A STUDY TO COMPARE THE RELATIVE EFFECTIVENESS OF SELECTED LABORATORY AND CLASSROOM PROGRAMS IN DRIVER AND TRAFFIC SAFETY EDUCATION Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY CHARLES E. .McDANIEL 1970 p439: LIBRARY Michigan State - University ,1 This is to certify that the thesis entitled A STUDY TO COMPARE THE RELATIVE EFFECTIVENESS OF SELECTED LABORATORY AND CLASSROOM PROGRAMS IN DRIVER AND TRAFFIC SAFETY EDUCATION presented by Charles E. McDaniel has been accepted towards fulfillment of the requirements for Ph 0 D 0 degree in Education 62 , Liz/UT #LLL “L7—r/ Major professor DateAUGUST 14 , 1970 0-169 §' BINDING In “5 ”OAR & SONS' BOOK BIIIOIRY "IC.’ LIBRADF HN' ' HS mw". - ~ ;A\:_ JAN 1 '2 gong Fm}; mod .35? 1 1 2009 \ 9911L9 ABSTRACT A STUDY TO COMPARE THE RELATIVE EFFECTIVENESS OF SELECTED LABORATORY AND CLASSROOM PROGRAMS IN DRIVER AND TRAFFIC SAFETY EDUCATION BY Charles E. McDaniel The purposes of this investigation are: 1. To compare the relative effectiveness of four selected driver and traffic safety education laboratory programs--Standard, Simulator, Range, and Four-Phase. 2. To compare the relative effectiveness of three selected driver and traffic safety education classroom programs: (a) thirty-hour classroom instruction; (b) thirty hours classroom instruction plus fifteen hours drivocator instruction, and (c) forty-five hours classroom instruction. 3. To determine whether the relative effective- ness of the laboratory and classroom program indicate differences between female and male students. The criterion measures used to evaluate student performance were divided into three areas: (a) driving knowledge, (b) driving performance, and (c) traffic analysis. Charles E. McDaniel The following is a summary of the findings where significant differences were found: 1. There was a significant gain in the mean scores from the Pre-Test to the Post—Test for Driving Knowledge attained by students in all categories. 2. On the Post-Test for Driving Performance, sig- nificance at the .05 level existed among the mean deduc- tion scores attained by students assigned to the four laboratory programs, in favor of a simulator program over a combination of a standard and a range program. 3. On the Post-Test for Driving Performance-- Part I, significance at the .05 level existed between the mean scores attained by students assigned to a standard and a simulator program, in favor of the standard program. 4. On the Post-Test for Driving Knowledge, signi- ficance at the .01 level existed among the mean scores attained by students assigned to the three classroom pro- grams. Significance at the .05 level existed between the mean scores attained by students assigned to a thirty hour classroom plus fifteen hour drivocator and a forty-five hour classroom program, in favor of the thirty hour class- room plus fifteen hour drivocator program. 5. Significance was found at the .10 level between the thirty hour classroom plus fifteen hour drivocator and the forty-five hour classroom program in favor of the thirty hour classroom plus fifteen hour drivocator program. Charles E. McDaniel 6. On the Post-Test for Driving Knowledge and Driving Performance, significance at the .001 level exis- ted between the mean scores attained by female and male students, in favor of the male students. 7. On the Post-Test for Driving Performance--Total, significance at the .01 level existed between the rejection and failure rates of female and male students taking the McGlade Road Test, with the female students having a higher frequency of rejection and failure than the male students. 8. On the Post-Test for Traffic Analysis, signifi- cance at the .025 level existed between the mean scores attained by female and male students, with the female stu- dents having a higher mean score than the male students. There were no interactions among the factors in the analysis, although a three-way interaction between laboratory, classroom and sex approached significance at the .05 level on the Post-Tests for Driving Knowledge and Driving Performance. The coefficients of intercorrelation among the six criterion measures were significant at the .05 level or higher except for the following: Post—Test for Driving Performance--Part II vs. Pre-Test for Driving Knowledge, Post-Test for Driving Performance--Part II vs. Post-Test for Driving Knowledge, and Post-Test for Traffic Analysis vs. Post-Test for Driving Performance-—Part I. Charles E. McDaniel An expected high correlation resulted between each part score and the total score on the Post—Test for Driv- ing Performance. A low correlation resulted between the Post-Tests for Driving Performance and the Pre- and Post-Tests for Driving Knowledge. A low correlation also existed between the Post-Test for Traffic Analysis and the Pre- and Post- Tests for Driving Knowledge. A STUDY TO COMPARE THE RELATIVE EFFECTIVENESS OF SELECTED LABORATORY AND CLASSROOM PROGRAMS IN DRIVER AND TRAFFIC SAFETY EDUCATION BY . .L‘ (i ‘ Charles E5 McDaniel A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education 1970 IN MEMORIAM While the study was in progress, Mr. Robert L. Smith, one of the instructors, died of cancer after a short illness. During the time Bob was engaged in the project, he gained the respect and admiration of his fellow workers and students. Bob was a very energetic and con- scientious teacher. He had a sound and comprehensive understanding of today's traffic problem. One of Bob's outstanding qualities was his belief that driver educa— tion was important to the student, and his teaching championed this belief. ii ACKNOWLEDGMENTS The Washington Driver and Traffic Safety Education Project received support and cooperation from a large num- ber of individuals and groups. Space prohibits the listing of all persons or groups who contributed to the study. However, the following individuals made outstanding con- tributions: Advisory Committee Dr. Dr. Dr. Dr. Dr. Dr. James E. Aaron, Coordinator, Safety Center, Southern Illinois University Richard W. Bishop, Associate Professor, Florida State University Leon Brody, Director of Research, Center for Safety, New York University William Covert, Instructor, Highway Traffic Safety Center, Michigan State University Robert O. Nolan, Associate Professor, Highway Traffic Safety Center, Michigan State University Thomas A. Seals, Curriculum Coordinator, Safety, San Diego County Board of Education Marland K. Strasser, Professor, San Jose State College Renton School District No. 403 Dr. Mr. Earl W. Hobbs, Superintendent Karl J. Weber, Principal, Renton High School Renton High School Faculty and Staff iii Mr. Wilbur Luft, Driver Education Department Chairman Lucile Mehus, Tenth Grade Counselor Mr. Charles Moore, Tenth Grade Counselor Mr. William Reynolds, Grounds Department Mr. William Belmondo and the Department of Trans- portation Mrs. Driver Education Instructors Mr. Mr. Mr. Mr. Mr. Mr. Mr. Eugene B. Bryan Mr. Lloyd R. Riddle Don M. Carnahan Mr. Eric Roberts Harold Conley Mr. Larry E. Schwitters Virgil Gross Mr. Robert L. Smith F. Joseph Koenig Mr. Gordon Wingard Jerry C. McVay Mr. Robert Wraith Helmer Paulson Driver and Safety Education Programs--Office of the Superintendent of Public Instruction Mr. Clifford Boyce, Supervisor Mr. William F. Hiblar, Associate Supervisor Department of Motor Vehicles Mr. Douglas Toms, Director Robert O. Koch, Assistant Director, Vehicle Services Mr. Driver License Examiners Mr. Mr. Mr. Mr. Mr. Mr. Mr. Mr. Mr. Mr. Mr. William L. Backman Donald L. Bonner John H. English Joe LeBlonde Edward R. Lindsey Lawrence W. Martin James H. Meier Charles B. Montgomery, Jr. Merle J. Schafer Ronald J. Selby Robert C. Updegraff iv Special appreciation is given to the National Com- mission on Safety Education for the preparation of the initial proposal for the project; the 1967 Washington Legislature for appropriating the funds to conduct the study; and the professional guidance given by the members of my doctoral committee--Dr. Robert O. Nolan, Chairman, Dr. Charles A. Blackman, Dr. William A. Mann, and Mr. Gordon H. Sheehe. Special recognition is given to Mr. Robert C. Munson for his assistance as Assistant Project Director and Mrs. Angie Dorian for her assistance as my secretary. An outstanding contribution to the study was per- formed by Dr. Alan Metcalf, Director of Research for the Office of the Superintendent of Public Instruction, through his guidance and support. Appreciation is expressed to Mr. Louis Bruno, State Superintendent, and the Office of the Superintendent of Public Instruction for their cooperation and willing- ness to let me use the information obtained from the Washington Driver and Traffic Safety Education Research Project for my dissertation. The following individuals contributed to the pro- ject in the consulting capacity: Michigan State University Dr. Andrew C. Porter Mr. J. Robert Shinn Sister Thomas More Emery, O. P. V University of Washington Dr. Gilbert Sax Dr. Alan J. Klockars National Commission on Safety Education Dr. Ray F. Wahl Data Processing Department--Office of the Superin- Dr. tendent ofdfublic Instruction Mr. Francis Flerchinger Mr. Maurice Whitney Mrs. Iris Cline Francis McGlade, Safety Division, Department of the Army During the fall, 1967, an in-service education program was conducted for the Renton High School driver education instructors. Appreciation is given to the following people who participated: Mr. Dr. Mr. Mr. Mr. Mr. William Covert, Michigan State University Roy B. Sawhill, University of Washington Lewis Ross, Metropolitan Planning Commission Elmer Hawley, Pacific Indemnity Insurance Co. Ellis Lind, Western Insurance Information Service Glenn L. Crawford, Department of Motor Vehicles Sergeant John E. Buff, Renton Police Department Trooper Fred L. Pilon, Washington State Patrol Mr. Mr. Mr. Mr. Mr. Mr. Richard Boyer, Allstate Insurance Company Lawrence A. Bewley, Allstate Insurance Company Richard Gauthier, Aetna Life and Casualty Jack E. Rounds, Raytheon Learning Systems Edward J. Steinberg, U-Haul Company of Wash- ington Ron Frank, U-Haul Company of Washington vi TABLE OF CONTENTS IN MEMORIAM . . . . . . . ACKNOWLEDGMENTS . . . . . LIST OF TABLES . . . . . LIST OF FIGURES . . . . . CHAPTER I. II. III. INTRODUCTION . . . Background . . . . Statement of the Problem . . . . Criterion Measures Hypotheses . . . . Definitions . . . . Delimitations . . . Basic Assumptions . Significance . . . Tasks (Sub-Problems) REVIEW OF LITERATURE PROCEDURES . . . . Sub-Problem One Selection of Study Sub-Problem Two In-Service Education of Instructors Location . vii Page ii iii xi xvi 12 18 18 20 21 22 61 62 66 Chapter IV. Sub-Problem Three Development of Standardized Procedures and Lesson Content . . . . . . . . . . Sub-Problem Four Provision of Necessary Facilities and Equipment 0 O O O O O O O I O O O O O Sub-Problem Five Design of Study . . . . . . . . . . . Sub-Problem Six Assignment of Students . . . . . . . . Sub-Problem Seven Assignment of Instructors . . . . . . Sub-Problem Eight Selection and Development of Valid and Reliable Measurement Instruments . . . Sub-Problem Nine Collection and Statistical Treatment Of Data 0 O O O O O I O I I O O O O O ANALYS IS OF DATA 0 O O O O O O O O O O O Pre-Test for Driving Knowledge . . . . . Post-Test for Driving Knowledge . . . . . Gain from Pre-Test to Post-Test for Driving Knowledge . . . . . . . . . . . . Post-Test for Driving Performance--Total Post-Test for Driving Performance-~Part I Post-Test for Driving Performance--Part II Post-Test for Traffic Analysis . . . . . viii Page 68 73 76 81 82 85 97 101 102 104 111 114 134 138 143 Chapter Linear Correlations . . . . . . Summary of Findings . . . . . . V. SUMMARY AND CONCLUSIONS . . . . Criterion Measures . . . . . . Results of Pre-Test for Driving Discussion of Findings . . . . Conclusions . . . . . . . . . . Recommendations . . . . . . . . Suggested Research . . . . . . BIBLIOGRAPHY . . . . . . . . . . . . APPENDIX . . . . . . . . . . . . . . A. Laboratory Lesson Sequence Simulator Lesson Sequence . On-Street Lesson Sequence . On-Street Lesson Plan . . . B. Evaluation Instruments . . Knowledge . . . . . . . . . Driving Performance . . . . Traffic Analysis . . . . . Knowledge C. Instructional Treatment Groups . . . Number of Students . . . . Teaching Experience . . . . Instructor Assignments . . Range Drawing . . . . . . . ix Page 148 150 159 160 164 164 181 183 185 191 200 200 201 203 205 207 207 215 216 225 225 227 229 230 Chapter Page Instructors and Per Pupil Costs . . . . . . 231 D. Supplemental Analysis--Post-Test for Driving Knowledge . . . . . . . . . . . . . 233 Table 1. 10. 11. LIST OF TABLES Means and Standard Deviations of Scores on the Pre-Test for Driving Knowledge among the Four Laboratory Programs . . . . . . . . . . . . . . Means and Standard Deviations of Scores on the Pre-Test for Driving Knowledge among the Three Classroom Programs . . . . . . . . . . . . . . . Means and Standard Deviations of Scores on the Pre-Test for Driving Knowledge Between Female and Male StUdentS I I I I I I I I I I I I I I I Analysis of Variance, Pre-Test for Driving KnOWledge I I I I I I I I I I I I I I I I I I I Means and Standard Deviations of Scores on the Post-Test for Driving Knowledge among the Four Laboratory Programs . . . . . . . . . . . . . . Means and Standard Deviations of Scores on the Post-Test for Driving Knowledge among the Three Classroom Programs . . . . . . . . . . . . . . . Means and Standard Deviations of Scores on the Post-Test for Driving Knowledge between Female and Male Students . . . . . . . . . . . . . . . Analysis of Variance, Post-Test for Driving KnOWIedge I I I I I I I I I I I I I I I I I I I Scheffé's Test for Multiple Comparisons, Post- Test Scores for Driving Knowledge--Classroom Programs I I I I I I I I I I I I I I I I I I I I Summary, T-Test for Significance of Gain between Means of Pre-Test and Post-Test Scores for Driving Knowledge . . . . . . . . . . Chi—Square Test for the Rejection of Students on the McGlade Road Test among the Laboratory Programs I I I I I I I I I I I I I I I I I I I I xi Page 102 103 103 105 106 106 107 108 110 112 115 Table Page 12. Chi-Square Test for the Rejection of Students on the McGlade Road Test among the Classroom Programs I I I I I I I I I I I I I I I I I I I I 116 13. Chi—Square Test for the Rejection of Students on the McGlade Road Test between the Sexes . . . 117 14. Chi-Square Test for the Rejection of Female Students on the McGlade Road Test among the Laboratory Programs . . . . . . . . . . . . . . 119 15. Chi—Square Test for the Rejection of Male Students on the McGlade Road Test among the Laboratory Programs . . . . . . . . . . . . . . 120 16. Chi—Square Test for the Rejection of Female Students on the McGlade Road Test among the Classroom Programs . . . . . . . . . . . . . . . 121 17. Chi-Square Test for the Rejection of Male Students on the McGlade Road Test among the Classroom Programs . . . . . . . . . . . . . . . 122 18. Chi-Square Test for the Failure of Students on the McGlade Road Test among the Laboratory Programs . . . . . . . . . . . . . . . . . . . . 123 19. Chi-Square Test for the Failure of Students on the McGlade Road Test among the Classroom Programs I I I I I I I I I I I I I I I I I I I I 124 20. Chi-Square Test for the Failure of Students on the McGlade Road Test between the Sexes . . . 125 21. Chi-Square Test for the Failure of Female Students on the McGlade Road Test among the Laboratory Programs . . . . . . . . . . . . . . 126 22. Chi-Square Test for the Failure of Male Students on the McGlade Road Test among the Laboratory Programs . . . . . . . . . . . . . . 127 23. Chi-Square Test for the Failure of Female Students on the McGlade Road Test among the Classroom Programs . . . . . . . . . . . . . . . 128 24. Chi-Square Test for the Failure of Male Students on the McGlade Road Test among the Classroom Programs . . . . . . . . . . . . . . . 129 I xii Table 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. Means and Standard Deviations of Deduction Scores on the Post—Test for Driving Per- formance among the Four Laboratory Programs . . . . . . . . . . . . . . . . . . Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Per- formance among the Three Classroom Programs . . . . . . . . . . . . . . . . . . Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Per- formance between Female and Male Students . Analysis of Variance, Post-Test for Driving Performance . . . . . . . . . . . . Scheffé's Test for Multiple Comparisons, Post-Test Deduction Scores for Driving Performance--Laboratory Programs . . . . . . Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Per- formance--Part I, among the Four Laboratory Programs . . . . . . . . . . . . . . . . . . Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Per- formance--Part I, among the Three Classroom Programs . . . . . . . . . . . . . . . . . . Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Per- formance--Part I, between Female and Male Students . . . . . . . . . . . . . . . Analysis of Variance, Post-Test for Driving Performance--Part I . . . . . . . . . . . . Scheffé's Test for Multiple Comparisons, Post-Test Deduction Scores for Driving Per- formance--Part I--Laboratory Programs . . . Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Per- formance--Part II, among the Four Laboratory Programs . . . . . . . . . . . . . . . . . . xiii Page 130 130 131 132 134 135 136 136 137 139 140 Table Page 36. Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Per- formance--Part II, among the Three Classroom Programs . . . . . . . . . . . . . . . . . . . . 140 37. Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Per- formance--Part II, between Female and Male Students . . . . . . . . . . . . . . . . . . . . 141 38. Analysis of Variance, Post-Test for Driving Performance--Part II o o o o o o o o o o o o o o 142 39. Means and Standard Deviations of Scores on the Post-Test for Traffic Analysis among the Four Laboratory Programs . . . . . . . . . . . . 143 40. Means and Standard Deviations of Scores on the Post-Test for Traffic Analysis among the Three Classroom Programs . . . . . . . . . . . . 144 41. Means and Standard Deviations of Scores on the Post-Test for Traffic Analysis between Female and Male StUdentS I I I I I I I I I I I I I I I 144 42. Analysis of Variance, Post-Test for Traffic AllaIYSiS O I o o o o o o o o o o o o o o o o o I 146 43. Scheffé's Test for Multiple Comparisons, Post-Test Scores for Traffic Analysis-- Classroom Programs . . . . . . . . . . . . . . . 147 44. Intercorrelations among the Six Criterion Measures . . . . . . . . . . . . . . . . . . . . 149 45. Parallel between Mean Deduction Scores on the Post-Test for Driving Performance and Number of Hours of Instruction in the Automobile among the Laboratory Programs . . . . . . . . . 168 46. Difference in the Number of Instructors and Per Pupil Cost for a Standard Program, a Simulator Program, a Range Program, and a Four-Phase Program . . . . . . . . . . . . . . . 171 47. Means and Standard Deviations of Scores on the Post-Test for Driving Knowledge among the Four Laboratory Programs . . . . . . . . . . 233 xiv Table Page 48. Means and Standard Deviations of Scores on the Post-Test for Driving Knowledge among the Three Classroom Programs . . . . . . . . . 234 49. Means and Standard Deviations of Scores on the Post-Test for Driving Knowledge between Female and Male Students . . . . . . . . . . . 234 50. Analysis of Variance, Post-Test for Driving Know1edge I I I I I I I I I I I I I I I I I I 2 36 51. Mean Scores in the Two-Way Interaction between Classroom and Sex . . . . . . . . . . 237 52. Scheffé's Test for Multiple Comparisons, Post-Test Scores for Driving Knowledge-- Classroom Programs . . . . . . . . . . . . . . 238 XV LIST OF FIGURES Figure Page 1. Time Allotment and Scheduling Order of Various Phases of Driver and Traffic Safety Education . . . . . . . . . . . . . . . 52 Increase or Decrease in Scores from the Pre-Test for Driving Knowledge to the Post-Test for Driving Knowledge . . . . . . . . 113 xvi CHAPTER I INTRODUCTION Background During the past three decades this nation has witnessed a mammoth increase in the comprehensiveness and complexity of its transportation network. The jet air- craft of today makes any place in the United States acces- sible within a matter of hours. The motor vehicle, also, has played a most important part in the development of this nation into a mobile society. An ever-increasing network of streets, highways, and expressways is making every part of this country accessible to the motoring public. Today there are 3,704,914 miles of streets and highways in the United States.1 There has also been a companion growth in the number of motor vehicles and licensed drivers. There are 103,000,000 licensed drivers operating 79,700,000 motor vehicles.2 In the State of 1U. S. Department of Transportation, Highway Sta- tistics 1967 (Washington: U. S. Government PrintIng Office), p.3167. 2National Safety Council, Accident Facts (1968 Edition; Chicago: National Safety CounciIY, p. 40. Washington 1,852,000 motor vehicles are operated by 1,705,000 licensed drivers on 72,424 miles of highways.3 The increasing dependence of American society upon this convenient form of mobility, however, has been accom- panied by the increasingly complex and severe problem of accidents. Traffic accidents basically involve an inter- relationship between the road, the motor vehicle, and the operator, plus additional factors such as other highway users. It appears logical, therefore, that prevention efforts should deal with each of these three components. Implementation of engineering improvements in con- struction and design of trafficways and motor vehicles may result in fewer accidents as well as reduced severity of injury to occupants of vehicles in case an accident occurs. Without due attention to the human factor, however, sub- stantial gains in traffic accident prevention appear to be improbable. The greatest potential may lie in the area of improved driver behavior. When the total driver population exhibits desirable operational performance, backed by in- depth knowledge and understanding of the complex traffic environment, a possible ultimate may be reached in safe and efficient use of motor vehicle transportation. Many traffic safety authorities feel that education has a positive effect on driver and pedestrian behavior. 3U. S. Department of Transportation, op. cit., cover, pp. 43, 167. The school should provide meaningful and relevant learning experiences in traffic safety education for all its stu- dents. Moreover, these experiences should be provided at all grade levels beginning with pedestrian performance in the primary grades and culminating in a high school driver and traffic safety education course consisting of both classroom and laboratory instruction, taught by competent, qualified and certified instructors. William Haddon, Jr., Director, National Highway Safety Bureau, has said: I think it is the unique opportunity, responsibil- ity, and privilege of the nation's driver education programs, and of all the dedicated people who carry them out, to have a greater influence than has any other single group related to highway safety. Perhaps what I really mean is that there are ac- tually two goals for driver education. First, the program should provide basic instruction in driving techniques, a knowledge of how to handle a car in special circumstances, environments, and emergencies, and a knowledge of local and state motor vehicle and traffic laws and ordinances. Second, and just as important, we should be turning out a far more knowledgeable breed of citizen who will know enough about highway safety to demand and support higher and higher standards all along the line, in relation to each of the three phases [pre-crash, crash, and post-crash] of this continuing and tragic national problem.4 4William Haddon, Jr., "Haddon on Highway Safety," Amalo (Northbrook, Illinois: Allstate Insurance Company, Winter, 1968), p. 6. While driver and traffic safety education had its start during the 1930's, quantitative and qualitative ex- pansion was slow during the early years. For example, during the 1941-42 school year only 7,500 students re- ceived a course in driver and traffic safety education in the United States.5 In the State of Washington, at that time, it is estimated that not more than a half dozen public high schools offered such courses. As late as 1962, approximately 3,000 students were enrolled in an approved course in driver and traffic safety education in the secondary schools of Washington. An additional 1,000 students were enrolled in substandard programs. After World War II, increased use of automobiles indicated a growing need for more and better driver educa- tion. Efforts by national and state support organizations, traffic officials, school administrators, lay leaders, and legislators combined to intensify the attention to the serious public problem of accident prevention and the develOpment of better traffic citizens. These coordinated efforts resulted in a most comprehensive growth. By the 5Earl Allgaier, "Results of the Driver Education Program - 1936-65" (No. 3616; Washington: American Auto- mobile Association, October, 1965) . 6Results of survey in the fall of 1962 conducted by'the Supervisor of Driver and Safety Education Programs, Office of the Superintendent of Public Instruction, State of Washington . 1966-67 school year, 74 per cent of the nation's public high schools were providing courses with an enrollment figure which represent 56 per cent of beginning drivers in the United States.7 Since 1962, growth of driver and traffic safety education in the State of Washington has been greatly accelerated. During 1966-67, enrollment in approved school courses exceeded 30,000 of a potential of about 55,100 who reached the minimum legal driving age during that year.8 The enrollment trend is now rising sharply as a result of existing special financial support legislation and recognition of course values. In the State of Wash- ington, such impetus was provided by the 1967 Legislature. RCW 46.20.100, Chapter 167, Laws of 1967, State of Wash- ington, requires all students under the age of eighteen years to successfully complete an approved course in driver education as a prerequisite to licensing. Nationwide, the Highway Safety Act of 1966 (P. L. 89-564) is effecting a similar positive influence on program expansion and im- provement. In order for a state to be eligible for federal funding under this Act, it must have a comprehensive 7Insurance Institute for Highway Safety, 20th Annual Driver Education Achievement Program, 1966-67 School Year TWashington: Insurance Institute for Highway SafetYY. PP. 21-23. 81bid. highway safety program approved by the Secretary of Trans- portation. Such approval may not be legally granted unless the state's comprehensive program includes continuing, improved, and/or expanded driver education administered by appropriate school officials for all school-age youth. Since states without approved highway safety programs are subject to loss of funds made available through the Act, and in addition, are subject to a 10 per cent reduction in federal subsidy for highway construction, new attention is being focused on the need for more and better driver edu- cation in the secondary schools. Meanwhile driver and traffic safety education, as indeed all disciplines taught in schools at one time or another, has been subjected to criticism. Even though critical of this subject field, Moynihan states: Now, at the hopeful beginnings of a new era, it becomes necessary to give a new cast to driver educa- tion. Although there is no conclusive proof as to the comparative effectiveness of various driver education techniques or, for that matter, the whole of present driver education practice, there is even less proof of the efficacy and value of any alternatives to present practices for communicating to the young person the rudiments of how to handle a car in modern traffic, and the associated social responsibilities. But op- erational driver education programs must continue. The problem is no different in principle than that for education in general. We have to continue with present systems even while recognized needed improvements are being studied. One would hardly advocate a moratorium on all schooling while looking for proof of better methods. 9Daniel P. Moynihan (Chairman), A Report of the Secretary's Advisory Committee on Traffic Safety (Washing- t£H1: Department of Health, Education andIWelfare, February 29, 1968), PP. 118-119., The Office of the Washington State Superintendent of Public Instruction, desiring to provide for the best possible program of instruction, and recognizing both the criticism and its responsibility under new legislation to find effective and efficient means for program expansion and improvement, proposed a research investigation designed to evaluate the relative effectiveness of selected labora- tory and classroom programs in driver and traffic safety education. The 1967 Legislature appropriated $151,000 for its conduct, described in the ensuing pages. Statement of the Problem 1 The purposes of this investigation are as follows: 1. To compare the relative effectiveness of four selected driver and traffic safety education lab- oratory programs (one-third of the students in each laboratory program receive one of the three classroom programs described on page 79): a. Standard Program (Groups IV-VI)--each student receives classroom instruction, plus six hours on-street driving instruction and twelve hours in-car observation. b. Simulator Program (Groups VII-IX)--each stu- dent receives classroom instruction, plus twelve hours simulator instruction, three hours on—street driving instruction, and six hours in-car observation. c. Range Program (Groups X-XII)--each student receives classroom instruction, plus six hours off-street multiple car driving range instruc- tion (eight lessons), three hours on—street driving instruction, and six hours in-car observation. d. Four-Phase Program (Groups I-III)--each student receives classroom instruction, plus eight hours simulator instruction, six hours off- street multiple car driving range instruction (eight lessons), two hours on-street driving instruction, and four hours in-car observation. 2. To compare the relative effectiveness of three selected driver and traffic safety education classroom programs: a. Thirty hours classroom instruction b. Thirty hours classroom instruction plus fifteen hours drivocator instruction c. Forty-five hours classroom instruction. 3. To determine whether the relative effectiveness of the laboratory and classroom programs indicates differences between female and male students. Criterion Measures The following criterion measures, described in detail in Chapter III, were used: Pre-Test for Driving Knowledge Post-Test for Driving Knowledge Post-Test for Driving Performance--Total Post-Test for Driving Performance-—Part I Post-Test for Driving Performance--Part II Post-Test for Traffic Analysis Hypotheses l. 3. There will be no significant differences among the mean scores attained on each of the criterion measures by students randomly assigned to one of four types of laboratory programs-—standard, simu- lator, range, and four-phase. Hl: A significant difference will exist among the mean scores attained on each of the criterion measures by students as- signed to the laboratory programs. There will. be no significant differences among the mean scores attained on each of the criterion measures by students randomly assigned to one of three types of classroom programs--thirty hours classroom instruction, thirty hours classroom instruction plus fifteen hours drivocator instruc- tion, and forty-five hours classroom instruction. Hzi=x=3€ H1: A significant difference will exist among the mean scores attained on each of the criterion measures by students as- signed to the classroom programs. There will be no significant differences between mean scores attained on each of the criterion measures by female and male students. 10 H1: A significant difference will exist between the mean scores attained on each of the criterion measures by female and male students. 4. There will be no significant differences among the four laboratory programs in the number of stu- dents rejected on the McGlade Road Test. H1: A significant difference will exist among the four laboratory programs in the number of students rejected on the McGlade Road Test. 5. There will be no significant differences among the three classroom programs in the number of stu- dents rejected on the McGlade Road Test. H1: A significant difference will exist among the three classroom programs in the number of students rejected on the McGlade Road Test. 6. There will be no significant differences between female and male students in the number of students rejected on the McGlade Road Test. H : X = X H1: A significant difference will exist between female and male students rejected on the McGlade Road Test. 7. There will be no significant differences among the four laboratory programs in the prOportion of students who pass and fail the McGlade Road Test. l: A significant difference will exist among the four laboratory programs in the proportion of students who pass and fail the McGlade Road Test. 8. There will be no significant differences among the three classroom programs in the prOportion of students who pass and fail the McGlade Road Test. x n xl n xl H0: H1: A significant difference will exist among the three classroom programs in the proportion of students who pass and fail the McGlade Road Test. 9. There will be no significant differences between female and male students in the proportion of stu- dents who pass and fail the McGlade Road Test. XI u x H : H1: A significant difference will exist between female and male students in the prOportion of students who pass and fail the McGlade Road Test. 10. There wilf) be no significant gain in mean scores from pre-test to post-test for driving knowledge within any of the three classroom programs, four laboratory programs, female students, or male students. H : i. = Y 0 To provide additional information regarding the interrelationships among the criterion variables, product- moment correlations were computed. Significance of 12 intercorrelations were tested against the following pro- jected null hypothesis. 