not (#13, 4.6 vs. 4.1); this survey prompt drew the single strongest response overall from the successful group. The groups were also alike in saying difficult study didnt bother them if they were interested in something (#11, 4.0 vs. 3.4). A cause of concern for these unsuccessful honors students was the statistically significant difference regarding unsuccessful
67 students having difficulty understanding what they read, which link s literacy to readiness (#14, 2.0 vs. 2.9, U = 89.5, p = 0.0278). Notable differences were evident between successful and unsuccessful students in the area of physics problem solving and alignment with expert views about physics (Prompts # 17, 17 20, 222 5). Successful students were more able to operate at the conceptual level rather than the operational level. Successful students said that more than half the time they would thoroughly analyze a few problems in detail (# 24, 3.4 vs. 3.1) and the derivati on of equations (#23, 2.7 vs. 3.1) to better understand physics. Successful students were more likely to approximate answers and assess their reasonableness (#2, 4.3 vs. 3.6). Successful students felt there was usually more than one approach to solving a physics problem (#18, 2.0 vs. 3.1). Unsuccessful students relied more heavily on equations than concepts (#25, 3.0 vs. 2.4) and felt they needed more problem solving practice (#3, 3.3 vs. 3.7). Unsuccessful students strongly believed the teacher must sh ow them sample problems like those on the test (#7, 3.7 vs. 4.3); this prompt elicited the single strongest response from the unsuccessful pool but was not statistically significant (U = 82.5, p = 0.0819). Successful students derived more enjoyment from problem solving (#22, 3.0 vs. 2.0). Both groups showed moderately strong need for collaborative learning (#17, 3.9 vs. 4.0). Successful and unsuccessful students were similar in their views about tutors (see Table 44). The average Likert scores for acces sibility and quality of the support provided by the expert tutor was (3.5 vs. 3.3 for successful vs. unsuccessful), peer tutor was (3.6 vs. 3.4), and computer tutor (masteringphysics) was (3.3 vs. 2.6). In the
68 free response section, students stated that accessibility was an advantage of peer tutors and quality was an advantage of the expert tutor. The post survey asked students, If you had a choice of two tools or resources to use to learn physics, what would you pick and why? Responses by successful and unsuccessful students are compared in Table 45 and appear similar. For both groups the number one preferred resource was masteringphysics (50% of successful students chose and 60% of unsuccessful students chose). Students stated masteringphysics w as valuable because it makes you practice concepts and forces me to learn to figure out what I did wrong so I can keep trying to get the right answer. Both groups chose instructor support as their second favorite resource (50% vs. 27%); however, unsuc cessful students also chose physicsclassroom.com as equally important (27%). Successful students also valued their textbook (34%). The final section of the survey asked students for feedback on the self directed mastery learning environment. The prompt w as, Comment on the mastery learning approach, which includes frequent tests with feedback from instructor and peers, followed by taking corrective actions before retesting. Do you learn more in this approach than in traditional courses? Overall, 19 students were positive to enthusiastic about LFM, two were neutral, and three were negative. Splitting responses between successful and unsuccessful groups, 86% of successful students were positive and 76% of unsuccessful students were positive. One success ful student was neutral. All negative opinions about the self directed mastery learning environment fell within the unsuccessful group. The most commonly stated reason for preferring LFM over traditional instruction was the retaking of tests because this gave the students a chance
69 to really learn concepts. Some selected successful student comments include: Yes I learn more I like this style and feedback on retaking my tests really helps; I like the mastery learning approach because you have to keep practicing until you understand the concept; and Its more hands on and I have to actually work to learn. One final comment from a successful student: I love having all the different resources. Having the book, masteringphysics, PhET, and other resources reinforces what I have already learned. I usually learn something different in every resource. In the traditional classroom you move on more quickly, but with mastery learning you can retake the tests until you completely understand every problem, if you want. Therefore I learn more from mastery learning. Of the neutral comments, one unsuccessful student said I feel like I learn the same amount, it is just performed using a different method. I would like a little more time in the classroom where the teacher could explain concepts. Of the negative comments, one unsuccessful student wrote Mastery learning approach = frustrating; frequent tests = frustrating. Another criticism was, I do not like the approach. The approach may have been more ef fective if the students knew how to work the problems. One final criticism was I like being able to retake, but I always feel like Im scrambling before every test. Summary In conclusion, the two physics classes were split into successful and unsuccess ful groups according to whether the students cleared the unit test in four tries, and seven of 24 students accomplished this. Both groups of students were positive about the self directed mastery learning environment (86% of successful students were posit ive and 76% of unsuccessful students were positive). The main reason given for preferring the self directed mastery learning environment over traditional instruction was that formative
70 testing helped the student learn from their mistakes. The favorite resource of all students was the computer tutor, masteringphysics, followed by tutoring with the instructor. Students allocated twice the time to peers than anything else. Successful students invested an average of 25% more time than unsuccessful students in SDL activities, particularly in retesting, tutoring with the instructor, and doing homework. Many notable differences between the perceptions of successful and unsuccessful students were indicated by post survey results in the areas of SDL readiness, problem solving approaches, and alignment with expert views about physics. These comparisons between successful and unsuccessful students will be more thoroughly discussed in Chapter 5.
71 Table 41 VOLO activities selected by percentage, and average educat ional value and enjoyment scores % of students selecting this activity VOLO Activity Average Educational Value (1=low, 5=high) Average Enjoyment Score (1=low, 5=high) 100% Masteringphysics 2.8 2.8 100% PBB Lab 2.5 2.5 100% PhET Energy Skate Park 2. 5 2.5 58% KE PE Quiz 3.4 3.1 25% Outline chapter 3.4 3.0 4% PhET Spring Virtual Lab 4.5 3.5 50% Roller Coaster Tycoon virtual roller coaster 2.9 4.3 17% Extra p endulum experiment 4.2 4.2
72 Table 42. Comparison of post survey average Likert responses of successful versus unsuccessful students (N=24). Question # Successful Average Likert Score Unsuccessful Average Likert Score Question/Prompt # 1 2.0 2.4 A sig ni the information I need to know. # 2 4.3 3.6 When I am solving a physics problem, I try to decide what would be a reasonable value for the answer. # 3 3.3 3.7 It is useful for me to do lo ts and lots of problems when learning physics. # 4 3.1 2.0 I think about the physics I experience in everyday life. # 5 2.3 2.7 Knowledge in physics consists of many disconnected topics. # 6 4.0 2.9 When studying physics, I relate the important inform ation to what I already know rather than just memorizing it the way it is presented. # 7 3.7 4.3 To do well in this physics course, I need to see sample problems like those on the tests. # 8 3.4 4.2 I cannot learn physics if the teacher does not explain things well in class. # 9 3.7 3.1 I am developing general probl em solving techniques in this course. # 10 2.9 2.6 In a learning experience, I prefer to take part in deciding what will be learned and how rather than having the teacher tell me what to do. # 11 3.9 3.6 Difficult study doesn't bother me if I'm interested in something. # 12 4.0 3.4 No one but me is truly responsible for what I learn. # 13 4.6 4.1 I can tell whether I'm learning something well or not. # 14 2.0 2.9 Understanding what I re ad is a problem for me. # 15 3.1 2.6 If I discover a need for information that I don't have, I know where to go to get it. # 16 3.4 2.0 If I experience a difficulty while studying physics, I try to figure it out on my own. # 17 3.9 4.0 To understand phy sics I discuss it with friends and other students. # 18 2.0 3.1 There is usually only one correct approach to solving a physics problem. # 19 2.3 2.5 I do not expect physics equations to help my understanding of the ideas; equations are for doing calcul ations.
73 Table 42. Continued Question # Successful Average Likert Score Unsuccessful Average Likert Score Question/Prompt # 20 3.1 2.4 The first thing I do when solving a physics problem is make a drawing to represent it. # 21 3.9 3.1 I am not satis ed until I understand why something works the way it does. # 22 3.0 2.0 I enjoy solving physics problems. # 23 2.7 3.1 I do not invest time understanding the derivations of equations. # 24 3.4 3.1 I have found carefully analyzing only a few problems in detail is a good way for me to learn physics. # 25 3.0 2.4 It is possible to explain physics ideas without mathematical formulas. # 26 1.9 1.8 I prefer to learn from our Glencoe textbook rather than Internet resources. # 27 1.7 3.0 I prefer tradition al lecture with a weekly lab compared to the self paced VOLO work. # 28 2.7 2.7 I invested significant hours outside of class because I did not have sufficient time during class to learn this material. # 29 3.6 2.4 I have mastered the key ideas about co nservation of energy. # 30 4.0 2.2 I am satisfied with my grade for this unit. # 31 4.1 3.6 Retaking the unit test helped me figure out my w eak spots and focused my learning. # 32 3.1 2.2 I developed a learning strategy that worked for me for this unit. # 33 3.3 2.6 Masteringphysics was very helpful to my learning process. # 34 2.6 2.4 PhET Energy Skate Park was very useful to my learning process. # 35 2.9 2.3 The Pendulum Box Bash experiment was very useful to my learning process. # 36 3.6 3.2 Thi s unit had the right balance of whole class, small group, and individual learning activities. # 37 3.6 3.1 I had all the support I needed whether from the teacher, peers, or course resources to learn this material. # 38 3.4 3.4 Some class periods I a ccomplished very little. # 39 4.0 3.6 I have a very productive team in this course. # 40 2.3 2.3 I had a plan for what I wanted to accomplish each day BEFORE I entered the physics classroom.
74 Table 43. MannWhitney U test indicated post survey Likert r esponses were statistically different between successful and unsuccessful groups ( = 0.05) Individual survey questions indicated to have statistically different responses between successful/unsuccessful groups #8 (U = 91, p = 0.0209) #14 (U = 89.5, p = 0.0278) #27 (U = 90.5, p = 0.0232) #30 (U = 15.5, p = 0.0076) Advantages of statistical method Non parametric method applicable to ordinal data type with nonnormal distribution and small sample sizes Operates by comparing sum of ranks between two gro ups and determines lik elihood (p) of the differences (Huck, 2008). Limitations of statistical method None (Huck, 2008). H o REJECTED: T he average rank s of the successful and unsuccessful groups were the same. Table 44. Likert ratings by success ful and unsuccessful students of accessibility and quality of three tutor supports Tutor Type Successful Unsuccessful Expert 3.5 3.3 Peer 3.6 3.4 Computer 3.3 2.6
75 Table 45. Comparison of successful vs. unsuccessful group responses to the post surv ey question, If you had a choice of two tools or resources to use to learn physics, what would you pick and why? Group % that selected Tool/Resource SUCCESS FUL (N=7 ) 50 Masteringphysics 50 Instructor 34 Textbook 17 Physicsclassroom.com; peers; la b; internet UNSUC CESSFUL (N=16 ) 60 Masteringphysics 27 Instructor 27 Physicsclassroom.com 14 Textbook; internet 7 Rube Goldberg; practice quizzes; lab; retaking tests; peers
76 Figure 41. The tally of SDL activity choices indicates all students completed the required activities and all used expert and peer tutors. Figure 42. Average minutes invested per SDL activity per student. Time data based on Time Tickets and masteringphysics time logs
77 Figure 43. The number of students who cleared the unit test in four tries is shown. Seven students cleared the unit test within four tries and 17 students did not. Figure 44 SDL activities evaluated based on minutes invested by successful and unsuccessful groups. Overall, successful s tudents spent 25% more time on their activities than unsuccessful students did.
