Citation |

- Permanent Link:
- http://ufdc.ufl.edu/AA00022214/00001
## Material Information- Title:
- The projection of social trends using time series indicators : methodology and application in educational planning
- Creator:
- Nelson, Jane Counihan, 1941-
- Publication Date:
- 1977
- Language:
- English
- Physical Description:
- xiii, 124 leaves : ill. ; 28 cm.
## Subjects- Subjects / Keywords:
- Analytical forecasting ( jstor )
Educational research ( jstor ) Forecasting techniques ( jstor ) Linear regression ( jstor ) Polynomials ( jstor ) Social research ( jstor ) Statistical forecasts ( jstor ) Statistics ( jstor ) Time series ( jstor ) Time series forecasting ( jstor ) Dissertations, Academic -- Educational Administration and Supervision -- UF ( lcsh ) Educational Administration and Supervision thesis Ph. D ( lcsh ) Educational planning -- Mathematical models ( lcsh ) Social indicators ( lcsh ) Time-series analysis ( lcsh ) - Genre:
- bibliography ( marcgt )
non-fiction ( marcgt )
## Notes- Thesis:
- Thesis--University of Florida.
- Bibliography:
- Bibliography: leaves 117-123.
- General Note:
- Typescript.
- General Note:
- Vita.
- Statement of Responsibility:
- by Jane Counihan Nelson.
## Record Information- Source Institution:
- University of Florida
- Holding Location:
- University of Florida
- Rights Management:
- Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
- Resource Identifier:
- 022249428 ( ALEPH )
04290057 ( OCLC )
## UFDC Membership |

Downloads |

## This item has the following downloads: |

Full Text |

TIlE PROJECTION OF SOCIAL TRENDS IISJNG TIME SERIES INDICATORS: METHIOIDO LOGY AND APPLE ICATION IN EDUCATIONAL PLANNING By JANE COUNIttAN NELSON A DISSERTATION PRESENTED) TO TIHE GRADUATE COUNCIL OF TIlE I UNIVERSITY (F FLORIDPA IN PARTIAL, FIULFII,LIENT OF TIlE RVI RIlrNTS FOR TIlE l)EGREE OF DOCTOR OF PHILOSOPY UNI VERSITY OF FLORIDlA 1977 Copyright 1977 by Jane Counihan Nelson ACKNOWLEDGMENTS I wish to express my sincere appreciation to my Supervisory Committee: to Dr. Michael Y. Nunnery, chairman, for his candor and understanding whether acting as counselor or critic; to Dr. Phillip A. Clark and Dr. Gordon D. Lawrence, for providing valuable suggestions and support. A special thank you goes to Dr. Arthur J. Lewis, director of the DOE social forecasting project, for his leadership, receptivity to new ideas, and especially for his confidence in me; and to my colleagues on the project, Dr. Robert S. Soar and Hs. Linda Troup, for their substantial contribution to the conceptual framework presented in this study. For carefully reviewing the statistical portions of this manuscript, I am indebted to my friend, Dr. Azza S. Guertin. I am especially grateful to my husband Edward for his love and encouragement and for sharing with me hi enthu- siasm for scientific inquiry. iii TABLE OF CONTENTS ACKNOWLEDGIENTS .... LIST OF TABLES ....... .................. LIST OF FIGURES ....... .................. ABSTRACT ......... .................... CHAPTER I INTRODUCTION .... ............... Page . iii . vii . ix * x 1 Background and Significance of the Study ... ...... 1 The Social Context of Education .... ......... 1 The Futures Perspective in Educational Planning ......... ................... 4 Forecasting Trends in Social Variables ...... 5 The Need for Research ....... .............. 7 The Problem ......... .................... 9 Delimitations and Limitations ..... ........... 9 Definition of Terms ...... ................ 11 Procedures ...................................... 13 The Selection and Operational Definition of Variables ...... ................. .. 13 Collection of Time Series Indicator Data .... 14 Comparison of Extrapolative Methods Using Time Series Indicators .... ............ 15 Development of Implications for Educational Planning ....... ................... .. 18 CHAPTER II RATIONALE FOR SELECTION OF VARIABLES/ TIME SERIES INDICATORS ... .......... The Social Indicator Movement .. ........... Historical Development .... ............ Definition and Use of Social Indicators . Data Base for Social Indicators ........... Educational Implications ... ........... Selection of Variables/Time Series Indicators . The Variables ....... ................. Bronfenbrenner's Ecology of Education Model. Operational Definition of Variables as Time Series Indicators ...... ............... CHAPTER III RATIONALE FOR EXTRAPOLATIVE METHODS SELECTED FOR COMPARISON ... ........ * 19 20 * 20 24 * 26 27 . 28 28 29 * 31 * 38 TABLE OF CONTENTS (continued) Page Overview of Extrapolative Forecasting Methods 38 Economic and Business Forecasting .......... ..39 Technological Forecasting .... ............ 41 Educational Forecasting .... ............. ...42 Extrapolative Methods in Other Areas ....... ..43 Applicability of Reviewed Extrapolative Methods for Study ........ .................... .. 44 The Pattern of the Data ..... ............ 44 The Class of Model ....................... ..44 Description of Methods to be Compared ........ ..46 The General Linear Model .... ............ 46 The Assumptions of the Linear Model ......... ...48 Criteria for Comparison of Methods ........ ..49 Method 1: Simple Linear Regression ......... ...52 Method 2: Log-linear Regression .. ........ ..52 Method 3: Polynomial Regression .. ........ ..54 CHAPTER IV COMPARISON OF EXTRAPOLATIVE METHODS USING SELECTED TIME SERIES INDICATORS ......... ...57 Presentation of Results ..... .............. 60 Indicator 1 ....... ................... 60 Indicator 2 ......... .... ............. ...62 Indicator 3 ....... ................... .. 69 Indicator 4 ....... ................... .. 73 Indicator 5 ....... ................... .. 77 Indicator 6 ....... ................... .. 79 Indicator 7 ....... ................... .. 86 Indicator 8 ....... ................... 90 CHAPTER V DISCUSSION ....... .................. ..95 The Variables ....... ................... 95 Selection ........ .................... 95 Bronfenbrenner's Ecology of Education Model. 96 Operational Definition of Variables ......... ...97 The Extrapolative Methods .... ............ .. 98 Statistical Considerations ... ........... 98 Practical Considerations ..... ............ .101 CHAPTER VI SUMMARY, CONCLUSIONS, AND IMPLICATIONS OF STUDY ....... .................. 102 Summary ......... ...................... .102 The Variables ........ .................. ..103 The Methods ........ ................... .103 Results ......... ..................... .104 Conclusions ........ .................... .104 Suggestions for Future Research .... .......... .105 Implications for Planners and Policy Makers 106 TABLE OF CONTENTS (continued) Page APPENDIX .......... ........................ .109 REFERENCE NOTES ......... ..................... .117 REFERENCES ......... ....................... .118 BIOGRAPHICAL SKETCH ........ ................... ..124 LIST OF TABLES Page Table 1 Time Series Indicators of Social Variables Affecting Outcomes of Education ......... Table 2 Indicator 1: Summary Statistics for Prediction Equations by Method ........ Table 3 Indicator 1: Observed Y's and Predicted Y's by Method ...... ................ Table 4 Indicator 2: Summary Statistics for Prediction Equations by Method ........ Table 5 Indicator 2: Observed Y's and Predicted Y's by Method ...... ................ Table 6 Indicator 3: Summary Statistics for Prediction Equations by method. ......... Table 7 Indicator 3: Observed Y's and Predicted Y's by Method ...... ................ Table 8 Indicator 4: Summary Statistics for Prediction Equations by Method ........ Table 9 Indicator 4: Observed Y's and Predicted Y's by Method ...... ................ Table 10 Indicator 5: Summary Statistics for Prediction Equations by Method ........ 