11. There will be no significant linear correlations between scores on the six criterion measures. For the purpose of this study, statistical signif- icance equal to or less than the .05 level is accepted as sufficient evidence for the rejection of the null hypo- theses. Definitions Approved School Course--a course in driver and traffic safety education approved by the Washington State Office of the Superintendent of Public Instruction, con- sisting of a minimum of thirty clock hours of classroom instruction, an equivalent of six hours of behind-the-wheel 10 instruction and twelve hours of observation time. Auto Trainer--a type of driving simulator consist— ing of a driver's compartment with regular controls, a continuous canvas belt (mounted in front of the compart— ment) which moves in response to the accelerator and brake, and a miniature car which can be steered over a roadway painted on the belt.11 10Office of the Superintendent of Public Instruc- tion, 1964 Driver Education Guide (Olympia, Washington: Office of the Superintendent of Public Instruction, 1965), p. 15. 11James H. Fox, Driver Education and Driving Simu- lators (Washington: National Education Association, Na- tionaI Commission on Safety Education, 1960), p. 51. 13 Certified Instructor--a driver and traffic safety education instructor certified by the state educational certification agency as having satisfied the minimum re- quirements for certification in driver and traffic safety education. Classroom Instruction--instruction given groups of students covering the subject content in the areas of traffic citizenship, laws and regulations, characteristics of drivers, role of government, automobile use, and traffic problems. (Each student involved in this investigation was assigned randomly to one of the three classroom pro- grams: (1) thirty hours classroom instruction, (2) thirty hours classroom instruction plus fifteen hours drivocator instruction, and (3) forty-five hours classroom instruc- tion.) Driving Procedures--a set of procedures or in- structions relating to the methods to be followed in the performance of a particular skill or driving action. Driver and Traffic Safety Education--selected formal learning experiences consisting of classroom and laboratory instruction, designed to help students become good traffic citizens and use motor vehicles safely and efficiently. Drivocator--a completely automated multi-media teaching device that utilizes motion picture, filmstrip, magnetic tape, individual student responder, and a portable l4 master console control unit. Student responses are re- corded on the master control unit which gives the instruc- tor an immediate indication of student comprehension, both as individuals and as a group, of the content presented and the questions asked. Dual Control Car--a practice-driving automobile equipped with an automatic transmission, a safety belt for each occupant, and an extra brake pedal for use by the teacher during an emergency situation. Examiner--the instructors and/or driver examining supervisors of the Washington Department of Motor Vehicles involved in administering the McGlade Road Test. Failf-those students either rejected for reasons indicated in the ensuing definition of that term or reg- istering deductions of fifty-five points or more on the McGlade Road Test. In-Car Observation--the learning experiences a student receives (as a passenger) in a dual—control car exclusive of actual driving (behind the wheel). Instructional Treatment Group--a prescribed com- bination of different types and amounts of classroom and laboratory learning experiences. Knowledge Test--a test for determining the driving knowledge acquired by each student during the course. Laboratory Instruction--an extension of classroom instruction which provides students with teacher-supervised 15 traffic experiences under real and/or simulated condi- tions.12 McGlade Road Test-—a road post-test to determine the performance level of each student in operating a motor vehicle under traffic conditions.. Off-Street Multiple Car.Driving Range--a hard- surfaced area on which eight to twelve student-operated, practice-driving vehicles are used simultaneously to pro- vide a portion of laboratory instruction under the direct supervision of a teacher. The area provides environmental conditions designed for development of fundamental driving skills; road surfaces wide enough for two-way and multiple- lane traffic; intersections, curves, lane markings and signs; expressway entrance and exit, hills, and parking. Each car utilized on this facility is equipped additionally with an FM radio as a means of communication between the teacher and the student operators., On-Street Driving Instruction--se1ected student learning experiences while actually operating a dual-con- trol car on public streets and highways under the direction of a qualified and certified teacher of driver and traffic safety education seated to the right of the student driver. (To satisfy the objectives of this report, each student 12Thomas A. Seals, "An Evaluation of Selected Driver and Traffic Safety Education Courses" (unpublished doctoral dissertation, Florida State University, August, 1966), p. 13. 16 also received two hours of in-car observation for each hour of on-street driving instruction.) Passe-those students completing the McGlade Road Test with maximum deductions of fifty-four points. Perception--that facet of the driving task which involves the decision making abilities of the driver and is based on discriminating or interpretating the visual stimuli that are received on a continuous basis while op- erating a motor vehicle; the identification of incoming stimuli, especially those related to the driving task; and the organization of the incoming stimuli for identification responses or patterns of behavior.1 Project--this is construed to reflect the sum total of investigative activities from September, 1967, through June, 1968. Qualified Instructor--an instructor in this study who was certified (as defined) to teach driver and traffic safety education in the State of Washington and who was involved in the pre-study in-service education program as described in Chapter III. Reject--those students whose Road Test was termi- nated because they were involved in an accident, committed 13James E. Aaron and Marland K. Strasser, Driver and Traffic Safety Education--Content, Methods and Organ— ization (New York: The Macmillan Company, 1966), pp. 74: 75"— ' 17 a dangerous action, committed a clear violation of any traffic law, exhibited a lack of cooperation, or refused to perform as instructed by the examiner.14 Simulation--a teaching method employing both films and electromechanical devices, designed to represent the driver's compartment of the automobile for student devel- opment of prOper judgment and behavior responses as well as manipulative skills.15 Studyf-that portion of the project from January, 1968, through June, 1968. Teaching Points--a predetermined group of instruc- tions, i.e., items of emphasis, utilized by each instructor to supplement the driving procedures. "You Are the Jury? Traffic Analysis Test--a post- test administered each student to ascertain his ability to analyze a traffic accident, identify the causes of the accident, and suggest how the accident could have been prevented. l4Francis S. McGlade, A New Road Test for Use in. Driver Licensin ,-Edugation and Employment (New York: New York University, Center for Safety Education, 1961), p. 22. 15Fourth National Conference on Driver Education, Policies and Practices for Driver and Traffic Safety Edu- cation (Washington: National Education Association, Na- tional Commission on Safety Education, 1964), p. viii. 18 Delimitations This investigation is limited to an examination of driving knowledge, ability to analyze a traffic accident in determining how the accident could have been prevented, and driving performance possessed by student groups which have completed one of four types of high school driver and traffic safety education laboratory programs, and one of three types of classroom programs. It is further limited to include only those sopho- more students at Renton High School, Renton, Washington. It does not attempt to measure behavioral aspects of drivers, nor is any attempt made to predict the driving habits or behavior of a person. Neither is any attempt made to prove that any one of the instructional conditions treated in this investiga- tion comprises an ideal program in driver and traffic safety education. This project is further limited by the amount of time available. Basic Assumptions 1. The McGlade Road Test16 is accepted as a valid and reliable instrument for determining the level of driving performance by student groups. 16McGlade, loc. cit. (see Appendix C). 19 2. The Knowledge Test developed by Brodyl7 is accepted as a valid and reliable instrument for measuring driving knowledge possessed by students. 3. The "You Are the Jury" Traffic Analysis Test18 is accepted as a valid and reliable instrument for measuring a student's ability to analyze a traffic accident, to identify the cause of the accident, and to determine how the accident could have been prevented. 4. The procedural training provided for testing per- sonnel19 is assumed sufficient for administration of knowledge, traffic analysis, and driving per- formance tests. 5. Standardization of instruction is assumed as having been attained through development of driving pro- cedures and teaching points, the preplanning of all classroom and laboratory lessons and their correlation to each other; the conduct of an in— tensive in-service education program; weekly meet- ings of the total staff, and periodic meetings of teachers concerned with a specific phase of instruction. l7National Test in Driver Education (Special Form:' New York: New York University, Center for Safety Educa- tion, 1967) (see Appendix C). 18"You Are the Jury" Traffic Analysis Test, devel— oped as a part of this investigation (see Appendix C). 19As described in Chapter III, pp. 66-67. 20 6. Group scores attained on the driving knowledge, road performance, and traffic analysis tests are assumed to constitute an acceptable qualitative measure of instruction. 7. The random assignment technique employed is ac- cepted as assuring that the comparison groups achieved balance with respect to factors which might have affected the results. Significance Field experience in many states indicates that the employment of innovative techniques of instruction in driver and traffic safety education constitutes a valuable contribution to such education for young drivers. A de- terrent to progress in many states is the reluctance of administrators and teachers to accept the results achieved in simulation, range, and four-phase programs as comparable to the results of the standard or traditional driver edu- cation program. However, there has also been insufficient research to clearly indicate which type of driver and traffic safety education program is superior. Since this investigation concerns not only the relative effectiveness of selected laboratory and classroom programs, but also the differential effect that these prOgrams may have on female and male students, the outcomes may have a signif- icant influence on the future structure of driver and traf- fic safety education courses in Washington and elsewhere. 21 Tasks (Sub-Problems) The following tasks are identified as sub—problems and serve as an outline for the description of procedures in Chapter III of this report: 1. Selecting a school large enough in student popula- tion to provide a statistically sound study sample. Providing a staff of instructors with the necessary competencies to provide equivalent and quality in- struction in the different instructional treatment groups. Developing standardized procedures and lesson con- tent for student instruction. Providing the necessary facilities and equipment to conduct4the study. Creating an experimental design to compare the relative effectiveness of four selected laboratory programs and three selected classroom programs, and to determine the significance of the sex factor. Assigning students randomly to each of the differ- ent instructional treatment groups. Assigning instructors to the different phases of instruction. Selecting and/or developing valid and reliable measurement instruments and standardizing proced- ures for administering the tests. Collecting and applying statistical treatment of the data. CHAPTER II REVIEW OF LITERATURE This investigation, the Washington Driver and Traffic Safety Education Study, was concerned with the. analysis and evaluation of school programs of both class- room and laboratory instruction in driver and traffic safety education. Another component aspect comprised the difference in achievement between female and male students assigned to selected laboratory and classroom programs. A review of existing literature reveals that very little investigation has been performed in either the classroom phase of instruction or the comparative achievement of female and male students in driver and traffic safety education programs. There have been several which have evaluated the relative effectiveness of different combina- tions of laboratory programs. This chapter presents the data available and the parts of these investigations which are pertinent to this study. Although inadequate at the present time, the need for development of more effective techniques of instruction has been voiced. The Wisconsin Department of Public In- struction conducted a three-day workshop on the classroom 22 23 curriculum in June, 1967. Its concern was expressed in the following excerpt from the outline/schedule distributed to the sixteen participants prior to the workshop. The classroom instruction is the key to success in a driver education program, regardless of the labora- tory methods employed. It is this phase of the total instructional program that determines the foundation for which all other phases must depend. Because of this, and the fact that too little investigation into this area has been attempted in the recent past, it is imperative that the classroom curriculum be evaluated. The objectives, content, methods, and presentation technique must be scrutinized to see what changes (modifications) in philosophy regarding these areas should be undertaken. A well-balanced program (within the time allowed) of classroom instruction to meet the needs of today's students will be the primary purpose of this workshop study. Today's students--tomorrow's drivers--demand an up-to-date classroom instructional prOgram utilizing recent advances in this curriculum. A number of states have developed driver education guides which have included sections on classroom instruc- tion. In referring to such instruction, one of these, the Florida Driver Education Guide, states: In the classroom phase learning experiences empha- size personal and social problems related to the safe and efficient movement of traffic. One major aim is to emphasize the desirable role of the pedestrian and driver in traffic, and another is to develop the knowledge and attitudes needed for safe use of traffic facilities. 1Wisconsin Department of Public Instruction, Wis- consin Clgssroom Currigulum Instructional Driver Education ‘Workshop Proceedin 3, June 15-17, 1967 (Madison: Wisconsin Department of Public Instruction), p. 40. 2Florida Department of Education, A Guide, Driver gfiucation in Florida Secondary Schools (Bulletin No. 6; firaIlahassee: State Department of Education, 1963), p. 26. 24 The Washington Driver Education Guide has expressed a similar view: Classroom instruction is designed to give the stu- dent a preliminary and theoretical knowledge of all the problems and skills a driver must have while he is undergoing actual behind-the-wheel training outside of class. The complete program of classroom instruction should include the following six units: The Traffic Problem The Driver The Automobile Laws and Regulations Driving Skills 3 The Pedestrian and Other Highway Users The Fourth National Conference on Driver Education recommended that a complete prOgram of classroom and lab- oratory instruction include: TRAFFIC CITIZENSHIP: responsibility to other drivers and highway users . . . community, family, self, etc. . .-. attitudes of safe living . . . courtesy and manners . . . support of public officials . . . traffic control devices LAWS AND REGULATIONS AND THEIR ENFORCEMENT BY COURTS: uniform traffic laws and ordinances, state motor vehicle laws, Uniform Vehicle Code and Model Traffic Ordinance . . . official safety agencies CHARACTERISTICS OF DRIVERS: mental, emotional, phys- ical, and physiological SOCIETY AND DRIVING: effects of alcohol and drugs . . . psychology and driving . . . our culture and driving 3Office of State Superintendent of Public Instruc- tion, 1964 Driver Education Guidg (Olympia: Office of State Superintendent of Public Instruction, 1965), pp. 26-33 I 25 DRIVING SKILLS: basic habits and maneuvers . . . driving in the city, on the highway, on expressways . . . hazardous conditions and meeting emergencies . . . efficient driving DEVELOPMENT OF JUDGMENTS:. vision and perception . . knowledge and analysis of traffic situations . . . making decisions . . . reaction time . . . physical laws that affect drivers and pedestrians THE MOTOR VEHICLE: history and development . . . eco- nomics of vehicle ownership . . . trip planning . . mechanics of the vehicle . . . safety devices . . . vocational driving TRAFFIC ACCIDENTS: causes . . . human and economic loss . . . what to do in case of an accident . . . built-in re3ponsive systems for meeting the unex- pected ENGINEERING: automotive . . . highway . . . traffic. Research on the relative effectiveness of different programs in driver and traffic safety education for female and male students is also limited. However, a survey of nearly two hundred driver education instructors conducted by the Aetna Drivotrainer staff, ". . . revealed that most instructors polled believe that boys learn to drive more quickly than girls. When it comes to courtesy, how- ever, they felt that the girls leave the boysbehind."5 Project TALENT provided additional information on this subject although it was not directly related to 4Fourth National Conference on Driver Education, Policies and Practices for Driver and Traffic Safety Edu— cation TWashington: National Education Association, Na- tionaI Commission on Safety Education, 1964), pp. 4-5. 5"Teachers Surveyed on Drivotrainer System," Drivo- trainer Digest, Vol. VI, No. 1 (May, 1963), p. 12. 26 driver and traffic safety education. It was reported that: . . . Boys seemed to acquire significantly more infor- mation than girls in many areas, including mathematics, physical science, aeronautics and space, electricity and electronics, mechanics, and sports; they also had significantly larger score gains than girls on several aptitude tests, including Creativity, Mechanical Reas- oning, Visualization in Three Dimensions, and Abstract Reasoning. Crancer, Washington State Department of Motor Vehicles, conducted a study which analyzed the accident and violation rates for four different types of driver training programs. His investigation revealed a wide range in the number of accidents and violations from 1961 to 1967 in the State of Washington between females and males, with the females having the lesser number of acci- dents and violations per one hundred drivers.7 Loft conducted a study which considered the ef- fects of a driver education course in driving knowledge and attitudes of high school seniors. Students were given the General Test on Traffic and Driving Knowledge and the Siebrecht Attitude Scale. The investigation revealed the following findings that were pertinent to this study: 6"Cognitive Growth During High School," A National Longitudinal Study of American Youth—-Project TALENT, BuI- letin No. 6 TApril, 1967), p. 1. 7Alfred Crancer, An Evaluation of Driver Training Based on Accident and Violation Rates (Report 004; Olympia, Washington: State Department of Motor Vehicles, May, 1967). PP. 4-9. 2. 27 Girls and boys who had driver education scored higher on the knowledge test and attitude scale. A combination of girls who had received and who had not received a course in driver education scored higher on the knowledge test than the boys combined. Driver education girls received higher scores at the .05 level of confidence on the knowledge test than non-driver education girls, non-driver educa- 8 tion boys, and the driver education boys. Two of the conclusions which were drawn as a result of the study were: Driver education courses should consist of specific learning experiences for boys so that attitudes and adequate knowledge will be a realistic outcome of such a course. Students in driver education should be required to take a screening test at the beginning of the course as a means of determining their needs.9 One of the recommendations which resulted from the investigation was that "a study be made to determine if 8Bernard I. Loft, "The Effects of Driver Education on Driver Knowledge and Attitudes in Selected Public Secondary Schools," Traffic Safety Research Review (June, 1960). pp. 13-14. 9Ibid., p. 15. 28 driver education courses should have any different content and/or methodology for girls and/or boys."lo In the article, "The Challenge from Within," Boyer comments on meeting the needs of students: . . . The emphasis is too often on subject matter to be learned, not on the needs of the learner. This can be shown by the use of available textbooks. No authors have ever considered their work to be the last word, but a lot of instructors never deviate from the printed text. Each student is given the same information re— gardless of his ability, background or need. In other disciplines, those same students are provided with 11 opportunities to interpret and relate to their needs. Toms, Director of the Washington State Department of Motor Vehicles, voiced certain observations concerning classroom instruction at the 1967 Illinois Annual State Conference for Driver and Traffic Safety Education: I. The classroom teacher should utilize more of the following approaches: A. Non-directive teaching. B.‘ Individual problem approach. C. We as teachers subject our youth to too much of our morals, likes and dislikes in trying to teach them. D. We must be able to communicate with youth in order to teach them and to change or influence their behavior. E. How do we assess risks? loIbid. 11Richard G. Boyer, "The Challenge from Within," CALDEA Calendar, Vol. XV, No. 1 (October, 1967), p. 7. 29 1. It is impossible to teach every specific risk or accident situation. II. We must change our approach in teaching driver education and take a closer look at the content we are teaching. A. We must consider such newer approaches as: 1. Programmed instruction. 2. Team teaching.12 The traditional classroom presentation in driver and traffic safety education has been the lecture-textbook approach. Some imaginative and energetic instructors have incorporated films, filmstrips, slides, transparencies and overlays, 8mm films, team teaching, small group discus- sions, and other teaching techniques into their presenta- tion. A comparatively recent innovation in driver and traffic safety education is the use of multi-media in classroom instruction. An example of such is the EDEX Learning System, more commonly known as the Aetna Drivo- cator System, which has received favorable support from many instructors who have used it. Class interest and student participation in many instances has increased. In the article, "Multimedia--A New Classroom Con- cept," Cook proposes the following advantages of the Drivocator: 12 . Segments ofva summary of an address given by Douglas Toms at the 1967 Illinois Annual State Conference for Driver and Traffic Safety Education. 30 The film series is designed to take beginning stu- dents through driver education's most essential learn- ing phase. Each lesson is complete in itself, since the Drivocator units are designed to correlate readily with textbook materials and thus provide the teacher as much flexibility in his course content as he may desire. Early reports reaching us indicate that the Drivo- cator system lends desirable elements of both flexi- bility and uniformity to classroom driver education. Surprisingly, for all its technical capability, the system actually demands more of the teacher—-princip- ally because it draws greater response from the student.1 However, no research concerning the Drivocator's effectiveness as part of the high school driver and traffic safety education curriculum has been reported. Some, though, which contain implications for driver and traffic safety education has been reported on its use in other areas. In an article, "A Special Report on Technology for Education," Herbert discusses the development of teaching machines, programmed learning, and other forms of innova— tion in educational technology. Among other observations, he states: The greatest involvement of a student in the teaching process and perhaps the method by which he learns most is by interaction with a data source-- extracting information, testing hypotheses, making 13Dean R. Cook, "Multimedia--A New Classroom Con- cept," CALDEA Calendar, Vol. XV, No. 1 (October, 1967), p. 24. 31 right or wrong decisions and learning by immediate detection and correction of his errors. . . . . . . The criterion for progress is not how much ma- terial is covered, but how well it is learned. Such programmed material changes the teacher's role in a special way. . . . Where feedback to the student and teacher is utilized, as in the Drivocator, it is helpful to the in- structor in revising and improving the program presenta- tion.15 The Western Greyhound Corporation examined the driving records of 1,500 of their drivers who participated in the Drivocator program, and 1,500 of their drivers who did not participate in the program. The results showed a significantly measurable reduction in the number of acci- dent-producing incidents for the group of drivers who par- ticipated in the program.16 During 1966 and 1967, the New Jersey Police Train- ing Commission evaluated different types of teaching environments and materials which affect the learning impact on their personnel. Among other things, they found a more l4Evan Herbert (Associate ed.), "A Special Report on Technology for Education," International Science and Technology (August, 1967), p. 31. lsIbid., p. 32. 16"Something New in Safety," EDEX Teaching Systems (Mountain View, California: EDEX CorporatiBfi), p. 2: 32 favorable attitude rating among those who had received multi-media instruction instead of the lecture-discussion method of teaching. The study.also revealed that class environment is of the utmost importance if police training programs are to be effective when measured in terms of learning and student satisfaction.17 Studies involving Internal Revenue personnel indi- cate that EDEX training has contributed to improved per- formance effectiveness.18 Instructors were also able to identify quickly high and low scores in the program, thus enabling the instructor to give closer observation and assistance to low scoring students.19 A news release from the Air Defense Command, United States Air Force, concerning the implementation of a driver training program, announced: All personnel in Air Defense Command under 26-years of age will be required to take a mandatory 18-hour drivers training course when the Air Force implements a new Multi Media Traffic Safety Teaching System, Ground Safety officials announced here today. 17Charles C. Drawbaugh, "Evaluating the Concept of Mobile Police Training," The Police Chief (August, 1968), p. 60. 18"Report on EDEX—ADP Training"--for the Internal Revenue Service, p. 13. lgIbid., p. 20. 20"Driver Training To Be Implemented," United States Air Force New Release (Colorado Springs, CoIorado: Air Defense Command, USAF, May 26), p. 1. 33 Barcus, Hayman, and Johnson conducted a study which compared programmed texts, teaching machines, and conventional classroom instruction using teachers with varying amounts of training and experience. They found: . . . with proper conditions and at least with the rather mechanical, non—creative type learning involved in this study, automated instruction can be as effec- tive as the more traditional teacher-directed method. In fact, the teaching machine results suggest that automated instruction can be superior, though condi- tions for this superiority are uncertain. Bridgeman conducted a study utilizing an individual student response system and counter which was placed in view of the instructor but not the students. The students would respond to a battery of multiple choice questions built around major concepts covered in lectures and lab- oratory periods.22 He concluded: In general, the student profits by the improved structuring of the information he receives. The in- structor retains control of the flow of information, exercises his prerogatives as a personality, and in- vokes his own style of lecturing. But by using this electronic response device as.a learning tool he can accurately sense the needs of a large group of students and freely adjust his presentation in the tutorial manner. 21An Abstract of Pertinent Research Related to EDEX EducationaIiSystems (Mountain View, California: EDEX Corporation), pp. 5-6, citing Delbert Barcus, John L. Hay- man, and James T. Johnson, Jr., "Programmed Instruction in Elementary Spanish," Phi Delta Kappan, March, 1963. 22Ibid., p. 8, citing Charles Bridgeman, "A Lecture Response Device: A Preliminary Report on a Key Aspect of . a Co-ordinated Teaching Program in Anatomy," Journal of Medical Education, February, 1964. 23Ibido ' pp. 8-90 34 Lancaster, at the Pennsylvania State University, conducted a study utilizing immediate reinforcement to students. The theory of learning indicates that students would learn more efficiently if they were reinforced as soon as they made the correct response to a_new concept. A simple device (MARI) for giving students immediate reinforcement within a class period was. designed, built, and tested in actual classrooms. Not only would MARI reinforce the student when he made the correct answer but it would also indicate to the in- structor the percentage of the class responding cor- rectly. The merit of this teaching aid was evaluated in terms of the usual hour tests, by comparing the achievement of students in classes using MARI with control groups not using MARI. The first year the results were statistically highly significant in favor of using it. Later results were not. Yet it is strongly believed that some such device could be de- signed which would enhance classroom learning and that other experiments should be conducted. The School of Dentistry at Loma.Linda University used the EDEX Automatic Teaching System to evaluate the instructional efficiency of a programmed-group instruc- tional approach to teaching in dentistry. The non-EDEX group did not receive the traditional instructional pro- gram, but highly organized, programmed materials which would be presented in a similar manner to students using the EDEX System.25 24Ibid., p. 20, citing Otis E. Lancaster, "MARI: Motivator and Response Indicator," I.R.E. Transactions on Education, December, 1961. 25Edwin M. Collins, Earl C. Collard, and Deryck R. Kent, "Programmed-group Instruction in Dental Education," Journal of Dental Education, Vol. 31, No. 4, pp. 511-512. 35 In general, results reveal a slight, often non- statistically significant difference between the Edex and non-Edex treatments in favor of Edex. The differ- ence was so modest in the individual class sessions that significant differences were ngt_yielded in all but 1 of the 12 separate comparisons. . . . . . . there was no significant difference in the over- all quiz performance or midterm performance, although both results favor the Edex group. There was, however, a significant difference (p < .01) on the final exami- nation performance favoring the Edex group.2 Much more attention has been given to the investi- gation of the laboratory phase of instruction in driver and traffic safety education. A large part of such re- search has dealt with some type of simulation. The first simulator program was conducted at Lane Technical High School in Chicago. . . . it is likely that these "homemade" simulators were constructed and used, primarily, to give students practice in steering, shifting gears, and braking. In 1953, the Aetna Life Affiliated Companies pro- duced the Drivotrainer. . . . The first motion pictures were on black and white film, but color film was in- troduced during the early 1960's. . . . The Auto Trainer of the American Automobile Asso- ciation consisted of the usual automobile controls which determined the speed and direction of a miniature car operating on a moving roadway painted on an endless canvas belt. . . . The Allstate Insurance Company, in 1962, introduced the Allstate Good Driver Trainer which Operated in a manner similar to the Drivotrainer. It featured use 26Ibid., p. 514. 36 of wide-screen film and a system for immediate feed- back of student errors.27 The growth of simulation investigation has been considerable. The Insurance Institute for Highway Safety reported that more than 1,000 schools taught 334,000 stu— dents in simulators during the 1967-68 school year.28 Two policy statements in 1958 greatly influenced the growth of simulator instruction during the last decade. The first of these was a result of the Third National Conference on Driver Education held at Purdue University: In recognition of simulators, the value of which has been established by research, simulated driving experience can be used in lieu of half of the recom- mended ratio of four to one. Decision for approval of any simulator should be the responsibility of the individual State Department of Education. 9 A second is from the Driver Education Section of the Na- tional Safety Council: The National Safety Council supports and encourages the use of laboratory equipment to simulate behind- the—wheel driver experience in high school driver 27Herbert J. Stack, History of Driver Education in the United States (Washington: National Education Asso- ciation, NationaI Commission on Safety Education, 1966), pp. 36-37. 28Insurance Institute for Highway Safety, let Annual DrivetjEducation Achievement Program,il967-68 School Year (Condensed Report; Washington: Insurance Institute fOr Highway Safety, 1968). 