78 CHAPTER 5 FINDINGS, DISCUSSION, CONCLUSIONS, RECOM MENDATIONS, AND FUTURE WORK This chapter consists of six sections: Overview, Limitations, Findings, Discussion, Conclusions, and Recommendations and Future Work. The Overview and Limitations sections provide a condensed description of the design and implementation of the study in the light of its limitations and literature base. This section is followed by the Findings section in which the findings of Chapter 4 are summarized. The Discussion section ties the literature, research questions, and findings together, and Conclusions are drawn. This is followed by the Recommendations and Future Research section, which provides pr actical recommendations based on this research as well as ideas for future research efforts with a rationale for why these research areas are important. Overview of the Study This four week study captured data for one learning unit completed by all student s in two honors physics classrooms (N=24) in February 2011. The topic of the four week learning unit was conservation of energy. The major concepts presented were: Force, work, power, kinetic energy, and potential energy are related. The Law of Conservat ion of Energy relates different energy forms and some forms of energy, such as friction, are hard to determine. Pendulums and springs alternate between potential and kinetic energy with virtually no energy losses and provide good models for the study of conservation of energy. Roller coasters provide a real world application of conservation of energy because they alternate between potential and kinetic energy, but some energy is lost due to friction. It is nearly impossible to use conservation of energy law s to resolve inelastic collision problems; however, conservation of momentum laws may be used.
79 No training, logistical, nor technological challenges occurred. Wholeclass and self directed learning proceeded as envisioned and the VOLO activities were robust. Google Apps provided an infrastructure for the physics course and no technical issues occurred. Students p ersonal web pages enabled dynamic workflow management for the instructor Google Docs enabled collaborative development of lab reports. Two s ignificant whole class activities were important to concept development but were not included in the data analysis because the focus of this study was self directed learning. The two activities were the Rube Goldberg project and the Pendulum Box Bash Lab (PBB Lab). The Rube Goldberg project was a competition between the two physics classes and functioned as a wholeclass activity. The two teams greatly exceeded ex pectations for this project and it ran longer than expected (consumed a week). It is notewor thy that the requirements were to fold a piece of paper into quarters using at least five energy conversions Team 1 used 65 energy conversions and Team 2 used 50 energy conversions The competition became a well attended event and a video of it was feat ured on the schools morning announcements. The Rube Goldberg project received very favorable unsolicited comments in the post survey, such as Rube Goldberg 5 hands on learning, loved it! and The Rube Goldberg helped me to visualize physics. The second wholeclass activity that was important for conceptual development but not included in the capstone analysis was t he Pen dulum Box Bash Lab (PBB Lab), adopted from the work of Zou (2000) which involved 1.5 days of whole class hands on lab and discus sion. The core ideas presented in the lab were the conditions under which conservation of energy laws would NOT apply in contrast to conservation of
80 momentum laws that DO apply. Zou (2000) created test questions that tracked the same concept in different packaging and he analyzed how Ohio State University physics students applied their laboratory experience to new situations For this study, after three tries 82% of the students answered the PBB Lab questions correctly. Post survey responses gave the PBB Lab an educational value rating of 2.5 out of 5 on the Likert scale, which was one of the lowest scores (the only lower score being reading the Glencoe textbook, which scored 1.8 out of 5). This score translates to students feeling the PBB Lab was not very useful to their learning process, whereas the instructor viewed the embedded concept to be very important The first research question pursued in this capstone study was What SDL acti vities did honors physics students choose in their self directed mastery learning environment ? The SDL LFM instructional strategy was tied to physics concept development via the second research question: How many students achieved concept mastery when SDL activities were infused into the honors physics classroom for a given time period, and how did they spend their time? Finally, perception differences between successful and unsuccessful students were evaluated through the third research question: Did successful and unsuccessful students perceive the self directed mastery learning environment differently? Self direction was a focus because assuming responsibility for ones own learning is deemed an important college and career skill (Partnership for 21stcentury skills, 2006). Mastery learning was pursued becaus e it is a time tested method of achieving learning gains in the 0.5 1 sigma range, improving attitudes t owards school, and personalizing the learning experience (Bloom, 1968, 1984; Guskey, 2007, 2010;
81 Hymen, 2002; Kulik, Kulik, & Bangert Drowns, 1990) SD L and LFM are interrelated. LFM facilitate s SDL and concept development through feedback and prescriptive learning, and SDL facilitates LFM by freeing the instructor to spend more oneto one time with students. Mastery learning involves oneto one tutoring with the student (Bloom, 1968; Keller, 1968). To free the instructor for more oneto one time with students, VOLOs were developed and peer and computer tutors were added. During this time with students, the instructor used formative tests as the fou ndation for providing targeted feedback to the student, as well as making suggestions for corrective action. The student had up to four tries to pass the unit test and his/her best grade was retained. Frequent formative tests with feedback and correctives are the hallmark of mastery learning (Bloom, 1968, 1984; Keller, 1968). Self direction was accomplished through the VOLO structure, which provided a selection of learning activities that were tracked and analyzed. It was thought that the choice of lear ning activities would accommodate differences in learning styles, interests, and readiness as well as create a sense of ownership (Guglielmino, 2000; Tomlinson, 2003). The learning activity choices of successful and unsuccessful students were compared (suc cess was defined as scoring at least an 80% on the unit test within four tries.) Self directed learning environments require sufficient levels of support to be effective (Breslow, 2004; Chang, 2008; Hendry, 2009; Kirschner, 2006; Mayer, 2004; Taylor, 1995) Three types of support were available to students expert, peer, and computer tutors. The emphasis on tutors was consistent with the mastery learning methodology (Bloom, 1968, 1984). Students use of support structures was tracked by
82 minutes invested, and their opinions about the availability and quality of support were assessed through the post survey The time investment of successful students with tutors was compared to the time investment of unsuccessful students. By comparing the time allocation choice data of successful students and unsuccessful students, some conclusions were drawn about SDL effectiveness in general and particular SDL activities which yielded learning gains. Through a post survey, perception differences about SDL readiness, basic problem solving and development of expert thinking about physics, and opinions about the self directed mastery learning environment were explored between successful and unsuccessful groups. Coupling the perception data with the time allocation data all owed a preliminary profile to emerge of a self directed learner on the pathway to becoming an expert in physics. Limitations All data was collected as planned; however, Time Tickets were crucial to the analysis and their execution was somewhat flawed. Specifically, every student did not fill out a Time Ticket every day, because perhaps they were not near a computer at the close of class. Hence students sometimes filled out several Time Tickets retroactively, and this reduced the reliability of this instru ment with the exception of the computer tutor times that were accurately known. Also, three students filled out so few Time Tickets that they were not part of the time studies. Another limitation of the study was the inherent bias of the researcher due to the research being conducted within her classroom, and the fact that this was the second year that the instructor had taught these same students.
83 Findings RQ #1 Findings SDL Activities Chosen VOLO tallies determined that 96% of students completed at least four VOLO activities for the unit, and the most popular elective VOLO activity was Roller Coaster Tycoon, followed by the online KE PE quiz (Figures 41 & 4 2). On the basis of minutes invested in an SDL activity, peer tutoring consumed double the t ime of any other option (203 minutes ). Masteringphysics was a distant second in time consumed (118 minutes ) followed by instructor support (84 minutes ). Thus, the three support structures garnered the greatest time investment of the students. Retesting PhET Skate Park and Roller Coaster Tycoon each consumed approximately 40 minutes per student. Interpretation of this data must be tempered by the inconsistency with which students submitted their Time Tickets. Students allocated more time to required activities, whereas electives were often chosen based on the quick andeasy, or I thought it wouldbefun mindset, according to post survey responses. A clear cutoff wa s evident when one considers the minutes invested in electives. Roller Coaster Ty coon was the only elective that garnered much time investment from the students. The post survey comments conveyed the advantage of the peer tutor was availability whereas the advantage of the expert tutor was quality of feedback. For instance, one student wrot e about the peer tutor, [P eers] are good at giving answers but not too great at explaining. I usually let the teacher explain things to me. She has a very thorough knowledge of the material herself, so this helps. Another student commented, My peers are readily accessible but they do not know everything, which balanced the student who said, I wouldnt fully understand the concept when [the teacher] was finished explaining it. My peers were probably more helpful than the
84 teacher. I would say they were able to explain things better. Still another commented, My peers were very accessible; we worked together in class and worked outside of class through Skype several times. It proved very effective. We were able to support each other and t alk through difficult concepts. Regarding the expert tutor, one student said, I talked to her most of the time but sometimes shes a little busy and we cant get her attention. The compu ter tutor was cited frequently as a useful tool for learning phy sics: I like the way the information is presented and the hints work you through a problem instead of giving you the answer right away. In summary, the data supporting RQ #1 revealed students were skilled in attaining high VOLO grades, all completed t he required VOLO activities, and all accessed the three types of support. Elective activity choices generally reverted to those that were quick, easy, and fun. The greatest time investment of students was in peer support, which was double the time invest ed in masteringphysics, the next most timeconsuming activity. RQ #2 Findings Relationship between SDL Activities and Concept Mastery At the close of the unit, only seven students had cleared the unit test with at least an 80% given four tries and the average unit test score was a very poor 57%. The instructor did not consider the series of tests difficult (Appendix C shows a sample test). Care was taken to align test questions with masteringphysics (11 of 34 test questions came directly from this source), the PBB Lab PhET energy park, the pendulum experiment, and the textbook. The seven students who cleared the test were separated into the successful group and their time allocation choices and perceptions were evaluated separately from the unsuccessful students. When time investment in SDL activities of successful and unsuccessful groups was compared, it was evident that
85 successful students consistently invested more time in every activity, particularly their masteringphysics homework problem set (139 minutes vs. 97 minutes ) and tutoring with the instructor (97 minutes vs. 70 minutes ), for an overall average of 25% more time invested than their unsuccessful counterparts with the sole exception of the Roller Coaster Tycoon activity. Roller Coaster Tycoon was the fourth most significant time investment of unsuccessful students (26 minutes vs. 46 minutes ) and deemed the most enjoyable in the post survey. While unsuccessful students built virtual roller coasters with Tycoon, successful students invested time learning from their mistakes on formative tests (retesting was the fourth most significant time investment of successful students 66 minutes vs. 23 minutes ). Both groups invested the largest number of minutes in collaborative learning (211 minutes vs. 194 minutes ). To summarize, the data supporting RQ#2 revealed that successful students invested 25% more time in their learning activities than did unsuccessful students and triple the time in formative testing. In rank order, successful stu dents prioritized their time as follows: peer tutor, computer tutor, expert tutor, retesting, PhET Skate Park, find g lab, Tycoon, KE PE quiz. Unsuccessful students prioritized their time as follows: peer tutor, computer tutor, expert tutor, Tycoon, P hET Skate Park, retesting, find g lab, KE PE quiz. RQ #3 Findings Differences in Perception between Successful and Unsuccessful Students Patterns were sought in the areas of SDL readiness, basic problem solving and development of expert thinking about physics, and perceptions of the self directed mastery learning environment. Due to the small sample size (N=24), a factor analysis was not conducted for the post survey. Throughout this section, questions will be