33 * 61 63 66 * 67 * 70 * 71 * 74 75 * 78 Table 11 Indicator 5: Observed Y's and Predicted Y's by Method ..... ............... Table 12 Indicator 6: Summary Statistics for Prediction Equations by Method ....... Table 13 Indicator 6: Observed Y's and Predicted Y's by Method ..... ............... Table 14 Indicator 7: Summary Statistics for Prediction Equations by Method ....... Table 15 Indicator 7: Observed Y's and Predicted Y's by Method ..... ............... . 80 * 83 S. 84 . 87 . 88 vii LIST OF TABLES (continued) Indicator 8: Summary Statistics for Prediction Equations by Method ........ Indicator 8: Observed Y's and Predicted Y's by Method ...... ................ Indicator by Method.. Table 19 Indicator 2: by Method. . Table 20 Indicator 3: by Method. . Table 21 Indicator 4: by Method. . Table 22 Indicator 5: by Method. . Table 23 Indicator 6: by Method. . Table 24 Indicator 7: by Method. . Table 25 Indicator 8: by Method. . Table 16 Table 17 Table 18 Page * 91 * 92 . 109 . 110 . 112 . 113 . 114 . 115 . 116 viii ANOVA Summary Tables ANOVA Summary Tables ANOVA Summary Tables ANOVA Summary Tables ANOVA Summary Tables ANOVA Summary Tables ANOVA Summary Tables ANOVA Summary Tables LIST OF FIGURES Page Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Bronfenbrenner's ecological structure of the educational environment. (Based upon Bronfenbrenner's [1976] description, pp. 5-6) ....... .................. Indicator 1: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extra- polation) ....... .................. Indicator 2: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extra- polation) ....... .................. Indicator 3: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extra- polation) ....... .................. Indicator 4: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extra- polation) ....... .................. Indicator 5: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extra- polation) ....... .................. Indicator 6: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extra- polation) ....... .................. Indicator 7: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extra- polation) ....... .................. Indicator 8: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extra- polation) ....... .................. * 64 68 * 72 * 76 81 * 85 * 89 93 Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE PROJECTION OF SOCIAL TRENDS USING TIME SERIES INDICATORS: METHODOLOGY AND APPLICATION IN EDUCATIONAL PLANNING By Jane Counihan Nelson December 1977 Chairman: Michael Y. Nunnery Major Department: Educational Administration Educational planners and policy makers need adequate information about the societal context of education to make appropriate decisions about the future role and function of education. Some of this information may be provided through the use of conceptually sound social and educational vari- ables operationally defined as time series indicators coupled with an empirically sound basis for forecasting future trends in such indicators. As evidence of the need for developing such a social forecasting framework for education, states including Florida have provided grants for that purpose. This study was one aspect of such a grant. The problem in this study was (a) to select, using Bronfenbrenner's ecology of education model, and operation- ally define at least 10 variables that research has shown to be related to the outcomes of education; (b) to use these variables operationally defined as time series indicators in the comparison of three purely extrapolative forecasting methods; and (c) to derive implications for the use of an ecological model such as Bronfenbrenner's, time series indicators, and selected extrapolative methods for educa- tional planning. The study was conducted in the following phases: 1. Using Bronfenbrenner's ecology of education model, 10 variables that research has shown to be related to the outcomes of education were selected and where possible, were operationally defined as state and/or national time series indicators. Data were collected for these indica- tors; eight which met the criteria established in this study were used in the comparison of extrapolative techniques. 2. Three purely extrapolative techniques derived from the general linear model were compared according to statis- tical criteria and practical considerations derived from the literature in statistics, economics, time series analysis, and forecasting methodology. The methods were (a) linear regression, (b) curvilinear regression (quad- ratic and cubic forms), and (c) log-linear regression (de- pendent variable undergoes logarithmic transformation). Each method was applied to each time series indicator. Time in years was used as the independent variable; the annual measure of the indicator was treated as the dependent vari- able. Each data set was divided into thirds; two-thirds of the data points were used to establish the prediction equation. This equation was used to predict the remaining third of the data points. Predicted values were compared with actual values. 3. Implications for the use in educational planning of an ecological model such as Bronfenbrenner's, time series indicators, and selected extrapolative techniques were dis- cussed. Results of the method comparison were (a) no method was a superior predictor for all indicators; (b) each method was a superior predictor for at least one indicator; and (c) the summary statistics for the original regression were not consistently related to the accuracy of the extra- polated values. The following conclusions appear to be warranted by the results of this study: 1. The Bronfenbrenner model is a useful framework for considering the numerous factors impinging upon the learner. 2. Time series indicators provide a means to compare trends in an indicator over time or to compare different groups in relation to a specific indicator. 3. The general linear model is appropriate for the analysis and extrapolation of the selected time series indicators used in this study. 4. Each method is appropriate for use with some indicators but not with others. Measures of "best fit" such as r2 and the standard error of estimate are not reliable criteria for the selection of an extrapolative method. A combination of strategies such as graphic representation of original and predicted data, analysis of residuals, and xii knowledge of the social phenomena being studied may provide guidance as to the most appropriate method for a particular indicator. xii i CHAPTER I INTRODUCTION Background and Significance of the Study The Social Context of Education Educators have become increasingly cognizant of the myriad forces in society impinging upon various facets of the educational process. The influence of a number of these forces upon educational purposes, outcomes, and resources has been analyzed from several social science perspectives (Boocock, 1976; Henry, 1961; Gordon, 1974). Keppel (in Thomas & Larson, 1976) acknowledged one of the reasons for this continuing interest in societal trends by educational planners and policy-makers: the impetus for change in educational institutions, from the preschool through the university, is more likely to derive from changes in the wider society than from forces within the institutions. (Foreword) Additionally, Keppel noted that "educational policy must be formed in concert with other aspects of public policy and program development" (Foreword). Bronfenbrenner (1976) proposed an ecological structure of the educational environment which must be taken into account if "any progress in the scientific study of educa- tional systems and processes" (p. 5) is to be made. Bronfen- brenner stated: Whether and how people learn is a function of sets of forces, or systems, at two levels: a. The first comprises the relations between characteristics of learners and their sur- roundings in which they live out their lives (e.g., home, school, peer group, work place, neighborhood, community). b. The second encompasses the relations and interconnections that exist between these environments. (p. 5) Building on Lewin's theory of topological territories and employing a terminology adapted from Brim (1975), Bronfen- brenner further elaborated that the construct environment can be "conceived topologically as a nested arrangement of structures, each contained within the next" (p. 5). 