29Herbert J. Stack, "A Resume and Evaluation of Research on the Teaching Effectiveness.of Simulated Driving Experiences and Conventional Driver Education Methods," Traffic Safety Research Review, Vol. 3, No. 4 (December, 1959). p. 12. 37 education courses. Also, the Council encourages con- tinued research as to the manner in which such devices can be most effectively used in the teaching of driver education.30 These policy statements were based on the outcomes of previous research studies in the area of simulation. In 1953, LeVan conducted a study, as a part of a Master's Thesis at Temple University, which examined the use of the Auto Trainer as a possible way of reducing the amount of time spent in the on-street phase of instruction. "The study design called for two groups of students: one, the experimental group, received Auto Trainer experience; the other, the control group, received the regular instruc- tion."31 The author arrived at the following conclusions: 1. Approximately 148 minutes of instruction on the Auto Trainer will save 47.5 minutes' instruction time in the car. The road test scores were ap- proximately the same for the experimental and the control groups. 2. There were no statistically significant differences between the two groups in driving ability. 3. The basic skills required in the instruction task and the actual task seem to be similar, for the Auto Trainer scores of the experimental group, the control group, and the experienced group of drivers were not significantly different. 3oIbid. 31'James H. Fox, Driver Education and Driving Simu- lators (Washington: National Education Association, Na- tionaI Commission on Safety Education, 1960), p. 52, citing Paul S. LeVan, "The Value of an Auto Training Device in the Teaching of Driver Training" (Master's Thesis, Temple University, 1953). 38 4. Driving ability as measured by the road test scores was not related to driving ability as measured by the Auto Trainer scores.. Three of the test items (errors, steering, and reaction time) showed a positive relationship, while the fourth (steering movement) showed no relationship. It is thus con- cluded that the steering movement scores lack reliability. 5. The students indicated a definite interefit in the Auto Trainer for instructional purposes. A more recent study of the Auto Trainer's effec- tiveness was reported in 1959 by Allgaier and Yaksich. Two groups of students were used in the study. The control group received the traditional on-street program. The experimental group received Auto Trainer instruction as a substitute for part of behind-the-wheel instruction in a dual-control car. Two conclusions reported in the study were: (1) both groups did equally well on the final road test, and (2) Auto Trainer instruction made it possible to teach one-third more students per unit of instructional 33 time. The December, 1961, issue of Traffic Safety re- ported the results of a study by Heeter and Allgaier. In the study, forty-four students received eight hours in- struction in the car and no practice on the Auto Trainer 32Ibid., p. 54. 33Earl Allgaier and Sam Yaksich, Effectiveness of a Driving Simglator (Washington: American Automobile .Association, Traffic Engineering and Safety Department, 1959). p. 4. 39 (Group D) while one hundred twelve students received the following amounts of practice on the Auto Trainer but less practice in a car: Group A--Nine hours Auto Trainer, five hours car. Group B--Four hours Auto Trainer, four hours car. Group C--Four hours Auto Trainer, six hours car. At the conclusion of instruction, all students were given a road test. The following conclusions reached from this study and previous studies using the Auto Trainer seem justified: 1. Practice on a simulator can be substituted for a substantial amount of practice driving in a car, thus,reducing the cost per student trained. 2. Because of lack of precise measurements of perfor- mance, it is not possible to determine a precise substitution ratio that is the number of hours of practice on a simulator which are equivalent to an hour of practice in a car. 3. The law of diminishing returns apparently applies. After a certain number of hours of practice on the simulator, little is gained by additional practice. The results of the study also showed that scores made on the road test by males were slightly better than the scores by females but the difference was not statistically significant.34 In 1955, a study was conducted to determine if students trained in a program involving the Aetna Drivo- trainer would have the same driving ability as those trained by the present methods of the Los Angeles City Schools. The students were divided into two groups. The 34Lewis M. Heeter and Earl Allgaier, "Can a Simu- JLator Sub for a Car?" Traffic Safety (December, 1961), p. 21. 40 control group received six class hours of behind-the-wheel training. The experimental group received three class hours of behind-the-wheel training, plus sixteen class 35 hours in the simulator. The staff made the following conclusions: 1. Results indicate that practically the same progress in driving skill and knowledge will be experienced by a student trained by either the experimental method using the Aetna Drivotrainer plus three hours of on-the-road training, or the California State prescribed course. 2. While increment in driver knowledge as measured by the final test for Sportsmanlike Driving and M32. and the Motor Car texts was observed, there was no evidence that either the control or the experimen- tal group established a statistically significant advantage. 4. Both groups made significant improvement in driv- ing ability as measured by the Aetna Road Test. The experimental group showed significantly greater improvement. Familiarity with the machine may have influenced the results in this instance. 5. There is a very slight indication of possible dif- ference in driving ability, as measured by the Road Test, in favor of the control group. However, the level of significance for this difference was not great enough to warrant unqualified accept— ance. In 1956, a study was undertaken at Iowa State Teachers College to assess the worth of the Aetna 35An Evaluation ofithe TeachingEffectiveness of ‘the Aetna Drivotrainer (a Condensed Report of the Los .AngeIes Study; Hartford, Connecticut: Aetna Casualty and Surety Division, 1955), p. 5. 361bid., p. 17. 41 Drivotrainer as a replacement for part of the on-street instruction in a high school driver education program. The instruments used to evaluate the effectiveness of student performance were the Iowa Driver's License Exam- ination Score Sheet and a Final Drivers Test. Three hours of dual-control car instruction plus fifteen class periods of Drivotrainer instruction (fifty minute periods) were found to produce results at least as good as those pro- duced by six hours of dual-control-car instruction.37 In 1958, Bernoff completed a study designed to investigate specifically the following problems: 1., Do students trained with the Drivotrainer possess a driving attitude comparable to that of students trained in a minimum conventional course? 2. Do students trained with the Drivotrainer possess as much driving knowledge as students trained in a minimum conventional course? 3. Do students trained with the Drivotrainer possess as much skill in manipulating an automobile as students trained in a minimum conventional course? 38 His Drivotrainer course consisted of sixteen hours in the Drivotrainer, supplemented with three hours behind- the-wheel and six hours observation in a dual-control car. 37Gordon J. Rhum, Bertram L. Woodcock, and Tom A. Lamke,_Tte Effectiveness of the Aetna Drivotrainer in Driver Education (Hartford, Connecticut: Aetna Casualty and Surety Company, July, 1956), pp. 3-20. 38Louis I. Bernoff, An Experimental Study of the Teaching Efficiency of the Aetna Drivotrainer.System_ ‘ THartford, Connecticut: Aetna Life and Casualty, June, 1958), p. 5. 42 The conventional course consisted of six hours behind-the- wheel instruction and twelve hours of observation in a dual-control car. of the study: 11. 39 The following is a summary of the major findings Bernoff investigation which are pertinent to this Before training, the Drivotrainer boys exceeded the Drivotrainer girls in general driving knowledge scores. Before training, boys in both groups exceeded the girls in their respective groups in specific driv— ing knowledge scores. After training, no significant differences in driving attitude scores existed between or within the Drivotrainer and dual control groups. After training, no significant differences in gen- eral driving knowledge scores existed between or within the Drivotrainer and dual control groups. After training, the Drivotrainer and dual control groups were comparable with respect to specific driving knowledge scores. After training, Drivotrainer boys exceeded signif- icantly the Drivotrainer girls in specific driving knowledge scores. In general, both Drivotrainer and dual control groups changed significantly in a positive direc- tion in mean specific driving knowledge scores. However, the mean change for boys was far less significant than for girls. 39Ibid., p. 6. 43 12. The median road test scores between total Drivo— trainer and dual control groups was not signif- icant. However, the Drivotrainer boys outperformed the dual control boys on this road test at the 10 per cent level of significance; and the Drivo— trainer boys also surpassed the Drivotrainer girls on this test at the 10 per cent level of signif- icance. The conclusions reached as a result of the inves- tigation which are pertinent to this study are: l. The Drivotrainer is a device which can be used successfully to train students as adequately, or possibly better, than they are now being trained by conventional means. 5. The Drivotrainer definitely offers a given staff the opportunity to train approximately thirty per cent more students in a school year than can now be trained by conventional means. This could result in tremendous savings to the taxpayer.41 A 1959 report of a study in the New York City Schools by Forlano and Wrightstone evaluated the effec- 'tiveness of the Aetna Drivotrainer in a sixteen-period driver education course. The criterion measures used were: (1) a paper and pencil test used to investigate such areas as driver knowledge, emergency judgment, and driver attitude; and (2) a performance test to measure the student's ability to start a car, drive for 150 yards, and stOp the car. The authors concluded that (1) no signif- icant difference in driver knowledge, driver attitude, and 4OIbido' pp. 7-80 41Ibid., p. 9. (I ’I‘ E (A ‘ufi ifikshfib DIV Lkfinl. 1.. 44 emergency judgment as measured by paper and pencil tests appeared between various control and Drivotrainer groups; (2) generally, success on the Road Readiness Test tended to vary directly with length and intensity of exposure to the Drivotrainer apparatus; and (3) the Drivotrainer pro- gram is more economical.42 Bishop conducted a study in 1963 "to determine whether sufficient transfer of learning occurs from simu- lator instruction to actual driving to permit substitution of such experience for part of the required behind-the- wheel teaching."43 Four groups of fifty students each were selected in a random manner from 1,250 tenth and eleventh grade students at Andrew Jackson High School, Jacksonville, Florida. A summary of hours for the four groups is as follows:44 Simulator On-the-Road Car Instruction Instruction Observation Group A 0 6 6 Group B 12 3 6 Group C 0 3 6 Group D 0 0 0 42 George Forlano and J. Wayne Wrightstone, Ag Evaluation of the Aetna Drivottainer in Selected New York City High Schools (Divisional BuIleEin No. 3; New York: State Board’of Education, Bureau of Educational Research, October, 1959). p. 13. 43Richard W. Bishop, Evaluating Simulator Instruc- tion for Accomplishing Driver Education Objectives (Talla- hassee: Fiorida State Department of Education, July, 1963). p. l. 44 Ibido' p. 4. 45 The evaluation instruments used in the study were: (1) a knowledge test for use with the Sportsmanlike Drivipg 45 (2) the Siebrecht Attitude Scale,46 and (3) 47 textbook, the McGlade Road Test for driving performance. A significant difference in knowledge improvement at the .01 level was found between the simulator group (Group B) and each of the other groups in favor of Group B.48 Also, there was a significant difference in Road Performance at the .05 level between Group A and Group C in favor of Group A. No significant difference at the .05 level was found between Group A and Group B, although Group B had a lower mean score that approached significance in favor of Group B.49 50 An item analysis of the McGlade Road Test indicated that Group B's most significant transfer occurred in three very important factors affecting 45Driving Knowledge Test (Washington: American Automobile Association, 1956). 46Elmer B. Siebrecht, Siebrecht Attitude Scale (New York: New York University, Center fOr Safety Educa- tion, 1941). - 47Francis S. McGlade, A New Road Test for Use in Driver Licensing, Education and Empigyment (New York: New York University, Center for Safety Education, 1961). 48 Bishop, 0p. cit., p. 9. 491bid., pp. 11-12. 50McGlade, loc. cit. 46 the driving task: (1) intersection observance, (2) inter- section speed, and (3) speed control. 51 The following general conclusions reached as a result of the investigation were: The findings of this study indicate that a program combining 12 hours of simulator experience as applied in this study with 3 hours behind-the-wheel taught by a competent instructor compares favorably with the conventional 6 hours behind-the-wheel also taught by experienced instructors.52 In 1965, Hayes, Porter, Saja, and Stehman reported on a study in Pennsylvania which compared two types of simulator (Allstate Good Driver Trainer) courses with the traditional type of program. The study had the following objectives: 1. Will the driving skills, driving knowledge, and attitude toward driving differ significantly be- tween students taught by a simulator centered course and those taught by a dual-control car centered course? In which specific areas of learning does the simu- lator centered course produce significantly dif- ferent results compared to a dual—control car centered course? Will there be a significant difference in results due to sex?53 51Bishop, op. cit., p. 17. 521bid., p. 18. 53Robert B. Hayes and others, Immediate Standard-.. ized Legtning Reinforcement to a Complex Mental-Motor Skill. (Driver Training) Using Electronicdlleroordinated Motion Pictures (Abstract, Title VII Project No. 1090; Washington: U. S. Department of Health, Education and Welfare, 1965), p. l. 47 The investigation showed: It was indicated that the integrated simulator approach did a significantly better job of preparing students for their fourth and final periods of dual- control car training than did either the non-integrated simulator or the non-simulator methods. In the fore- going area of driving skills the difference in means was not significant between students taught by the non-integrated simulator method and those taught by the standardized, conventional car program (non—simu- lator). Overall there was no significant difference among treatments in passing the State license examina- tion or in traffic accidents for the first six months of driving. The integrated simulator method produced significantly better results than did the non-simulator approach in the specific skills involving road posi- tion, turn signals and intersections and in knowledge of stopping, night speed, headlights, and passing. The differences between final means on a locally- devised knowledge test were statistically significant in favor of both the integrated simulator and the non- simulator treatments compared with the non-integrated simulator approach. The integrated simulator group had the highest means on both the 1962 AAA Knowledge Test and the Siebrecht Attitude Scale but these re- sults were not statistically significant compared with the means of the other two groups. . . . Girls taught by the integrated simulator method made significantly fewer errors in the car training phase than did girls taught by the other two methods. Also the integrated simulator girls had sig- nificantly less skills error variance than did the non-simulator girls. Sex was the most important pre- experimental variable. The average girl apparently commences driver training at a much lower point on the skills continuum than does the average boy. It further indicated: With an integrated teaching approach the new simu- lator with its immediate learning reinforcement can be used to substitute nine hours of simulated training for three of six hours of behind-the-wheel and three of six hours of observation in a dual-control car. 54Ibid., p. 2. 48 With prOper scheduling to take advantage of an improved student-teacher ratio the new simulator could increase by 60 percent the number of students trained per in- structor per year.55 Many instructors with experience in simulation have become advocates of its use and report that benefits, such as recognition of hazards, decision making, and per- formance, have a positive influence on the driving task. Research strongly supports the premise that a transfer of learning occurs from the simulator environment to the actual task.56 The off-street multiple car driving range first came into existence in 1936 at Lane Technical High School in Chicago.57 Under the leadership of Graham, the Detroit School District developed its first such facility at Pershing High School in 1944.58 Michigan State University developed its range in 1956. While the off-street multiple car driving range program was slow to achieve recognition, its utilization in the last decade has increased consid— erably. The latest report from the Insurance Institute SSIbid. 56Richard W. Bishop, “Questions and Answers About Driving Simulators," Safetnyducation (December, 1964), p. 10. 57Automotive Safety Foundation, The Multiple Car Method (Washington: Automotive Safety Foundation, March, 1967), p. 1. 58Herbert J. Stack, History of Driver Education in the United States (Washington: National Education Asso- ciation, NationEI Commission on Safety Education, 1966), p. 35. 49 for Highway Safety shows that 464 schools in 28 states. utilized this method of instruction to teach nearly 154,000 students during the 1967-68 school year.59 The off-street multiple car driving range shOuld not be confused with parking lots, athletic fields, or similar areas which do not provide adequate space or ex- periences. These areas are sometimes used in early lessons by dual—control cars each occupied by an instructor and students. This does not mean that parking lots, athletic fields, and similar areas cannot be useful and effective areas for instruction. Many excellent ranges are so lo- cated, but they provide the necessary space, design, and experiences that students need. They also provide for the following distinctive features of the true multiple car method: "(a) more cars than teachers are functioning at the same time, and (b) students are instructed by a teacher positioned outside the vehicle."60 Range instruction offers a broad potential to schools desiring to strengthen their driver and traffic safety education programs. Properly taught, it transfers the emphasis from teaching to learning. In an address to 59Insurance Institute for Highway Safety, 21st Annual Driver Education Achievement Program, 1967-68 §g§ool Year (Condensed Report; Washington: Insurance Institute for Highway Safety, 1968). 60 Automotive Safety Foundation, op. cit., p. 4. 50 the Seventh Annual Conference of the American Driver and Traffic Safety Education Association, Nolan said: . . . perhaps the outstanding feature of the multiple car, off-street driving range is that it places high priority on the acceptance of responsibility and on cooperation with one's fellows. In addition, young- sters learn to adjust to various road and weather conditions. A student driver must communicate with other student drivers while on the range. When the student is alone in the car, we have a real situation in which to test his driving behavior.6 An examination of schools which have adopted range instruction as one component of their driver and traffic safety education programs reveals a wide variation among the number of hours of on-street instruction required in addition to that on the range, as well as the number of hours of range instruction deemed equivalent to on-street instruction. These differences result from variances among philosophies on range instruction, size and design of the range, different combination of programs, state standards, and other factors. The Fourth National Confer- ence on Driver Education recognized these variances in recommending that: . . . experience on a multiple-car driving range be supplemented by one or more hours of practice driving under real traffic conditions in a dual-control car. In the absence of a sufficient amount of investigation and experience in this area, it is not feasible to 61National Commission on Safety Education, Seventh .Anntgl Conference Proceedipgs of American Driver and Traf- 'fic Safety Education Association, June 26-28, 1963 (Wash- Efington: National Education Associatidn, National Commis- sion on Safety Education, 1964), pp. 35-36. 51 recommend a definite ratio between time on a multiple- car driving range and time for on-street practice in a dual-control car.62 The Four-Phase Program, as commonly designated, combines instruction in classroom, simulators, off-street multiple car driving ranges, and on-street instruction into a correlated and integrated program extending through a school semester. The first evidence of such a program in operation was during the Spring Term of 1957 at Michigan State University.63 The program provided approximately . forty-five hours classroom, five hours simulation, six hours range, and one and one—half hours on-street instruc-. 64 White Plains High School in New York has offered 65 tion. a similar prOgram since 1959; however not all the course components were offered during the same school term. The term, Four-Phase PrOgram, was initially applied by the Brevard County Public Schools in Florida in conjunction 62Fourth National Conference on Driver Education, Policies and Practices for Driver and Traffic Safety Edu- cation (Washington: National Education Association, Na- tionaI Commission on Safety Education, 1964), p. 24. 63Information obtained from a conference with Robert O. Nolan, Highway Traffic Safety Center, Michigan State University. 64Robert O. Nolan, "East Lansing Combines Drivo- trainer and Range Plan Program in Second Year," Drivotrainer Digest, Vol. III, No. 2 (April, 1960), pp. 3-4. 65"City Officials Visit White Plains High School," .Drivotrainer Digest, Vol. VI, No. 1 (May, 1963), pp. 8—9. 52 with its development of a program embodying these compo- 66 At about this same time, the Jefferson nents in 1963. County Public Schools in Colorado developed a similar program.67 Currently there are 150 four-phase programs in sixteen states.68 In 1965, Brazell reported on a study which evalu- ated accident and violation records of students in Dear- born, Michigan, who had received instruction in one of four different laboratory programs-Pall on-street instruc- tion, all range instruction, range plus one hour on-street instruction, or simulator plus on-street instruction.69 Three findings of the Dearborn investigation were: 1. The on-street method, generally considered to be the most expensive, did not yield superior driving results in terms of the criteria of this study. 66Charles E. McDaniel, "A Four-Phase Driver Edu- cation Program, 1964 National Safety Congress Transactions (Vol. XXIII; Chicago: National Safety Council, 1964), pp. 51-55. 67"Jefferson County, Colorado Combines Driving Range, Behind-the-Wheel and Simulators in Unique Driver Education Program," Rockwell Safety News, Vol. II, No. 1 (Winter, 1964), pp. 10-13. 68Insurance Institute for Highway Safety, 20th Annual-Driver-Aghievement Program, 1966-67 School Year (Washington: Insurance Institute for Highway Safety), p.-27. 69Robert E. Brazell, "A Follow-Up Study of Public School Driver Trainess, Relating Driving Performance Records to Selected Academic and Training Factors" (un- published doctoral dissertation, University of Michigan, 1961). p. 19. 53 2. Off-street methods, usually considered to be the least expensive, produced at least comparable results compared with other methods. 3. Students trained under simulator methods showed approximately equal accident records but more . moving-violation points per 1,000 months of driving exposure, when compared with students trained under non-simulator methods.70 BishOp in a 1965 study, which compared the effec- tiveness of on-street and multiple car driving range in- struction, reported: Four groups of students, each receiving different instructional treatments, were compared in this proj- ect. Group A received the traditional 6 hours on street; Group B received 6 hours instruction under the multiple-car driving range method; Group C received 1 hour on-street and 6 hours of multiple-car driving range experience; and Group D received 2 hours on- street and 6 hours on the multiple-car driving range.71 . . . all students received a minimum of 30 hours of classroom instruction.72 The results of the investigation indicate that under the conditions of the study, a traditional program of six hours on-street instruction is not significantly superior to either of the three off-street multiple car driving range programs.73 70Ibid., p. 95. 71Richard W. Bishop, "Comparing the Effectiveness of Various Combinations of On-Street and Multiple Car Driving Range Instructional Hours" (unpublished research study, Florida State Department of Education, Tallahassee, 1965): P. 9. 721bid., p. 11. 73Ibid., p. 17. 54 A 1965 study by Nolan evaluated the relative ef- fectiveness of students taught to drive by means of the Drivotrainer and those taught to drive on the multiple car off-street driving range. A driver-attitude test, driver- knowledge test, and driving skill test were used as cri- terion measures.74 Students in the simulator group re- ceived ten clock hours of Drivotrainer instruction supplemented by three hours behind-the—wheel and six hours observation in a dual-control car. Students in the range group received ten clock hours of off-street multiple car driving range instruction plus two hours behind-the-wheel and four hours observation time in a dual-control car.75 The following are results of his investigation which are pertinent to this study: 4. After instruction no significant differences in general driving knowledge scores existed between the Drivotrainer and multiple car groups. 5. After instruction the Drivotrainer and multiple car boys had significantly better general driving knowledge scores than the Drivotrainer and multiple car girls. 74Robert O. Nolan, "A Comparative Study of the (Teaching Effectiveness of the Multiple Car Off-Street Driving Range and the Aetna Drivotrainer" (synopsis of an 'unpublished doctoral dissertation, Michigan State Univer- sity, 1964), pp. 1-2. 751bid., pp. 8-9. 11. 12. 13. 14. 17. 55 After instruction the Drivotrainer and multiple car groups were comparable with respect to spe- cific driving knowledge scores. Both Drivotrainer and multiple car groups made significant positive mean changes in general driv- ing knowledge. . Both Drivotrainer and multiple car groups changed significantly in a positive direction in mean specific driving knowledge scores. However, the mean change for the second semester Drivotrainer and multiple car groups was far more significant than that for the first semester Drivotrainer and multiple car groups. After instruction the differences in mean road test (vehicle handling) scores between the Drivo- trainer and multiple car groups were not signif- icant. However, the Drivotrainer boys out per- formed the Drivotrainer girls at the .02 level of significance. After instruction the differences in mean road test (road problems) scores between the Drivo- trainer and multiple car groups were not signif- icant. However, the Drivotrainer boys again out performed the Drivotrainer girls at the .02 level of significance. The low correlations obtained for the Drivotrainer. . and multiple car groups when comparing intelligence. and final road test scores were not significant.76 A study by Gustafson in 1965 compared the effec- tiveness of instruction in the Allstate Good Driver Trainer. and on the multiple car off-street driving range with the :multiple car off-street driving range. The purpose of his investigation was: 761bid. ’ pp. 12.-l4. 56 To determine whether students who were taught to drive by means of a combination program involving use of the Allstate Good Driver Trainer and the multiple car off-street driving range were compar- able in driving knowledge, attitude, and skill with students who were taught on the multiple car off-street driving range exclusively; and To determine whether students who were taught to drive by means of a combination program involving use of the Allstate Good Driver Trainer and the multiple car off-street driving range with some additional on-the-street instruction were compar- able in driving knowledge, attitude, and skill with students who were taught on the multiple car off-street driving range, with the same additional on-the-street instruction. The combination simulator/range program (experi- mental group) consisted of six hours of instruction in the simulator correlated with four hours of instruction on the range. This instruction was followed by two hours of observation and two hours of instruction on- the-street. The range program (control group) con- sisted of ten hours of instruction on the range fol- lowed by two hours of observation and two hours of instruction on-the-street. to this 1. Findings of the Gustafson investigation pertinent study are: On the pre-tests (before the simulator and/or range instruction) there were no significant dif- ferences between the experimental and control groups in general driving knowledge, specific driving knowledge, or in driving attitude. 77Robert E. Gustafson, "A.Study to Compare the. Effectiveness of Instruction in the Allstate Good Driver Trainer and on the Multiple Car Off-Street Driving Range with the Multiple Car Off-Street Driving Range" (an ab- stract of an unpublished doctoral dissertation, Michigan State University, 1965), p. l. 781bid., p. 2. 57 2. On the mid-tests (after the simulator and/or range instruction) there were no significant differences between the experimenta1.and control groups in general driving knowledge, specific driving knowl- edge, driving attitude, or on the simulator driving test. On the range skill tests, however, there was a significant difference in favor of the con— trol group. 3. Between the pre- and mid-tests both the experi- mental and control groups showed significant im- provement in general driving knowledge, specific driving knowledge, and on the Siebrecht Attitude Scale. No significant improvement was shown for either group on the Mann Personal Attitude Survey. 4. On the post-tests (after the on-the-street instruc- tion) there were no significant differences between the experimental and control groups in general driving knowledge, specific driving knowledge, driving attitude, or on the traffic problems and road problems sections of the final road test. There was a significant difference, however, on the vehicle handling section of the final road test in favor of the control group. 5. Between the mid- and post-tests neither the ex- perimental or control groups showed significant improvement in general driving knowledge, on Part I of the specific driving knowledge test, or on the Mann Personal Attitude Survey. Significant improvement was shown by the control group on Part II of the specific driving knowledge test, and by the experimental group in the Siebrecht Attitude Scale. 6. Between the pre- and post-tests both the experi— mental and control groups showed significant im- provement in general driving knowledge, specific driving knowledge and on the Siebrecht Attitude Scale. Neither group showed significant im rove- ment on the Mann Personal Attitude Survey. Seals in a 1966 study compared a traditional course (Group I), a three-phase course--c1assroom, 791bid.. pp. 3-4. 58 simulator, on-street (Group II), and two variations of a four-phase course (Groups III and IV) in driver and traffic safety education (see Figure l for a time allotment and scheduling order of various phases of driver and traffic safety education) in terms of student achievement on driv- ing knowledge and road performance tests.80 A pre- and post-knowledge test, and a road per- formance test, were used as the evaluation criteria. The following conclusions were reached as a result of the investigation: 1. Pre-test for driving knowledge. a. The student groups assigned to the four in- structional treatments did not differ signif- icantly in mean pre-test scores for driving knowledge. 2. Post-test for driving knowledge. a. Each of the four instructional treatments apparently produced a significant improvement in driving knowledge as measured by an objec- tive test. b. There was no significant difference between mean scores made by Group I and Group II. c. There was no significant difference between mean scores made by Group I and Group III. d. There was no significant difference between mean scores made by Group I and Group IV. e. There was no significant difference between mean scores made by Group III and Group IV. 80Thomas A. Seals, "An Evaluation of Selected Driver and Traffic Safety Education Courses" (unpublished doctoral (iissertation, Florida State University, August, 1966), pp. 36-37. 