86 referred to as #12 for Question #12 (s ee Table 42) and Likert scores will be shown as (5.0 vs. 4.3) for (mean of successful Likert ratings vs. mean of unsuccessful Likert ratings). The Likert scale ranged from 1 = I never feel this way to 5 = There are very few times when I dont feel thi s way. When a question had a statistically significant outcome as determined by the MannWhitney U Test ( = 0.05), this is noted in the narrative. SDL readiness Considering the prompts related to SDL readiness, one emergent theme was that unsuccessful students expected more direct instruction from the instructor. Unsuccessful students felt it was essential for the instructor to explain things well in class and to see sample problems exactly like those on the test. One of the students who gave a negativ e vote to LFM wrote, I prefer for the teacher to teach us everything we need to know. Unsuccessful students were less confident to resolve problems on their own and less convinced of their responsibility to manage their own learning. They were markedly disappointed in their inability to develop a successful learning strategy for the unit. Successful students, in contrast, embraced the freedom provided by the SDL framework and were pleased with their learning outcomes. They felt they were responsible t o manage their own learning so the instructor was less essential. They were more confident of their literacy skills and knew where to go to get information. Basic problem solving strategies and development of expert thinking about physics. Development o f expert thinking requires time and successful construction of knowledge ( Bransford, 2000). Ultimately, the expert becomes capable of pattern recognition and categorization of problems, which guides limited cognitive functions
87 towards problem resolution ( Bransford 2000; Gagne, 1988; Larkin, 1980). Novices, in contrast, remain tangled in the details, consuming their limited cognitive capacity with minutia, thereby making it difficult to elevate their thinking to solve the problem ( Bransford, 2000; Larkin, 1980). Physics experts often use drawings to compose their thinking and are more contemplative in using equations than a novice, operating at a more conceptual level (Halloun, 2001; Halloun & Hestenes, 1998). Notable differences emerged between successf ul and unsuccessful students in their perceptions about ways to solve physics problems. Post survey results indicated successful students showed evidence of expert thinking about physics whereas the unsuccessful students were lagging (Halloun, 2001; Hallou n & Hestenes, 1998). For instance, successful students gave a strong response to post survey Prompt #6, When studying physics, I relate the important information to what I already know rather than just memorizing it the way it is presented (#6, 4.0 vs. 2.9). This prompt related to the pattern formation that is characteristic of experts. On the flip side, unsuccessful students were more likely NOT to see patterns, as evidenced by their somewhat elevated response to the prompt, Knowledge in physics cons ists of many disconnected topics (#5, 2.3 vs. 2.7). Successful students were more likely to relate physics to their everyday world (#4, 3.1 vs. 2.0), which was again a testament to their identification of patterns and themes. Perceptions of the self di rected mastery learning environment The average Likert scores for successful and unsuccessful groups recorded in Table 42 for questions pertaining to the learning environment (Prompts # 26 40) appear similar with two notable exceptions. Prompt #27 cont rasted opinions between traditional to SDL,
88 specifically, I prefer traditional lecture with weekly lab compared to the self paced VOLO work. The average score of successful students was 1.7 (corresponding to I almost never feel this way) and the aver age score of unsuccessful students was 3.0 (corresponding to I feel this way half the time.) This was a statistically significant result (U = 90.5, p = 0.0232). Unsuccessful students did not feel they mastered the key ideas of the unit whereas success ful students did (#29, 3.6 vs. 2.4). Unsuccessful students were highly dissatisfied with their grades for the unit, in contrast to successful students (#30, 4.0 vs. 2.2); this was a very significant outcome (U = 15.5; p = 0.0076). Certain prompts elicit ed similar responses from successful and unsuccessful students. Both groups agreed they would rather use the Internet to learn than the textbook (#26, 1.9 vs. 1.8); they appreciated the opportunity to retake tests to focus their learning (#31, 4.1 vs. 3.6); they felt they had adequate support from teacher, peers and course resources (#37, 3.4 vs. 3.1); and they felt that the unit had the right blend of wholeclass, small group, and individual learning activities (#36, 3.6 vs. 3.2). Free response section Three specific areas were probed in the freeresponse section of the survey. The first profiled students opinions about the accessibility and quality of tutors, and this has already been presented in Table 44. Both groups were similar in their ratings of the three supports (an approximate score of 3). The second area probed was students favorite two resources to learn physics. As shown in Table 4 5, both groups valued masteringphysics most highly (50% vs. 60% mentioned), followed by instructor support (50% vs. 27% mentioned). It is relevant to contrast these responses with the time allocation data, which emphasized that the number one resource used by students was peer support, accessed twice as frequently as any other
89 resource. Perhaps time alloca ted to peer support wasnt always focused on learning physics, but time devoted to masteringphysics clearly was. The final freeresponse section of the survey provided students with an opportunity to review the mastery learning approach. Some 86% of succ essful students were positive about LFM, and 76% of unsuccessful students were positive. One successful student was neutral. All negative opinions about the self directed mastery learning environment fell within the unsuccessful group. The most commonly stated reason for preferring LFM over traditional instruction was the retaking of tests because this gave the students a chance to really learn concepts. Discussion The overall class atmosphere was positive collaborative, and industrious Only once in fo ur weeks did the instructor need to speak to a student to get to work. The VOLO structure worked well as an SDL framework with clear rubrics and due dates (Guglielmino, 2000) Students clearly understood they needed to complete five VOLO tasks to receive the full grade, and they usually chose a partner(s) to complete the VOLO task. Teachers can create supportive environments in which collaboration is frequent and socio cognitive construction of knowledge is possible (Buschman, 2003). Good instructional design encourages active learning. "Learning is not a spectator sport.... [Students] must talk about what they are learning, write about it, relate it to past experiences and apply it to their daily lives. They must make what they learn part of themselves" (Chickering & Gamson, p. 66). Mastery learning fosters a collaborative learning environment because it is based on expert and peer interaction and dispels with competition for high grades (Bloom, 1978). The mastery learning environment appeared a nurturing space for students to receive feedback and learn from their
90 mistakes in their efforts to achieve concept mastery; however, future efforts must drive towards greater learning outcomes in addition to positive perceptions. The most significant time investment, by a 2:1 margin, was time invested with peer tutors. It was unclear whether peer support always translated to physics help and dialog or whether some of that time was social (post survey results attributed less quality to peer support than expert support). Merrill & Gilbert (2008) found the best collaborative learning was problem centered. "Learning is promoted when learners are engaged in a task centered instructional strategy involving a progression of whole real world tasks" (Merrill & Gilbert, 2008, p. 200). The VOLO activities were clearly structured as to task but Merrill and Gilbert (2008) suggest students may have benefitted from more guidance as to how they should collaborate. For instance, Crouch and Mazurs (2001) concept tests in universi ty physics courses require students to convince their peers that they have the right solution to a problem. Merrill and Gilbert (2008) clarify the changing role of peers through the instructional process: Peer sharing in the activation phase, peer demons tration in the demonstration phase, peer collaboration in the application phase, and peer critique in the integration phase. This final phase of peer critique requires deeper processing for students to make their intent clear for collaborators (Merrill & Gilbert, 2001, p. 202). Specification of the nature of peer interactions may be an enhancement for future SDL course design. In general, students spent more time in required activities but enjoyed them less (# 41). The majority of students chose learning activities because they were quick and easy and fun, but then they could not pass the test. The successful students might
91 have chosen these same activities, but extracted learning in the process of doing them. For instance, one successful student c ommented, I always read at least part of a chapter, but usually all of it, because the book is easy to understand and provides a good introductory understanding of the concept. I built the [virtual] roller coaster because it was quick and easy to relate to after the Islands of Adventure trip. I took the KE PE quiz because working problems reinforces the concepts that I have already learned and checks my understanding. Successful students expressed metacognitive intention behind their choices, sometimes choosing a sixth activity to supplement their learning. Unsuccessful students, in contrast, seemed in a rush to complete their VOLO, unaware that this was their opportunity to learn the concepts. Successful students expressed the need to practice and appl y concepts to master them. Two issues with the VOLO structure were noted: 1) the selection of VOLO activities was critical and 2) some students hurried through a task to check it off the list, with little regard to the activitys learning potential. T he first issue naturally resolves over time as the instructor gains feedback about the learning content and student uptake of various activities. For instance, the capstone work clearly indicated the value of masteringphysics for this unit. As far as the second issue, the only way to change students behavior regarding rushing through SDL activities is to reinforce that they are responsible for their own learning. Idros (2010) and Taylor (1995) address this challenge and state that one of the most impor tant tasks of the SDL instructor is to raise student awareness of their responsibility to manage their own learning. Breslow (2004, 2007) and Gibbons (2002) point out that this shift in responsibility from the teacher to the student can raise discomfort f or both parties. Students, accustomed to passively sitting and receiving, are forced to manage their own learning, which Guglielmino (2000)
92 identifies as an important career survival skill. Instructors experience a loss of control that causes them discomf ort (Gibbons, 2002). Herbal Eisenman & Breyfogle ( 2005) and ( Pape, 2000) indicate that the transition in ownership can be smoothed through constructive discourse and modeling problem solving and Learning to Learn strategies. The majority of students were positive about the self directed mastery learning environment (#45, 86% vs. 76%). Research indicates several factors that were embedded in the capstone study that might have contributed to a positive learning climate: a choice of learning activity (Knowles 1975; Taylor, 1995; Tomlinson, 2003), peer interaction (Mazur, 2007; Merrill, 2008), personalized attention (Bloom, 1968, 1984; Chickering & Gamson, 1991), a lack of competition for high grades (Bloom, 1978; Chickering & Gamson, 1991), adequate support ( Kirschner, 2006), sufficient time to master concepts and complete the work (Bloom, 1968, 1984; Cavanaugh, 2009), a chance to learn from mistakes (Bloom, 1968, 1984), effective technology applications (Bates & Poole, 2003; Jonassen, 2003; Koohang, 2009), and hands on activities (Breslow, 2007; McDermott, 2001, 2002). The Google Apps infrastructure worked flawlessly and supported the SDLLFM learning environment. Regarding learning outcomes, the final average unit test score was very poor (57%) for the unit and only seven of 24 students had cleared the test. This was a disheartening outcome considering the effort made to create and manage the capstone course structure. A surprising epilogue to the capstone story followed the presentation of these results t o students after which they were given five days to prepare for an additional test. Everything changed. A remarkable 51% gain in test scores occurred,15 students cleared the test, and the class average jumped to 79% (Figure 5.1; gain was
93 calculated as [( Recent Score Previous Score) / (100 Previous Score)] ). Was this a grand example of procrastination? Four possible interpretations were (1) Yes, students procrastinated; (2) Students needed clearer feedback to determine what they needed to do to reach success; (3) SDL was not modus operandi and students still needed the teacher to drive their activities; and (4) Not enough time was allocated to the learning unit. The radical shift in test performance lent credence to Chickering and Gamsons (1991, p. 66) view that, "No feedback can occur without assessment. But assessment without timely feedback contributes little to learning." Feedback is vital to the learning process. Much of the current literature on discourse and metacognition is built on the pr inciple that feedback fuels the learning process (Cobb, Wood & Yackel, 1993; Herbal Eisenman & Breyfogle, 2005; Verschaffel,1999). Mandernach (2009) says the most important factor stimulating critical thinking is instructor involvement. Bloom (1968) desc ribes the expert tutor as a listener, who asks questions and represents concepts in a different way, trying to break through the learner's misconceptions or assist them in building schemas. "In other words, to decrease variation in results, the teacher must increase variation in their teaching" (Guskey, 2007, p.16). Guskey (2007) emphasizes early intervention yields the greatest gains: "The time used for correctives and enrichments in early units yields powerful benefits that later will make things easi er" (Guskey, 2007, p. 19). SDL does not translate to a lessened need for feedback but as Aubdullah (2001) and Hogg (2008) indicate is often associated with increased collaboration between the student, teacher, and peers. Timely and assessment based feedback should be an important component of future course designs.