1) A micro-system is an immediate setting containing the learner .. 2) The meso-system comprises the inter- relationships among the major settings containing the learner at a particular point in his or her life a system of micro-systems. 3) The exo-system is an extension of the meso-system embracing the concrete social structures, both formal and informal, that impinge upon or encompass the immediate settings containing the learner and, there- by, influence and even determine or delimit what goes on there. These structures in- clude the major institutions of society, both deliberately structured and spontane- ously evolving, as they operate at the local community level ... 4) Macro-systems are the overarching insti- tutions of the culture or subculture, such as the economic, social, educational, legal and political systems, of which local micro-, meso-, and exo-systems are the concrete manifestations. (pp. 5-6) (See Figure 1 for a representation of these ideas.) Figure 1. Bronfenbrenner's ecological structure of the educational environment. (Based upon Bronfen- brenner's [19761 description, pp. 5-6.) The Futures* Perspective in Educational Planning While Bronfenbrenner proposed his ecological structure primarily as a framework for learning research efforts, that is, for examining relationships among variables associated with learning, others (Harman, 1976; Webster, 1976) have dis- cussed the societal context of education as a framework for future-oriented educational planning. Indeed, this emphasis on future awareness has evolved into a significant movement within education referred to as educational futurism (Hencley & Yates, 1974; Pulliam & Bowman, 1974), educational futures (Marien & Ziegler, 1972), or alternative futures perspective (Webster, 1976). The primary purpose of futures research or future studies is "to help policy makers choose wisely--in terms of their purposes and values--among alternative courses of action that are open to leadership at a given time" (Shane, 1973, p. 1). According to Webster (1976), this requires that we attend to alternatives--to alternative assumptions, ends and means. It requires us to examine alternative plausible futures that might be rendered more or less possible by our planning and action; to identify un- intended as well as intended consequences for others of achieving the goals that seem desirable to us; to analyze alternative stra- tegies and tactics for achieving any desired future; and to anticipate the variety of potential consequences of our strategies, tactics, and short-run planning. Perhaps, most fundamentally it asks of us that we look hard at our basic premises about the nature of man and the world and consider implications and alternatives for the future. (p. 2) *"Futures" refers to the number of different possible views of what is ahead in subsequent time periods for society and, thus, for education. S Webster also noted: the futures perspective implies that we not just attend to alternatives in and for educa- tion, but also consider the societal context in more comprehensive fashion than is usual in educational planning. (p. 2) In order to assist decision makers in the selection of alternatives which have positive future consequences for society, educational planners at both national and state levels must take into account those societal forces which affect not only the outcomes of education, but also the pur- poses of education, and the human and material resources available to the educational process. To do this, however, the planner must delineate the societal factors or variables to be included in the planning process and develop a sound rationale based on research and theory for such inclusion. Then trends--past, present, and future--in these variables may be examined in order to derive implications for educa- tional planning and policy. Forecasting Trends in Social Variables Available to the educational planner in this undertaking are a number of predictive and heuristic devices to explore alternative futures which have been developed by government, industry, non-profit organizations, and futures consulting groups. These forecasting techniques can be categorized into exploratory forecasting methods and normative forecasting methods: Exploratory forecasting methods start from the present situation and its preceding history, and attempt to project future developments. Normative forecasts, on the contrary, start with some desired or pos- tulated future situation, and work back- wards to derive feasible routes for the transition from the present to the desired future. (Martino, 1976, p. 4) Exploratory forecasting methods, all of which are based upon extrapolation of some kind, include (a) purely extra- polative methods, (b) explanatory methods, and (c) auxiliary methods. Since forecasting of social phenomena is still in a highly intuitive developmental phase, there is a growing interest in examining those exploratory methods considered to be purely extrapolative, which are based upon time series data representing social and educational variables. These time series data, often called time series indicators, are defined measurements made at specified intervals over a period of time. By extrapolating identified patterns in the time series data into the future, planners may compare present, past, and future states of that indicator. Thus, a projec- tion of future societal trends can provide the impetus to examine present policy and to analyze the consequences of contemplated changes. This approach need not be only "pre- ventive" forecasting, in the sense used by Ziegler (1972) of preventing undesirable forecasts. It may also be extended to examine all consequences of action or intervention, in- tended or not. Purely extrapolative methods, when combined with auxiliary methods such as trend-impact analysis, cross- impact matrices, or scenarios, can provide a vehicle for ex- ploring the relationships among identified future patterns in society. While the use of purely extrapolative methods with time series data is fairly well defined in technological and economic areas, their application to social forecasting has not been the focus of significant definitive study. Indeed Harrison (1976) emphasized the need for such research, speci- fically the consideration of "each method in terms of some aspect of the social process it would likely be applied to" (p. 13). For, as Harrison explained, while some problems in regression and time series analysis which remain unresolved are currently the concern of statisticians and mathematicians, "it appears that resolution might best lie in terms of inves- tigation in concrete application cases" (p. 14). In social forecasting there is a great need in almost all the known extrapolative methods for an explicit statement of the algorithmic, theoretical, and empirical weaknesses or sensitivities of such procedures. Such a discussion, as noted, would be more mean- ingful if carried on in the context of an analysis of some specific aspect or aspects of social process. (Harrison, 1976, p. 17) Only through empirical study of the performance of various extrapolative methods applied to particular social phenomena will a basis for selection of appropriate and accurate tech- niques be formulated. The