59 can Ho>wuo mo mommnm msowum> mo 0mm om \ OUHQH* «cowumooom >HOMMm owmmmuu Hmcuo mcwadomnom com ucmEuoHHm mEHBII.H madman muson om musoc om mason om musoc ow Hmuos mason m.N mnoon h mnson h musoc m wosum ommw>uomom mnson m mason m mason m muooc NH cowuosuumcH umonumlco .coaum>uwmno HMOIGH musoc m.m noon H Moon a mnsoc m Hmmnzaonpuocflcwm coauoouumcH pmmuumlco mason m muooc m muson m muson o omcmm mcH>HHQ HMO mamfluasz ummuumummo .GOHpm>Hmmno HMUIQH mason m mason m musoc m mnsoc o mmcmm mcfl>wuo HMO mamfluasz ummuumummo mason oa mason oa muooc o mason o usmEmflsvm coHudeEflm musos om mason om mason ow muooz oe coaposnumcH Eooummmau >H HHH HH H coauonuumzH mo mmmnm Honesz ucmfiummua HMGOfluoauumcH 60 There was a significant difference between mean scores made by Group II and Group III in favor of Group III. There was a significant.difference between mean scores made by Group II and Group IV in favor of Group IV. Post-test for road performance. a. b. There was no significant difference between mean scores made by Group I and Group III. There was no significant difference between mean scores made by Group I and Group IV. There was no significant difference between mean scores made by Group III and Group IV. There was a significant difference between mean scores made by Group I and Group II in favor of Group I. There was a significant difference between mean scores made by Group II and Group III in favor of Group III. There was a significance between mean scores made by Group II and Group IV in favor of Group IV. Correlation between post-test scores for driving knowledge and post-test scores for road perfor- mance. ' a. No significant correlation existed between the study population's post-test scores on the driving knowledge examination and the study population's post-test scores on the McGlade Road Test.81 81 Ibido’ pp. 62—64. CHAPTER III PROCEDURES The results of any investigation are dependent to a large measure on the procedures adopted. Well-defined procedures give direction to the researcher. They guide him in making decisions during the course of the investi- gation. Procedures should be meaningful, based on sound principles and clearly stated. ‘The procedures adopted for this study are pre- sented as "tasks" or "sub-problems." The following is a brief description of each task or sub-problem: 1. Selecting a school large enough in student popula- tion to provide a statistically sound study sample. 2. Providing a staff of instructors with the necessary competencies to provide equivalent and quality instruction in the different instructional treat- ment groups. 3. Developing standardized procedures and lesson content for student instruction. 4. Providing the necessary facilities and equipment to conduct the study. 61 62 Creating an experimental design to compare the relative effectiveness of four selected laboratory programs and three selected classroom programs, and to determine the significance of the sex factor. Assigning students randomly to each of the dif- ferent instructional treatment groups. Assigning instructors to the different phases of instruction. Selecting and/or developing valid and reliable measurement instruments and standardizing proce- dures for administrating the tests. Collecting and applying statistical treatment of the data. gflub-Problem One--Selecting a school large enough in student pepulation to provide a statistically sound study sample. Several possible locations were investigated and exraluated. Answers to the following questions were con- Sixflered the evaluative criteria for selection of the study Siste. 1. Is the school district willing to cooperate with the Office of the Superintendent of Public Instruc- tion in conducting the study? Does the school district offer driver and traffic safety education as a part of the regular school day proqram? 10. 11. 63 Does the school district require driver and traffic safety education for graduation? Does the school district have a school or schools in close proximity to each other which have enough students in the tenth grade to provide a statis- tically sound study sample?. Does the school district encourage the use of innovative techniques of instruction in driver and traffic safety education? Does the school district have simulators, a drivo- cator system, an off-street multiple car driving range, and other necessary facilities and equipment to provide for effective instruction? If the school district does not have this equipment. or these facilities, would space be made available for them? Is the school district able.to provide an adequate. number of automobiles, textbooks, and other needed equipment and supplies for the study? Is the school district able to provide an adequate number of instructors to conduct the study? Is the school district able to provide adequate classrooms and office space to conduct the study? Is the school district willing to schedule all tenth grade students into driver and traffic safety education from January 1, 1968, until the end of that school year? 64 Renton High School, Renton, Washington, was se- lected for the study. Renton was the largest high school in the state and had a sophomore class of over nine hundred students. One of the major considerations in the selection of a school or school district for the study was the availability of a large number of tenth grade students for driver and traffic safety education after January 1, 1968. A large number of students was necessary to provide a sta- tistically sound study sample because of the number of different instructional treatment groups being considered. It was felt that an absolute minimum of fifty students should be required for each instructional treatment group, although seventy-five students for each instructional treatment group would be desirable. Renton High School offered driver and traffic safety education during the school day and required suc- cessful completion of the course for graduation. Renton School District had also shown a willingness to use dif- ferent innovative techniques of instruction in its driver and safety education programs.. The district had purchased a mobile simulator and a forty—place drivocator system, in addition to other types of audio-visual equipment. The district was further planning to construct an off-street uniltiple car driving range in the near future. Adequate space for equipment, classroom, planning, and offices was sindlarly made available. Arrangements were made for an 65 adequate number of automobiles, textbooks, and supplies. The school district was very receptive to the idea of cooperating with the Office of the Superintendent of Pub- lic Instruction and pledged its support of the project. The district was able also to provide the.additional teachers needed to conduct the study. The school district also agreed to free all driver and traffic safety education instructors from September, 1967, to January, 1968, for the purpose of participating in an in-service education prOgram. The district further agreed to schedule all tenth grade students into driver and traffic safety education from.January.l, 1968, until the end of the school year. This was important for two reasons: (1) time was needed to conduct the in-service education program, to develop and agree on procedures and lesson plans, to construct an off-street multiple car driving range, to purchase a second simulator, and ready all equipment to be used in the investigation; and (2) it was important that all students involved receive instruc- tion during the same period of time.to reduce the number of variables and insure as much uniformity of instruction as possible. In addition to being the largest high school in tflie State of Washington, Renton High School is located :near the Boeing Aircraft Company whose employees represent. . (all.parts of the country, as well as various socio-economic 66 levels. This situation gave the study sample a broad cross-section of students from various types of homes and with varied experiences, which typifies the population of the state. Sub-Problem Two--Providing a staff of instructors with the necessary competencies to provide equivalent and quality instruction to the different instructional treatment groups. Renton High School possessed a staff of six driver and traffic safety education instructors. However, a total of thirteen instructors was needed to conduct the study. The additional instructors were needed because of the num- ber of different instructional treatment groups in the study, and because all tenth grade students would be taught during the same period of time. .The additional instructors were provided by the personnel office of the Renton School District. There was no control in their selection by the Research Division of the Office of the Superintendent of Public Instruction. The instructors were found to possess a wide range of teaching experience, from four to forty years, of which zero to twenty-two years were spent teaching driver educa- tion (see Appendix C). All met the certification require- :ments for teaching driver education in the State of 67 Washington.1 All instructors had no, or limited, exper- ience in the teaching techniques used in simulators, off- street multiple car driving ranges, and the drivocator system. During the time from September 6, 1967, to Decem- ber 21, 1967, the Project Director and Assistant Director conducted an in-service education program to provide the instructors with the necessary philosophical background and teaching techniques to perform this phase of the in- vestigation. Each received both instruction and practical application of the techniques to be used in the simulator, off-street multiple car driving range, and drivocator phases of instruction. Renton junior and senior students who had previously taken the classroom phase of the program. were used in the laboratory phases of the in-service edu- cation program. Some junior and senior students also received some classroom instruction with emphasis in the techniques used in drivocator instruction. Consultants from the Aetna and Allstate Insurance Companies, which developed and produced the simulator and multi-media films, were used during the in-service simulator and drivocator instruction phase of the program. 1Certification requirements obtained from the Certification Division, Office of the Superintendent of Public Instruction, Olympia, Washington. 68 The in-service education phase of the prOgram also included the development of and agreement on techniques and procedures to be used by students and instructors in the various phases of the laboratory program. During this time, consensus of content sequence in the different phases of instruction was also reached by the instructional staff. Sub-Problem Three--Developing standardized procedures and lesson content for student instruction. This was vital in order to reduce the number of variables as much as possible. The development of a set of driving procedures was necessary for standardization of instruction, since a driving guide published by a textbook company was not used. It was the opinion of the staff that the driving procedures in the student textbook were not adequate since some discrepancies.appeared in the procedures used in the two simulators. The prior teaching experience and practices of the Renton High School driver and traffic safety education staff also differed from those expressed in the student text. Committees of in- structors were used to develop draft copies of the pro- cedures. These were critically reviewed and revised by the entire staff until agreement was reached. The driving . procedures were then duplicated.and given to the instruc- ‘tors and students prior to their use in the classroom and laboratory phases of instruction. 69 A set of teaching points was developed to supple- ment the driving procedures. These provided additional information and cited certain aspects of the driving pro- cedures which should be stressed to students, and were used only by instructors. Beginning drivers often have problems in effecting a transfer between the classroom and laboratory exper- iences, especially if they have not had a similar exper- ience in the laboratory phase of instruction. Students are sometimes confronted with situations in which they have had no previous knowledge or information on the sub- ject imparted to them. These situations present obstacles to the students because they are unable to establish mean- ingful relationships between the different phases of instruction. In an effort to make the course more interesting and meaningful to students and instructors, a novel ap- proach to classroom and laboratory content was used in the Washington Driver and Traffic Safety Education Project. The course content was designed to provide (1) a thorough integration and correlation of learning experiences from one phase of instruction to another, and (2) a systematic analysis of the essential elements pertaining to different types of driving experiences. The correlated and integrated approach was imple- rmented throughout the instruction. Students were 70 alternated between classroom and laboratory instruction during treatment of a particular type of driving environ— ment. They received classroom instruction concerning a particular skill, concept, or driving situation prior to, or concurrently with, the laboratory phase of instruction. A progression of understanding and skills, manipulative and perceptual, was developed through a process from classroom to simulator, to the range, and finally to the on-street phase of instruction. Students in the different laboratory programs followed a sequential progression of skills, regardless of the type of laboratory program to which they were assigned. A progression of laboratory lessons, which included the necessary understanding and skills, was developed to correlate with the classroom lessons (see Appendix A for a progression of lessons for the four different laboratory programs). The classroom phase of instruction was developed around (1) an overview of the traffic problem--how it re- lates to and affects the student, both today and in his future driving; (2) five different types of driving exper- iences--residential, light city, highway, heavy city, and expressway; and (3) a summary focusing on the role of the ‘well-informed traffic citizen. A systematic analysis of ‘the essential elements of driving was applied to the dif- ferent types of driving experiences. In the light of this, gnxblished driver and traffic safety education course 71 content was classified and reviewed. Those content seg- ments which best applied to a particular driving experience were then extracted and placed in their proper context within the most pertinent type of driving experience. Selected and treated within the respective experimental categories, as most applicable, were content subjectival facets such as engineering, enforcement, laws of nature, man—made laws, stopping distances, driving procedures, perceptual skills, health, effects of alcohol, and driving emergencies. For example, force of impact and centrifugal force were discussed in the highway driving unit instead of the unit on residential driving, because they were more applicable to the higher speeds and types of roadways. The principles of traffic engineering were discussed in all units. Other content aspects of driving, such as trip planning and preventive maintenance, were placed in only one or two of the units. Thus, an instructional unit for each of the five types of driving experiences was developed. Three individual sets of on-street lesson plans were developed for instructional application to the stand- ard (6 hours), simulator and range (3 hours), and four- phase (2 hours) laboratory programs, respectively. All, however, were designed to provide similar driving experi- ences. The lessons and their routes were engendered 'through staff group process and duplicated for each in- structor (see Appendix A). They were coded by a program 72 designation--standard "B," simulator and range "C," and four-phase "A"; and a trip number "1-18." This code was necessary for scheduling and specifying the specific on- street lesson to be taught a particular student. Three students were assigned to a driving group. If one was absent, the two remaining students drove for the entire period.' When the absentee returned, he was required to make up both the lesson(s) missed and the driving time lost. Instructors were not permitted to take only one student out for a lesson. In like manner, the simulator films (see Appendix A) were evaluated for content and selections were made for the two types of simulator programs. The simulator and on-street programs utilized all or part of thirteen films to be shown in the twelve lessons. The four-phase program utilized eleven films to be shown in the eight lessons. Similar films from the two types of simulators were matched. After the simulator instructors were selected, they critically analyzed and evaluated the films to de- termine where planned stops of the films should be made, what supplemental information, if any, should be present- ed, and/or what concepts should be treated during the film stops. The simulator films were coded in a similar.manner .as the on-street lessons, except for an "S" to designate a simulator lesson on the daily lesson schedule. 73 The off-street multiple car driving range lessons were developed to provide a progression of skills and concepts. Whenever possible, a particular skill was taught on the range immediately following instruction in. the simulator; but in all cases was taught prior to on- street instruction. For example, students were shown the instructional film on passing in the simulator prior to practicing the passing maneuver on the range, and prior to passing other cars in the on—street phase of instruction, except for situations in residential and light city traf- fic. Range lessons were coded similar to the on-street lessons, except for an "R" used on the daily schedule to designate that it was a range lesson. Sub-Problem Four-~Providing the necessary facilities and equipment to conduct the study. A considerable variety of facilities and equipment . was needed. Excellent cooperation in providing the neces-. sary facilities and equipment was lent by both the Renton School District and the Office of the State Supervisor of Driver and Safety Education Programs. The Renton School District was responsible for providing and/or making arrangements for the following facilities and equipment: l. 74 A c0py of the text, Sportsmanlike Driving,2 Fifth Edition, for each student. Twenty-two automobiles, all of which were provided (nine Oldsmobiles, four Fords, four Dodges, three Chevrolets, and two Ramblers). Each was a standard size, four-door sedan equipped with power steering, power brakes, heater and defroster, two side-view mirrors, hydraulic dual brake, and a seatbelt for each occupant. Eighteen of the automobiles were equipped with AM-FM radios for use on the range. Maintenance, refueling, and storage of the auto- mobiles were also provided. A number of different size and types of rooms were required. Office and work space were provided in a portable classroom. Three classrooms and storage space also were provided in portable classrooms. One of these classrooms was equipped with a forty- place drivocator installation. A large, one- hundred-seat, study hall was provided for driver and traffic safety education students not scheduled for classroom or laboratory instruction. Office space for the Project Director and Assistant Di- rector was provided at the Renton Vocational- Technical School. Drivin Company , 2American Automobile Association, S ortsmanlike (Fifth Edition; Washington: McGraw-HiII Book 1965). 4. 75 A sixteen—place Allstate Good Driver Trainer simu- lator, equipped with a complete set of films, in- stant error identification panels on units, and a printer, was likewise made available. The Office of the Superintendent of Public Instruc- tion was responsible for providing and/or making arrange- ments for the following facilities and equipment: 1. A fire extinguisher, first-aid kit, two flares, and an inside instructor's mirror for each car in the on-street phase of instruction. Two sets of the above equipment were provided for use on the range. Two simulator installations were needed for the study. A twelve-place Aetna Drivotrainer simu- lator, equipped with a complete set of films, instant error identification panel on units, and a printer was leased for the study. The Office of the Superintendent of Public Instruc— tion constructed an off-street multiple car driving range on land owned by the Renton School District. The range was designed by the Project Director. Bidding was handled by the Renton School District. The range is 320 feet wide, 457 feet long, with an additional area for hill and expressway exercises. The range affords a wide variety of skill develop- mental and simulated traffic experiences, as 76 described elsewhere. It is of a size and embodies a number of experiences adequate for accommodating twenty cars at one time although only twelve cars were used simultaneously in the conduct of this aspect of the investigation. The instructor com- municated with each automobile via an FM trans- mitter and was received in each car through an FM radio. 4. One of the experiences planned for each student was that of pulling a trailer in one or two of the on-street lessons. Five trailers and trailer hitches were provided by the U-Haul Company of Washington. Trailer operating booklets were also provided for each student. 5. Range equipment such as signs, car numbers, cones, and barricades, was provided jointly by the Renton School District and the Office of the Superintend- ent of Public Instruction. A number of different types of audio-visual equipment and the supplies which were utilized were similarly supplied. Sub-Problem Five--Creating an experimental design to com- pare the relative effectiveness of four selected laboratory programs and three selected classroom programs, and to determine the significance of the sex factor. 77 The laboratory programs consisted of (l) a standard program with six hours on-street instruction; (2) a simu— lator program with twelve hours simulator and three hours on-street instruction; (3) a range program with six hours range (eight lessons) and three hours on-street instruc- tion; and (4) a four-phase program with eight hours simu- lator, six hours range (eight lessons), and two hours on- street instruction. Selection of the laboratory programs was based on the following rationale. The program to which the term, standard, is commonly applied, consisting of six hours on- street instruction in a dual-control automobile, is the program currently most prevalent in schools. It therefore required selection, per se. The 1964 report of Policies. and Practices for Driver and Traffic Safety_Education was used as a criterion for determining the simulator program.3 As recommended in this source, twelve hours simulation were given in lieu of three of the six hours on-street instruction. In the consideration of establishing the four-phase and range programs, another source, The Multiple- Car Method, has suggested a combination of eight hours simulation, six hours range, and two hours on-street 3Fourth National Conference on Driver Education, Policies and Practices for Driver and Traffic Safety Edu- cation (Washington: National Education Association, Na- tionaI Commission on Safety Education, 1964), p. 23. 78 instruction as a reasonable and practical four-phase pro- gram.4 In order to compare the relative effectiveness of range instruction and simulator instruction, the range program also used three hours on-street instruction to keep both programs constant in this respect. Six hours range instruction--identical to the amount of range in- struction hours provided in the four—phase program--was similarly retained to further reduce the number of vari- ables to a minimum. No ratio of range to on-street hours was employed for determining the number of hours in the range program. Conversely, effort was made to avoid any ratio. Off-street multiple car driving range instruction, it is felt, does not appear to lend itself to the establishment of ratios. If minimum standards for course approval need be estab- lished, it is deemed well to consider as criteria such. factors as size of the range, number and types of differ- ent experiences (both skill and perceptual), space and distance provided for various maneuvers and simulated traffic experiences, safety features embodied in the design of the range, number of vehicles the area can ef- fectively use, adequacy of the communication system, and effectiveness of the instructor operating the range. In 4Automotive Safety Foundation, The Multiple-Car Inethod (Washington: Automotive Safety Foundation, March, 'T5—197 , p. 6. 79 assessing the effectiveness of the range instructor the following factors are felt to require consideration: 1. Does he have control of the range? 2. Are his instructions to students simple and clear? 3. Does he have lesson plans which provide for a se- quential progression of skills? 4. Does he have empathy for the problems of students on the range? 5. Does he consider the safety factors involved in range operation? There have been a number of studies which have examined laboratory instruction, but very little research which examines the classroom phase of instruction has been conducted. This investigation, in part, treats this most important part of instruction. The classroom programs selected for consideration consisted of (l) thirty hours classroom instruction, (2) thirty hours classroom instruc- tion plus fifteen hours drivocator instruction, and (3) forty-five hours classroom instruction. Selection procedures for the three classroom pro-. grams embodied a rationale similar to that of the labora- tory programs. The thirty hours classroom program was selected because it is currently the most prevalent in schools. Many instructors of driver and traffic safety education have voiced the opinion that thirty hours class- :room instruction is not adequate to cover the material 80 which students need to become safe and efficient motor vehicle operators in today's complex traffic environment. An increase in the number of classroom hours raises a num- ber of questions. How can this time best be spent? Is an increase in the number of hours sufficient? Should the additional time be spent on the.same material, or should new material be considered? Should a different teaching technique be used? Could this time be better utilized through the use of a multi-media learning system? Although it was impossible to treat all the factors indicated by the queries, as many as possible were incorporated into the study design. It was decided that the basic content for all three programs should be the same. The second classroom program added fifteen hours drivocator instruc- tion to the basic thirty hours classroom program. The third program expanded the thirty hours teacher-centered, text—oriented instruction to forty-five hours similar in- struction with no new content given to the students.‘ The only difference was the extended amount of time given to each of the units of instruction. Twelve different instructional treatment groups (see Appendix C) were formed, thus affording an opportunity to compare the respective programs separately or collec- tively for significant differences which might result ‘through an analysis of the results obtained from the evaluation criteria . 81 Sub-Problem Six--Assigning students randomly to each of the different instructional treatment groups. Renton High School employed computers to determine class schedules. Utilizing this method, students were divided as evenly as possible among the six periods of the school day for driver and traffic safety education classes. Those SOphomore students who were eliminated from partici- pation were either not 15-1/2 years old by January 9, 1968; had previously received classroom or laboratory instruction in driver and traffic safety education; were unable to obtain a driver's license because of a previous traffic violation or parental refusal to sign for the driving permit: had a physical or mental impairment which would prevent them from being able to progress satisfactorily in a regular class (this included students who had been classified as special education students by the school, and those students who had casts which would prevent nor- mal operation of an automobile): or knew they were moving and would be in the program for only a short period of time. The remainder of the sophomore students were placed in alphabetical order, last name first, from A to Z. Each was then assigned a number from 1 to 12, starting with the first through the twelfth student. The thirteenth on the list was again numbered "1": the fourteenth, "2," etc., until all were assigned a number from 1 to 12. These 82 numbers coincided with the twelve different instructional treatment groups (see Appendix C) and constituted the determining factor for assigning students to their class- room and laboratory programs. At the conclusion of this portion of the investi- gation, the information and data on those students who moved during the study, or who did not have complete files of evaluation criteria were eliminated from the sample. This resulted in a total study sample of 801 students-- 402 boys and 399 girls (see Appendix C). Sub-Problem Seven--Assigning instructors to the different phases of instruction. The Office of the Superintendent of Public Instruc- tion did not participate in the selection of instructors. The Renton Personnel Office employed the additional number of instructors needed to conduct this phase of the investigation. The wide range of teaching experience, age, and physical limitations were factors considered in assigning instructors to the different phases of instruction. These factors, it was felt, would not seriously affect their performance in all areas of instruction, but that they could do so in certain phases of instruction. The assignment of range instructors represented one such serious concern. Range instruction requires a 83 great deal of mobility and good physical condition. Two of the instructors had back and foot conditions which lim- ited their effectiveness on the range. In addition, three instructors had no prior teaching experience in driver and traffic safety education, while still another had only taught driver and traffic safety.education for two months of summer school. Although not required, teaching experi- ence in the on-street phase of instruction is recommended prior to teaching on a range. The efficient, safe, and successful operation of a range is dependent upon the in— structor being able to identify and evaluate the actions of several drivers simultaneously. He should also be able to anticipate the actions that a student might make in his driving and be prepared to react instantaneously. Thus, range instructional competence is greatly enhanced by prior teaching experience in the on-street phase of instruction. A second difficulty in the assignment of instruc- tors resulted from an emergent complication in structuring the instructional procedure. The investigation of three different types of classroom programs required one more instructor than was anticipated. Furthermore, one of the instructors was available to the project for only three of the six periods each day. These factors made it necessary for the Assistant Director to teach two periods of the day. 84 To eliminate as many variables in instruction as possible, it was decided that two instructors would each teach three periods of simulation in the Aetna simulator and two instructors would each teach three periods in the Allstate simulator (see Appendix C). Two periods of in- struction on the range were assigned to each of the three instructors (see Appendix C). Each of the simulator and range instructors taught all students receiving simulator and range instruction in the periods they were teaching in that phase of instruction. Five instructors were assigned to teach in the classroom phase of instruction. To the degree of schedul- ing possible, each instructor had an equal effect on all instructional treatment groups (see Appendix C). One of the instructors'became seriously ill and died after the investigation was underway. His classes were assigned to the Assistant Director and to other in- structors who had planning periods at the time, but were teaching the particular phase of instruction. The on-street phase of instruction was taught by all instructors (see Appendix C). Instructors who taught in the other phases of instruction were also assigned stu- dents for on-street instruction. Where possible, on-street instructors were assigned students from all instructional treatment groups. 85 From September 5, 1967, to December 21, 1967, the Project Director and Assistant Director conducted an in- tensive in-service education program. During this time, all instructors were observed and evaluated in terms of their teaching strengths and weaknesses, suitability for each phase of instruction, and their adaptability to the techniques of instruction involved. Instructors were as- signed so that no one phase of instruction or instructional treatment group would receive instruction superior or in- ferior to the others. The desires of the instructors for teaching assignments were considered, but the equal effect on all instructional treatment groups by instructors was the primary criterion used in their assignment. Sub-Problem Eight--Selecting and/or developing valid and reliable measurement instruments and standardizing pro— cedures for administering the tests. Here initial consideration was given to the devel- opment of a series of tests--knowledge and road perform- ance--which would evaluate the various segments of the driving task. However, the time available for the devel- opment of instruments of this magnitude was inadequate. Similarly, the time necessary for the administration of a series of road performance tests to all students was also considered too great. The decision was made to use one existing test and to develop two other tests. 86 For testing driving performance, A New Road Test5 was selected to determine the performance level which students had attained at the conclusion of their driver and traffic safety education program. The correlated ap- proach of all phases of instruction made it possible for all students to complete their classroom and laboratory instruction at approximately the same time. The McGlade Road Test, as it has become known, was developed from a study of existing driver licensing tests from forty-six states. The preliminary information was screened and evaluated by a panel of safety experts. The road test, thus devised, was then put through a series of trials to determine whether it was reliable and valid, and could be readily administered. Reliability correlation ranged from .77 to .93. The McGlade Road Test was also sensitive enough to discriminate between experienced» drivers, inexperienced drivers, and chronic violators. Significance at the .01 level of confidence was found to exist between students who had received practice driving instruction and those who had not.6 The McGlade Road Test is a.twenty-eight item test (see Appendix C) consisting of the following: 5Francis S. McGlade, A New Road Test for Use in Driver Licensing, Education and Employment (New York: New York University, Center for Safety Education, 1961). 