94 The third explanation for the radical post study turnaround in test scores, specifically that students were not equipped or prepared for SDL and were accustomed to teacher direction, has merit. A gradual shift from teacher direction to student direction has been advocated by several authors (Bolhuis, 2003; Gibbons, 2002). Whereas the capstone study was preceded by a semester long pilot study, SDL is a characteristic of the learner and not the instructional strategy, thus learners differ in their capacity to assume control of their learning even after a semester of practice (Bolhuis, 2003). Teaching should provide students with constructive frictions that require them to exercise a little mor e self direction than they had in the past (Bolhuis, 2003, p. 339). Modeling is a method whereby the instructor makes the process of learning transparent for the learner and is a recommended technique for building SDL skills (Bolhuis, 2003; Verschaffel, 1 999). The fourth explanation regarding inadequate time points to the essential idea that time on task relates directly to learning (Cavanaugh, 2009; Chickering & Gamson, 1991). Time Ticket data revealed over the four week period the average student whether successful or unsuccessful invested only 23 hours in physics homework Focusing on allocation of class time, therefore, successful students allocated 25% more time to learning activities (633 minutes vs. 500 minutes). The emphasis on retesti ng may have been one of the most defining elements of success. As a percentage, retesting was the activity eliciting the greatest time difference between successful and unsuccessful students. On average, successful students invested almost triple the time in retesting than did unsuccessful students (66 minutes vs. 23 minutes ). In addition, expert tutor time was often associated with correcting and reviewing for tests, and
95 successful students spent more time with the expert tutor than unsuccessful students (97 minutes vs. 70 minutes). In the post survey, both successful and unsuccessful students acknowledged the importance of formative testing (see survey results under RQ #3, Question #31 average Likert score of 4.1 for successful students and 3.6 for unsuccessful students), but successful students took greater advantage of it and gleaned more feedback from the instructor. Successful students expressed greater metacognition and intention in time usage on the post survey; specifically, they targeted certain activities to provide practice with concepts and minimized time investment in less educational activities (Roller Coaster Tycoon). One unsuccessful capstone student, whose split of minutes for expert/peer/computer tutoring time was 10/260/90, analyzed that the reason she did not do well on the unit was because she did not work example problems with the teacher, which usually helped her a lot. These results are consistent with Abar & Lokens (2010) evaluation of 205 high school students. They found hi ghly self regulated students (as determined from measures of metacognition, effort, efficacy, and positive attribution, among others) studied more material for a longer time than less self regulated students. In addition, highly self regulated students had a mastery goal orientation and used appropriate resources to attain their goals (Abar & Loken, 2010). Did the SDL structure allow the instructor to engage in more oneto one tutoring? A review of Time Ticket data (Figure 42) shows that the expert tutor spent an average of 167 minutes per student over 10 class periods (four weeks total, minus four FTF days, minus five Rube Goldberg days, minus test taking days). This then reduces to an average of 16.7 minutes per student per day, which is not an impressi ve number except
96 when one considers a typical class period, if divided by the number of students, allowed approximately 4 minutes per student. The Time Ticket data was faulty, and these impressions require further investigation; however, it appeared that s tudents gleaned more personal time with the instructor and this concurred with the instructors view that lecture time was converted to oneto one interactions that were more targeted to the students needs. Conclusions A profile of a self directed second ary honors physics student emerged through this research. A successful self directed student invested more time learning from activities rather than simply completing them, intentionally selected activities to fill in gaps of knowledge and practice concepts, actively constructed knowledge into a cognitive framework, focused on learning concepts more than rote operations, engaged in academic discourse with instructor and peers as they made repeated attempts to master content and pass the test given construc tive feedback, used a wide variety of learning resources, and managed their workload to meet deadlines. Considering the interesting epilogue to the capstone story, i.e. given an extra week eight additional students cleared the unit test, one could conclude that SDL readiness was the characteristic lacking in the unsuccessful students. Perhaps their original poor test performances could be attributed to insufficient time and intention for schema development, or insufficient feedback, or both, which result ed in underdeveloped problem solving ability. Once they were given assessment based feedback, reviewed the material again with the instructor and prepared properly using their older tests, they were able to master the concepts. They always had the abilit y to master the concepts, but lacked the SDL skills and motivation to manage the learning process.
97 Knowles (1975) defines a self directed learner as an individual who takes initiative, with or without the help of others, diagnoses his/her learning needs, formulates goals, identifies resources, chooses and implements appropriate learning strategies, and evaluates learning outcomes. The profile that emerged from the capstone study is consistent with the SDL definition provided by Knowles. In this study, the seven successful students differed from the 17 unsuccessful students in SDL readiness according to post survey results. In Blooms (1968) terms, the unsuccessful students were still highly dependent upon the expert tutor, which translates to the instruc tor needing to modify instructional strategies to make better connections for the students. This is a teacher led view. Unsuccessful students agreed with statements such as, I cant understand physics if the teacher doesnt explain things well in class (#8, 3.4 vs. 4.2), in contrast to successful students who agreed, No one but me is truly responsible for what I learn (#12, 4.0 vs. 3.4). Unsuccessful students preferred traditional, teacher led instruction (#27, 1.7 vs. 3.0). By separating successful and unsuccessful students and evaluating their behaviors and perceptions, the instructor gained a clearer perspective about how to create, support, and sustain an effective self directed mastery learning classroom. Effective captures the goal of improv ed learning outcomes (and attitudes not a goal of this work but often a sideproduct of mastery learning (Bloom, 1978)). An instructor can take steps to foster SDL skills, problem solving ability, and expert thinking. From Blooms perspective (1968, 19 84), more time with the expert tutor, more feedback, and more correctives are recommended. Pape (2003) and Verschaffel (1999) make similar recommendations by promoting modeling with discourse, but they extend into self -
98 direction by recommending explicit i nstruction in Learning to Learn strategies as well. Because students differ in their readiness, a gradual transition is recommended (Abdullah, 2001; Bolhuis, 2003; Breslow, 2003; Gibbons, 2002). This parallel instruction in content and process is the fun damental recommendation emerging from this study. Mastery learning principles coupled with modeling in self direction appear mutually reinforcing and, when more explicitly approached, should yield dual benefits in concept mastery as well as self efficacy. Recommendations and Future Work This capstone study found that parallel instruction in content and SDL skills could be important for improving learning outcomes and better equipping secondary honors physics students for college and life in general. Self direction skills are important in a know ledgebased society and prepare students to persist, manage their time and resources, use logic to construct their knowledge, argue their views, collaborate, and hopefully, find meaning through the learning process (Partnership for 21stcentury skills, 2006). As a secondary physics student assumes greater responsibility to diagnose their learning needs, to strategize on how to acquire necessary knowledge, and to set personal learning goals, they are developing criti cal college and career survival skills supplemental to the mastery of foundational physics concepts (Guglielmino, 2000). Recommendation 1: As Bloom (1968) so eloquently conveyed, time is the variant to mastery, not ability. Successful students invested triple the time in formative testing versus unsuccessful students. Chickering and Gamson (1991, p. 66) said it well, "There is no substitute for time on task. Learning to use ones time well is critical for students and professionals alike. By choosing easier tasks and hurriedly completing them, unsuccessful honors students compromised their learning. The first important
99 parallel instruction centers on the importance of teaching time management skills with goal setting in addition to the regular curriculum, and helping students realize that schema development is a process and takes time (Bransford, 2000; Gagne, 1988). Recommendation 2: DeCorte (2000) states the degree of students self regulation correlates strongly with academic achievement. Capstone results indicated that successful students exhibited SDL readiness, while unsuccessful students showed less metacognition and commitment to the learning process (as evidenced by 25% less time investment and fewer attempts at retaking the unit test). The 1 7 unsuccessful honors students would have benefited from explicit instruction in SDL attitudes and behaviors. The core attitude must be that the student is ultimately responsible for and controls their learning (Bolhuis, 2003; Bransford, 2000; Breslow, 2003; Gibbons, 2003). Pape (2000) encourages instructors to explicitly teach Learning to Learn strategies, such as using feedback to grow, actively listening and reading, maintaining a success orientation, avoiding procrastination, and preparing for exams. To foster positive attitudes towards learning and motivation, Pape, Bell & Yetkin (2003) and Weiser (2005) promote the value of student journals and post test surveys to help students associate effort and disposition with grades. Verschaffel (1999) recomm ends that instructors model problem solving strategies for students and requires group practice as part of the process. The second important parallel instruction involves teaching Learning to Learn and other SDL strategy techniques in concert with content instruction. Recommendation 3: Bolhuis (2003) effectively distills these parallel instruction ideas into striking a balance between content oriented and process oriented instruction. Bolhuis (2003) states instruction will foster SDL if four principles ar e addressed. The
100 first is the need to maintain a healthy constructive friction and move gradually to SDL, because learners differ in readiness (p. 339). A continuum exists between teacher directed learning and self directed learning for each learner (B reslow, 2004; Gibbons, 2002). The second is to focus on building content knowledge by making the learning more meaningful, particularly through the use of problem based learning. Several authors recommend anchoring instruction around a real world problem because it establishes a framework and relevance for the learning (Jonassen, 2003; Merrill, 2007; Merrill & Gilbert, 2008). The third principle is to pay attention to the emotional aspects of learning by providing positive feedback, helping students to e mbrace uncertainty, fostering motivation, and modeling the reward of solving hard problems. The final principle is to treat learning as a social phenomenon and facilitate social skills and effective cooperative learning. Merrill and Gilbert (2008) emphas ize that learning is a social phenomenon and delineate helpful peer interactions for the different stages of learning. Bolhuis (2003) content and process oriented instruction provides a multidimensional perspective to teaching. Future work will implem ent modeling, more intentional peer interactions, and more explicit SDL strategy instruction in parallel with content instruction and the outcomes will be evaluated.
101 Figure 51. Epilogue: Gains in average unit test scores over five tests Gain was ca lculated as [( Recent Score Previous Score) / (100 Previous Score)]
102 APPENDIX A TIME TICKET
103 APPENDIX B POST STUDENT SURVEY Post Student Survey Conservation of Energy Unit Instructions: This is a questionnaire designed to gather data on preferences and attitudes towards learning physics. After reading each item, please indicate the degree to which you feel that statement is true of you. There are no right or wrong answers. Please read each choice carefully and choose the response which best express es your feeling. Your answers are CONFIDENTIAL and will only be presented as pooled data. There is no time limit for the questionnaire. Try not to spend too much time on any one item; however, your first reaction to the question will usually be the most accurate. ________________________________________________________________ PART A: These items generally relate to learning physics. Responses 1 = I hardly ever feel this way. 2 = I feel this way less than half the time. 3 = I feel this way about half the time. 4 = I feel this way more than half the time. 5 = There are very few times when I don't feel this way. Items 1. A signi information I need to know. 2. When I am solving a physics problem, I try to decide what would be a reasonable value for the answer. 3. It is useful for me to do lots and lots of proble ms when learning physics. 4. I think about the physics I experience in everyday life. 5. Knowledge in physics consists of many disconnected topics. 6. When studying physics, I relate the important information to what I already know rather than just memorizing it the way it is presented. 7. To do well in this physics course, I need to see sample problems like those on the tests. 8. I cannot learn physics if the teacher does not explain things well in class.