Need for Research Since there are no widely-accepted planning models in- corporating quantitative data on social variables, the edu- cational planner who wants to utilize such information is confronted with a number of questions related to (a) the identification of social variables to be included, (b) the operational definition of social variables in terms of time series indicators, (c) the selection of a purely extrapolative technique which will yield the most accurate forecast for a specific indicator, and (d) the utilization of these forecasts in the planning process. Answers require futures research which is derived from a conceptually sound framework and is pursued with methodological vigor. As evidence of the impor- tance of such investigation to the educational planner, the State of Florida through the Office of Strategy Planning in the Department of Education funded in 1976 a social fore- casting project (STAR Project No. R5-175) at the University of Florida for the second year. The study described herein was part of that effort to forecast social trends affecting education in Florida. To summarize: Educational planners and policy makers need adequate information to make appropriate decisions about the role and function of education in creating improved quality of life for citizens of the future. The State of Florida, in funding STAR Project No. R5-175 of which this study is a part, acknowledged that need. Part of this in- formation may be provided through the use of conceptually sound social and educational variables operationally defined as time series indicators coupled with an empirically sound basis for forecasting future states of such indicators. The Problem The problem in this study was (a) to select, using Bronfenbrenner's ecology of education model, and operationally define at least 10 variables that research has shown to be related to the outcomes of education; (b) to use these variables operationally defined as time series indicators in the comparison of three purely extrapolative forecasting methods; and (c) to derive implications for the use of an eco- logical model such as Bronfenbrenner's, time series indicators, and selected extrapolative methods for educational planning. Delimitations and Limitations The Bronfenbrenner ecology of education model was used primarily as a framework for the selection of social and educational variables and was not evaluated itself in this study. Ten variables (e.g., socio-economic status of family, peer group characteristics) were selected to be operationally defined, where possible, in terms of national and/or state level time series indicators (e.g., median family income, juvenile crime rates). Of these identified indicators, eight which met the following criteria were used in the comparison of extrapolative techniques: (a) the indicator was readily available, (b) the data were available for a 10 year or greater time span, and (c) the indicator was a reasonably reliable and valid measure of one aspect of the social or educational variable that it represented. It should be noted that the selection of the eight indicators used in this study was in many cases influenced more by data availability than the logic or appropriateness of the indicator to represent a specific social variable. Thus, the eight indicators are examples of the type of data that might be employed to operationally define the variables; utilization in a specific planning situation would require evaluation of the appropri- ateness of the indicators presented in this study and the addition and/or substitution of other indicators. In this study only the variables related to the outcomes of education were used. As previously noted, this study was part of a larger social forecasting and educational planning effort which also included the status of education 1976-77, social trends affecting the purposes of education, and social trends affecting the resources for education. While the literature in mathematics, statistics, and economics was reviewed and considered in preparation for the selection and use of the three extrapolative techniques (linear, log-linear, and curvilinear regression), there was no attempt to present the comparison of these techniques in the detail desired by these disciplines. Rather the compari- son was made in such a way as to be most relevant to the planner in education. There was no attempt to write or adapt computer programs for various techniques. Instead, an effort was made to identify and utilize computer programs and statistical pack- ages which had already been adapted for use at the North East Regional Data Center's computer facilities. Additionally, the projection of specific trends per se was not of interest in this study. Rather the focus of this study was the development of the conceptual framework and methodology for such projection. Also, there has not been any attempt to forecast educational outcomes from the operationally defined social and educational variables. The present work may be considered an initial step in determining the feasibility of developing such a mathematical forecasting model. Definition of Terms Extrapolative forecasting. The procedure consists of identifying an underlying historical trend or cycle in social processes that can be extrapo- lated by means as varied as multiple regression analysis, time series analy- sis, envelope curve fitting, three-mode factor analysis, correlational analysis, averages, or any other method that takes current and historical data as the prin- cipal basis for estimating future states in a given variable. (Harrison, 1976, p. 3) Indicator, educational. Educational indicators are statistics that enable interested publics to know the status of education at a particular moment in time with respect to some selected variables, to make comparisons in that status over time and to project future status. Indicators are time-series statistics that permit a study of trends and change in education. (Gooler, 1976, p. 11) Indicator, social. "The operational definition or part of the operational definition of any one of the concepts central to the generation of an information system descrip- tive of the social system" (Carlisle, 1972, p. 25); "time- series that allow comparisons over an extended period which permit one to grasp long-term trends as well as unusually sharp fluctuations" (Sheldon & Freeman, 1970, p. 97); "a statistic of direct normative interest which facilitates concise, comprehensive and balanced judgments about the con- dition of major aspects of a society" (U.S. Department of Health, Education,& Welfare, 1970, p. 97). Outcomes of education. Those measures of performance, such as achievement test scores, or utilization, such as employment rates, which appear to be the result of partici- pation in the formal educational process. Regression, linear. Most common type of regression in which the objective is to locate the best-fitting straight line through a scattergram based on interval-level variables (Nie, Hull, Jenkins, Steinbrenner, & Bent, 1975, p. 278). Regression, log-linear. As used in this study, a least squares regression method in which a geometric straight line is located through a scattergram plotted on semi-logarithmic paper; also called exponential curve or trend curve. Regression, polynomial or curvilinear. Regression method for fitting a curve to a set of data using the cri- terion of least squares distances (Nie et al., 1975, p. 278). Time series. "A set of observations generated sequen- tially in time" (Box & Jenkins, 1970, p. 23). Procedures The study proceeded in the following phases: (a) using Bronfenbrenner's ecology of education model, 10 variables that research has shown to be related to the outcomes of education were selected and, where possible, were operational- ly