6 Ibido I pp. 1-20 87 A. Road Test Items 1. Skill Tests a. Prior to start b. Starting c. Backing d. Turnabout e. Park, uphill f. Start on upgrade g. Parallel park 1) between standards 2) by curb 2. Road Problems a. Right turns b. Left turns c. Traffic lights d. STOP signs e. Other type signs f. Lane changes B. Driving Situations l. Uncontrolled intersection a. Straight through b. Turning 2. Blind intersection C. Road Test Area Should Include: 1. Restricted area for skill tests 2. Residential area 3. Multiple-lane roads The course covered a minimum distance of two miles and took about twenty minutes to drive.7 For the purpose of this investigation, the test was evaluated as Driving Performance Test--Total and Driv- ing Performance Test--Parts I and II. Part I included items 1 to 10, 16, 21, 26 to 28, and were considered 7Ibido' pp. 3‘4. 88 primarily as manipulative skill items. Part II included items 11 to 15, 17 to 19, 22 to 25, and were considered as perceptual skills. A team of ten Washington State driver licensing supervisors, one state driver licensing examiner, twelve driver and traffic safety education instructors, and the associate supervisor of driver and safety education pro- grams, Office of the Superintendent of Public Instruction, administered the Driving Performance Test. A two-day workshop was conducted to familiarize the examiners with the different parts and instructions included in the test manual. When questions and/or.points of disagreement arose which could not be answered by the Project Director, a telephone call was made to the author of the instrument to answer these questions. Examiners were given a typed copy of the test routes and went over each route in a car with- out students. The examiners were then divided into two and three-man teams, consisting of both driver and traffic safety education instructors and driver licensing person- nel. Each team evaluated a student not involved in the study while he drove over the test route. After returning to school, each group compared evaluations in an effort to resolve any possible differences. Each group then eval- uated a second student not involved in the study and com- pared evaluations at the end of the trip. A standard evaluation sheet was used by all examiners (see Appendix 89 B). The entire group of examiners reassembled to resolve any questions which might still be present concerning any part of the test. A random assignment of students in each instruc- tional treatment group to examiners was used. Half of the students in each instructional treatment group were exam- ined by the driver license examiners and the other half were examined by the driver and traffic safety education instructors. The examiners were not told which instruc- tional treatment group a student had received. They were further instructed not to ask the student what group or type of program to which he had been assigned. No driver and traffic safety education instructor examined a student he had taught in the on-street phase of instruction. The examiners were assigned students from all instructional treatment groups. To evaluate student learning, it was decided to employ the National Test in Driver Education.8 This is a true-false test consisting of seventy questions (see Ap- pendix B). Sixty-four of the questions covered material usually included in driver and traffic safety education classes and textbooks. These questions relate to the various aspects of the driving task. The remaining six 8National Test in Driver Education (Special Form: New York: New York University, Center for Safety Educa- tion, 1967). 90 questions apply particularly to the State of Washington. The material contained in these questions is based pri- marily on information contained in the 1966 State of Wash- 9 ington Driver's Guide. This test was constructed on the basis of face validity. Reliability was established by a test, re-test procedure of driver and traffic safety education students in New York and Washington. A reliability coefficient of .60 to .67 was obtained. The instrument was pre-tested in the usual manner. The test was administered to all students partic- ipating in the study before beginning (as a Pre-Knowledge Test) and after completing the course in driver and traffic safety education (Post-Knowledge Test). The directions appearing on the test booklet were read to all students and the time limit was observed. The Post-Knowledge Test was administered to the students prior to the Traffic Analysis Test and the McGlade Road Test. The Pre-Knowledge Test should show if any partic—v ular instructional treatment group possessed a signif- icantly greater amount of knowledge pertaining to the material included in a driver and traffic safety education program prior to the course. The Post-Knowledge Test 9Department of Motor Vehicles, State of Washin ton Driver's Guide (Olympia, Washington: Department of Motor Vehicles, 1966). 91 should show if any particular instructional treatment group possessed a significantly greater amount of knowledge pertaining to driver and traffic safety education after taking a course in driver and traffic safety education. The purpose of a traffic analysis test has been construed to ascertain the ability of a person to analyze a traffic accident and determine how it could have been prevented. This process involves three basic steps: (1) the person has to analyze the factors contributing to the accident, (2) the person has to determine the causes of the accident (this indicates his ability to examine the accident objectively), and (3) the person has to be able to offer a solution as to how the accident could have been prevented. To perform this function, then, a Traffic Analysis Test was constructed by the investigator. Ahmann and Glock have recommended that essay test 10 The items be used for measuring pupil understanding. essay test is usually based on broad principles and rela— tionships. Properly constructed, it also measures import- ant and vital educational objectives, such as the abilities to select, organize, relate, synthesize, and apply infor- mation. Essay type questions were therefore utilized in the development of the test. 10J. Stanley Ahmann and Marvin D. Glock, Evaluating Pupil Growth (Second Edition; Boston: Allyn and Bacon, InC., 1963) ’ pp. 177-1780 92 The Post-Test for Traffic Analysis was developed to obtain more than a simple feedback of knowledge. The purpose of the test is to measure a person's ability to apply_his knowledge of driver and traffic safety to the driving task. The selected accident situations treated in the instrument also provide an opportunity for the person to indicate his attitude, especially in responses required as to causes of the accidents. Specifically, in terms of the Washington Driver and Traffic Safety Education Study, the Post-Test for Traffic Analysis was developed to deter- mine if any differences resulted in a person's ability to analyze a traffic situation as a possible result of being assigned to one of four laboratory programs or one of three classroom programs. The Traffic Analysis Test, as it emerged, consti- tuted an extension of the "You Are the Jury" radio series which emanated from the Highway Traffic Safety Center, Michigan State University, between 1956 and 1960. In 1966, Emery evaluated program tapes of the series and selected twelve of the most common accident situations and condi- tions in which a car and driver became involved.]Tl These programs also concerned the types of accident situations llSister Thomas More Emery, O. P., "A Critical Review of Selected 'You Are the Jury' Tapes" (Unpublished independent study, Michigan State University, Highway Traffic Safety Center, August, 1966). 93 that driver education students are most interested in dis- cussing. From these twelve, four programs were.chosen for treatment in the Traffic Analysis Test, i.e., "Case of the Missing Links," "Case of the Night Freight," "Case of the Invisible Ice," and tCase of the Busy Intersection." The basic consideration in selection of the four programs was their treatment of most types of driving situations, time of day, location, and weather conditions with which an automobile driver is confronted. The most notable excep- tion was the lack of treating an expressway driving situation. That portion of each of the four selected programs which described the accidents was dubbed onto one tape. A test booklet (see Appendix B) was prepared. It described briefly the procedure to be followed while taking the test, provided a drawing of each accident location, and listed the names of the people or objects involved in the acci- dent. It then asked two questions about each accident: Question A: How could the accident have been prevented? Question B: Who do you feel is primarily respon- sible for the cause of the accident? Each student was given the following instructions, "Answer in essay form for 9332 of the people involved in the accident. Make your answers as thorough as possible in the time allotted. If additional space is needed, 94 continue on the back of the page." Students were given seven minutes to respond to each tape. Transparencies were prepared and projected on a screen while the descrip- tion of the accident was being played and while the stu- dents were answering the questions. To insure uniformity, a tape of the directions for taking the test was made (see cover sheet of test booklet --Appendix B). A set of instructions for the person ad- ministering the test was also provided. The same person gave the test to all of the students.* Content validity of the test instrument was estab- lished by submitting a typed verbatim copy of the taped narrative description of each accident, plus the criteria to be used in evaluating the student responses (see Ap- pendix B) to each member of the Project Advisory Committee. Their responses attested the validity of the instrument. The evaluation criteria for each program were based on an analysis of the accident by the Highway Traffic Safety Center Staff12 and additional factual material included in the narrative portion of each tape. Two points were as- signed to those factors which were considered primary causes of the accident. One point was assigned to those factors which contributed to the cause of the accident but were not considered primary. 12Professional Staff of the Highway Traffic Safety Center, Michigan State University, East Lansing, Michigan. 95 The unreliability, lack of consistency, of the scoring of the pupil's response is one of the major limi- tations encountered when pupil achievement is measured by means of an essay test. However, it is possible to vastly improve scores or reader reliability if certain conditions are observed. First, the responses being scored must have been elicited by carefully framed test items which present the examinee with a well-defined task. Secondly, individ- uals preparing scoring systems using the analytical or rating method must master the method thoroughly and apply it carefully. Third, ample time must be allowed for the scoring.13 The Encyclopedia of Educational Research, Third Edition, has also given some factors which affect reader reliability. Some of these are as follows: The training of the reader--highly trained readers are more reliable than untrained ones. The specificity of the grading or marking criteria employed by the readers--the more precise and detailed the marking criteria, the higher the reliability. The extent to which the question is structured for the examinee--the greater the structuring, the higher the reliability. The heterogeneity of the population of candidates from which the essays to be marked are drawn--marks of paper drawn from homogeneous population are much less reliable than the marks of those drawn from hetero- geneous population. 13Ahmann and Glock, op. cit., p. 178. 96 The familiarity of the marker [reader] with the stu- dent--markers [readers] well acquainted with the work of the student are more reliable than those who do not know the student.14 The conditions listed above for improving reader reliability were all met to the greatest extent possible in scoring the responses to the Traffic Analysis Test. Reader reliability has been defined as "the extent to which individuals can agree on the score which should be assigned a particular essay question."15 The two ques- tions on each accident situation were structured to give the examinee a well-defined task, i.e., a statement as to how the accident described might have been prevented. The group of students involved in the study was heterogeneous since it involved most SOphomore students at Renton High School. Three scorers were used in evaluating the student responses. The scorers were instructors in the program and were familiar with both the program and students. Instructional treatment groups were not identified with students prior to or during the scoring of the papers. The three scorers went through a training session with the Project Director and the Director of Research, Office of the Washington State Superintendent of Public Instruction. l4Chester W. Harris (ed.), Encyclopedia of Educa- Eional Research (Third Edition; New York: The Macmillan Company, 1960), p. 1504. 15 Ibid. 97 Detailed evaluation criteria (see Appendix B) and score point values assigned were distributed to the scorers and were reviewed critically. Each was then given three test booklets, randomly selected, and was asked to evaluate them. Scores were not placed on the test booklet. After each, individually, had scored the three test booklets, the results were compared and differences in scoring analyzed. Following this analysis, three additional test booklets were selected randomly and treated in the same manner as described above. This process continued until twelve tests had been scored. The twelve tests were then returned to their original places. The scorers approached a more equal assignment of points as the scoring progres- sed. Forty-eight tests, eight from each period, were then selected randomly for determining a reliability coefficient of the scorers. Each reader scored the forty-eight tests independently. Reliability coefficients derived were .790, .864, .896, which are relatively high for this type of scoring. The remaining tests were divided into the twelve instructional treatment groups by period. Each reader scored one-third of each instructional treatment group.» Adequate time was provided for the scoring of all tests. Sub-Problem Nine--Collecting and applying statistical treatment of the data. 98 Scores from the pre and post National Test in 16 17 Driver Education (special form), McGlade Road Test, and Traffic Analysis Test18 were collected, matched, and placed in individual folders. A student information card containing the student's age, sex, instructional treatment group, classroom group, laboratory group, driver license number, and other pertinent information was also placed in the folder. These data were then key punched on IBM cards and transferred to a tape file for an IBM System 360 Com- puter.19 A program was then set up to produce the neces- sary information to test our hypotheses. The following statistical treatments were applied to the data: 1. A three—way factorial (4 X 3 x 2) unweighted means analysis of variance was applied to each of the seven criterion variables to test the basic hy- potheses concerning the relative effectiveness of the four laboratory programs, the three classroom programs, female and male students, and their 16National Test in Driver Education (Special Form: New York: New York University, Center for Safety Educa- tion, 1967). 17Francis S. McGlade, A New Road Test for Use in Driver LicensingL_Education and Emplo ent (New York: New York University, Center for Safety E ucation, 1961). 18"You Are the Jury" Traffic Analysis Test--devel- Oped as a part of this investigation as described in pp. 92-97 0 19The Washington State Office of the Superintendent of Public Instruction has installed an International Busi- ness Machines (IBM) System 360, Model 30 Computer. 99 various interactions. .Unequal cell frequencies indicated the use of an unweighted_means analysis as outlined by Winer. 2. Scheffé's Test for Multiple Comparisons was ap- plied in those instances where analysis of variance resulted in an F-value significant at a level equal to or less than .05.2 3. The McGlade Road Test22 provided, in part, for the rejection of the examinee.and termination of the test. As a result, a certain portion of the total sample of pupils returned incomplete and, there- fore, unusable deduction scores for the Road Test —-Parts I, II, and Total. The analysis of variance described above was applied to deduction scores of only those subjects who completed the Road Test. 4. Chi square was used to determine whether the fre- quency of rejection on the Road Test differed sig- nificantly among the four laboratory programs, among the three classroom programs, and between the sexes. 5. Chi square was used to test the hypotheses regard- ing the relative number of rejects and the propor—' tion of pass versus fail which were made by students on the McGlade Road Test.24 6. The Pearson Product-Moment Coefficient of Correla— tion25 was computed to determine whether a 20B. J. Winer, Statistical Principles in Experi- mental Desi n (New York: McGraw HiiiiBook Company, 1962), pp. - I 374-378. 21Allen L. Edwards, Experimental Design in Ps cho- lo ical Research (Revised EditiOn; New York: Holt, Rine- Hart, and Winston, May, 1962), pp. 154-156. 22 McGlade, loc. cit. 23Quinn McNemar, Ps cholo ical Statistics (Third Edition: New York: John WiIey & Sons, Inc., 1962), pp. 228-229. 24 McGlade, loc. cit. 25William L. Hays, Statistics for Psychologists (New York: Holt, Rinehart, and Winston, 1963), pp. 496- 510. 100 relationship existed between the study population's scores on the Pre-Driving Knowledge Test, Post- Driving Knowledge Test, Traffic Analysis Test, and Road Performance Test. Significance of coefficient of correlation was determined by the procedure explained in McNemar. A t-test was employed to determine if a significant gain in mean scores from Pre-Test to Post-Test for Driving Knowledge existed within any of the three classroom programs, four laboratory programs, female students, or male students. 7 26McNemar, op. cit., p. 137. 271bid., pp. 101-102. CHAPTER IV ANALYSIS OF DATA The ensuing pages deal with the statistical analy- sis of the data obtained on each student during the investigation. A three-way factorial (4 x 3 X 2) un- weighted means analysis of variance;1 Scheffé's Test for Multiple Comparisons;2 Chi-square;3 the Pearson Product- Moment Coefficient of Correlation;4 and t-test comprises the statistical instruments applied to the data. To insure optimal treatment, the chapter is divided into the following subdivisions: Pre-Test for Driving Knowledge Post-Test for Driving Knowledge Post-Test for Driving Performance - Total 1B. J. Winer, Statistical Principles in Experi- mental Design (New York: McGraw Hill Book Company, 1962), pp. 222-224, 374-378. 2Allen L. Edwards, Experimental Design in Psycho- logical Research (Revised Edition; New York: Holt, Rinehart, and Winston, May, 1962), pp. 154-156. 3Quinn McNemar, Psychological Statistics (Third Edition, New York: John Wiley & Sons, Inc., 1932), pp. 228-229. 4William L. Hays, Statistics for Psychologists (New York: Holt, Rinehart, and Winston, 1963), pp. 4§6-510. 5McNemar, op. cit., p. 101. 101 102 Post-Test for Driving Performance - Part I Post-Test for Driving Performance - Part II Post-Test for Traffic Analysis Linear Correlations Pre-Test for Driving Knowledge The National Test in Driver Education (Special Form)6 was given to all students in the study prior to instruction. The means and standard deviations of scores on the Pre-Test for Driving Knowledge are recorded below for the four laboratory programs (Table 1), the three classroom programs (Table 2), and for each sex (Table 3). TABLE 1. Means and Standard Deviations of Scores on the Pre-Test for Driving Knowledge among the Four Laboratory Programs. Laboratory Unweighted Standard Program N Mean Mean* Deviation Four-Phase 210 49.94 49.836 4.850 Standard 211 49.75 49.733 5.088 Simulator 196 49.86 49.862 4.377 Range 184 50.20 50.143 4.643 *Since an unweighted means analysis of variance was used, both means are given in these tables. No significant differences among these means occurred at the .05 level. 6National Test in Driver Education (Special Form; New York: New York University, Center for Safety Educa- tion, 1967). 103 TABLE 2. Means and Standard Deviations of Scores on the Pre-Test for Driving Knowledge among the Three Classroom Programs. Classroom N Mean Unweighted Standard Program Mean* Deviation 30 hour Class- room 268 49.99 49.958 4.671 30 hour Class- room plus 259 50.03 49.996 4.384 15 hour Drivo- cator 45 hour Class- room 274 49.78 49.728 5.157 *No significant differences among these means occurred at the .05 level. TABLE 3. Means and Standard Deviations of Scores on the Pre-Test for Driving Knowledge between Female and Male Students. Unweighted Standard Sex N Mean Mean* Deviation Female 399 48.94 48.903 4.539 Male 402 50.92 50.885 4.756 *No significant differences among these means occurred at the .05 level. A three-way factorial (4 X 3 X 2) unweighted means analysis of variance was used to test the null hypotheses that no significant differences would probably exist among the mean scores attained by students assigned to 104 one of four laboratory programs, one of three classroom programs, or between female and male students. Since the computed values of F (Table 4) were less than the critical value of F, the null hypotheses of no difference among the four laboratory programs; among the three classroom pro- grams; or between female and male students was accepted at the .05 level of significance. Post-Test for Driving Knowledge The National Test in Driver Education (Special Form)7 was again administered to all students at the completion of the course. The means and standard devia- tions of scores on this Post-Test for Driving Knowledge are recorded below for the four laboratory programs (Table 5), the three classroom programs (Table 6), and for each sex (Table 7). A three-way factorial (4 X 3 X 2) unweighted means analysis of variance was used to test the null hypotheses that no significant differences would probably exist among the mean scores attained by students assigned to one of four laboratory programs, one of three class- room programs, or between female and male students. The computed value of F (Table 8) for the laboratory program was less than the critical value of F. Therefore, the null hypothesis of no difference among the four 71bid. 105 momm.m nun mHHmo casuaz oc Hv ommm.a m ommm.n coauom IkucH mmzum oc Hv mmho. m ommm. xmm x mnoumuonmq oc av Hmme.a m Nonm.~ xmm x Eooummmao on av macs. m aesm.m snoumuonmq x Eoonmmmao 0: av mama. N 55mm. Eooummmau oc mm.m mvmm.m mamm.mm H mamm.mm xmm on Hv mvmfl. m mmmm. suopmuonmq Amo.vm m Eoommum wocmoHMHcmHm mo madm> mo moam> mmmwvm mo mommme HMOHDHHU omudmsoo z moummo m m .mmomHBOGM ocw>wuo How ummsuwum .mocmflhm> mo mammamcd .v mqméa 106 TABLE 5. Means and Standard Deviations of Scores on the Post-Test for Driving Knowledge among the Four Laboratory Programs. Laboratory N Mean Unweighted Standard Program Mean* DeViation Four-Phase 210 53.91 53.757 5.128 Standard 211 53.70 53.682 5.369 Simulator 196 54.28 54.312 4.993 Range 184 54.52 54.513 4.151 *No significant differences among these means occurred at the .05 level. TABLE 6. Means and Standard Deviations of Scores on the Post-Test for Driving Knowledge among the Three Classroom Programs. - —— Classroom N Mean Unweighted Standard Program Mean* Deviation 30 hour Class- 268 54.03 53.995 4.479 room 30 hour Class- room plus 15 hour Drivo- 259 54.69 54.736 4.821 cator 45 h°ur Class‘ 274 53.57 53.466 5.455 room *Significant differences among these means occurred at the .01 level. See Tables 8 and 9. 107 TABLE 7. Means and Standard Deviations of Scores on the Post-Test for Driving Knowledge between Female and Male Students. Unweighted Standard Sex N Mean Mean* Deviation Female 399 53.05 53.038 5.271 Male 402 55.11 55.094 4.392 *Significant differences among these means occurred at the .001 level. See Table 8. laboratory programs was accepted at the .05 level of significance between female and male students. The com- puted value of F (Table 8) for the laboratory program was less than the critical value of F. Therefore, the null hypothesis of no difference among the four laboratory programs was accepted at the .05 level of significance. The computed value of F (Table 8) for the differ- ence among mean score of students assigned to one of three classroom groups is greater than the critical value (F at the .01 level = 4.64) of F (Table 8). Therefore, the null hypothesis of no difference is rejected at the .01 level of significance. The computed value of F (Table 8) for the differ- ence between mean score attained by female and male stu- dents is greater than the critical value (F at the .001 level = 10.83) of F (Table 8). Therefore, the null hypothesis of no difference was rejected at the .001 level 108 moon. hub mHHmU aflnuflz oc HH.N ~mmo.m ommv.a m anom.m cofiuomumucH mmzum o: mm.m nova.a omom. m Hmo¢.m xmm x muoumnonmq on No.m Hmmm.a Hmmm.a N momh.m xmm x Eooummmao on av ammm. m 54am.m xsmwwmmmmmw Ho. mo.m mmvm.v mmmm.m m mHHm.m Eoonmmmau Hoo. mm.m mov~.mm mmnm.mm H mmnm.m~ xmm on as.a ommv.a naoo.d m ~m~o.m muoumuonmq mUGmOAMHcmfim mew9wm> mo wsam> mMMMWm Eowwmnm mwmm“Wm HMOHUHMU omusmeoo omummo .wmomH3OGM mca>fluo How unmelumom .mocmaum> mo mammamc< .m mamas 109 of significance, with the male student having a higher mean score than the female student. There were no significant interactions among the three factors in the analysis, although the three-way interaction did approach significance at the .05 level. Scheffé's Test for Multiple Comparisons8 was ap- plied to isolate the classroom programs having significant differences. Because the computed and critical values of F were so close, it was impossible to isolate the differ— ences at the .01 level. However, significance between classroom groups at the .05 level was found between the thirty hour classroom plus fifteen hour drivocator program and the forty-five hour classroom program in favor of the thirty hour classroom plus fifteen hour drivocator program (Table 9). Edwards states that Scheffé's Test is more conservative and larger differences will be required for significance.9 A further examination of the data revealed that five extreme scores (all female students) on the Post-Test for Driving Knowledge did not fall within the normal range of the other scores. A supplemental analysis, after eliminating the five extreme scores, produced no change in the level of significant differences among the classroom 8Edwards, loc. cit. 91bid., p. 154. 110 TABLE 9. Scheffé's Test for Multiple Comparisons, Post Test Scores for Driving Knowledge--Classroom Programs. Classroom Unweighted Comparisons Program Mean C1 C2 30 hour Classroom 53.995 0 -l 30 hour Classroom plus 54.736 +1 +1 15 hour Drivocator 45 hour Classroom 53.466 -1 0 di = 1.270 .741 Zaz = 2 2 Sd = .4184 .4184 i t = 3.035 1.771 Mean Square Error = .7003 Critical Value of t (at the .05 level) = 2.458 N = 8 programs or between female and male students. However, the elimination of the extreme scores did result in significance at the .05 level of confidence in the two- way interaction between classroom and sex and a three-way interaction between classroom, laboratory and sex (see Appendix D for the supplemental analysis). lll Gain from Pre-Test to Post-Test for Driving Knowledge The National Test in Driver Education (Special Form)lo was given to all students in the study prior to instruction and again at the completion of the instruc- tional period. Mean and standard deviations of gain scores were computed for the total sample, laboratory pro- grams, classroom programs, female students, and male stu- dents (Table 10). A t-test11 was employed to determine if a significant gain in mean scores from the Pre—Test to the Post-Test for Driving Knowledge occurred. There was a significant gain in mean scores at the .001 level of confidence for the total sample, for each laboratory program, for each classroom program, and for both female and male students. Further evaluation revealed that 80.3% of the students made some gain from the Pre-Test to the Post-Test for Driving Knowledge, 6.5% of the stu- dents did not have a change in their scores from the Pre- Test to the Post-Test, and 13.2% of the students had a decrease in scores from the Pre-Test to the Post-Test for Driving Knowledge. Figure 2 provides additional informa- tion on the amount of increase or decrease in scores from the Pre-Test to the Post-Test. 10National Test in Driver Education, loc. cit. llMcNemar, op. cit., pp. 101-102. 112 Hoo. mmo.a mmnm.ma Hoe momm.v ma.v Nov mam: Hoo. mmm.a mvmm.ma mmm Homo.m Ha.v mmm mamfimm xmm Hoo. mmm.a ooom.ma mum omo.m mh.m whm EoonmmmHU upon me Houmoo>fluo noon ma Hoo. mmm.a mmhm.ma mmm w>>.¢ mo.v mmm moan EooummmHo Moon om Hoo. mmo.a mmmm.ma mom mma.v vo.v mom Eooummmao H50: om Emumonm Eoonmmmao Hoo. mmm.a HHmm.mH mma ohm.v mm.¢ «ma mmcmm Hoo. mmm.a Hmmm.ma mma mvm.v av.¢ mad HoumHsEHm Hoo. mmm.a nmmm.aa oam mmb.v mm.m Ham pumocmum Hoo. mmm.a voom.aa mom mmm.v hm.m cam mmmnmuusom Emnmoum huoumuonmq Hoo. mmm.a noma.mm oom mom.v ma.v How mamemm Hmuoa Amo.v mmuoom monoom wocmo u no mwamm Eoowwum camm mo came 2 -amfiamflm msam> aoflumfi>mo no HMUflano omusmfiou mmmummo pumpcmvm c002 .mmomazocx mcfl>wuo MOM monoom ummalumom can unmelmum mo mammz cmmzumn swam mo mocmowmwcmwm How “moans .mumsfiom .oH mqmda .mmoma3ocx mcw>wuo How unmanumom may on mmoma3ocm mcfl>fluo How ummBImMm may Eonm mmuoom cw mmmmnomo can mmmmuocH .m musmfim 113 mmdeUZH mmmoum Emma mmdmmumn om+ mH+ 0H+ m+ 0 ml CHI mal oml .lflfi44Vfl4IL/PL.. .. . .. 5... «.. . . V1144AV1flHJ!o om om 00H mBZmDDBm m0 mmmEDz 114 Post-Test for Driving Performance--Tota1 The McGlade Road Test was given to all students at the completion of the course. This instrument provides for the immediate rejection of students at any time during the road test if they, (1) are involved in an accident, (2) make a dangerous action, (3) clearly violate any traffic law, or (4) show a lack of cooperation or refuse to perform as instructed.12 As a result, a number of students were rejected during the test. Their scores are not usable and have not been included in this statistical treatment. Chi-square was used to test the hypotheses that there would probably be no significant differences among the four laboratory programs, the three classroom programs, or between the sexes in the number of students rejected on the McGlade Road Test. The computed value of x2 for the laboratory pro- grams (Table 11) and the computed value of x2 for the classroom programs (Table 12) were both less than the critical value of x2. Therefore, the null hypotheses of no differences among the four laboratory programs and among the three classroom programs were accepted at the .05 level of significance. 12Francis S. McGlade, A New Road Test for Use in Driver Licensing, Education and Employment (New York: New York University, Center for Safety Education, 1961), p. 22. 115 TABLE 11. Chi-Square Test for the Rejection of Students on the McGlade Road Test among the Laboratory Programs. . Non- 2 222.222 — —— Four-Phase 24 186 210 .8857 164.7429 Standard 27 184 211 .8720 160.4550 Simulator 31 165 196 .8418 138.9031 Range _39 gig 1§4 .8587 135.6739 Total 108 693 801 599.7749 -599.5618 .2131 2 _ (801)2 _ 641,601 _ X ‘ 108 x 693 ‘ ‘717812‘ ‘ 8°5725 x2 = 8.5725 x .2131 x2 1.8268 The critical value of x2 at the .05 level of significance with three degrees of freedom is 7.815. 116 TABLE 12. Chi-Square Test for the Rejection of Students on the McGlade Road Test among the Classroom Programs. 2 . Non- Classroom Reject . B B Program A RefieCt A+B Kih 3:8 30 hour Classroom 32 236 268 .8806 207.8209 30 hour Classroom plus 34 225 259 .8687 195.4633 15 hour Drivocator 45 hour Classroom 42 232 274 .8467 196.4380 Total 108 693 801 599.7222 -599.5618 .1604 2 = (801)2 641,601 8 5725 X 108 x 693 7 ,844 ° x2 = 8.5725 x .1604 x2 = 1.3750 The critical value of x2 with two degrees of freedom is 5.991. at the .05 level of significance 117 The computed value of x2 (Table 13) for the differ- ences between the number of female and male students re- jected on the McGlade Road Test is greater than the critical value (x2 with one degree of freedom at the .01 level = 6.635) of x2 (Table 13). Therefore, the null hypothesis of no difference was rejected at the .01 level of significance with the female student having a higher frequency of rejection than the male student. TABLE 13. Chi-Square Test for the Rejection of Students on the McGlade Road Test between the Sexes . Non- 2 Reject . B B Sex A Regect A+B AIB’ £13- Female 69 330 399 .8271 272.9323 Male 39 363 402 .9030 327.7836 Total 108 693 801 600.7159 -599.5618 1.1541 2 2 _ (801) = 641,601 X ‘ 108 x 693 ‘737813 8°5725 x2 = 8.5725 x 1.1541 x2 = 9.8935 The critical value of x2 at the .05 level of significance with one degree of freedom is 3.841. 118 A further analysis of differences on the McGlade Road Test13 was pursued as a result of finding significance of rejection on the McGlade Road Test between the sexes. Chi-square was used to determine whether significant dif- ferences existed among the three classroom programs or among the four laboratory programs when female and male samples were analyzed separately. The computed values of x2 for female and male students in the laboratory programs (Tables 14, 15) and the computed values of x2 for female and male students in the classroom programs (Tables 16, 17) who were rejected on the McGlade Road Test were all less than the critical values of x2. These results suggest that no significant classroom X sex or laboratory X sex interaction exist. On the basis of past experimentation, a maximum allowable deduction of 54 points appears reasonable as a 14 For the purpose of this report, minimum passing score. then, those students who had deductions of more than 54 points and/or were rejected by the examiner failed the Road Test. Chi-square was also used to test the hypotheses that there would probably be no significant differences among the four laboratory programs, the three classroom programs, or between the sexes in the number of students who pass or fail the McGlade Road Test. l3Ibid. 14Ibid., p. 23. 