104 9. I am developing general problem solving techniques in this course. 10. In a learning experience, I prefer to take part in deciding what will be learned and how rather than having the teacher tell me what to do. 11. Difficult study doesn't bother me if I'm interested in something. 12. No one but me is truly responsible for what I learn. 13. I can tell whether I'm learning something well or not. 14. Understanding what I read is a problem for me. 15. If I discover a need for information that I don't have, I know where to go to get it. 16. If I experience a difficulty while studying physics, I try to figure it out on my own. 17. To understand physics I discuss it with friends and other students. 18. There is usually only one correct approach to solving a physics problem. 19. I do not expect physics equations to help my understanding of the ideas; equations are for doing calculations. 20. The first thing I do when solving a physics problem is make a drawing to represent it. 21. I am not satis 22. I enjoy solving physics problems. 23. I do not invest time understanding the derivations of equations. 24. I have found carefully analyzing only a few problems in detail is a good way for m e to learn physics. 25. It is possible to explain physics ideas without mathematical formulas. PART B: These items are specific to the conservation of energy mastery learning environment. 26. I prefer to learn from our Glencoe textbook rather than Internet r esources. 27. I prefer traditional lecture with a weekly lab compared to the self paced VOLO work. 28. I invested significant hours outside of class because I did not have sufficient time during class to learn this material. 29. I have mastered the key ideas about conservation of energy.
105 30. I am satisfied with my grade for this unit. 31. Retaking the unit test helped me figure out my weak spots, and focused my learning. 32. I developed a learning strategy that worked for me for this unit. 33. Masteringphysics was very helpful to my learning process. 34. PhET Energy Skate Park was very useful to my learning process. 35. The Pendulum Box Bash experiment was very useful to my learning process. 36. This unit had the right balance of wholeclass, small group, and individual learning activ ities. 37. I had all the support I needed whether from the teacher, peers, or course resources to learn this material. 38. Some class periods I accomplished very little. 39. I have a very productive team in this course. 40. I had a plan for what I wanted to accomplish each day BEFORE I entered the physics classroom. ________________________________________________________________ PART C: Open response questions about the conservation of energy mastery learning environment. 41. Three VOLO activities were required masteringphysics, PhET Energy Park, and the Pendulum Box Bash experiment. Explain why you chose the other two VOLO activities that you did, identify them, rate how much you enjoyed them, and rate their usefulness to your learning. Use the scale of 1 (low) to 5 (high). 42. Describe how accessible your teacher was in meeting your learning needs, and rate the quality of that support on a scale of 1 (low) to 5 (high). 43. Describe how accessible your peers were in meeting your learning needs, and rate the quali ty of that support on a scale of 1 (low) to 5 (high).
106 44. If you had a choice of two tools or resources to use to learn physics, what would you pick and why? 45. Comment on the mastery learning approach, which includes frequent tests with feedback from instruc tor and peers, followed by taking corrective actions before retesting. Do you learn more in this approach than in traditional courses?
107 APPENDIX C SAMPLE TEST Honors Physics Conservation of Energy Test For each problem or question on the right, show your work on the left, clearly mark your answer, and remember significant figures and units. Answer Problem / Question 1 1. A spring with a spring constant of 4.0 N/m is compressed with a force of 1.2 N. What is the total elastic potential energy stored in this compressed spring? 2. 2. Suppose in Problem #1, you were not given the value of the spring constant. Explain how you can determine the value of the spring constant. 3. 3. Two blocks of ice, one four times as heavy as the other, are at rest on a f rozen lake. A person pushes each block the same distance. Ignore friction and assume that an equal force is exerted on each block. Compared to the speed of the heavier block, how fast does the light block travel?
108 Answer Problem / Question 4 4. A person is standing in an elevator, which is moving upward at a constant speed. Consider the person as a system. As the elevator goes between the 3rd and 4th floor, (a) The total work done on the person by the elevator and the Earth is positive. (b) The total work done on the person by the elevator and the Earth is zero. (c) The total work done on the person by the elevator and the Earth is negative. (d) Not enough information to determine v 5a. 5 A roller coaster car has a mass of 290 kg. Starting from rest, the car acquires 3.1 3E5 J of kinetic energy as it descends to the bottom of a hill in 5.3 seconds. a. Calculate the height of the hill. 5b. 5b. What is the speed of the car at the bottom of the hill? 6. 6. A pendulum experiment is conducted with a 0.35kg bob that is moving to the right at 2.7 m/s. What is the bobs speed at point A?
109 APPENDIX D IRB Appendix G UFIRB 02 Social & Behavioral Research Protocol Submission Form This form must be typed. Send this form and the supporting documents to IRB02, PO Box 11225 0, Gainesville, FL 32611. Should you have questions about completing this form, call 352-3920433. Title of Protocol: Use of Google Apps platform to create a mastery learning environment Principal Investigator: Wendy Athens UFID #: 94591189 Deg ree / Title: M.S., Ed.D candidate in Educational Technology Mailing Address: (If on campus include PO Box address): 11507 Timberline Circle Fort Myers, FL 33966 Email: firstname.lastname@example.org Department: College of Educa tion Department of Educational Technology Telephone #: 2394708900 Co Investigator(s): UFID#: Email: Supervisor (If PI is student) : Dr. Kara Dawson UFID#: Degree / Title: Associate Professor of Educational Technology Mailing Address: ( If on campus include PO Box address): P.O. Box 117048 Gainesville, FL 326117058 Email : email@example.com u Department: College of Education School of Teaching & Learning Department of Educational Technology Teleph one #: 3523929191 x261 Date of Proposed Research: January February 2011 Source of Funding (A copy of the grant proposal must be submitted with this protocol if funding is involved):
110 Scientific Purpose of the Study: This study will be cond ucted in my 11th & 12th grade honors physics classroom (N = 25) at Evangelical Christian School in Fort Myers, FL, for a designbased research effort that will serve as my Ed.D. capstone project. The purpose of this study is to assess learning gains withi n a 21st century mastery learning environment over a six week period in January and February 2011. More specifically, I seek to identify expert/peer/computer tutor time required by each student to meet his or her individual learning goals. To do this, I employ Google Apps a nd web tools Students learn through a largely self paced program, and the time the student spends with the expert, peer, and computer tutor will be tracked. I will correlate these three times with performance on the Force Concept Inventory test (FCI, Attachment 1), a validated instrument that probes conceptual understanding, and also course performance. I will collect weekly learning blog data from the students. I will also collect data from the validated Values about Science Surve y (VASS, Attachment 2). An auxiliary investigation is the correlation between student learning styles and learning tasks selected. Describe the Research Methodology in Non Technical Language: ( Explain what will be done with or to the research partici pant. ) During three learning units of approximately six weeks duration students will track the tutorial resources they use to learn, whether expert, peer, or computer. The learning unit structure is designed as follows: two days of whole class, faceto face instruction followed by the balance of self paced work. The whole class phase introduces the major concepts for the unit, engages the students in an inquiry activity, and demonstrates and models the problem solving skills to be developed. Students are then presented with a list of nine learning options of which they must complete five to receive the highest grade. Students submit a learning contract to the instructor by placing the digital contract on their personal web page. Some learning options are required, such as the masteringphysics homework tutorial problem set, a type of computer tutor, which requires typically 1.5 hours to complete. Students then begin self paced work, encompassing approximately seven days. Approximately halfway through this time students take a formative assessment. This is a prescriptive assessment which provides necessary feedback, and the instructor suggests corrective pathways. This is the heart of mastery learning: assessment feedback corrective cycles. Students access all course resources, manage their eportfolio, and track their time through the class website, which is a Google Apps for Education website. All student s have a personal web page within the class website, which holds their learning contracts, physics learning blog, and an electronic file cabinet to collect their work (eportfolio). Every student has editor capability for the expert tutor calendar to schedule oneonone time with the teacher in advance of class. Every student has Google Gro ups and Gmail capability, which facilitate peer collaboration. Towards the end of the learning unit, students take a summative assessment. This assessment takes approximately 25 minutes to complete (10 questions), and if the student does not score at least an 80%, he or she must retake a similar assessment tool The assessments are graded oneto one with the instructor to receive essential feedback and target areas for improvement. The learning unit concludes when the student s have completed their learning contract s and scored at least an 80% on the unit test. Grades are allocated as follows: 45% learning contract completion, 45% test, and 10% blogs. The conservation of energy unit includes a project (Rube Goldberg contraption), which counts for a test. At the conclusion of the three learning units, students complete the post FCI and VASS.
111 Describe Potential Benefits: The purpose of this study is to demonstrate viability of 21st century technologies to enable mastery learning, a proven instructional practice with learning gains typically in the 12 sigma (standard deviation) range. The 21st century technologies to be investigated include Google Apps, computer tutors (including masteringphysics), and web tools (including PhET from Univer sity of Colorado at Boulder and MOPS from Glenbrook South High School in Glenview, IL.) By tracking which tutor resources the student uses expert, peer, and computer a correlation can be drawn between concept mastery and the form of tutor support util ized, if at all. Since the ideal is to simulate a oneto one tutor relationship in a wholeclass setting, correlations between learning success and the amount of individual tutoring time can be explored, as well as the success of substituting a computer t utor for a human tutor. The overall goal of this study is to improve performance on the FCI compared to performance of similar classes in two prior years. An auxiliary study involves a chi square analysis between students' learning styles (students compl eted a learning style analysis earlier in the school year) and the learning options selected within the unit learning contracts. A third benefit anticipated from this study is the contribution to two bodies of literature, namely physics education research and blended secondary learning environments. Describe Potential Risks: ( If risk of physical, psychological or economic harm may be involved, describe the steps taken to protect participant.) There are no risks to this study. Students will enjoy working i ndepe ndently and in small groups and with more oneonone time with the teacher. No individual identities will be revealed in the reporting of this research. Describe How Participant(s) Will Be Recruited: All students in ECS Honors Physics will be part icipate in the mastery learning environment, and at the conclusion of the two learning units, a consent letter will be sent to parents and students for permission to include their anonymous data in the capstone document and publications. Maximum Number of Participants (to be approached with consent) 25 Age Range of Participants: 1617 Amount of Compensatio n/ course credit: N/A pertaining to the study itself, but the learning units covered during the study do constitute partial fulfillment of 1.5 credits of science Describe the Informed Consent Process. (Attach a Copy of the Informed Consent Document. See http://irb.ufl.edu/irb02/samples.html for examples of consent.)