defined as time series indicators; (b) data were collected for these time series indicators, eight of which were used in the comparison of the selected extrapolative techniques; (c) using the selected time series indicators, three purely extrapolative techniques were compared according to statis- tical criteria and practical considerations derived from the literature; and (d) implications for the use in educa- tional planning of an ecological model such as Bronfen- brenner's, time series social and educational indicators, and selected extrapolative techniques were derived. The Selection and Operational Definition of Variables The work by Collazo, Lewis, and Thomas (1977), completed during the first year of STAR Project No. R5-175, on fore- casting selected educational outcomes from social variables was utilized. Since the variables selected by these inves- tigators were derived from a review of the research litera- ture and were acknowledged to be appropriate for the stated social forecasting purposes by a panel of experts in various disciplines, they appeared to fulfill the requirements of this study. Additionally, each of the 10 variables selected for use was described and classified according to Bronfen- brenner's ecology of education model. For each variable an attempt was made to identify one or more types of time series indicators which might logically represent the variable. For some variables several indica- tors were identified, while for others, no indicator could logically be identified or no time series data were avail- able for the indicator at the time of the study. This phase of the study is explained further in Chapter II. Collection of Time Series Indicator Data Sources of needed time series data at both the national and state level'were identified in several ways. The expan- ding literature on social trends (e.g.,U.S. Department of Health, Education, & Welfare, 1970) and specifically the literature on these social trends operationalized as social indicators (e.g., Executive Office of the President, Office of Management & Budget, 1973) was reviewed. Furthermore, examination of initial efforts in using time series indica- tors related to education by the Office of Technology Assess- ment for the United States Congress (Coates, Note 1) and several state departments of education (e.g., Oregon, Penn- sylvania, & Florida) yielded additional sources. Published sources of data such as U.S. Census Reports and Florida Statistical Abstracts were consulted. When data did not appear to be available in suitable form or for desired time periods, inquiries and requests were directed to appropriate sources. Any apparent limitations in the data such as known measurement error due to sampling technique were noted. After data collection was completed, eight indicators which met the criteria outlined in a previous section were selected for inclusion in the next phase of the study. Comparison of Extrapolative Methods Using Time Series Indicators The following steps were involved in this phase of the study: (a) initial identification and testing of methods using data similar in form to selected indicators, (b) recon- sideration and testing of additional available methods, (c) selection of three methods to be used for comparative extra- polations, (d) derivation of specific criteria and practical considerations from the literature, (e) application of three methods to each data set, (f) extrapolation of identified trend into future using equation generated in (e), and (g) comparison of actual versus predicted values of indicators. From a preliminary review of the literature in statis- tics, economics, time series analysis, and forecasting methodology, the following four methods were tentatively identified for comparison: (a) linear regression (computer program by Nie et al., 1975), (b) curvilinear or polynomial regression (computer program by Nie et al., 1975), (c) Box- Jenkins time series analysis (computer program by Cooper, Note 2), and (d) FIT curve-fitting with weighted data (com- puter program by Stover, Note 3). An initial analysis of the methods using trial sets of data combined with a visual analysis of the general form of the data to be used revealed that two of the methods under consideration were inappropriate. The Box-Jenkins procedure, while an extremely powerful tool for time series analysis of data which are characterized by seasonal or cyclic variation (usually resulting in autocorrelation of observations and residuals), did not seem suitable for the social indicator data collected. (Should subsequent tests reveal autocorrela- tion and hence a violation of the assumptions of the linear model, Box-Jenkins could then be appropriately employed.) The FIT curve-fitting procedure utilizing a weighted data principle was rejected because the computer program required extensive modification to yield necessary comparative statis- tics and reliable output. Theoretical justification for the weighting formula and data transformations employed was unavailable. Thus, two of the four methods tentatively considered were rejected. Since the comparison phase was to involve three methods, the literature was again searched for other appropriate methods. The most promising of these was a curve fitting technique which utilizes an exponential function to describe a constant growth rate. This method, called log- linear regression in this study, can be described in terms of the general linear model and solved by least squares pro- cedures when the dependent variable undergoes logarithmic transformation. Since social phenomena sometimes exhibit what appears to be a constant growth rate, log-linear regres- sion seemed to be an appropriate method to include in this study. The three methods finally selected for comparison were (a) linear regression (without data transformation), (b) curvilinear or polynomial regression, and (c) log-linear regression. The mathematical properties of each are pre- sented in Chapter III. All three approaches to trend extra- polation were executed by using variations of SPSS subpro- grams SCATTERGRA4 and REGRESSION and that system's data transformation capabilities (Nie et al., 1975). Each of the three methods was applied to each of the eight selected time series indicators. Time in years was used as the independent variable; the annual measure or index of the indicator was treated as the dependent or response variable. Each data set was divided into thirds; two-thirds of the data points were used to establish the prediction equation. This equation was then used to predict the remaining third of the data points. Predicted values were then compared with actual values. Thus, in this phase of the study three prediction equations (one for each method) were generated for each of the eight time series indicators. Statistical criteria de- rived from the literature were used to evaluate the "good- ness of fit" of the regression line derived from the pre- diction equation to the data. The distribution of error (residuals) about the regression line was also examined to determine if the data satisfied the assumptions of the sta- tistical model. Results of the method comparison phase are reported in Chapter IV. Development of Implications for Educational Planning In Chapter V methodological strategies involved in the selection and operational definition of variables are analyzed in terms of viability for future use. Results of the technique comparison phase are analyzed according to the statistical criteria and practical considerations derived from the literature in forecasting methodology and statis- tics. In Chapter VI a summary of the study and conclusions warranted by the results of the study are presented. Future directions for research suggested by the results of this study are