119 TABLE 14. Chi-Square Test for the Rejection of Female Students on the McGlade Road Test among the Laboratory Programs. Non- 2 22:22 —. — Four-Phase 16 79 95 .8315 65.6947 Standard 20 85 105 .8095 68.8095 Simulator 16 84 100 .8400 70.5600 Range 11 ‘_§g _gg .8282 67.9191 Total 69 330 399 272.9833 —272.9323 .0510 x2 = —..22:§ = —22;:22 x2 = 6.9917 x .0510 x2 = 3.566 The critical value of x2 at the .05 level of significance with three degrees of freedom is 7.815. 120 TABLE 15. Chi-Square Test for the Rejection of Male Students on the McGlade Road Test among the Laboratory Programs. . Non- 2 nggggzgry RegeCt Regect A+B X28 323 Four-Phase 8 107 115 .9304 99.5565 Standard 7 99 106 .9340 92.4623 Simulator 15 81 96 .8437 68.3437 Range _3’ _1g _§§ .8941 67.9529 Total 39 363 402 328.3154 -327.7836 .5318 2 (402)2 161,604 x = 39‘2“??? = "117157 = 11.4151 x2 = 11.4151 x .5318 x2 = 6.0706 The critical value of x2 at the .05 level of significance with three degrees of freedom is 7.815. 121 TABLE 16. Chi-Square Test for the Rejection of Female Students on the McGlade Road Test among the Classroom Programs. . Non- 2 Classroom Reject . B B Program A RegeCt A+B A+B A+B 30 hour Classroom 21 108 129 .8372 90.4186 30 hour Classroom plus 21 116 137 .8467 98.2190 15 hour Drivocator 45 hour Classroom 21_ 12§_ 133 .7970 84.4812 Total 69 330 399 273.1188 -272.9323 .1865 2 _ (399)2 _ 159,201 _ x - gg—g—ggfi - —§77775 — 6.9917 x2 = 6.9917 x .1865 x2 = 1.3040 The critical value of x2 at the .05 level of significance with two degrees of freedom is 5.991. 122 TABLE 17. Chi-Square Test for the Rejection of Male Students on the McGlade Road Test among the Classroom Programs. . Non- 2 Classroom Reject . B B Program A Regect A+B A+B A+B 30 hour Classroom 11 128 139 .9209 117.8705 30 hour Classroom plus 13 109 122 .8934 97.3852 15 hour Drivocator 45 hour Classroom 15 126 141 .8936 112.5957 Total 39 363 402 327.8514 -327.7836 .0678 2 = (402)2 _ 161,604 z 11 4151 X 39 x 363 ‘ ‘IZTIE7’ ° x2 = 11.4151 x .0678 x2 = .7739 The critical value of x2 at the .05 level of significance with two degrees of freedom is 5.991. 123 The computed value of x2 for the laboratory pro- grams (Table 18) and the computed value of x2 for the classroom programs (Table 19) were both less than the critical values of X2. Therefore, the null hypotheses of no differences among the four laboratory programs and among the three classroom programs were accepted at the .05 level of significance. TABLE 18. Chi-Square Test for the Failure of Students on the McGlade Road Test among the Laboratory Programs. Laboratory Fail Pass A+B _B_ _§i Program A B A+B A+B Four—Phase 56 154 210 .7333 112.9333 Standard 45 166 211 .7867 130.5972 Simulator 56 140 196 .7143 100.0000 Range _4gl 153 1§4 .7717 109.5870 Total 199 602 801 453.1175 -452.4395 .6780 x2 = 5.3556 x .6780 x2 = 3.6311 The critical value of x2 at the .05 level of significance with three degrees of freedom is 7.815. 124 TABLE 19. Chi-Square Test for the Failure of Students on the McGlade Road Test among the Classroom Programs. _—7 I _ - — — f -_ — Classroom Fail Pass A+B B B2 Program A B A+B A+B 30 hour Classroom 64 204 268 .7612 155.2836 30 hour Classroom plus 70 189 259 .7297 137.9189 15 hour Drivocator 45 hour Classroom _65_ 222_ 214_ .7628 159.4197 Total 199 602 801 452.6222 —452.4395 .1827 x2 = r.—.- =n—6222: = x2 = 5.7389 x .1827 x2 = 1.0485 The critical value of x2 at the .05 level of significance with two degrees of freedom is 5.991. The computed value of x2 (Table 20) for the dif- ferences between the number of female and male students who failed the McGlade Road Test is greater than the critical value (x2 with one degree of freedom at the .01 level = 6.635) of x2 (Table 20). Therefore, the null hypothesis of no difference was rejected at the .01 level of significance with the female student having a higher frequency of failure than the male student. 125 TABLE 20. Chi-Square Test for the Failure of Students on the McGlade Road Test between the Sexes. *2 Fail Pass B B sex A B A+B A¥B 318 Female 131 268 399 .6717 180.0100 Male 68 334 402 .8308 277.5025 Total 199 602 801 457.5125 _452.4395 5.0730 2 2 _ (801) _ 641,601 X ' 199 x 602 ‘ II97798 5'3556 x2 = 5.3556 x 5.0730 x2 = 27.1690 The critical value of x2 at the .05 level of significance with one degree of freedom is 3.841. A further analysis of differences on the McGlade Road Test15 was pursued as a result of finding signifi- cance of failure between the sexes in its application. Chi-square was used to determine whether significant dif- ferences existed among the three classroom programs or among the four laboratory programs when female and male samples were analyzed separately. The computed values of x2 for female and male students in the laboratory programs (Tables 21, 22) and lsIbid. 126 TABLE 21. Chi-Square Test for the Failure of Female Students on the McGlade Road Test among the Laboratory Programs. Laboratory Fail Pass A+B _B__ _§3 Program A B A+B A+B Four-Phase 28 67 95 .7053 47.2526 Standard 29 76 105 .7238 55.0095 Simulator 41 59 100 .5900 34.8100 Range _33_ _§§_ _99 .6667 44.0000 Total 131 268 399 181.0721 -l80.0100 1.0621 x2 = 4.5346 x 1.0621 x2 = 4.8162 The critical value of x2 at the .05 level of significance with three degrees of freedom is 7.815. the computed values of x2 for female and male students in the classroom programs (Tables 23, 24) who failed the McGlade Road Test were all less than the critical values of x2. These results suggest that no significant class- room X sex or laboratory X sex interaction exist. 127 TABLE 22. Chi-Square Test for the Failure of Male Students on the McGlade Road Test among the Laboratory Programs. Laboratory Fail Pass A+B _§__ pg: Program A B A+B A+B Four-Phase 28 87 115 .7565 65.8174 Standard 16 90 106 .8491 76.4151 Simulator 15 81 96 .8438 68.3438 Range _9 _Z§ _85 .8941 67.9529 Total 68 334 402 278.5292 -277.5025 1.0267 x2 = (4)02)2 = 161,604 = 7.1154 X2 = 7.1154 X 1.0267 x2 = 7.3054 The critical value of x2 at the .05 level of significance with three degrees of freedom is 7.815. 128 TABLE 23. Chi-Square Test for the Failure of Female Students on the McGlade Road Test among the Classroom Programs. Classroom Fail Pass A+B B B2 Program A B A+B A+B 30 hour Classroom 39 90 129 .6977 62.7907 30 hour Classroom plus 47 90 137 .6569 59.1241 15 hour Drivocator 45 hour Classroom _45_ _88_ _33_ .6617 58.2256 Total 131 268 399 180.1404 -180.0100 .1304 2 (399)2 159,201 x = x 2 8 = ‘357I08 = 4.5346 X2 = 4.5346 X .1304 x2 - .5913 The critical value of x2 at the .05 level of significance with two degrees of freedom is 5.991. 129 TABLE 24. Chi-Square Test for the Failure of Male Students on the McGlade Road Test among the Classroom Programs. Classroom Fail Pass A+B B B2 Program A B A+B A+B 30 hour Classroom 25 114 139 .8201 93.4964 30 hour Classroom plus 23 99 122 .8115 80.3361 15 hour Drivocator 45 hour Classroom 29_ 131_ 141_ .8582 103.8369 Total 68 334 402 277.6694 -277.5025 .1669 2 _ (402)2 _ 161,604 _ x - gfi‘i—ggz' — ‘2277I2 — 7.1154 x2 = 7.1154 x .1669 x2 = 1.1876 The critical value of x2 at the .05 level of significance with two degrees of freedom is 5.991. The means and standard deviations of deduction scores on the Post-Test for Driving Performance are re- corded below for the four laboratory programs (Table 25), the three classroom programs (Table 26), and for each sex (Table 27). 130 TABLE 25. Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Performance among the Four Laboratory Programs. Laboratory N Mean Unweighted Standard Program Mean* DeVlatlon Four-Phase 186 42.29 42.846 21.250 Standard 184 40.01 40.144 22.963 Simulator 165 46.89 46.831 24.866 Range 158 41.53 41.303 22.395 *Significant differences among these means occurred at the .05 level. See Tables 28 and 29. TABLE 26. Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Performance among the Three Classroom Programs. Classroom N Mean Unweighted Standard Program Mean* Deviation 30 hour Classroom 236 40.51 40.944 21.968 30 hour Classroom plus 225 42.55 42.566 23.703 15 hour Drivocator 45 hour Classroom 232 44.79 44.833 23.099 *No significant differences among these means occurred at the .05 level. 131 TABLE 27. Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Perform- ance between Female and Male Students. fl Unweighted Standard Sex N Mean Mean* Deviation Female 330 46.95 46.937 24.485 Male 363 38.65 38.625 20.726 *Significant differences among these means occurred at the .001 level. See Table 28. A three—way factorial (4 X 3 X 2) unweighted means analysis of variance was used to test the null hypotheses that no significant differences would probably exist among the mean scores attained by students assigned to one of four laboratory programs, one of these classroom programs, or between female and male students. The computed value of F (Table 28) for the classroom program was less than the critical value of F. Therefore, the null hypothesis of no difference among the three classroom programs was accepted at the .05 level of significance. The computed value of F (Table 28) for the dif- ference among mean scores of students assigned to one of four laboratory programs is greater than the critical value of F (Table 28). Therefore, the null hypothesis of no difference is rejected at the .05 level of significance. The computed value of F (Table 28) for the dif- ference between mean score attained by female and male 132 Nmm 331.2 a 2.8 .22.qu mo. .uddm NH.~ ammo.m meom.om o aeo~.H~m ooHuomuouoH mmzum oc mo.m OHvH.H wmmH.om m wwmm.om xmm x anoumuonmq on H v mmmm.H m ommm.m xmm x EooummmHo oc NH.m movm.H mmmm.Hm m boom.HmH >HODTHOQMH x EooummMHu oc mo.m mm~>.H mmHm.om m momo.Hm EooummMHu Hoo. mm.m Hmmo.mm mbmm.VHv H mhhm.vHv xom mo. ~m.m oomm.m mOOH.Hm m vmom.mmH muoumnonmq TUGMOHMHcmHm memswm> mo wsHm> mmmwmm Eoowwnm mWMMWWM HMOHDHHU oousmfiou mmummo .mocmEHOMHmm wcH>Huo How unmalumom .mocmHum> mo mHmaHmcd .mm wands 133 students is greater than the critical value (F at the .001 level = 10.83) if F (Table 28). Therefore, the null hy- pothesis of no difference is rejected at the .001 level of significance, with the female student having a higher mean deduction score than the male student. There were no significant interactions among the three factors in the analysis, although the three-way interaction did approach significance at the .05 level. Scheffé's Test for Multiple Comparisons was applied to isolate the laboratory programs having significant dif- ferences. No significant difference was found at the .05 level between the widest range of mean scores for the dif- ferent laboratory programs. "Scheffé suggests that with his test we might consider taking a = .10 rather than a = .05."16 On this basis, significance was found at the .10 level between the standard program and the simulator program in favor of the standard program (Table 29). If significance can not be found between the mean scores of the groups, Edwards suggests that the difference might lie in a comparison of different combinations of the treatments sums.l7 Significance was found at the .05 level between the simulator program and a combination of the standard program and range program in favor of the 16Edwards, op. cit., p. 154. 17Ibid., pp. 154-155. 134 combination standard program and range program. No other significant differences were found between comparison of different combinations of treatment sums (Table 29). TABLE 29. Scheffé's Test for Multiple Comparisons, Post- Test Deduction Scores for Driving Performance—- Laboratory Programs. Laboratory Unweighted Program Mean 1 2 5 Four-Phase 42.846 0 0 +1 0 +1 Standard 40.144 -1 0 -1 -l —2 Simulator 46.831 +1 +1 +1 +2 +1 Range 41.303 0 —1 -l -l 0 di '= 6.687 5.528 8.230 12.215 9.389 2a2 = 2 2 4 6 6 Sd == 2.425 2.425 3.430 4.201 4.201 = 2.758 2.280 2.399 2.908 2.235 17.6453 (at the .05 level) (at the .10 level) Mean Square Error Critical Value of Critical Value of N = 6 2.803 2.516 rl'rl'll Post-Test for Driving Performance--Part I For the purpose of this study, a further examina- 18 tion of the Post-Test for Driving Performance was made 18McGlade, loc. cit. 135 by dividing the test items into two parts. Part I included those items which were basically manipulative skills. The means and standard deviations of deduction scores on the Post-Test for Driving Performance--Part I, are recorded below for the four laboratory programs (Table 30), the three classroom programs (Table 31), and for each sex (Table 32). TABLE 30. Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Performance-- Part I, among the Four Laboratory Programs. Laboratory N Mean Unweighted Standard Program Mean* Dev1atlon Four-Phase 186 24.63 25.084 14.408 Standard 184 21.89 21.961 13.606 Simulator 165 26.73 26.725 16.497 Range 158 23.82 23.844 14.434 *Significant differences among these means occurred at the .05 level. See Tables 33 and 34. A three-way factorial (4 X 3 X 2) unweighted means analysis of variance was used to test the null hypotheses that no significant differences would probably exist among the mean scores attained by students assigned to one of four laboratory programs, one of three classroom programs, or between female and male students. The computed value of F (Table 33) for the classroom program was less than 136 TABLE 31. Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Performance-- Part 1, among the Three Classroom Programs. Classroom N Mean Unweighted Standard Program Mean* Deviation 30 hour Classroom 236 22.91 23.252 14.626 30 hour Classroom plus 225 24.61 24.697 14.999 15 hour Drivocator 45 hour Classroom 232 25.17 25.261 14.772 *No significant differences among these means occurred at the .05 level. TABLE 32. Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Perform- ance--Part 1, between Female and Male Students. Sex N Mean Unweighted Standard Mean* Deviation Female 330 26.82 26.844 15.543 Male 363 21.85 21.963 13.703 *Significant differences among these means occurred at the .001 level. See Table 33. the critical value of F. Therefore, the null hypothesis of no difference among the three classroom programs was accepted at the .05 level of significance. The computed value of F (Table 33) for the dif- ference among mean scores of students assigned to one of four laboratory programs is greater than the critical 137 Nmm amo¢.e wmm nHHmo erqu os NH.~ mwom.H moon... o mmHm.ao soHuomnousH smzum on H0 ~o~a.m m aoos.m xom x muoumuonmq os mo.m osem.H maaH.OH N mmam.o~ xom x soounano on NH.~ Hem~.H oaom.a o emem.mm suoumnonmq x Scounano oo mo.m mooH.H momm.m N mHmH.SH soonnnnHo Hoe. om.m HmHm.aH oemm.~eH H oema.me. xmm mo. mo.~ aoe~.m eHom.em m ommn.~e mnoumuonmq 2.2.2.222... .megwm2 .6 M2... .mmwmm amowwwmma .MM2mmm HmoHano oousmsoo H pummnuoocmEHOMHmm msH>Huo How unmauumom .mocmHum> mo anmHnsd .mm mamde 138 value of F (Table 33). Therefore, the null hypothesis of no difference is rejected at the .05 level of signifi- cance. The computed value of F (Table 33) for the dif- ference between mean scores attained by female and male students is greater than the critical value (F at the .001 level = 10.83) of F (Table 33). Therefore, the null hypothesis of no difference is rejected at the .001 level of significance, with the female student having a higher mean deduction score than the male student. There were no Significant interactions among the three factors in the analysis. Scheffé's Test for Multiple Comparisons was ap- plied to isolate the laboratory programs having signifi- cant differences. Significance at the .05 level of confidence (Table 34) was found between the standard program and simulator program in favor of the standard program. No other significant differences were found among the laboratory programs. Post-Test for Driving Performance--Part II For the purpose of this investigation, a further examination of the Post—Test for Driving Performance19 was made by dividing the test items into two parts. 19Ibid. 139 TABLE 34. Scheffé's Test for Multiple Comparisons, Post- Test Deduction Scores for Driving Performance-- Part I--Laboratory Programs. Laboratory Unweighted Program Mean C1 C2 C3 Four-Phase 25.084 0 +1 0 Standard 21.961 -1 -l 0 Simulator 26.725 +1 0 +1 Range 23.844 0 0 -1 d1 = 4.764 3.123 2.881 Ta2 = 2 2 2 Sd. = 1.571 1.571 1.571 1 t = 3.032 1.988 1.834 Mean Square Error = 7.4039 Critical Value of t (at the .05 level) = 2.803 N = 6 Part II includes those items which are primarily per- ceptual skills. The means and standard deviations of deduction scores on the Post—Test for Driving Performance-- Part II, are recorded below for the four laboratory pro- grams (Table 35), the three classroom programs (Table 36), and for each sex (Table 37). 140 TABLE 35. Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Perform- ance--Part II, among the four Laboratory Programs. Laboratory N Mean Unweighted Standard Program Mean* Deviation Four—Phase 186 17.66 17.762 11.947 Standard 184 18.12 18.183 13.985 Simulator 165 20.15 20.107 13.299 Range 158 17.71 17.458 13.351 *No significant differences among these means occurred at the .05 level. TABLE 36. Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Perform- ance--Part II, among the Three Classroom Programs. Classroom N Mean Unweighted Standard Program Mean* Deviation 30 hour Classroom 236 17.60 17.692 12.020 30 hour Classroom plus 225 17.94 17.869 13.684 15 hour Drivocator 45 hour Classroom 232 19.61 19.571 13.714 *No significant differences occurred at the .05 level. among these means 141 TABLE 37. Means and Standard Deviations of Deduction Scores on the Post-Test for Driving Perform- ance--Part II, between Female and Male Students. Unweighted Standard Sex N Mean Mean* Deviation Female 330 20.13 20.093 14.224 Male 363 16.80 16.662 11.916 *Significant differences among these means occurred at the .001 level. See Table 38. A three-way factorial (4 X 3 X 2) unweighted means analysis of variance was used to test the null hypotheses that no significant differences would probably exist among the mean scores attained by students assigned to one of four laboratory programs, one of three classroom programs, or between female and male students. The computed value of F (Table 38) for the laboratory program and the class— room program was less than the critical value of F. Therefore, the null hypotheses of no difference among the four laboratory programs and among the three classroom programs was accepted at the .05 level of significance. The computed value of F (Table 38) for the dif- ference between mean scores attained by female and male students is greater than the critical value (F at the .001 level = 10.83) of F (Table 38). Therefore, the null hy- pothesis of no difference is rejected at the .001 level 142 sesa.m moo nHHmo sHsqu os ~H.~ mmme.H maa~.oH o oama.Ho soHuomnousH mnzum oo mo.m mmae.. Hmon.OH m eemH.mm xmm x muoumuonoq os Hv momm.m m oooe.oH xom x soounano on H0 Hmaa.m o emmm.o~ muoumuonmq x soounano on mo.m MHae.H mHHo.m m mm-.aH soounano Hoo. om.m maHm.HH mono.oe H momo.os row 0: mo.~ mmme.H Hmom.m m «mHm.mm snoumuonnq .mo..m m -MWMMMH. wmommwww mwpmmwww .mmwmm .mowwmmma .WM2mwm .HH unmmllmocmEHOMHmm mcH>HHQ MOM ummBIumom .mocmwum> mo mHmaHmcé .mm mqmda 143 of significance, with the female student having a higher mean deduction score than the male student. There were no significant interactions among the three factors in the analysis. Post-Test for Traffic Analysis The Post-Test for Traffic Analysis20 was given to all students at the completion of the course. The test deals with the analysis of the traffic situation. The means and standard deviations of scores on the Post-Test for Traffic Analysis are recorded below for the four laboratory programs (Table 39), the three classroom pro- grams (Table 40), and for each sex (Table 41). TABLE 39. Means and Standard Deviations of Scores on the Post-Test for Traffic Analysis among the Four Laboratory Programs. Laboratory N Mean Unweighted Standard Program Mean* Dev1atlon Four-Phase 210 12.21 12.28 3.691 Standard 211 12.34 12.32 3.705 Simulator 196 12.11 12.11 3.498 Range 184 12.74 12.71 3.487 *No significant differences among these means occurred at the .05 level. 20 pp. 78-820 "You Are the Jury" Traffic Analysis Test, developed as a part of this investigation as described in Chapter III, 144 TABLE 40. Means and Standard Deviations of Scores on the Post-Test for Traffic Analysis Among the Three Classroom Programs. Classroom N Mean Unweighted Standard Program Mean* Deviation 30 hour Classroom 268 12.34 12.379 3.475 30 hour Classroom plus 259 12.72 12.698 3.463 15 hour Drivocator 45 hour Classroom 274 11.99 11.985 3.825 *Significant differences among these means occurred at the .10 level. See Tables 42 and 43. TABLE 41. Means and Standard Deviations of Scores on the Post-Test for Traffic Analysis between Female and Male Students. Sex N Mean Unweighted Standard Mean* Deviation Female 399 12.65 12.64 3.621 Male 402 12.03 12.06 3.563 *Significant differences among these means occurred at the .025 level. See Table 42. 145 A three-way factorial (4 X 3 X 2) unweighted means analysis of variance was used to test the null hypotheses that no significant differences would probably exist among the mean scores attained by students assigned to one of four laboratory prOgrams, one of three classroom programs, or between female and male students. The computed value of F (Table 42) for the difference among mean scores of students assigned to one of four laboratory programs was less than the critical value of F (Table 42). Therefore, the null hypothesis of no difference among the four labora- tory programs was accepted at the .05 level of significance. The computed value of F (Table 42) for the differ- ence among mean scores of students assigned to one of three classroom programs was less than the critical value of F (Table 42). The null hypothesis of no difference among the three classroom programs was accepted at the .05 level of significance. However, the computed value of F is significant at the .10 level of confidence. The computed value of F (Table 42) for the differ- ence between mean scores attained by female and male stu- dents iS'greater than the critical value (F at the .025 level = 5.09) of F (Table 42). Therefore, the null hy— pothesis of no difference is rejected at the .025 level of significance, with the female student having a higher mean score than the male student. 146 «mam. use nHHoo sHsqu on ~H.~ Heam.H ammo. o mmme.m soHuomuousH Hmsnm on H0 ammo. m NaoH. xom x snoumuonmq os Hv mmHH. m onN. xom x soounano on Hv ommH. o emHm. muoumuonoq x soounano oH. mo.m amam.m HaHo.H N ~mmo.m soonnnoHo mmo. om.m HHoH.m memo.m H memo.~ row on Hv omhm. m HhMH.H wuoumuonmq -Mwnmm.. .a.w..m. .. 2.... .HMMW. ..WM... MM.HMW. . . . HMOHuHHU oopsmfioo mmumwo .mHmemc¢ OHMMMHB Mom ummalumom .mocmHnm> mo mHmaHmcd .mv mqmda 147 There were no significant interactions among the three factors in the analysis. Scheffé's Test for Multiple Comparisons was ap— plied to isolate the classroom program having significant differences. Significance at the .10 level of confidence (Table 43) was found between the thirty hour classroom plus fifteen hour drivocator program and the forty-five hour classroom program in favor of the thirty hour class- room plus fifteen hour drivocator program. No other significant differences were found among the classroom programs. TABLE 43. Scheffé's Test for Multiple Comparisons, Post- Test Scores for Traffic Analysis--C1assroom Programs. Classroom Unweighted C C Program Mean 1 2 30 hour Classroom 12.379 0 +1 30 hour Classroom plus 12.698 +1 0 15 hour Drivocator 45 hour Classroom 11.985 -1 -l d. = .713 .394 1 2a2 = 2 2 Sdi = .313 .313 t = 2.2278 1.258 .3922 (at the .05 level) (at the .10 level) MeSn Square Error Critical Value of Critical Value of N = 8 2.458 2.168 rfrrll 148 Linear Correlations The coefficients of intercorrelation among the six criterion measures are shown in Table 44. An in- spection of the intercorrelation matrix shows that all intercorrelations are significant at the .05 level except the following: Post-Test for Driving Performance--Part II vs. Pre-Test for Driving Knowledge, Post-Test for Driving Performance--Part II vs. Post-Test for Driving Knowledge, and Post-Test for Traffic Analysis vs. Post- Test for Driving Performance--Part I. The two highest correlation coefficients (.842 and .796) represent an expected relationship between each part score and the total score on the Post-Test for Driv- ing Performance.21 The relatively low correlation coef- ficient of .343 between the two parts of the Post-Test for Driving Performance is an indication that those parts do, as defined, measure different components of the driv- ing task. The low intercorrelations between the Post-Test for Driving Performance (Total, Part I, Part II) and the 22 Pre- and Post-Tests for Driving Knowledge indicate no practical relationship between these two measures. The 21McGlade, loc. cit. 22National Test in DriveEEducation (Special Form; New York: New York University, Center for Safety Educa- tion, 1967). 149 mmmo. u R How u 2 «one. u H mmm u z Ho>mH mo. man no u m0 msHm> HMUHHHHU I mmo.u omo.| omo.| OHM. «om. Hom om.mH mHmmHmcd UHMMMHB unmalumom I was. mom. HHH.| MHH.| mmm moo.mv Hmuoalnoocmfiuomuwm mcH>Hno umoalumom I mom. mmo.| hmo.| mmo omm.mH HH unmmuumocmEHOMHom msH>HHo unmenumom I omH.| HvH.u mam mHm.em H unmmuuoocmEHOMHTm mcH>Huo ummsuumom - smm. How mo.em omoonosx msH>Huo umoeuumom . Hoo ma.ae moooHsosm ooH>Huo pmoauoum Id Ida add dad Ga Gd “N w 2.0 o.en. o.a.u .e=.o 2.0 1.1 m. 9 p.3 441s 1.15 1.19 Trs era 8 34 934 131 331 A3 A. u 3 _ +10 . o . o . 1.. T;1 a Ill 1.1 411.1 711.1 u.m u.e I «.8 m.e .1m.e m.e .b.o Ros s e s e.s p.s s 1 W.+ u.+ u.+ u.+ X24 X o o o u u 2 e.u .enu e.u o o T. . I _ 1 _ J M M A . I. .T. .T. .1 I S A A A a a I. T. I. T. p. .P s u u n. .b .b .b .b .b e e .mmuammoz coHkuHHo me msu macaw mcoHumHmHHooumusH .vv mqmma 150 .304 and .310 correlation coefficients between the Post- 23 and the Pre— and Post-Tests Test and Traffic Analysis for Driving Knowledge indicate a very minor overlap in these two factors. All negative correlation coefficients in the mat- rix result from the fact that deduction scores were used for the Post-Test for Driving Performance. Summary of Findings This investigation was developed to examine selected laboratory programs, classroom programs, and the differences between female and male students with respect to the six criterion measures. The summation below groups the findings revealed into the above three categories. A section also treats the intercorrelation among the six criterion measures. The following is a summary of the findings for differences among the four laboratory programs: four- phase, standard, simulator, and range. 1. On the Pre-Test for Driving Knowledge, no significant differences existed among the mean scores attained by students assigned to the four laboratory programs. 23 loc. cit. "You Are the Jury" Traffic Analysis Test, 151 On the Post-Test for Driving Knowledge, no significant differences existed among the mean scores attained by students assigned to the four laboratory programs. There was a significant gain in the mean scores from the Pre-Test for Driving Know- ledge to the Post-Test for Driving Knowledge attained by students assigned to the four laboratory programs. On the Post-Test for Driving Performance, no significant differences existed among the rejection rates of students assigned to the four laboratory programs. On the Post-Test for Driving Performance, no significant differences existed among the failure rates of students assigned to the four laboratory programs. On the Post-Test for Driving Performance, a significant difference at the .05 level of confidence existed among the mean deduction scores attained by students assigned to the four laboratory programs. Multiple compari- sons revealed that students assigned to a simulator program had significantly higher deduction scores than a combination of a standard program and a range program. A 152 significant difference at the .10 level of confidence existed between the mean deduction scores attained by students assigned to a standard program and a simulator program, in favor of the standard program. 7. On the Post-Test for Driving Performance-- Part I, a significant difference at the .05 level of confidence existed between the mean scores attained by students assigned to a standard program and a simulator program, in favor of the standard program. 8. On the Post-Test for Driving Performance-- Part II, no significant differences existed among the mean scores attained by students assigned to the four laboratory programs. 9. On the Post-Test for Traffic Analysis, no significant differences existed among the mean scores attained by students assigned to the four laboratory programs. The following is a summary of the findings for differences among the three classroom programs: thirty hour classroom, thirty hour classroom plus fifteen hour drivocator, and forty-five hour classroom. 1. On the Pre-Test for Driving Knowledge, no significant differences existed among the 153 mean scores attained by students assigned to the three classroom programs. On the Post-Test for Driving Knowledge, a significant difference at the .01 level of confidence existed among the mean scores at- tained by students assigned to the three classroom programs. Multiple comparisons were unable to isolate any significance at the .01 level, but significance at the .05 level of confidence existed between the mean scores attained by students assigned to a thirty hour classroom plus fifteen hour drivocator program and a forty—five hour classroom program, in favor of the thirty hour classroom plus fifteen hour drivocator program. There was a significant gain in the_mean scores from the Pre-Test for Driving Knowledge to the Post-Test for Driving Knowledge attained by students assigned to the three classroom programs. On the Post-Test for Driving Performance, no significant differences existed among the re- jection rates of students assigned to the three classroom programs. 154 On the Post—Test for Driving Performance, no significant differences existed among the failure rates of students assigned to the three classroom programs. On the Post-Test for Driving Performance, no significant differences existed among the mean deduction scores attained by students assigned to the three classroom programs. On the Post-Test for Driving Performance-- Part I, no significant differences existed among the mean deduction scores attained by students assigned to the three classroom programs. On the Post-Test for Driving Performance-- Part II, no significant differences existed among the mean deduction scores attained by students assigned to the three classroom programs. On the Post-Test for Traffic Analysis, no significant differences at the .05 level ex- isted among the mean scores attained by stu- dents assigned to the three classroom programs. However, significance was found at the .10 level of confidence between the thirty hour classroom plus fifteen hour drivo- cator program and the forty-five hour classroom 155 program in favor of the thirty hour class- room plus fifteen hour drivocator program. The following is a summary of the findings for differences between female and male students on the six criterion measures. 1. On the Pre-Test for Driving Knowledge, no significant difference was found between the mean scores attained by female or male stu- dents. On the Post-Test for Driving Knowledge, a significant difference at the .001 level of confidence existed between the mean scores attained by female and male students, in favor of the male students. There was a significant gain in the mean scores from the Pre-Test for Driving Knowledge to the Post-Test for Driving Knowledge at— tained by both female and male students. On the Post-Test for Driving Performance-- Total, a significant difference at the .01 level of confidence existed between the re- jection rates of female and male students 24 taking the McGlade Road Test, with the 24 McGlade, loc. cit. 156 female students having a higher frequency of rejection than the male students. Further analysis suggested that no significant class- room X sex or laboratory X sex interactions On the Post-Test for Driving Performance-- Total, a significant difference at the .01 level of confidence existed between the failure rates of female and male students 25 with the fe- taking the McGlade Road Test, male students having a higher frequency of failure than the male students. Further analysis suggested that no significant class- room X sex or laboratory X sex interactions On the Post-Test for Driving Performance-- Total, a significant difference at the .001 level of confidence existed between the mean scores attained by female and male students, with the female students having a higher mean deduction score than the male students. On the Post-Test for Driving Performance-- Part I, a significant difference at the .001 level of confidence existed between the mean exist. 5. exist. 6. 7. 25 Ibid. 157 scores attained by female and male students, with the female students having a higher mean deduction score than the male students. 8. On the Post-Test for Driving Performance-- Part II, a significant difference at the .001 level of confidence existed between the mean scores attained by female and male students, with the female students having a higher mean deduction score than the male students. 