112 (SIGNATURE SECTION) Principa l Investigator(s) Signature: Wendy Athens Date: 112410 Co Investigator(s) Signature(s): Date: Supervisors Signature (if PI is a student): Date: Department Chair Signature: Date:
113 28 February 2011 Dear Parent/Guardian: I am a doctoral candidate in the Educational Technology program at the University of Florida, conducting research on the use of technology to enable mastery learning under the supervision of Dr. Kara Dawson. This research was conducted in two sections of ECS Honors Physics i n January and February. The research ideal was to simulate a oneto one tutor relationship in a whole class setting, and correlations between learning success and the amount of indivi dual tutoring time was explored as well as the success of substituting a computer tutor for a human tutor. The overall goal of this study was to improve long term conceptual understanding in physics, and two measures of this were the course grade as well as performance on a nationally recognized conceptual physics test called the Force Concept Inventory. An additional aspect of this study involved seeking a relationship between students' learning styles (students completed a learning style analysis earlier in the school year) and the learning options chosen for each unit. All students participated in the mastery learning environment; however, only with your consent will I include anonymous data from your son/daughter in the publications associated with this research. The type of data to be included in publications includes: time spent with a computer/peer/computer tutor anonymous trends noted in students weekly physics learning blogs anonymous results to the Force Concept Inventory test anonymous course grades anonymous results to the Values about Sciences Survey pooled data only on learning styles in the classroom pooled data only on learning options chosen for each unit. Although I believe this study will help your student and future students learn physics better, a choice to not participate will not affect your student's grade. In fact, I waited until the end of the learning units so your response would in no way influence your students grade nor bias my records. There are no known risks or immediate benefits to the participants, and no compensation is offered for participation. Again, no student names will ever be reported. All results will be expressed as group data. This study will most likely be published in a science education journal and presented at a technology conference. Study results will be available after May 2011. To receive a copy of results or to ask questions, contact me at firstname.lastname@example.org. Questions or concerns about your child's rights as research participant may be directed to the IRB02 office, Univ ersity of Florida, Box 112250, Gainesville, FL 32611, (352) 3920433. Wendy Athens Wendy Athens
114 LIST OF REFERENCES Abar, B. & Loken, E. (2010). Self regulated learning and self directed study in a precollege sample. Learning and Individual Differences, 20 25 29. Abdullah, M. (2001). Self directed learning. Bloomington, IN: ERIC Clearinghouse on Reading, English, and Communication. [ERIC ED 459458]. Retrieved December 26, 2010 from http://www.indiana.edu/~reading/ieo/digests/d169.html Anderson, L. & Block, R. (1987). Values, evidence and mastery learning. Review of Educational Research, 57(2), 215223. Anglin, L., Anglin, K., Schumann, P.L. & Kaliski, J. (2008). Improving the efficiency and effectiveness of grading through the use of computer assisted grading rubrics. Decision Sciences Journal of Innovative Education, 6(1), 51 73. Arons, A. (1972). Toward wider public understanding of science. American Journal of Physics, 4 1 (6), 769 782. Arons, A. (1974). Toward wider public understanding of science: Addendum. American Journal of Physics, 42(2), 157 158. Ausubel, D. (1963). The Psychology of Meaningful Verbal Learning. New York: Grune & Stratton. Bates, A. & Poole, G. ( 2003). Effective Teaching with Technology in Higher Education: Foundations for Success. San Francisco, CA: Jossey Bass. Retrieved July 27, 2009 from http://batesandpoole.ubc.ca/ Beichner, R., Saul, J., A bbott, D., Morse, J., Deardorff, D., Allain, R., Bonham, S., Dancy, M. & Risley, J. (2000). The student centered activities for large enrollment undergraduate programs (SCALE UP) project. Proceedings of the 2000 Annual Meeting of the American Society for Engineering Education, Session 2380, 2000. Blake, C. & Scanlon, E. (2007). Considering simulations in science education at a distance: Features of effective use. Journal of Computer Assisted Living, 23, 491502. Bliuc, A., Goodyear, P. & Ellis, R. ( 2007) Research focus and methodological choices in studies into students experiences of blended learning in higher education. Internet and Higher Education, 10, 231244. Bloom, B. (1968) Learning for mastery. Evaluation Comment, 1(2): 1 5. Bloom, B. (1978). New views of the learner: Implications for instruction and curriculum. Educational Leadership, 35(7), 563 576.
115 Bloom, B. (1984). The search for methods of group instruction as effective as oneto one tutoring. Educational Leadership, 41(8), 4 17. Bloom, B. (1987). A response to Slavin's Mastery Learning Reconsidered. Review of Educational Research, 57(4):507 508. Bolhuis, S. (2003). Towards process oriented teaching for self directed lifelong learning: a multidimensional perspective. Le arning and Instruction, 13, 327347. Bonk, C., & Graham, C. (2006). The Handbook of Blended Learning. San Francisco, CA: John Wiley & Sons. Boote, D. & Beile, P. (2005). Scholars before researchers: On the centrality of the dissertation review process in research preparation. Educational Researcher, 34(6),3 15. Bransford, J. (2000). How people learn: Brain, mind, experience, and school. Washington, DC: National Academy Press. Breslow, L. (2004). Findings from ten formative assessments of educational initiatives at MIT. MIT Teaching and Learning Laboratory. Retrieved 10 September 2010 from http://web.mit.edu/annualreports/pres04/15.11.pdf Breslow, L. (2007). Lessons learned: Findings from MIT's efforts in educational technology (20012005). Journal of Science Education and Technology, 16(4), 283297. Retrieved 10 September 2010 from http://www.springerlink.com/index/D4358K0R200807NJ.pdf Brophy, S., Klein, S., Portsmore, M., & Rogers, C. (2008). Advancing engineering education in P 12 classrooms. Journal of Engineering Education, 97(3), 369 387. Buschman, L. (2003). Children who enjoy problem solving. Teaching Children Mathematics, 9, 539544. Candy, P. (1991). Self direction for lifelong learning: A comprehensive guide to theory and practice. San Francisco, CA: Jossey Bass. Cavanaugh, C. (2009). Getting students more learning time online: Distance educati on in support of expanded learning time in K 12 schools. Report to Center for American Progress. Chang, K., Chen, Y., Lin, H. & Sung, Y. (2008). Effects of learning support in simulation based physics learning. Computers & Education, 51, 1486 1498. Chickering, A. & Gamson, Z. (1991). Appendix A: Seven principles for good practice in undergraduate education. New Directions for Teaching and Learning, 47, 63 69.
116 Chou, P. & Chen, W. (2008). Exploratory study of the relationship between self directed learning and academic performance in a webbased learning environment. Online Journal of Distance Learning Administration, 11(1). Retrieved December 26 2010 from http://www.we stga.edu/~distance/ojdla/spring111/chou111.html Clark, D. (2009). Kirkpatrick's four level evaluation model. Retrieved June 10, 2010 from http://www.nwlink.com/~Donclark/isd/kirkpatric k.html Cobb, P., Wood, T., & Yackel, E. (1993). Discourse, mathematical thinking, and classroom practice. In E. A. Forman, N. Minick, & C. A. Stone (Eds.). Contexts for learning: Sociocultural dynamics in children's development. New York: Oxford Univeris ty Press. Cook, D. (1996). Reminiscences: Fred S. Keller: An appreciation. Behavior and Social Issues, 6 (1), 61 72. Corbett, A. (2001). Cognitive computer tutors: Solving the twosigma problem. User Modeling 2001 in Lecture Notes in Computer Scienc e Series, 2109(2010), 137147. Heidelberg: Spring Verlag. Cracolice, M. & Roth, S. (1996). Keller's old Personalized System of Instruction: A new solution for today's college chemistry students. Chemical Educator, 1(1), 1 18. Creative Commons (2010). Retrieved September 25, 2010 from http://creativecommons.org Crouch, C. & Mazur, E. (2001). Peer instruction: Ten years of experience and results. American Journal of Physics, 69(9), 970 977. Dana, N. & Yendol Hoppey, D. (2009). The Reflective Educators Guide to Classroom Research. Thousand Oaks, CA: Corwin Press. Second edition. DeCorte, E., Verschaffel, L., and Eynde, P. (2000). Self regulation: A characteristic and goal of mathematics education. In M. Boekoeris, O. Pintrich, & J. Zeidner (Eds.), Handbook of Self Regulation, (pp. 687 726). San Diego: Academic Press. Dick, W. & Carey, L. (1990). The Systematic Design of Instruction New York, NY: Harper Collins. Third edition. Dori, Y., Hult E., Breslow, L. & Belcher, J. (2007). How much have they retained? Making unseen concepts seen in a freshman electromagnetism course at MIT. Journal of Science Education and Technology, 16(4), 299 323.
117 Dori, Y. & Sasson, I. (2007). Chemical underst anding and graphing skills in an honors case based computerized chemistry laboratory environment: The value of bidirectional visual and textual representations. Journal of Research in Science Teaching, 45(2), 219 250. Downes, S. (2005). E Learning 2.0. Retrieved 1 October 2010 from http://www.elearnmag.org/subpage.cfm?section=articles&article=291 Efimova, L. (2004). Discovering the iceberg of knowledge work: A weblog case. Proceedings of the Fifth European Conference on Organisational Knowledge, Learning and Capabilities (OKLC 2004), Innsbruck, 23 April. Etkina, E., Karelina, A. Murthy, S. & Ruibal Villasenor, J. (2009). Using action research to improve learning and formative assessment to conduct research. Review Special Topics Physics Education Research, 5(010109), 115. Eyre, H. (2007). Keller's Personalized System of Instruction: Was it a fleeting fancy or is there a revival on the horizon? Behavior Analyst Today, 8 (3), 317 325. Forehand, M. (2010). Bloom's Taxonomy Emerging Perspectives on Learning, Teaching and Technology website of University of Georgia. Retrieved November 27, 2010 from http://projects.coe.uga.edu/epltt/index.php?title=Bloom%27s_Taxonomy Friedman, T. (2008). The World is Flat 3.0 MIT Open Courseware. Retrieved September 23. 2010 from http://mitwo rld.mit.edu/video/519 Gagne, R. (1988). Mastery learning and instructional design. Performance Improvement Quarterly, 1(1): 7 18. Gardner, H. (1993). The Unschooled Mind: How Children Think and How Schools Should Teach. New York, NY: Basic Books. G arrison, D., Anderson, T. & Archer, W. (2003). A theory of critical inquiry in online distance education. In M.G. Moore & W. Anderson (Eds.), Handbook of Distance Education. Mahweh, NJ: Erlbaum. Gates, B. (2007A ). A robot in every home. Scientific American, 296(1), 56 65. Gates, B. (2007B ). How to keep America competitive. Washington Post 25 February 2007. Gerace, W., Dufresne, R., Leonard, W. & Mestre, J. (1999). Minds on Physics: Materials for developing concept based problem solving skills in physics. [ERIC ED 471418]. Gibbons, M. (2002). The self directed learning handbook : Challenging adolescent student to excel San Francisco, CA: Jossey Bass.