discussed. Additionally, implications for the use in educational planning of an ecological model such as Bronfenbrenner's, time series social and educational in- dicators, and selected extrapolative methods are discussed. CHAPTER II RATIONALE FOR SELECTION OF VARIABLES/ TIME SERIES INDICATORS In the previous chapter, the need for educational planners and policy makers to have an awareness of the societal context of education was emphasized. To this end the Bronfenbrenner ecology of education model was proposed as a framework for the selection of social variables which affect the outcomes of the educational process. The selected social variables may then be operationalized as time series indicators; trends in these indicators can be identified and extrapolated into the future. Such information might then be incorporated into a planning model in order to assist planners and policy makers in making informed decisions about the role and function of education in the future. In order to place the use of time series indicators described in this study into perspective, in the first section of the present chapter social indicators are dis- cussed in relation to their historical development, defini- tion and use, and data base. Educational applications of indicators are briefly noted. In the second section the social variables selected for use in this study are presented in relation to the Bronfenbrenner model. These variables are then operationally defined as time series indicators, and the eight indicators selected for use in the comparison of the three extrapolative methods are listed. The Social Indicator Movement Historical Development Interest in societal trends by policy planners is not of recent origin in the United States. Indeed, in 1933 a presidential task force reported on social trends in a com- prehensive work documenting social change in the United States (President's Research Committee on Social Trends, 1933). The development of indicators, or measures, of social change, however, did not receive the sustained governmental support that was provided for indicators of the economic process. Thus, while the development of economic statistics during the 1930's and 1940's provided "a solid basis for economic analysis and economic reporting which eventually resulted in the establishment of the Council of Economic Advisors and the Economic Report" (U.S. Department of Health, Education, & Welfare, 1970, p. v), comparable development of social in- dicators was not undertaken. In the 1960's a renewed interest in statistics describing the social condition became apparent. Impetus for the de- velopment of social indicators was provided by social scientists in various disciplines, government policy makers, and business leaders in the private sector (Brooks, 1972, p. 1). While this early effort was not well defined as to membership, organization, or objectives, the participants in the social indicator movement "sensed great needs and oppor- tunities for change, [and] celebrated shared but necessarily ambiguous symbols" (Sheldon & Parke, 1975, p. 693). The following examples were drawn from the many mani- festations of interest in the development of social indica- tors during the period from 1965 through 1975 (over 1000 items were listed in a bibliography issued in late 1972 by Wilcox, Brooks, Beal, & Klongian): 1. The Russell Sage Foundation commissioned in 1965 and published in 1968 an independent study, Indicators of Social Change: Concepts and Measurements, on a number of aspects of structural change in society (Sheldon & Moore, 1968). 2. A study on ways to measure the impact of massive scientific and technological change on society (Bauer, 1966) was prepared by the American Academy of Arts and Sciences for the National Aeronautics and Space Administration. This work, Social Indicators, was an overview of the task of developing indicators as part of a feedback mechanism docu- menting social change. 3. President Johnson in March of 1966 directed the Secretary of Health, Education and Welfare "to develop the necessary social statistics and indicators to supplement those prepared by the Bureau of Labor Statistics and the Council of Economic Advisors" (U.S. Department of Health, Education, & Welfare, 1970, p. iii). The result of this directive, Toward a Social Report issued in 1969, was considered "a preliminary step toward the evolution of a regular system of social reporting" (U.S. Department of Health, Education, & Welfare, 1970, p. iii). 4. The Social Science Research Council established in 1972 the Center for Coordination of Research on Social Indicators, whose objective is "to enhance the contribution of social science research to the development of a broad range of indicators of social change" (World Future Society, 1977, p. 97). 5. The appearance of the landmark government publica- tion, Social Indicators 1973 (Executive Office of the Presi- dent: Office of Management & Budget, 1973),was heralded as a significant attempt to provide a collection of social sta- tistics describing quality of life in the United States. This work, which was scheduled to be up-dated every three years, was a compilation of statistics on eight major areas of social interest: health, public safety, education, em- ployment, income, housing, leisure and recreation, and popu- lation. (It may be noted that it is often impossible to strictly categorize what is social and what is economic, as almost all aspects of life are the result of interaction be- tween social and economic forces.) The idea of systematic collection and use of social indicators has not always mct a favorable reception. Bezold (Note 4) and Shostak (Note 5) documented the unsuccessful efforts, beginning in 1967, by Walter F. Mondale to earn congressional approval of a far-reaching plan for new government use of the applied social sciences. Mondale's blueprint for better collection and use of social intelli- gence involved two statutorily-mandated additions to the Executive Office of the President: (a) a Council of Social Advisors (CSA) comparable to the Council of Economic Advisors established in 1946, and (b) an annual Social Report of the President prepared by the CSA to parallel the annual Economic Report to the President. While aspects of Mondale's plan may be satisfied by such efforts as Social Indicators 1973 and several congressional provisions which required the develop- ment and application of social science techniques to the study of present and future national problems (Shostak, Note 5), the comprehensive nature of the Mondale plan is absent. The future of such governmental efforts at social accounting was in 1977 uncertain. Interest in social indicators has not been confined to the United States (Johnson, 1975; Sheldon & Parke, 1975). In 1973, for example, the Organization for Economic Coopera- tion and Development (OECD) to which the United States also belongs issued a list of social concerns shared by many member countries. The identification of concerns was a first step in the development of a set of social indicators designed explicitly to reveal, with validity, the level of well-being for each social concern in the list and to monitor changes in those levels over time. (OECD, 1973, p. 4) Additionally, international organizations such as the Conference of European Statisticians, the United Nations Research Institute for Social Development, and the United Nations Educational, Scientific,and Cultural Organization have been actively concerned with social indicators (Sheldon & Parke, 1975). Efforts to develop social indicators have been initiated in countries such as France, Great Britain, West Germany, Canada, Japan, Norway, Sweden, and Denmark (Brooks, 1972; Johnson, 1975). Definition and Use of Social Indicators The social indicator movement has been characterized by ambiguity of definition and purpose due, in part, to the heterogeneous nature of participants with their own back- grounds, skills, and interests, and also, to the necessary stages of evolution that such a movement experiences. These problems in definition and purpose of social indicators have been discussed by a number of critics (Land, 1971; Little, 1975; Plessas & Fein, 1972; Sheldon & Freeman, 1970; Sheldon & Land, 1972). Attempts have been made to resolve a number of these problems. Land (1971), for example, proposed the following social science-oriented definition of social indicators: social indicators refer to social sta- tistics that (I) are components in a social system model (including sociopsy- chological, economic, demographic, and ecological) or of some particular segment or process thereof, (2) can be collected and analyzed at various times and accumu- lated into a time series, and (3) can be aggregated or disaggregated to levels appropriate to the specifications of the model. .