9. On the Post-Test for Traffic Analysis, a significant difference at the .025 level of confidence existed between the mean scores attained by female and male students, with the female students having a higher mean score than the male students. There were no interactions among the factors in the analysis, although a three-way interaction between laboratory, classroom and sex approached significance at the .05 level of confidence on the Post-Test for Driving Knowledge and the Post-Test for Driving Performance. How- ever, on the supplemental analysis of the Post-Test for Driving Knowledge, after eliminating the five extreme scores,significance.at the .05 level of confidence did exist both in the two-way interaction between classroom and sex and in the three-way interaction between labora- tory, classroom and sex. 158 The coefficients of intercorrelation among the six criterion measures were significant at the .05 level of confidence or higher except for the following: Post— Test for Driving Performance--Part II vs. Pre-Test for Driving Knowledge, Post-Test for Driving Performance-— Part II vs. Post-Test for Driving Knowledge, and Post- Test for Traffic Analysis vs. Post-Test for Driving Per- formance--Part I. An expected high correlation resulted between each part score and the total score on the Post-Test for Driving Performance. The low correlation between the two parts of the Post—Test for Driving Performance is an in- dication that the two parts measure different components of the driving task. A low correlation resulted between the Post-Tests for Driving Performance and the Pre- and Post-Tests for Driving Knowledge. A low correlation also existed between the Post-Test for Traffic Analysis and the Pre- and Post- Tests for Driving Knowledge. CHAPTER V SUMMARY AND CONCLUSIONS As treated in the preceding chapters, this investi- gation was designed to examine the relative effectiveness of four selected driver and traffic safety education laboratory programs; three selected driver and traffic safety education classroom programs; and to determine if the relative effectiveness of the laboratory and classroom programs is different for female and male students. The largest high school in the State of Washington was selected for the experimental procedure to insure an adequate study sample. Precautions were taken to assure that no one phase of instruction or instructional treatment group would receive superior or inferior instruction. This was accomplished by randomly assigning students to the twelve instructional treatment groups, an intensive four-month in-service education program of the instructurs, the development of driving procedures and teaching points, and regularly scheduled staff meetings throughout the entire project. While applied specifically to the State of Wash- ington, hopefully the results of this study, together with the findings and conclusions of similar investigations, 159 160 will assist school administrations and instructors, both in Washington and in other states, in their quest for quality driver and traffic safety education programming. Criterion Measures This investigation examined selected laboratory and classroom programs, and the significance of the sex factor on the criterion measures. Criterion measures used for its evaluation were grouped into three evaluative categories. Knowledge Tests--The National Test in Driver Edu- cation (Special Form)1 was used for evaluating knowledge, and was given to all students in the study prior to and at the conclusion of the instructional program. The increase in mean scores from the Pre-Test for Driving Knowledge to the Post-Test for Driving Knowledge was significant at the .001 level of confidence for each laboratory program, each classroom program, female students and male students, although the increase in mean socres for each group was unexpectedly low. A review of the National Test in Driver Education (Special Form)2 and of the classroom lesson plans was made 1National Test in Driver Education (Special Form: New York: New York University, Center for Safety Educa- tion, 1967). 2 Ibid. 161 to determine the relevancy of the driving knowledge examin- ation questions to the subject material taught. An evalua- tion of the test items revealed that answers to three of the test questions were not covered in the instructional outline. Also, the answers to ten other questions were only covered as a general topic of information and were not spoken to directly, although they were generally covered in outside reading assignments. An evaluation of the classroom lesson plans revealed that most of the major areas were covered to some extent by the test questions, although not proportionate to the amount of time spent in the classroom phase of the pro- gram. Some of the topics that received inadequate atten- tion relative to the extent they were covered in the course were: (1) the various aspects of traffic engineering, (2) alcohol and the driver, (3) driving procedures, (4) load- ing and pulling a trailer, (5) travel planning, (6) various types of driving relative to the organizational structure of the program, (7) traffic citizenship, and (8) the test questions did not draw a relationship between the different instructional topics in the course. Driving Performance Test--The McGlade Road Test3 was utilized for evaluating driving performance. This 3Francis S. McGlade, A New Road Test for Use in Driver Licensing, Education and Employment (New York: New York University, Center for; Safety Education, 1961). 162 evaluation instrument was used as a Total Test and was also divided into two parts. Part I contained those items which were primarily designed to evaluate manipulative skills. Part II items pertained to perceptual skills. The test also provided for the rejection and/or failure of students. The Road Test was terminated if the student was involved in an accident, committed a dangerous action, committed a clear violation of any traffic law, exhibited a lack of cooperation, or refused to perform as instructed by the examiner.4 A student who received more than fifty-four deduction points failed the test. It should be remembered that deduction points work in a re- verse order--the lower the deduction points, the better the score. Traffic Analysis Test--Criterion measures are generally structured around true-false, completion, match- ing or multiple choice questions which ask the student to provide rote answers. These types of tests do not satis- factorily measure the student's ability to comprehend and apply the knowledge he has acquired. The Traffic Analysis Test was developed for this study to provide a measure of a student's ability to draw a meaningful relationship between the causes of an accident and how the accident might have been prevented. The premise on which this test 4Ibid., p. 13. 163 was developed was more than a feedback of knowledge which the student had acquired. Rather, it was an opportunity for the student to apply his knowledge to factors which contributed to the cause of the accident. Intercorrelations--Most of the coefficients of intercorrelation among the six criterion measures were significant at the .05 level of confidence. The two highest correlation coefficients (.842 and .796) represent an expected relationship between each part score and the 5 The rela- total score on the driving performance test. tively low correlation coefficient of .343 between the two parts of the driving performance test is an indication that those parts do, as defined, measure different com- ponents of the driving task. The low correlations between the driving perform- ance tests and the knowledge test6 indicate no practical relationship between these two measures. The .304 and .310 correlation coefficients between the Post-Test for Traffic Analysis7 and the Pre- and Post-Test for Driving Knowledge indicate a very minor overlap in these two factors. 51bid. 6National Test in Driver Education, loc. cit. 7"You Are the Jury" Traffic Analysis Test, developed as a part of this investigation as described in Chapter 111, pp. 78-82. ‘ 164 Results of Pre-Test for DriVing Knowledge The mean scores that students made on the Pre-Test for Driving Knowledge were not significantly different among the laboratory or classroom programs or between the sexes. The fact that the F value among the laboratory and classroom programs was <1 strongly supported the technique of randomly assigning students to the twelve instructional treatment groups. Discussion of Findings Laboratory Program--Laboratory techniques have re- ceived more research attention than any other phase of instruction in driver and traffic safety education. How- ever, many of these studies have been limited to driving performance in some type of laboratory or behind-the- wheel instruction, and have not concerned themselves with the effect that different laboratory programs have on knowledge. No significant difference among the mean scores on the Post-Test for Driving Knowledge was found among the laboratory programs. These results were similar to those obtained on the Post-Test for Driving Knowledge by 165 Nolan8 and Gustafson,9 but differed from the results ob- tained by Seals who found significant differences between the range and four-phase programs. Seals did not find a significant difference between the standard and range pro- grams.10 Perhaps a more sensitive knowledge test with a higher reliability coefficient would have produced a significant difference among the four laboratory programs in this investigation. A knowledge test developed around performance objectives or several knowledge tests developed with performance objectives as their base might produce a clearer insight into the effects that different laboratory programs have on knowledge. There were no significant differences among the laboratory groups in the rejection or failure rates on the McGlade Road Test which was utilized for evaluating driv- ing performace.11 8Robert O. Nolan, "A Comparative Study of the Teaching Effectiveness of the Multiple Car Off-Street Driv- ing Range and the Aetna Drivotrainer" (synopsis of an un- published doctoral dissertation, Michigan State University, 1964): PP. 12-14. 9Robert E. Gustafson, "A Study to Compare the Ef- fectiveness of Instruction in the Allstate Good Driver Train- er and on the Multiple Car Off-Street Driving Range with the Multiple Car Off-Street Driving Range" (an abstract of an unpublished doctoral dissertation, Michigan State Univer- sity, 1965), p. 2. 10Thomas A. Seals, "An Evaluation of Selected Driver and Traffic Safety Education Courses" (unpublished doctoral dissertation, Florida State University, August, 1966), pp. 62-64. 11McGlade, loc. cit. 166 As a result of the rejections on the McGlade Road Test, only 693 students had usable driving performance scores. An unweighted means analysis of variance of these mean deduction scores revealed significant differences among the driving performance of students in the labora- tory programs. Students in a standard program, or a com- bination of a standard and range program, had significantly fewer points deducted than students in a simulator program only. The mean deduction scores on both the Total and Part I (manipulative skills) of the McGlade Road Test for students in the simulator program were significantly larger than those for students in a standard program. Mean deduction scores for perceptual skill items were not significantly different among students assigned to any one of the four laboratory programs. The findings of this investigation are not in com- plete accord with results of previous studies comparing simulator programs with other laboratory programs. Other studies using driving performance tests have shown simula- tion programs to be equal to or possibly better than the standard program. However, Gustafson found that students enrolled in a range program, followed by two hours on- street instruction, scored significantly higher on the ve- hicle handling section of the final road test than students 167 enrolled in a combination simulator-range program followed by two hours on-street instruction.12 Proponents of simulation have cited the value of simulators in the development of perceptual skills. This investigation revealed no significant differences among the respective laboratory programs treated, with regard to this important aSpect of driving. The fact that differ- ences occurred on the Total and Part I of the McGlade Road Test is not easily explained in view of past research. The most credible explanation appears to lie in the inverse relationship between the mean deduction scores and the number of hours of driving instruction in an automobile, both as the Operator and as an observer, in either the on- street or range phase of the program (Table 45). Apparently, the additional instruction time in the automobile improves the manipulative skills needed in the operation of an automobile. The same parallel between the number of hours of driving instruction in an automobile and the mean deduction scores does not exist on Part II of the Post-Test for Driv- ing Performance (Table 45). Apparently, the experiences which students received in the standard program did not have as much influence on the perceptual skills as was evidenced in the simulator, range, and four-phase programs. 12Gustafson, op. cit., p. 4. 168 TABLE 45. Parallel between Mean Deduction Scores on the Post-Test for Driving Performance and Number of Hours of Instruction in the Automobile among the Laboratory Programs. Laboratory Mean Deduction Scores Number of Hours Pro ram of Instruction 9 Part I Part II Total in the Automobile Standard 21.89 18.12 40.01 18 Range 23.82 17.71 41.53 17 Four-Phase 24.63 17.66 42.29 14 Simulator 26.73 20.15 46.89 9 This suggests that simulation instruction may substitute adequately for behind-the—wheel experience in the attain— ment of perceptual skills, but perhaps not in the develop- ment of manipulative skills. A As in the knowledge test, a driving performance test developed around performance objectives or several driving performance tests based on performance objectives might represent a better evaluation of the driving task than the McGlade Road Test. Although the McGlade Road Test is considered to be one of the best of the driving performance tests presently available, there are certain aspects of the driving task which are not included. The criticism being focused on driver and traffic safety educa- tion today is partially a result of our failure to identify those elements or patterns of instructional activity in 169 simulation, classroom, multi-media, driving ranges or on- street instruction that contribute to those aspects of driving which are essential to the driving task. The Post—Test for Traffic Analysis did not reveal any significant differences among the mean scores of stu— dents assigned to the different laboratory programs. The results seem to indicate that students in all four labora- tory programs are equally capable of analyzing a traffic accident, determining the causes of the accident, and sug- gesting how the accident could have been prevented. Why all the bother and concern over the effective- ness of different laboratory programs? What difference does it make which type of laboratory program a student receives? First, and of primary importance, school adminis- trations and instructors desirous of improving their driver and traffic safety education programs are searching for new and additional knowledge which will give them the best possible program for their particular, and possibly uni- que, situation. A program which is highly successful at one school might not satisfy the requirements and character— istics of another school. The number of students, available space, qualifications of instructors, and a multitude of other variables dictate the type of driver and traffic safety education program a school should develop. ' The second question is one of economics. The lab- oratory program is the most expensive phase of instruction. 170 With an ever-increasing number of students taking driver and traffic safety education in schools, it is imperative that more efficient and economical methods of instruction be developed. The results of this investigation have sub- stantiated the findings of previous research which have indicated that range, simulator, and four-phase programs are as effective as the standard program. The major factor affecting the cost of the labora- tory program is the salary of the instructor. In the standard program, a one-to-one teaching ratio affects the number of instructors needed in the program. In a simula- tor, range, or four-phase program, a greater teaching ratio, often as high as one-to-twelve, will reduce the number of instructors or the equivalent of full-time in- structors needed to operate the program. Table 46 shows the difference in the number of instructors and cost per pupil between a standard program, a simulator program, a range program, and a four-phase program. (See Appendix C for explanation of procedure for arriving at number of instructors and cost per pupil.) Table 46 is used as an example only. The 360 students used in arriving at the number of instructors needed and the per pupil cost do not represent a minimum or ideal number of students for simulator, range, or four-phase programs. The number of class periods are based on fifty-four minute classes, 171 and they represent the number of periods needed to satisfy the hours of instruction for each of the programs. TABLE 46. Difference in the Number of Instructors and Per Pupil Cost for a Standard Program, A Simulator and a Four-Phase Program, a Range Program, Program. Pro ram Number of Per Pupil g Instructors Cost Standard 6 hours on-street (20 class periods) 3.3 $73.33 Simulator 12 hours simulation (13 class periods) 2.4 53.33 3 hours on-street (10 class periods) Range 6 hours range (8 class periods) 3 hours on-street (10 class 2.2 48.89 periods) Four-Phase 8 hours simulation (9 class periods) 6 hours range (8 class periods) 2.0 44.44 2 hours on-street (7 class periods) As shown in Table 46, simulator, range, and four- phase programs are all less expensive to operate than the standard program. The cost for the construction and/or purchase of a simulator and/or off-street multiple car driving range can be amortized over a three-to-five year period for a school having an annual enrollment of 172 approximately three-hundred-sixty students in a school- day driver and traffic safety education program. A school expending $50,000 for the purchase and construction of a simulator and range in the development of a four- phase program could amortize the cost in less than five years (see Appendix C). A school with more students, or a combination of several schools which would total more than three hundred sixty students taking driver and traffic safety education each year, could amortize the cost for purchase and/or development of facilities for a simulator, range, or four-phase program in less time. However, school administrations and instructors should not think only of simulator, range, or four-phase programs as means of reducing the cost of instruction. Their first and primary concern should be the improvement of their driver and traffic safety education program. Part of the instructor salary savings should be returned to the program for the purchase and development of audio- visual equipment, multi-media equipment, film libraries, transparencies and overlays, teaching aids, up-to-date textbooks, and other equipment needed for a quality pro- gram of driver and traffic safety education. Classroom Program--A1though the classroom phase of instruction in high school driver and traffic safety education has been virtually neglected in past research, it is encouraging to note interest which has recently been 173 given to this most important phase of the program. Exam- ples of this interest are: The Wisconsin Classroom Cur- riculum Instructional Driver Education Workshop,13 studies on the EDEX Learning System (Drivocator),14 and encouragement by Mann to utilize the technique of small group discussions as a method of instruction.15 However, that which probably contains the greatest implications for influencing classroom instruction in driver and traffic safety education is the Automotive Safety Foundation Driver Education Curriculum Study and Development Project.16 As stated earlier, this investigation was designed, in part, to investigate several questions concerning class- room instruction: Is additional time needed in the class- room phase of instruction in driver and traffic safety education? How can the additional time in the classroom best be spent? Complete answers to these two questions have not evolved. However, results of this investigation do provide new insight into the problem. 13Wisconsin Department of Public Instruction, Wis- consin Classroom Curriculum Instructional Driver EducaEIEn Workshop Proceedings, June 15-17, 1967 (Madison: WiSconsin Department of Public Instruction), p. 40. 14"Something New in Safety," EDEX Teaching Systems (Mountain View, California: EDEX Corporation), p. 2. 15William A. Mann, "Let's Talk It Over," Analogy (Charter Issue; Skokie, Illinois: Allstate Insurance Company), pp. 4-9. 16A Driver Education Curriculum Study and Develop— ment Project sponsored by the Automotive Safety Foundation. 174 An examination of the data reveals that students receiving thirty hours of classroom instruction had higher mean scores than the group that received forty-five hours of instruction, although the difference was not signifi- cant. It is impossible to determine the cause of this reduction. One possible explanation might be the lack of a challenge for the student. Too much time spent on an area might cause boredom. Another explanation might re- sult from the lack of student motivation for longer per- iods of time when the same technique and method are applied to the same content. This certainly indicates the need for different instructional techniques and content when the classroom hours in driver education are expanded. Results of this investigation indicate that fif- teen hours of additional time in the classroom, studying the same content, materials, and utilizing like techniques of instruction, did not result in significantly higher mean scores on the post-criterion measures. However, when fifteen hours of drivocator instruction were added to a thirty hour classroom program, significant differences on the Post-Test for Driving Knowledge and the Post-Test for Traffic Analysis were found in favor of the Drivo- cator Program. The results did not show any significant differ- ences when the criterion measures were applied to the thirty class hour classroom and the forty-five class hour 175 classroom programs. An examination of the data reveals that students receiving thirty class hours of classroom instruction had higher mean scores than the group that received forty-five class hours of instruction, although the differences were not significant. This disparity in score is difficult to determine. A possible explanation might be the lack of challenge to the student. Another explanation might result from the lack of student motiva- tion for longer periods of time when the same technique and methods are applied to the same content. This certain- ly indicates the need for different instructional tech- niques and content when the classroom hours in driver education are expanded. No significant differences among the classroom programs were revealed among the mean deduction scores nor the rejection and failure rates among the Post—Tests for Driving Performance. The results of this investigation indicate that the drivocator system made a significant contribution to classroom instruction within the design and controls of this investigation. The supplemental analysis of the Post-Test for Driving Knowledge data, after eliminating five extreme scores, did not result in a change in the level of signifi- cance among the classroom programs or between female and male students. However, the elimination of the extreme 176 scores did result in significance at the .05 level of con- fidence in the two-way interaction between classroom and sex and a three-way interaction between classroom, labora- tory and sex. An inspection of mean scores (Table E, Appendix G) in the two-way interaction between classroom and sex in- dicates that the difference between female and male stu- dents is less in the drivocator group than the thirty hour or forty-five hour classroom groups. This may suggest that the drivocator or multi-media approach to instruction benefits female students more than males. An inspection of mean scores in the three-way inter- action between classroom, laboratory, and sex did not pro- duce any viable explanation for this interaction. Female and Male Students-~Very little effort and investigation have hitherto been devoted to the relative needs of female and male students in driver and traffic safety education courses. One of the stated objectives of this investigation was to determine whether the relative effectiveness of the laboratory and classroom programs is different for female or male students. One of the most consistent results throughout this investigation was the highly significant differences (.001) in the post-test mean scores on the driving knowledge and driving perform- ance tests between female and male students in favor of the male students. 177 Other investigations have produced similar results. Hayes concluded, "the average girl apparently commences driver training at a much lower point on the skills con- tinuum than does the average boy."17 Bernoff found that drivotrainer boys exceeded drivotrainer girls significantly in specific driving knowledge and the boys surpassed the 18 Nolan found girls on the road test at the .10 level. that drivotrainer and multiple car boys had significantly better scores than drivotrainer and multiple car girls on a post-test for general driving knowledge. He also found that the drivotrainer boys out performed the drivotrainer girls at the .02 level of significance in mean road test scores for vehicle handling and road problems.19 A report from Project Talent indicated that boys seem to acquire significantly more information than girls in many areas.2 17Robert B. Hayes and others, Immediate Standard- ized Learning Reinforcement to a Complex Mental-Motor Skill (Driver Training) Using Electronically-Coordinated Motion Pictures (Abstract, Title VII Project No. 1090; Washington: U.S. Department of Health, Education, and Welfare, 1965), p. 2. 18Louis I. Bernoff, An Experimental Study of the Teaching Efficiency of the Aetna Drivotrainer System THartford, Connecticut: Aetna Life and Casualty, June, 1958). PP- 7-8. A 19Nolan, op. cit., pp. 1-2. 20"Cognitive Growth During High School," A Na- pional Longitudinal Study of American Youth--Project TALENT, Bulletin No. 6 (April, 1967), p. l. 178 Loft recommended that, "a study be made to determine if driver education courses should have any different content and/or methodology for girls and/or boys."21 The only criterion measure where the female stu— dents were significantly superior to the male students was in the Post-Test for Traffic Analysis where the female students had significantly higher mean scores at the .025 level of confidence. This result of the investigation represents a reversal in female-male differences on the post-test criterion measures. While there is no clear cut reason for this difference, several possible explanations exist. The Traffic Analysis Test measures more than a feedback of knowledge. It provides the student with a means of applying the knowledge he has acquired. The low intercorrelations between the Traffic Analysis Test and the knowledge and driving performance tests indicate that the Traffic Analysis Test is measuring some cognitive factors not included in the other criterion measures. The female students in this investigation were better able to apply their knowledge as it related to the possible prevention of the accident. If this were true generally, it could be a factor along with the amount of driving and differences in the time, place, and circumstances of 21Bernard I. Loft, "The Effects of Driver Educa- tion on Driver Knowledge and Attitudes in Selected Public Secondary Schools," Traffic Safety Research Review (June, 1960), p. 15. 179 driving, which result in women having fewer automobile accidents and deaths per miles driven.22 On the other hand, it is possible that the results could be an artifcact of the testing procedure, where fe- male students are better able to express in writing their analysis and accident prevention solutions. Although not conclusive, the results of this and previous investigations support the need for evaluating the present curriculum in driver and traffic safety educa- tion in an effort to ferret out the variables in driving knowledge and driving performance and female students in the analysis of traffic accidents which contribute to the greater achievement of male students. Research is needed to determine where differential content emphasis and time allocation for both female and male students is needed to compensate for any deficiences affecting the driving task. What different techniques of instruction or organi- zation of the various phases of instruction will reduce the differences on post-test criterion measures between female and male students? Would a large amount of class- room instruction early and laboratory instruction late produce different results than a large amount of laboratory instruction early and most of the classroom instruction . . 22National Safety Council, Accident Facts (1968 Edition; Chicago: National Safety Council, 1968), p. 55. 'L IL -I -l‘L 180 later in the course? Would either of these produce re- sults different than a totally correlated and integrated program in driver and traffic safety education? Answers to these questions are needed to determine the needs of both female and male students if a differential curriculum and/or time allocation in driver and traffic safety educa- tion is to be deve10ped. This could also provide informa- tion relative to a better type of classroom and laboratory organization for all students. The differences obtained in this investigation and the results of previous studies suggest the need for atten- tion to a differentiated curriculum for female and male students. In addition, there is an apparent need for the individualization of instruction in driver and traffic safety education for all students. Students commence driver education with different competencies relative to their knowledge and backgrounds of the subject. They have. different interests and needs. Each student is searching for different types of information which coincides with his area of interest. Why not capitalize on this motiva- tional factor?--Individualize the student's instructional program. Provide an Opportunity for him to delve into the various aspects of traffic safety that he feels are rele- vant to his needs. An individualized program in driver and traffic safety education can satisfy many of the needs of driver education students and can reduce much of the 181 needless duplication of material and information that the student already knows. Conclusions As a result of this investigation, the following conclusions have been reached: 1. The no significant difference result on the Post-Test for Driving Knowledge among the laboratory programs seems to indicate that: a. laboratory instruction in a standard, simulator, range, or four-phase program has no significant effect on the amount of driving knowledge attained by students; or b. the knowledge test used in the study was not sensitive enough to measure the in- fluence that the different laboratory pro- grams had on the attainment of driving knowledge by students. 2. The number of hours of instruction in an auto- mobile whether on the range or on-street seems to have a direct relationship to the develop- ment of manipulative skills. On the other hand, the results of this investigation sug- gest that perceptual skills may be developed through simulated as well as actual driving experiences. 182 The results of the Post-Test for Driving Per- formance suggest that simulation instruction as defined in this investigation may substitute adequately for a portion of behind-the-wheel instruction in the attainment of perceptual skills, but perhaps not in the development of manipulative skills. A simulator, range, or four-phase program, as defined in this investigation, provides as good a basis for program deve10pment as the standard program. Fifteen hours of additional classroom time, utilizing the same content, materials, and in- structional techniques, did not result in a significant difference in driving knowledge among the classroom groups. However, fifteen hours of additional classroom time utilizing the Drivocator System resulted in a significant difference in driving knowledge, in favor of the classroom plus Drivocator program. There- fore, the results of this investigation plus that of previous studies seem to indicate that the Drivocator System, using an immediate response-feedback concept, provides a quality base for the expansion of the classroom phase of driver and traffic safety education. 183 The results of this and previous investiga- tions seem to indicate a differential between female and male students in the acquisition of competencies related to the driving task. A need is also suggested for the individualiza- tion of instruction in driver and traffic safety education for all students. Simulation, range, and four-phase programs reduce the per-pupil cost of instruction. Recommendations As a result of this and previous investigations, the following recommendations should be considered: 1. School administrators and instructors of driver and traffic safety education should seriously consider simulator, off-street multiple car driving range, and/or four-phase laboratory programs in the expansion and im- provement of their driver and traffic safety education courses. In the expansion and improvement of the class- room phase of instruction, it is important to consider more than additional hours of instruc- tion. Consideration should also be given to the Drivocator System used in this study or other multi-media systems, small group 184 discussion, television, programmed texts, transparencies and overlays, magnetic tapes, time-lapse photography, films, and other audio-visual aids. Due to the many variables influencing off- street multiple car driving range instruction, no ratio of on-street to range hours is sug- gested. It is strongly recommended that each program be evaluated individually rather than a fixed ratio of on-street to range hours. Criteria for evaluation should consider the size of the range, type and number of differ- ent experiences (both skill and perceptual), number of vehicles the area can effectively use, space and distance provided for different maneuvers and simulated traffic experiences, adequacy of the communication system, and the effectiveness of the instructor operating the range. School administrators and instructors of driver and traffic safety education should critically evaluate the experiences that students in their classroom and laboratory programs are receiving. This is especially true in on-street instruction when it is supplemented by laboratory experi- ences received in simulators and/or off-street 185 multiple car driving ranges, so that unneces- sary duplication of driving experiences can be prevented and gaps filled which relate to the driving task. 5. Consideration should be given to the develop— ment of programs in driver and traffic safety education which will provide the additional time, content, and experiences to compensate for the differences in knowledge and driving skill (manipulative and perceptual) between female and male students. 