118 Ginns, P., & Ellis, R. (2007). Quality in blended learning: Exploring the relationships between online and faceto face teaching and learning. Internet and Higher Education, 10(1), 53 64. Guglielmino, L. (1977). Development of the self directed learning readiness scale Unpublished doctoral dissertation. The University of Georgia, Athens, GA. Guglielmino, L. (1989). Development of an adult basic education form of the Self Directed Learning Readiness Scale. In the H. B. Long and Associates, Self directed learning: Emerging theory and practice. Norman OK: Oklahoma Research Center for Continuing Professional and Higher Education, University of Oklahoma. Guglielmino, L., Guglielmino, P., & Durr, R. (2000). Learning contracts: A learning technique and a developmental process. In Piskurich, G., Beckschi, P., & Hall, B. (Eds.). American Society for Training and Development handbook of training design and delivery. (396 414). NY: McGraw Hill. Guskey, T. (1987). Rethinking mastery learning reconsidered. Review of Educational Research, 57(2), 225 22. Guskey, T. (2007). Closing achievement gaps: Revisiting Benjamin S. Bloom's "Learning for Mastery." Journal of Advanced Academics, 19(1), 8 31. Guskey, T. (2010). Lessons of mastery learning. Educational Leadership, 68(2), 52 57. Hake, R. (1998). Interactive engagement versus traditional methods: A six thousandstudent survey of mechanics test data for introductory physics courses. American Journal of Physics, 66(1), 64 74. Hake, R. (2007). Six lessons from the physics education reform effort. Latin American Journal of Physics Education, 1(1), 24 31. Halloun, I. (2001). Student Views about Science: A Comparative Survey. Monograph. Beirut, Lebanon: Lebanese University. Halloun, I. & Hestenes, D. (1985). Common sense concepts about motion. American Journal of Physics 53, 10561065. Hallo un, I. & Hestenes, D. (1985). The initial knowledge state of college physics students. American Journal of Physics, 53, 10431055. Halloun, I. & Hestenes, D. (1998). Interpreting VASS dimensions and profiles for physics students. Science and Education, 7 (6): 553577.
119 Hart, G. & Rudman, W. (2001). Excerpts of road map for national security: Imperative for change. The Phase III Report of the U.S. Commission on National Security/21st Century. Retrieved November 13, 2010 from http://www.senate.gov/~govtaff/032901_Excerpts.pdf Hayes, B. (2008). Cloud computing. Communications of the ACM, 51(7), 9 11. Heller, P., Keith, R. & Anderson, S. (1991). Teaching problem solving through coop erative grouping. American Journal of Physics, 60(7), 627 644. Hendry, G. & Ginns, P. (2009). Readiness for self directed learning: Validation of a new scale with medical students. Medical Teacher 31(10), 918 920. Herbal Eisenman, B. A., & Breyfogle, L (2005). Questioning our patterns of questioning. Mathematics Teaching in the Middle School, 10, 484 489. Hestenes, D. & Halloun, I. (1995). Interpreting the Force Concept Inventory. The Physics Teacher, 33, 502 506. Hestenes, D. & Wells, M. (1992). A mechanics baseline test. The Physics Teacher, 30, 159165. Hestenes, D., Wells, M. & Swackhamer, G. (1992). Force Concept Inventory. The Physics Teacher, 30, 141 158. Hill, D. (2009 May). Can cloud computing ever replace traditional desktop computi ng? Partial fulfillment for degree of Master of Science in Management Information Systems, Bowie State University. Retrieved September 19, 2010 from http://faculty.ed.umuc.edu/~ sdean/ProfPaps/Bowie/S09/Hill.pdf Hill, D., Rapoport, A., Lehming, R. & Bell, R. (2007). Changing U.S. output of scientific articles: 19882003. Special Report for the National Science Foundation. Retrieved September 5, 2010 from http://www.nsf.gov/statistics/nsf07320/. Hofmann, J. (2006). Why blended learning hasn't (yet) fulfilled its promise. In C. Bonk & C. Graham (Eds.), Handbook of Blended Learning. San Francisco, CA: John Wiley & So ns. Hogg, K. (2008). Development of an instrument to measure tendencies toward self directedness in learning within a workplace setting. Unpublished doctoral dissertation. The University of Georgia, Athens, GA. Huck, S. (2008). Reading Statistics and Research Boston, MA: Pearson. 5th Ed. Hyman, J. & Cohen, S. (2002). Learning for mastery: Ten conclusions after 15 years and 3,000 schools. Educational Leadership, 37(2): 104109.
120 Idros, S., Mohamed, A., Esa, N., Samsudin, M. & Daud, K. (2010). Enhancing self directed learning skills through eSOLMS for Malaysian learners. Procedia Social and Behavioral Sciences, 2, 698 706. Infinity Project. Retrieved October 10, 2010 from http://www.infinity project.org/infinity/infinity_hist.html Jackson, J. (1996). Overview of VASS. Retrieved May 14, 2011 from http://modeling.asu.edu/R&E/Research.html Johnson, D.W., Johnson, R.T., & Stanne, M.B. (2000). Cooperative learning methods: A meta analysis. Retrieved September 25, 2010 from http://www.clcrc.com/pages/cl methods.html Jonassen, D. (2001). Exclusive interview w ith Professor David Jonassen. Retrieved May 14, 2011 from http://www.elearningpost.com/articles/archives/exclusive_interview_with_profess or_davi d_jonassen Jonassen, D. (2003). Learning to Solve Problems with Technology. Upper Saddle River, NJ: Merrill Prentice Hall. Katz, S. (2008). Assessing a hybrid format. Journal of Business and Technical Communications, 22(1), 92 110. Retrieved May 31 2009 from http://jbt.sagepub.com/cgi/content/abstract/22/1/92 Keller, F. (1968) GoodBye, Teacher.... Journal of Applied Behavior Analysis, 1, 79 89. King, R. (2009 August 4). H ow cloud computing is changing the world. Retrieved September 19, 2010 from Businessweek website http://www.businessweek.com/technology/content/aug2008/tc2008082_445669.h tm Kirschner, P., Sweller, J. & Clark, R. (2006). Why minimal guidance during instruction does not work: An analysis of the failure of constructivist, discovery, problem based, experiential, and inquiry based teaching. Educational Psychologist, 41(2), 75 86. Knowles, M. S. (1975) Self Directed Learning. A guide for learners and teachers Englewood Cliffs: Prentice Hall/Cambridge. Koohang, A. (2009). A learner centered model for blended learning design. International Journal of Innovation and Learning, 6(1), 76 91. Kulik, C. & Kulik, J. & Bangert Drowns, R. (1990). Effectiveness of mastery learning programs: A metaanalysis. Review of Educational Research, 60(2): 265 299.
121 Larkin, J., McDermott, J., Simon, D., & Simon, H. (1980). Expert an d novice performance in solving physics problems. Science, 208(4450), 1135 1342. Lemak, D. J., Shin, S. J., Reed, R., & Montgomery, J. C. (2005). Technology, Transactional Distance, and Instructor Effectiveness: An Empirical Investigation. Academy of Management Learning and Education, 4(2), 150159. Lipson, A. (2001). A threeschool comparative analysis of student usage patterns and attitudes towards PIVOT. Teaching and Learning Laboratory. MIT. Retrieved September 24, 2010 from http://relate.mit.edu Lockman, P., Gaasch, J., Borges, K., Ehlo, A. & Smith, Q. (2008). Innovations in teaching: Using WebCT to implement a basic science competency education course. American Journal of Pharmaceutical Education, 72(2), 1 8. Ma, Y. & Harmon, S. (2009). A case study of designbased research for creating a vision prototype of a technology based innovative learning environment. Journal of Interactive Learning Research, 20(1), 75 93. Mandernach, B., Forrest, K., Babutzke, J., & Manker, L. (2009). The role of instructor interactivity in promoting critical thinking in online and faceto face classrooms. MERLOT Journal of Online Teaching and Learning, 5(1), 49 62. Mayer, R. (2004). Should there be a threestrikes rule against pure discovery learning? American Psychologist, 59, 14 19. Mayo, M. (2007). Games for science and engineering education. Communication of the ACM, 56(7), 31 36. Mazur, E. (1997). Peer instruction. Upper Saddle River, NJ: PrenticeHall. McDermott, L. (2001). Oersted Medal Lecture 2001: Physics education research: The key to student learning. American Journal of Physics, 69(11), 11271138. McDermott, L., Shaffer, P. & Physics Education Group. (2002). Tutorials in Introductory Physics. U pper Saddle River, NJ: PrenticeHall. McTighe, J. & Wiggins, G. (2004). Understanding by Design Professional Development Workbook. Alexandria, VA: Association for Supervision and Curriculum Development. Meltzer, D. & Manivannan, K. (2002). Transform ing the lecturehall environment: The fully interactive physics lecture. American Journal of Physics, 70(6), 639654. Merrill, M. (2002). First principles of instruction. Educational Technology Research and Development, 50(3), 4359.
122 Merrill, M. (20 07). A task centered instructional strategy. Journal of Research on Technology in Education, 40(1), 33 50. Merrill, M. & Gilbert, C. (2008). Effective peer interaction in a problem centered instructional strategy. Distance Education, 29(2), 199 207. Merrill, M. D. (2009). Finding e3 (effective, efficient and engaging) Instruction. Educational Technology, 49(3), 15 26. Mintzes, J. (2007). Teaching science for understanding at MIT: Engaging the best and the brightest. Journal of Science Education a nd Technology, 16(4), 365368. Moore, M. & Anderson, W. (2003). Handbook of Distance Education. Mahwah, N.J.: Lawrence Erlbaum Associates, Inc. Morote, E. & Pritchard, D. (2009). What course elements correlate with improvement on tests in introductor y Newtonian mechanics? American Journal of Physics, 77, 746753. Morote, E., Warnakulasooriya, R. and Pritchard, D. (n.d.) Effectiveness of tutorials in web based physics tutor. Retrieved June 20, 2008 from http://tinyurl.com/shortathenscapstone Morris on, G., Ross, S. & Kemp, J. (2007). Designing effective instruction. Hoboken, NJ: John Wiley & Sons. Fifth edition. Najjar, M.. (2008). On scaffolding adaptive teaching prompts within virtual labs. International Journal of Distance Education Technologies, 6 (2), 35 54. National Assessment of Educational Progress. (2005). The nation's report card Retrieved 5 September 2010 from http://nationsreportcard.gov/science_2005/ National Science Boar d. (2006). Science and Engineering Indicators 2006 Vol. 1 NSB 061, Vol. 2 NSB 0601A. Arlington, VA: National Science Foundation. National Survey of Student Engagement (NSSE). (2009). Assessments for improvement: Tracking student engagement over time. Retrieved October 10, 2010 from http://nsse.iub.edu/_/?cid=41. Nellman, S. (2008). A formative evaluation of a high school blended learning biology course Ed.D. dissertation, University of Southern C alifornia, United States -California. Retrieved July 19, 2009, from Dissertations & Theses: Full Text. (Publication No. AAT 3325075). Nelson, L. (2009). Minority students needed in math and science to combat "brain drain," professors say. Chronicle of Higher Education, 22 September 2009. Retrieved October 10, 2010 from http://chronicle.com/article/Minority Students Neededin/48568/