*. The important point is that the criterion for classifying a social statistic as a social indicator is its informative value which derives from its empirically verified nexus in a con- ceptualization of a social process. (p. 323) Part of the confusion over definition is the result of disagreement over purposes, or uses, of social indicators. These purposes, or uses, have been considered under a number of overlapping, sometimes synonomous, headings: (a) descrip- tions reporting, (b) policy planning, (c) social accounting, (d) program evaluation, (e) social modeling, (f) social fore- casting, and (g) social engineering. While the ultimate ob- jective of guiding social policy is rarely disputed, the form of this guidance is still debated. Social scientists are more likely to be concerned with the analysis and pre- diction of social change, while public administrators and legislators are often more concerned with uses of indicators related to public program evaluation and agency goal setting. Sheldon and Parke (1975) in acknowledging these concerns, said: It is apparent that many different types of work go on under the rubric of social indicators. What is impor- tant is that the field be seen as an arena for long-term development, as an effort of social scientists to push foreward developments in concepts and in methodology that promise payoffs to both science and public policy. (p. 698) To underscore this point, Sheldon and Parke (1975) selected an observation by Duncan: The value of improved measures of social change. is not that they necessarily resolve theoretical issues concerning social dynamics or settle pragmatic issues of social policy, but that they may permit those issues to be argued more productively. (p. 698) Data Base for Social Indicators Various efforts have been undertaken to improve the data base for social indicators. Among the efforts in the early 1970's were basic surveys on crime and education as well as replications of previous social science studies and surveys (Sheldon & Parke, 1975). Most social statistics, available Drimarily from govern- ment sources, are objective in nature; that is, they measure the frequency of occurrence of an attribute or commodity in the population. Numbers of births, deaths, marriages, years of schooling, and percent of occupied housing with television sets could thus be considered objective measures. (Some would disagree, however, with the objectivity of these measures, see Andrews & Withey, 1976, p. 5.) Several researchers (Andrews & Withey, 1976; Campbell, Converse, & Rogers, 1976) have attempted to measure people's perceptions of their well-being, their quality of life. Such measures collected on a regular basis are expected to be valuable supplements to the usual objective quality of life indicators. (See for examples of the latter: Liu, 1976; Thompson, 1976b, 1977.) Creation of a social indicator data base is not without conceptual and methodological problems. Various aspects of the social measurement problem have been acknowledged in the literature (see, for example, de Neufville, 1975, pp. 175-179; Etzioni & Lehman, 1969; Social Measurement, 1972). While de- tailed discussion of measurement dysfunction (in the termi- nology of Etzioni & Lehman, 1969) is beyond the scope of this study, the following observation might be kept in mind: Increased investment, intellectual as well as financial, no doubt can go a long way to increase the efficacy of social measurements and to reduce much of the likelihood of dysfunctions. But, in the final analysis, these problems can never be eliminated en- tirely. Here, the client of systematic measurement and accounting should be alerted to the limitations of social indicators, both to make his use of them more sophisti- cated and to prevent him from ultimately rejecting the idea of social accounting when he encounters its limitations. (Etzioni & Lehman, 1969, p. 62) Educational Implications Educational indicators, a subset of social indicators, have traditionally been measures of the educational system's inputs and outputs stated in such terms as numbers of tea- chers, per pupil expenditures, and achievement test scores. There have been attempts, however, to broaden this base of educational statistics to include both objective and sub- jective indicators under the categories of access, aspirations, achievement, impact, and resources (Gooler, 1976, p. 15). There have also been attempts to link indicators of social processes (e.g., divorce rates, voting rates) to educational goals and thus to establish accountability measures, albeit remote, external to the educational system (Clemmer, Fairbanks, lall, Impara, & Nelson, 1974; Collazo, Lewis, & Thomas, Note 6; Grady, 1974). The use and abuse of indicators in an edu- cational setting, however, remained in 1977 a matter of debate (Impara, Note 7) and cautious optimism (Hall, Note 8). Hope- fully, investigations of the problem, such as that described in this study, will provide some guidance as to the most promising applications of social indicators to education. Selection of Variables/Time Series Indicators The Variables In the first year (Sept. 1975-June 1976) of Florida Department of Education STAR Project R5-175 on social fore- casting for educational planning, trends in five indicators of educational outcomes were forecast. In order to do this, it was necessary to identify variables that influence the outcomes of education. Through a review of the research and theoretical literature, a number of social variables were identified. This list was refined by an interdisciplinary panel of experts at the University of Florida to the following 10 variables: (a) socio-economic status; (b) family expectations, attitudes, and aspirations; (c) student's self- concept; (d) student's general ability; (e) student's sense of fate control; (f) student's attitudes and motivation; (g) peer group characteristics; (h) teacher expectations; (i) teacher behavior in the classroom; and (j) administrative leadership style. Collazo et al. (1977) said that only the variables (a) and (d) received strong support from research; a number of the other variables, while "identified as impor- tant in the theoretical literature. .had inconclusive support from research" (p. 298). (See Collazo, Lewis, & Thomas, Note 9, for a review of the research literature on variables affecting educational outcomes.) The panel of experts was further utilized to forecast the future trends of these variables and their effect on specified performance and utilization measures of the out- comes of education. Cross-impact analysis, a computer assist- ed modification of the Delphi forecasting technique, was then used by the panel to generate the future trends in the five outcome indicators. The