6. There is a need for the development of a highly sensitive driving knowledge and driving per- formance test or a series of tests based on performance objectives to adequately evaluate certain types of driving situations and exper- iences. Suggested Research The results of this investigation have suggested the need for additional research: 1. A study to investigate the relative effective- ness of the Drivocator System as the basis of a classroom program of instruction. 2. A study to evaluate the advantages of a cor- related program of classroom and laboratory 186 experiences based on the performance objectives for various driving situations such as residen— tial, city, highway, expressway, and driving emergencies. A study to determine at what stage in the de- velopment of student competencies, the advan- tages of simulation can best be realized. As a result of the relationship shown in this investigation between the development of mani- pulative skills and the number of hours of driving instruction in an automobile, a study should be conducted to further determine the nature of this relationship and the amount of driving time needed in the development of these manipulative skills. A study to evaluate the present curriculum in driver and traffic safety education in an effort to ferret out the variables which con- tribute to differences in achievement between female and male students. A study to determine where differential con- tent emphasis and time allocation for both female and male students is needed to compen- sate for any deficiencies affecting the driv- ing task. 187 7. A study to evaluate different organizational structures used in driver and traffic safety education courses, both classroom and labora— tory instruction. 8. A study to further refine and explore the ap- plicability of the Traffic Analysis Test as an evaluation instrument in driver and traffic safety education. 9. A follow-up study of this investigation util- izing this study sample should be conducted to determine, among other things, the retention of knowledge, driving performance, and the ability to analyze traffic accidents among students in the different classroom and labora- tory programs and between the sexes, and the accident and violation records among students in the different classroom and laboratory pro- grams and, between the sexes. Observations At the conclusion of the investigation, an evalua- tion of methods and procedures used in the study indicated the need for changes in subsequent studies. Perhaps these alterations or changes would simplify the assigned task. A full year should be provided for the development and procurement of facilities and equipment, the training 188 of teachers, and the development of detailed lesson con- tent. A pilot study should be conducted prior to the investigation to possibly eliminate problems often en- countered in a research effort of this nature. The introduction of too many new innovations or teaching techniques should be limited. Many teachers are inexperienced in the teaching techniques and methods utilized in the drivocator, simulator, and off-street multiple car driving ranges. The concept of a totally correlated and integrated driver education program and detailed lesson plans are also unfamiliar to many driver education teachers. A thorough evaluation should be made of the background and competencies of the staff. The ex- tent of the new materials anc concepts introduced into the curriculum should be based on this evaluation. An adequate number of staff should be provided. The estimate of staff needs in the early stages of the project did not consider accurately enough the number of teachers who would be required by the research design, nor did it consider the possibility of extended absences by staff members. When a shortage of teachers was apparent, additional teachers should have been employed or there should have been a reduction in the number of students taught. 189 As stated previously, there is a need for more sensitive and comprehensive testing instruments for class- room instruction. The present usable instruments are es- tablished on textbook content and do not adequately cover the various aspects of traffic engineering, alcohol and drugs, traffic citizenship, and the broad area of traffic safety programming. Until these instruments have been developed, an adequate evaluation of student achievement in a driver and traffic safety education program similar to the one described in this study, will be difficult to measure. A similar need exists for an instrument to mea- sure achievement in road performance. Testing instruments used in the evaluation of student achievement should be obtained well in advance of the start of the study. The development of the knowledge test used in this investigation was not completed until a few days prior to the administering of the pre-test for driving knowledge. The delay in obtaining the knowledge test resulted in the need to develop lesson plans and course organization without knowledge of the specific areas contained in the test. The test, therefore, did not adequately measure student achievement of the knowledge and material contained in the curriculum. 190 The content and material contained in the curricu- lum, although not adequately treated in the test, consti- tuted a most important part of the knowledge needed by young drivers in today's complex traffic environment. BIBLIOGRAPHY BIBLIOGRAPHY A. BOOKS Aaron, James E. and Marland K. Strasser. Driver and Traffic Safety Education--Content, Methods and Organization. New York: The MacMillan Company, 1966. Ahmann, J. Stanley and Marvin D. Glock. Evaluating Pupil Growth. Second Edition. Boston: Allyn and Bacon, Inc., 1963. American Automobile Association. Sportsmanlike Driving. Fifth Edition. Washington: McGraw-Hill Book Company, 1965. . Teaching Driver and Traffic Safety Education. New York: McGraw-Hill Book Company, 1965} Edwards, Allen L. Experimental Design in Psychological Research. Revised Edition. New York: Holt, Rinehart, and Winston, May, 1962. Harris, Chester W. (ed.). Encyclopedia of Educational Research. Third Edition. New York: The Mac- Millan Company, 1960. Hays, William L. Statistics for Psychologists. New York: Holt, Rinehart, and Winston, 1963. McNemar, Quinn. Psychological Stapisticg. Third Edition, New York: John Wiley and Sons, Inc., l962. Winer, B. J. Statistical Principles in Experimental Desi n. New York: McGraw-Hill Book Company, I962. 191 192 B. PUBLICATIONS AND STUDIES An Abstract of Pertinent Research Related to EDEX Educa- tional Systems. Mountain View, Califorfiia: EDEX Corporation. Allgaier, Earl. "Results of the Driver Education Program 1936-65." No. 3616. Washington: American Automobile Association, October, 1965. , and Sam Yaksich. Effectiveness of a Driving Simulator. Washington: American Automobile AssociatiOn, Traffic Engineering and Safety De- partment, 1959. American Automobile Association. Driving Knowledge Test. Washington: American Automobile AssoCiation, 1956. . Driving Ranges--Multiple Car Plans for Driver Education Courses. No. 3677-D. Washington: American Autom5bile Association, Traffic Engineer- ing and Safety Department, April, 1961. Arthur D. Little, Inc. The State of the Art of Traffic Safet . Cambridge, Massachusetts: Arthur D. Little, Inc., June, 1966. Automotive Safety Foundation, The Multiple-Car Method. Washington: Automotive Safety Foundation, March, l967. Bernoff, Louis I. An Experimental Study of the Teaching Efficiency of the Aetna Drivotrainer System. Hartford, Connecticut: Aetna Life and Casualty, June, 1958. A Condensed Report of the Los Angeles Study of the Aetna Drivotrainer. Hartford, Connecticut: Aetna Casualty and Surety Division, 1955. Crancer, Alfred. An Evaluation of Driver Traininngased on Accident and Violation Rates. Report 004. Olympia, Washington: State of Washington Depart- ment of Motor Vehicles, May, 1967. Department of Motor Vehicles. State of Washington Driver's Guide. Olympia, Washington: Department of Motor Vehicles, 1966. 193 Driver Behavior--Cause and Effect. Second Annual Auto- mobile Insurance Industry Traffic Safety Research Symposium, Allstate Plaza, Northbrook, Illinois, March 19-21, 1968. "Driver Training to be Implemented." United States Air Force News Release. Colorado Springs, Colorado: Air Defense Command, USAF (May 26). Florida Department of Education. A Guide, Driver Educa- tion in Florida Secondary Schools. Bulletin No. 6. Tallahassee: State Department of Education, 1963. Forlano, George and J. Wayne Wrightstone. An Evaluation of the Aetna Drivotrainer in Selected New York City High Schools. Divisional Bulletin No. 3. New York: State Board of Education, Bureau of Educational Research, October, 1959. Fourth National Conference on Driver Education, Policies and Practices for Driver and Traffic Safety Educa- tion. Washington: National Education Association, National Commission on Safety Education, 1964. Fox, James H. Driver Education and Driving Simulators. Washington: National Education Association, National Commission on Safety Education, 1960. Hayes, Robert B. and others. Immediate Standardized Learning Reinforcement to a Complex Mental-Motor Skill (Driver Training) Using Electronically: Coordinated Motion Pictures. Abstract, Title VII. Project No. 1090. Washington: U. S. De- partment of Health, Education, and Welfare, 1965. Insurance Institute for Highway Safety, 20th Annual Driver Education Achievement Program, 1966-67lSCHOOl Year. Washington: Insurance Institute for Highway Safety. . 21st Annual Driver Education Achievement Program, 1967-68 School Year. Condensed Report. Washing- ton: Insurance Institute for Highway Safety, 1968. Key, Norman. Status of Driver Education in the United States. Washington: National Commission on Safety Education, National Education Association, 1960. 194 McGlade, Francis S. A New Road Test for Use in Driver Licensing, Education and Employment. New York: New Yofk University, Center for Safety Education, 1961. Moynihan, Daniel P. (Chairman). A Report of the Secre- taryjs Advisory Committee on Traffic Safety. Washington: Department of Health, Education and Welfare, February 29, 1968. National Commission on Safety Education, Seventh Annual Cpnference Proceedings of American Driver and Traffic Safety Education Association, June 26-28, 1963. Washington: National Education Associa- tion, National Commission on Safety Education, 1964. National Safety Council. Accident Facts. 1968 Edition. Chicago: National Safety Council, 1968. National Test in Driver Education. Special Form. New York: New York University, Center for Safety Education, 1967. New Horizons for Highway Safety Through Driving§Simulation. A Report Based on Discussions and Findings of the National Conference on Driving Simulation. Santa Monica, California, February 27-March l, 1961. Office of State Superintendent of Public Instruction. 1964 Driver Education Guide. Olympia, Washington: Office of Public InstructiBn, 1965. "Report on EDEX-ADP Training" - for the Internal Revenue Service. Rhum, Gordon J., Bertram L. Woodcock, and Tom A. Lamke. The Effectiveness of the Aetna Drivotrainer in DriVer Education. Hartford, Connecticut: Aetna Casualty and Surety Company, 1956. Siebrecht, Elmer B. Siebrecht Attitude Scale. New York: New York University, Center for Safety Education, 1941. "Something New in Safety," EDEX Teaching Systems. Mountain View, California: EDEX CorporatiOn. 195 Stack, Herbert J. History of Driver Education_in the United States. Washington: National Education Association, National Commission on Safety Educa- tion, 1966. Stratemeyer, Clara G. Accident Research for Better Safety Teachin . Washington: National Education AssociEF tion, National Commission on Safety Education, 1964. U. S. Department of Transportation. Highway Statistics 1967. Washington: U. S. Government Printing Office, 1967. Wisconsin Department of Public Instruction. Wisconsin Classroom Curriculum Instructional Driver Educa- tion Workshop Proceedings, June 15-17,ll967. Madison: Wisconsin Department of Public Instruc- tion. C. PERIODICALS Asher, J. William. "Do Driver Training Courses Produce Better Drivers?--An Alternative Hypothesis," Traffic Safety Research Review (March, 1968), pp. 2-60 Barcus, Delbert, John L. Hayman, and James T. Johnson, Jr. "Programmed Instruction in Elementary Spanish," Phi Delta Kappan (March, 1963). Bishop, Richard W. "Questions and Answers About Driving Simulators," Safety Education, Vol. 44, No. 4 (December, 1964), pp. 8-11. Boyer, Richard G. "The Challenge From Within," CALDEA Calendar, Vol. XV, No. 1 (October, 1967), pp. 7-8. . "Feedback Makes the Difference," Analogy (Winter 1966-67), PP. 10-15. . "The Triple Influence," Drivotrainer Digest, V01. III, NO. 3 (May, 1960Y, pp. 5-70 Bridgeman, Charles. "A Lecture Response Device: A Pre- liminary Report on a Key Aspect of a Co-ordinated Teaching Program in Anatomy," Journal of Medical Education (February, 1964). 196 Brody, Leon. "A Study of the Learning of Selected Driving Skills Through Exposure to a Specially Produced Motion Picture Film," Traffic Safety Research Re- view (June, 1961), pp. 25-29. "City Officials Visit White Plains High School," Drivo- trainer Digest, Vol. VI, No. 1 (May, 1963), pp. g-g. "Cognitive Growth during High School," A National Longi- tudinal Study of American Youth--Project Talent. Bulletin No. 6. (April, 1967). Collins, Edwin M., Earl C. Collard, and Deryck R. Kent. "Programmed-group Instruction in Dental Education, Journal of Dental Education. Vol. 31, No. 4, pp. Cook, Dean R. "Multimedia--A New Classroom Concept," CALDEA Calendar, Vol. XV, No. 1 (October, 1967), pp. 9, 24. Drawbaugh, Charles C. "Evaluating the Concept of Mobile Police Training," The Police Chief (August, 1968), pp. 56-60. "Driver Education Explosion in California," Rockwell Safety News, Vol. 1, No. 2 (April, 1963), pp. l—3. "The Driving Range: Instruction and Construction," Safety Education, Vol. 44, No. 9 (May, 1965), pp. 6-10, 19. Finch, Alfred C. "A Following Distance You Can Count On," Traffic Safety, Vol. 67, No. 11 (November, 1967), pp. 12-14’ 34-35. "The First 10 Years," Drivotrainer Digest, Vol. VI, No. 2 (December, 1963), pp. 9, 12-17. Haddon, William, Jr. "Haddon on Highway Safety," Analogy (Winter, 1968), pp. 4-7. Heeter, Lewis M. and Earl Allgaier. "Can a Simulator Sub for a Car?" Traffic Safety (December, 1961), pp. 20-21. 197 Herbert, Evan (Associate ed.). "A Special Report on Technology for Education," International Science and Technology (August, 1967), pp. 28-49} Horne, Virgil L. "Driving Range--Creation and Cost," Safety, Vol. IV, No. 2 (March-April, 1968), pp. 16-17. "Jefferson County, Colorado Combines Driving Range, Behind— the—Wheel and Simulators in Unique Driver Educa- tion Program," Rockwell Safety News, Vol. II, No. 1 (Winter, 19647, pp. 10-13. Kaywood, Richard. "Coordinating Drivotrainer System and Dual Control Car," Drivotrainer Digest, Vol. V, No. 3 (November, 1962), pp. 3-6. "Lakewood Study Convinces Educators," Drivotrainer Digest, V01. VII, NO. 2 (June, 1965), pp. 5-7. Lancaster, Otis E. "MARI: Motivator and Response Indi- cator," I.R.E. Transactions on Education (Decem- ber, 19617: Loft, Bernard I. "The Effects of Driver Education on Driver Knowledge and Attitudes in Selected Public Secondary Schools," Traffic Safety Research Review (June, 1960), pp. 12-15. Mann, William A. "Let's Talk It Over," Analogy (Charter Issue), pp. 4-9. Matthias, Howard. "New Concept for Range," Safety, Vol. IV, No. 4 (September-October, 1968), pp. 20-21. McDaniel, Charles E. "A Four-Phase Driver Education Pro- gram," 1964 National Safety Congress Transactions. Vol. XXIII. Chicago: National Safety Coundil, 1964, pp. 51-55. . "What Road Tomorrow?" Safety, Vol. 2, No. 2 (November-December, 1966), pp. 8-11. Nolan, Robert 0. "East Lansing Combines Drivotrainer and Range Plan Program in Second Year," Drivotrainer Digest, Vol. III, No. 2 (April, 1960), pp. 3-5. "Oklahoma City Agents Place Drivotrainer Classrooms in City's 10 High Schools," Drivotrainer Digest, Vol. VII, No. 2 (June, 1965?, pp. 11—12. 198 "Ottawa Township High School Plans Ahead to Keep Pace With Increasing Enrollments," Rockwell Safety News, Vol. 1, No. 2 (April, 1963), pp. 4-5. "Public Law 89-564, an Act," Safety, Vol. 2, No. 2 (Novem- ber - December, 1966), pp. 13-16. Ribicoff, Abraham. "Time for a New Look," Safet , Vol. 2, No. 2 (November-December, 1966), p. 7. Stack, Herbert J. "A Resume and Evaluation of Research on the Teaching Effectiveness of Simulated Driv- ing Experiences and Conventional Driver Education Methods," Traffic Safety Research Review, Vol. 3, No. 4 (December, 1959), pp. 12-14. "Teachers Surveyed on Drivotrainer System," Drivotrainer Digest, Vol. VI, No. 1 (May, 1963), pp. ll-12. D. UNPUBLISHED MATERIALS Bishop, Richard W. "Comparing the Effectiveness of Various Combinations of On-Street and Multiple Car Driving Range Instructional Hours." Un- published research study, Florida State Depart- ment of Education, Tallahassee, 1965. . "Evaluating Simulator Instruction for Accom- plishing Driver Education Objectives." Unpub- lsihed research study, Florida Institute for Continuing University Studies, Tallahassee, 1963. Brazell, Robert E. "A Follow-Up Study of Public School Driver Trainees, Relating Driving Performance Records to Selected Academic and Training Fac- tors." Unpublished doctoral dissertation, Uni- versity of Michigan, 1961, p. 19. "College Instructors Workshop Report, Summer, 1965," (Part 1 includes reports on simulation and off- street multiple car driving ranges), Michigan State University, Highway Traffic Safety Center. "Committee Reports of the College Driver Education In- structors Workshop, August l7-28, 1964," (a series of reports on simulator and off-street multiple car driving ranges), Michigan State University, Highway Traffic Safety Center. 199 Emery, Sister Thomas More, 0. P. "A Critical Review of Selected 'You Are the Jury' Tapes." Unpublished independent study, Michigan State University, Highway Traffic Safety Center, August, 1966. Gustafson, Robert E. "A Study to Compare the Effective- ness of Instruction in the Allstate Good Driver Trainer and on the Multiple Car Off-Street Driv- ing Range with the Multiple Car Off-Street Driving Range." Abstract of an unpublished Doctoral dissertation, Michigan State University, 1965. LeVan, Paul S. "The Value of an Auto Training Device in the Teaching of Driver Training." Unpublished Master's thesis, Temple University, 1953. Nolan, Robert O. "A Comparative Study of the Teaching Effectiveness of the Multiple Car Off-Street Driving Range and the Aetna Drivotrainer." Synopsis of an unpublished Doctoral dissertation, Michigan State University, 1964. Seals, Thomas A. "An Evaluation of Selected Driver and Traffic Safety Education Courses." Unpublished Doctoral dissertation, Flordia State University, August, 1966. APPENDICES APPENDIX A Laboratory Lesson Sequence Simulator Lesson Sequence On-Street Lesson Sequence On-Street Lesson Plan 200 mIU.mIm muo.nnm.mum nno.mum muo.muo.vum «no.muo.mnm NIU.HIU.NIM.HIm 8V Emnwoum mmcmm mIU.AUVNHIm mHIm mlm.Advmlm mIU .Aocflaum.xovoaum Baum.maum.maum 0-4.sum.mum.xmvnum muo.xovmum.xovm-m «Hum.maum m-«.m-m.xavmum muo «Hum.HHum muo.xovnum.xovmnm .oaum.mum.mum.num q-¢.vum .Aavmnm quo.xovmnm.xovvum mum.mum m-¢.mum.xavvum Nuo ~u<.Amm mmzmmmnmxm muwo >>wmm hasnmflm suflo unmfiq amaucmowmmm mcfi>fiua mo mama 201 WASHINGTON TRAFFIC SAFETY EDUCATION PROJECT SIMULATOR LESSON SEQUENCE SIMULATOR PROGRAM Film Titles Evaluation Checks "AETNA" (Groups 28, 30) 1. "You and the Drivotrainer System" 2. "A Drive in an Automatic Shift Car" #1 - Residential 3. "Backing Safely" 4. "Angle Parking and Turning Maneuvers" 5. "Blending in Traffic" #2 - Light City 6. "Perfect Passing" 7. "Safe Highway Driving" #3 - Highway 8. "ABC's of Parallel Parking" 9. "Traffic Strategy" #4 - Heavy City 10. "Expressway Excellence" ll. "Expressway Excellence" #5 - Expressway 12. "Road Check" "ALLSTATE" (Groups 29, 31) 1. "Start of Good Driving" 2. "The Good Turn" #1 - Residential 3. "In Reverse" 4. "City Driving" "Parking" (angle parking sequence) 5. "City Driving" #2 - Light City 6. "Highway Driving" 7. "Highway Driving" #3 - Highway 8. "Parking" 9. "Advanced City Driving" #4 - Heavy City 10. "Expressways Are Different" 11. "Expressways Are Different" #5 - Expressway 12. "Let's Review" 202 WASHINGTON TRAFFIC SAFETY EDUCATION PROJECT SIMULATOR LESSON SEQUENCE FOUR-PHASE PROGRAM Film Titles Evaluation Checks "AETNA" (Groups 20, 22) 1. "You and the Drivotrainer System" 2. "A Drive in an Automatic Shift Car" #1 3. "Backing Safely" 4. Objective Sequence of "Angle Parking and Turning Maneuvers" "Blending in Traffic" (ck) #2 5. "Perfect Passing" "Safe Highway Driving" (ck) #3 6. "Traffic Strategy" (ck) "ABC's of Parallel Parking" #4 7. "Expressway Excellence" #5 8. "Road Check" "ALLSTATE" (Groups 21, 23) 1. "Start of Good Driving" 2. "The Good Turn" #1 3. "In Reverse" 4. "City Driving" #2 5. "Highway Driving" #3 6. "Advanced City Driving" (ck) "Parking" (parallel parking sequence)#4 7. "Expressways are Different" #5 8. "Let's Review" Residential Light City Highway Heavy City Expressway Residential Light City Highway Heavy City Expressway Lesson B-l B-9 B-10 B-11 B-12 B-13 B-14 B-15 B-16 B-17 B-18 Trip Trip Trip Trip Trip Trip Trip Trip Trip Trip Trip Trip Trip Trip Trip Trip Trip Trip #1 #2 #2 #3 #4 #5 #6 #6 #7 #7 #8 #8 #9 #9 #10 #11 #11 #12 203 ON-STREET LESSONS Standard Program WASHINGTON TRAFFIC SAFETY EDUCATION PROJECT Evaluation Check Starting to Drive (area adjacent to school) Residential Driving (turns) Residential Driving (turns) Residential Driving (land changing and use of one-way streets) Light City Driving #1 Light City Driving #2 Rural Driving (county, secondary state roads and highways) Rural Driving (county, secondary state roads and highways) Highway Driving Highway Driving Trailer Trip #1 Trailer Trip #2 Heavy City Driving Driving Hills Driving #1 Heavy City Driving on Expressway Expressway Driving #2 Review #1 #3 #4 ' 204 On—Street Lessons Page 2 Lesson Four-Phase Program Evaluation Check A-l Trip #2 - Residential Driving (turns) A-2 Trip #3 - Residential Driving #1 (lane changing and use of one-way streets) A-3 Trip #5 - Light City Driving #2 #2 A-4 Trip #8 — Trailer Trip #3 A-5 Trip #9 - Heavy City Driving #4 A—6 Trip #11 - Expressway Driving #5 Lesson Range or Simulator Program Evaluation Check C-1 Trip #2 - Residential Driving (turns) C-2 Trip #3 - Residential Driving (lane changing and use of one-way streets) C-3 Trip #3 - Residential Driving #1 (lane changing and use of one-way streets) C-4 Trip #5 - Light City Driving #2 #2 C-5 Trip #7 - Highway Driving #3 C-6 Trip #8 - Trailer Trip C-7 Trip #9 - Heavy City Driving #4 C-8 Trip #11 - Expressway Driving #5 C-9 Trip #12 - Review 205 WASHINGTON TRAFFIC SAFETY EDUCATION PROJECT Lesson B-l II. TRIP #1 - STARTING TO DRIVE (Area Adjacent to School) Objectives: A. To become familiar with the driver's compartment of a specific vehicle (gauges, controls, and safety devices). B. To develop the ability to perform basic proce- dures involved in pre-ignition, starting, putting car in motion, steering, stopping, and securing the automobile. C. To experience the feel of moving the automobile forward and backward. Experiences: A. General 1. Pre-entry checks. 2. Explain and discuss driving procedures: pre- ignition, starting, putting car in motion, steering (when moving forward and backward), stopping, and securing the automobile. 3. Hand-over-hand steering. 4. Procedures to follow when changing drivers. 5. Stop signs, stop lines, crosswalk (with sign), crosswalk (no sign). 6. Presence of other vehicles. 7. Left and right turns. 8. Backing from one-way to two-way. 206 Trip #1 - Starting to Drive Page 2 III. B. Specific (may present hazard) l. Entering Logan from student parking lot where visibility is often limited and traffic may be moderately dense. 2. Parked vehicles on narrow streets, and ve- hicles emerging from parking lots and business areas. Route: After having the students inspect conditions around vehicle for obstructions, condition of tires, etc., point out the gauges, controls, and safety devices on the automobile. Leave school via the student parking lot. Turn left onto Logan and proceed north to Tobin. Turn left onto Tobin and proceed to Shattuck. Use the bus loading area, Tobin, Lake, Tillicum and Shattuck as a practice area. Return to the school via Tobin, Logan and the east student parking lot. APPENDIX B Evaluation Instruments Knowledge Driving Performance Traffic Analysis 207 THE CENTER FOR SAFETY SCHOOL OF CONTINUING EDUCATION NEW YORK UNIVERSITY NATIONAL TEST IN DRIVER EDUCATION (Special Form) Fill in the following blank spaces, read the directions for answering the test questions, and then start immedi- ately with the test. You have up to thirty (30) minutes, beginning now. NAME CITY STATE AGE SEX DATE SCHOOLING (indicate grade level completed or nearly com- pleted) HAVE YOU HAD A DRIVER EDUCATION COURSE IN HIGH SCHOOL? HAVE YOU RECEIVED ANY DRIVING INSTRUCTION FROM: Parents Friends Private school DIRECTIONS: Some of the following statements are generally true; others are false or inaccurate. Read each statement carefully. If you think it is TRUE, place an X on the blank line in the TRUE column. If you consider it FALSE, place an X on the blank line under FALSE. Please do ngp ask any ques- tions regarding interpretation of any items. EXAMPLE: TRUE FALSE Drivers under 20 years of age have proportionately fewer accidents than those in their forties. x 208 National Test in Driver Education Page 2 10. 11. 12. TRUE FALSE Most skids are due to circumstances beyond the driver's control. On multi-lane highways, slow drivers should use the extreme right lane. Bicycles should be ridden on the left side of the roadway facing traffic. When a car's speed is doubled, its force of impact is four times as great. It is all right to cross the center line on a curve, provided you can see at least 400 feet ahead. If the right front tire blows out, apply brakes hard and steer to the left. When driving at night in a heavy fog, it is best to use the lower headlight beam. Studies show that alcohol is a factor in about 50 per cent of fatal accidents at night. When driving at 50 mph with a moderate curve ahead, it is best to apply the brakes before entering it. It is not necessary to signal when you wish to make a left turn if you see no car following you. A left turn at an intersection of multiple-lane streets should be made from the lane nearest the center line. Sunglasses are not recommended for night driving glare. 209 National Test in Driver Education Page 3 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. TRUE FALSE If your car is forced off the road to the right on a soft shoulder, apply brakes hard and steer sharply to the left. Most accidents cannot be prevented because "chance" brings together the conditions for a collision. If a car starts to skid, gradually release the pressure on the accelerator. More accidents are due to errors on the part of the driver rather than mechanical defects on the car. The "show-off" driver is one who lacks the skill necessary to handle a car. Because of greater traffic conges— tion, it is more hazardous to drive during the day than at night. When backing a car, you do not have the right of way even though you proceed slowly. Defective headlights contribute to more accidents than defective brakes. When it is necessary for pedestrians to walk on highways, they should always face the oncoming traffic. It is good practice for drivers to focus their eyes immediately in front of the car rather than on objects some distance ahead. Snow tires give less traction on ice than do chains. 210 National Test in Driver Education Page 4 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. TRUE FALSE Pedestrians should be given the right of way even though they are crossing against the lights. A car approaching an intersection should yield the right of way to one that has already entered the intersection from a cross street. Liability insurance protects a driver against damage to his own car. The differential permits the rear wheels to turn at different speeds while the car goes around a corner. Orders by a policeman should be obeyed even if the orders are in conflict with an operating signal light. PeOple with orthopedic disabilities, as a group, have established accept- able driving records. The steering system of most modern automobiles makes use of rods which move the front axle in various directions. The presence of "bluish" smoke from an exhaust pipe is an indication that the level of oil in the cylinder head is low. Diamond-shaped yellow signs with black letters warn of potential danger ahead. Three times the number of fatal traffic accidents occur on rural highways as on urban streets. 211 National Test in Driver Education Page 5 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. TRUE FALSE Every motor vehicle should be safety inspected by a thorough, competent mechanic at least two times each year. The basic purpose of driver licensing is the collection of revenue to build newer and better highways. Visual acuity is a term used to identify the ability of a person to judge the relative distance between two or more objects. If the accelerator sticks to the floor when driving a gearshift transmission car, the first move should be to push the clutch pedal down. When overtaking and passing other vehicles on a limited access high- way, it is permissible to exceed the posted speed limit while in the act of passing. In case of a skid in which the rear of a car is moving to the driver's left, the steering wheel should be turned to the driver's left. Since the parking brake has more holding power than the brake pedal, it should be used when a quick stOp is necessary. The ability to accelerate is less important than the ability to decelerate. Lowering the height of cars im- proves cornering ability. A variation of one degree in head- light adjustment will cause the beam to be five feet out of line at a distance of 300 feet. 212 National Test in Driver Education Page 6 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. TRUE FALSE Nothing can be done about emotions that may interfere with safe and efficient driving. About 25 feet are needed to stOp a car from a speed of 25 mph. Establishment of minimum speed limits is an unfair invasion of human rights. When the foot brake fails, chances are that the parking brake has also failed. An emergency situation exists prior to every collision in traffic. Nondriver pedestrians are involved in proportionately more auto- pedestrian accidents than pedes- trians who possess a driver's license. Although expressways have eliminated cross traffic and have separated opposing lanes of traffic, a lower accident rate has not yet been attained. Vehicle-actuated traffic signals are set to follow a fixed cycle. A flashing red light at an inter- section means the same as a stop sign. Selective enforcement refers to the practice of careful selection of policemen for traffic duty. In case of a traffic accident, the limit of liability is the amount a court decides is ade- quate to compensate for damages. 213 National Test in Driver Education Page 7 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. TRUE FALSE Statistics show that about one half of all accidents occur within 25 miles of drivers' homes. When leaving an expressway, there is no difficulty in adjusting to lower speed requirements on ordinary roadways. Just before entering a two—way inter- section, the driver should look first to the left and then to the right for cross traffic, regardless of whether he is going to make a left turn or a right turn. When parking downhill where there is no curb, the driver should have the front wheels of his car turned to the right. For most pe0p1e, driving three hours in a single stretch presents no special difficulty. Tailgating another vehicle is all right if you know for a fact that you have quick reflexes. It is permissible to use another person's license if you have his authorization in writing. Begin your right or left turn signals at least 100 yards before making the turn. There is no excuse for ignorance of laws and regulations applying to each area in which a person may be driving. The function of a carburetor is to mix gasoline and air as fuel for the engine. 214 National Test in Driver Education Page 8 *65. *66. *67. *68. *69. *70. TRUE FALSE On a two-lane, two-way raodway, it is permissible to pass if there is a yellow line on your side of the road- way, parallel to the center line. The law requires that a driver have his vehicle lights on a half hour after sunset. It is permissible, but not desirable, for drivers to offer rides to hitch- hikers. Parking is prohibited within twenty feet of a crosswalk at an inter- section. A driver must stop when approaching a school bus with flashing red lights and a stop sign showing, unless his path around the bus is unobstructed. Tire failure is a major cause of expressway accidents. *Items applying particularly to the State of Washington 215 HIM lApelirebl. Number .0 Mel. Gina in Pennttuuest Date of Road Test SCORE FAIR IAO Prior to Start (1) .e' III-er poster. 6.. I ulIlO' OI DOORS ADIL 57 IN MONCN Start, Pull Out (11 n..tro 94 Ly. I'cvt ha d>.~.e goo muumu SiGNAl tnmc CHECK neuazulf N01 thustc 10 Backing, 50 Feet (11 1L" " not wen». i 're‘ loot t-u.r ." shave. no.5 YHAN 15 MN wiui um; 'HAN ONE 0001 USES muons ”~11 Turnabout (1) (30 ft. street) ‘vU‘YIle, bug-g r ‘ .f-- 1,. "a: n!.,'ov.| w-th trg't.. lM‘. 'i'PLE‘ “0“ “‘AP‘ "‘PEE ' vEN' (lllfils CU'I SYARYS ON (“7 £11pr . Parallel Park (11 (H: tu'hcl an“ no no' r--te ..~ (the rum. . m.» o in ct in”: ("Id 1 "m. . 1A" Ciwfls C.“ \NDCH ”Cd CV“. 2 AtYEM'Ts MO" WAN e (a 1‘ H.” Al CL“ DISY‘Nrf 1»: ~09: YHAN . v V Shifting Gears . 0“. M. (NARC: IODAAIC .[VL’SE 't vi;'.()~ Use of Accelerator 'C A" Otg F. v.1 ~o Ac E.S‘-A‘E on pp; ”Jute ‘15th ~5Cf$