123 Ogilvie, C. (2001). Effectiveness of di fferent course components in driving gains in conceptual understanding. Department of Physics, MIT Internal Report No. 01. Retrieved October 6, 2010 from http://relate.mit.edu/ogilvie.pdf Oishi, L. (2007). Working together: Google Apps goes to school. Technology & Learning, 27(9), 2. ONeil, H., ed. (2005). What Works in Distance Learning: Guidelines Greenwich, CT: Information Age Publishing. O'Reilly, T. (2007). What Is Web 2.0: Design patterns and business models for the next generation of software. Munich Personal RePEc Archive Communications & Strategies, 61(1), 7 29. Retrieved September 19, 2010 from http://www.ttivanguard.com/ttivanguard_cfmfiles/pdf/dc05/dc05session4003.pdf Ozmen, H., Demircioglu, H., Demircioglu, G. (2008). The effects of conceptual change texts accompanied with animations on overcoming 11th grade students' alternative concept ions of chemical bonding. Computers & Education, 52, 681 695. Pape, S. (2005). Intervention that supports future learning: Developing self regulated learners. In S. Wagner (Ed.), PRIME: Prompt intervention in mathematics education (pp. 7798). Colum bus, OH: Ohio Resource Center for Mathematics, Science, and Reading and Ohio Department of Education. Pape, S., Bell, C. & Yetkin, I. (2003). Developing mathematical thinking and self regulated learning: A teaching experiment in a seventhgrade mathemat ics classroom Educational Studies in Mathematics, 53, 179 202. Park, S., Lee, G., Kim, M. (2009). Do students benefit equally from interactive computer simulations regardless of prior knowledge levels? Computers & Education, 52 649 655. Parkhurst, R. Moskal, B, Downey, G., Lucena, J., Bigley, T. & Elber, S. (2008). Engineering Cultures: Online versus inclass. MERLOT Journal of Online Teaching and Learning, 4(4), 438 445. Retrieved June 22, 2009 from http://jolt.merlot.org/vol4no4/moskal_1208.htm Partnership for 21stCentury Skills Science Map. (2009). Retrieved December 23, 2010 from ht tp://www.p21.org/index.php?option=com_content&task=view&id=82&Itemid=40 #maps Patsula, P. (1999). Applying learning theories to online instructional design. Retrieved April 24 2010 from http://www.patsula.com/usefo/webbasedlearning/tutorial/learning_theories_full_v ersion
124 Piaget, J. (1926). The Child's Conception of the World. London: Routledge and Kegan Paul Ltd. Prain, V. (2006). Learning from wr iting in secondary science: Some theoretical and practical considerations. International Journal of Science Education, 28(2 13), 179201. Pritchard, D. (2010). Masteringphysics introductory video. Retrieved September 20, 2010 at http://masteringphysics.com/assets/site/res/swf/Pritchard_Video/index.html Program for International Student Assessment (PISA). 2009. National Center for Education Statistics. Retrie ved December 22, 2010 from http://nces.ed.gov/surveys/pisa/pisa2009highlights_2.asp Reddish, E. & Steinberg, R. (1999). Teaching physics: Figuring out what works. Physics Today, 52, 24 30. Reeves, T. (2000). Enhancing the worth of instructional technology research through "design experiments" and other development research strategies. Paper presented at the American Educational Research Association, New Orleans, LA. Rollett, H., Lux, M., Strohmaier, M., Dosinger, G. & Tochtermann, K. (2007). The Web 2.0 way of learning with technologies. International Journal of Technology, 3(1), 87106. Rovai, A., Ponton, M., Wighting, M. & Baker, J. (2007). A comparative analysis of stud ent motivation in traditional classroom and elearning courses. International Journal of E Learning, 6(3), 413 433. Science Commons (2010). Retrieved September 25, 2010 from http://sciencecommons.org Siemens, G. (2005 A ). Connectivism: A learning theory for the digital age. International Journal of Instructional Technology and Distance Learning, 2(1), 3 10. Siemens, G. (2005 B ). Learning Development Cycle: Bridging learning design and modern knowledge needs. Retrieved October 4, 2010 from http://www.cedma europe.org/newsletter%20articles/misc/Learning%20Development%20Cycle%20 (Jul%2006).pdf Siemens, G. (2008). Learning and knowing in networks: Changing roles for educators and designers. Presented to University of Georgia ITFORUM on January 27, 2008. Retrieved October 4, 2010 from http://www .tskills.it/userfiles/Siemens.pdf
125 Siemens, G. & Tittenberger, P. (2009). Handbook of Emerging Technologies for Learning. Retrieved September 29, 2010 from http://posterous.com/getfile/files.posterous.com/mpeersinformation/sOmrJmLWc FSRIkam1KMbaTsAg4AyAZf0LncTA6FpEZD871wAlkm6FqnpDqBG/13__HETL.pdf Singh, H. (2003). Building effective blended lear ning programs. Educational Technology, 43(6), 51 54. Retrieved July 27, 2009 from http://asianvu.com/digital library/elearning/blendedlearning by_Singh.pdf Skill, T. & Young, B. (2002). Embracing the hybrid model: Working at the intersections of virtual and physical learning spaces. New Directions for Teaching and Learning, 92, 23 32. Slavin, R. (1987A ). Mastery learning reconsidered. Review of Educational Research, 57(2): 175213. Slavin, R. (1987B ). Taking the mystery out of mastery: A response to Guskey, Anderson and Burns. Review of Educational Research, 57(2): 231 235. Steen, H. (2008). Effective e learning design. MERLOT Journal of Online Teac hing and Learning, 4(4): 526 532. Sweeney, J., O'Donoghue, T., & Whitehead, C. (2007). Traditional faceto face and web based tutorials: A study of university students' perspectives on the roles of tutorial participants. Teaching in Higher Education, 9 (3), 311323. Taylor, B. (1995). Self directed learning: Revisiting an idea most appropriate for middle school students. Paper presented at the Combined Meeting of the Great Lakes and Southeast International Reading Association, Nashville, TN, Nov 111 5. [ERIC ED 395287]. Thornburg, D. (2008). Have we hit a "Sputnik" moment again? Thornburg Center for Space Exploration. Retrieved October 9, 2010 from www.tcse k12.org Timothy, T., Chee, T., Beng, L., Sing, C., Linga, K., Li, C. & Mun, C. (2010). The self directed learning with technology scale (SDLTS) for young students: An initial development and validation. Computers & Education, 55(4),17641771. Toback, D., Mershin, A. & Novikova, I. (2005). Integrating w ebbased teaching tools into large university physics courses. Physics Teacher, 43 594 597. Tomlinson, C. (2003). Fulfilling the promise of the differentiated classroom. Danvers, MA: Association for Supervision and Curriculum Development. Tomlinson, C & McTighe, J. (2006). Integrating Differentiated Instruction and Understanding by Design: Connecting Content and Kids. Alexandria, VA: Association for Supervision and Curriculum Development.
126 Tough, A. (1967). Learning Without a Teacher. A study of tasks and assistance during adult self teaching projects Toronto: Ontario Institute for Studies in Education. Twigg, C. (2003). Improving learning and reducing costs: New models for online learning. Educause Review Sep/Oct, 28 38. US Department of Edu cation. (2007). Report of the Academic Competitiveness Council. Retrieved October 10, 2010 from http://www.ed.gov/about/inits/ed/competitiveness/acc mathscience/r eport.pdf US Department of Education. (2009). Evaluation of EvidenceBased Practices in Online Learning: A MetaAnalysis and Review of Online Learning Studies. van Joolingen, W., de Jong, T. & Dimitrakopoulou, A. (2007). Issues in computer supported inquiry learning in science. Journal of Computer Assisted Living, 23, 111119. van Rens, L., Pilot, A., & van der Schee, J. (2010). A framework for teaching scientific inquiry in upper secondary school chemistry. Journal of Research in Science Teachin g, 47 (7), 788 806. Verschaffel, L., De Corte, E., Lasure, S., Van Vaerenbergh, G., Bogaerts, H., & Ratinckx, E. (1999). Learning to solve mathematical application problems: A design experiment with fifth graders. Mathematical Thinking and Learning, 1, 195229. Wagner, E. (2006). On designing interaction experiences for blended learning. In C. Bonk & C. Graham (Eds.), Handbook of Blended Learning. San Francisco, CA: John Wiley & Sons. Wang, F. & Hannafin, M. (2005). Design based research and technology enhanced learning environments. Educational Technology Research and Development, 53(4), 5 23. Warnakulasooriya, R., Palazzo, D. & Pritchard, D. (2005). Evidence of problem solving transfer in webbased Socratic tutor. Published in the Proceedings of the 2005 Physics Education Research Conference, pp. 4143. Weiser, E. (2008). Students control their own learning: A metacognitive approach. Teaching Children Mathematics, 15, 90 95. Wells, M. & Hestenes, D. (1995). A modeling method for high sch ool physics instruction. American Journal of Physics, 63(7), 606 619. Wells, M., Hestenes, D. & Swackhamer, G. (1995). A modeling method for high school physics instruction. American Journal of Physics, 63(7), 606619. Retrieved June 20, 2008 from http://modeling.asu.edu
127 Wetzel, D. (2009). Science blogs in middle and high school science. Retrieved March 29, 2009 from http://Suite101.com at http://teacherstipstraining.suite101.com/article.cfm/science_blogs_in_high_and_ middle_school_classes Whitehouse Press. (2010). President Obama to announce major expansion of "Educate to Innovate" campaign t o improve science, technology, engineering and math (STEM) education. Retrieved October 9, 2010 from http://www.w hitehouse.gov/thepress office/2010/09/16/president obamaannouncemajor expansioneducateinnovatecampaignimpro Wieman, C. (2008). Oersted Medal Lecture 2007: Interactive simulations for teaching physics: What works, what doesn't, and why. America n Journal of Physics, 76(4), 393 399. Xu, Z. (2008). When hybrid learning meets blended teaching: Online computer mediated communication (CMC) discourse and classroom faceto face (FTF) discourse Learning and Teaching chapter in Lecture Notes in Computer Science. 5169. Retrieved June 24, 2009 from http://www.springerlink.com/content/118706t0778t6872/ .analysis Zou, X. (2000). The use of multiple representations and visualizations in student learning of introductory physics: An example from work and energy. Unpublished doctoral dissertation. Ohio State University, Columbus, OH. Zou, X. (2006). The sandbag collision. The Physics Teacher, 44, 448 450.
128 BIOGRAPHICAL SKETCH Inn ovation comes from having two or more specialties and applying the framework of one specialty to the framework of another. ( Friedman, 2008) r In his The World is Flat 3.0 lecture at MIT, Thomas Friedman (2008) encouraged 'mash ups' of skill set s as a way to contribute to society. Wendy Athens mash up of expertise combines science (specifically chemistry and physics ), technology, and education within a worldview of social constructivism. Her expertise in secondary science stems from her undergraduate and graduate work in chemistry, seven years of experience in the chemical instruments and computer business while at Hewlett Packard, and seven years in secondary chemistry and physics education. She refined her understanding of effective integration of technology in the classroom by conducting her own classroom research while simultaneously completing the doctoral program in educational technology within the School of Education at the University of Florida Her interest in self directed mastery learning orig inated from her motivation to develop [big view] America's talent in science and technology by [small view] better developing her students. Blooms original writing on mastery learning was eloquent and compelling, and its implementation was proven effecti ve over decades of evaluation ( Bloom, 1968; Gagne, 1988). Wendy hoped that t echnolog ical innovation would make the integration of mastery learning protocols in her classroom easier and thereby free her to invest more oneto one time with students thus providing more individualized feedback and instruction. This led to the need for SDL activities to productively occupy others while she tutored, which birthed the VOLO. The VOLO idea developed while reading differentiated instruction literature that encouraged choice in learning activities.
129 Wendy believes the Google Apps/VOLO infrastructure of this capstone project was a complete success and she will continue to use it. She will finetune the program with the selection of VOLO activities, the FTF/online balance, the wholeclass/small group/individual balance, and by tweaking the allocation of points to stress concept development. But refining the infrastructure is easy compared to the real challenge of incorporating explicit instruction in SDL strategies into the curriculum. She believes the contrast between successful and unsuccessful students in this capstone made it very clear: There are no shortcuts to learning. Time plus effort equals learning (Chickering & Gamson, 1991, p. 66). To better equi p students to be successful in her physics courses and beyond, she must simultaneously teach on a supplemental plane to science; specifically, she must foster development of SDL skills.