framework for looking at the future established during these first year project activities is utilized in the present study. Previous forecasting activities were based primarily on the subjective judgment of panel participants. In this study, however, the feasibility of using time series data, where available, as the basis for forecasting future trends in the 10 variables affecting educational outcomes is examined. In addition, the use of a model containing the selected variables is considered. Bronfenbrenner's Ecology of Education Model In the previous section, the 10 variables affecting educational outcomes which were derived from the research literature were presented. flow can these variables be put into perspective as social forces influencing what the stu- dent learns? The Bronfenbrenner (1976) model which was presented in Chapter I (pp. 1-3) is a multi-dimensional ecological struc- ture of the educational environment. At the center of the interacting meso-, exo- and macro-systems is the micro-system, "the immediate setting containing the learner" (Bronfenbrenner, 1976, p. 5). The meso-system is actually a system of micro- systems; that is, it "comprises the inter-relationships among the major settings containing the learner at a particular point in his or her life" (Bronfenbrenner, 1976, p. 5). Some of the social variables that were identified previously could be considered as part of the meso-system. The home, for ex- ample, is represented by socioeconomic status and family expectations, attitudes, and aspirations; the peer group by peer group characteristics; and the school by teacher expec- tations, teacher behavior in the classroom, and administra- tive leadership style. The other variables: student's self- concept, student's general ability, student's sense of fate control, and student's attitudes and motivation are all di- rectly related to the learner. Bronfenbrenner (1976) proposed that learning is a func- tion of (a) the dynamic relationship between characteristics of the 'learners and their various surroundings (meso-system) and (b) the interaction between these various environments (e.g., home, school, peer group). The Bronfenbrenner ecology of education model thus appears to provide the necessary framework to support use of the presently identified variables and to generate directions for future forecasting research. Operational Definition of Variables as Time Series Indicators In previous sections 10 variables affecting educational outcomes were presented and then classified according to the Bronfenbrenner ecology of education model. In order to iden- tify trends in these variables and to extrapolate these trends into the future, it was necessary to operationally define these variables as time series measures, or indicators. Since some of these variables were expressed in general terms, it seemed necessary to try to represent each by a number of measures and thus avoid "fractional measurement" which is often a concern when operationally defining a social concept (Etzioni F, Lehman, 1969). Several problems became apparent in operationalizing the variables: 1. A number of indicators were identified for the variables (a) socioeconomic status; (b) family expectations, attitudes, and aspirations; and (c) peer group characteristics. For some indicators, however, data were not collected annually; for others, measures were not comparable over time due to a different basis for measurement. 2. For the variables related to the school and student characteristics (except student attitudes and motivation), no time series data which met the criteria for selection were available. 3. Operational definitions were in many cases influenced by the availability of indicators rather than the logic or appropriateness of the indicator to measure the social concept it represented. The social variables, examples of indicators that might be used to operationally define these variables, and sources of the available time series data are presented in Table 1. The following eight indicators which met the criteria estab- lished for this study (see p. 9) were selected for use with the three extrapolative methods described in Chapter III: 1. Median family income in the United States ex- pressed in 1971 constant dollars. 2. Number of families in the United States headed by women expressed as a percentage of total families. 3. Number of wives in the labor force expressed as a percentage of total wives in the United States. 4. Number of marriages in Florida expressed as rate per 1,000 population in Florida. 5. Number of dissolutions of marriage in Florida ex- pressed as rate per 1,000 population in Florida. 6. Number of resident live births in Florida ex- pressed as rate per 1,000 population in Florida. 7. Number of 3 to 5 year olds enrolled in nursery school and kindergarten expressed as percentage of total children 3 to 5 years old in the United States. 8. Number of children involved in divorce or annulment expressed as rate per 1,000 children under 18 years old in the United States. While rates or percentages are used for forecasting purposes, the magnitude of the actual numbers should be kept in mind before interpretation of an identified trend is Table 1 Time Series Indicators of Social Variables Affecting Outcomes of Education Available-time series data Social variables Indicatora Years b) State U.S. Source Socioeconomic status of family Median family income Employment rate-total labor force Unemployment rate Husband-wife families with two workers or more Female headed families 1947- 1971 1947- 1976 1947- 1975 1950- 1975 1950, 1955- 1957 1940, 47,50 55,60 65,70- 75 Fla. X U.S. Dept. of Commerce, Bureau of the Census in Social Indicators 1973 X U.S. Dept. of Labor, Bureau of Labor Statistics in Employment and Earnings, Dec. 1976 X U.S. Dept. of Labor, Bureau of Labor Statistics in Employment and Earnings, Dec. 1976 Fla. Dept. of Commerce, Division of Employment Security Research and Statistics X U.S. Dept. of Labor, Bureau of Labor Statistics in Special Report 189 X U.S. Dept. of Labor, Bureau of Labor Statistics in Special Report 190 Table 1 continued Available time series data Social variables Indicator a Yearsb State U.S. Source Labor force participation rates of wives Children living in poverty Family expecta- tions, attitudes, aspirations Marriage rate Divorce rate Birth rate 1950, 55-75 1959, 62,65 68,71 75,76 1930, 40,50 60,63- 75 1930, 40,50 60,63- 75 1964- 1975 Births to unwed mothers By age 10-18 1957- 1975 Fla. X U.S. Dept. of Labor, Bureau of Labor Statistics in Special Report 189 X U.S. Dept. of Commerce, Bureau of the Census and 1976 Survey of Income and Education State of Florida, Dept. of Health and Rehabilitative Services, Division of Health in Florida Statistical Abstract 1976 (Thompson, 1976a) Fla. Fla. Fla. X U.S. Dept. of Commerce, Bureau of the Census in Florida Statistic Abstract 1976 Public Health Statistics Section, Fla. Dept. of Health & Rehabilitative Services Table 1 continued Available time series data Social variables Indicator a Yearsb State U.S. Source Peer group characteristics Student's self- concept By race By age & race Public atti- tude toward education 3 to 5 year olds enrolled in nursery school and kindergarten Children under 18 involved in divorce Suspected offenders for four violent crimes, by age NAc 1930, 40,50, 56-75 1956- 1975 1969- 1976 1964- 1975 Fla. Fla. 1953- 1975 1958- 1972 Public Health Statistics Section, Fla. Dept. of Health & Rehabilitative Services Public Health Statistics Section, Fla. Dept. of Health & Rehabilitative Services X Gallup Poll (published annually in Phi Delta Kappan, 1969-77 X HEW, National Center for Educational Statistics in Advisory Committee on Child Development (1976) study X 1953-67: 1968-75: Ferriss (1970) Bane (1977) X See Social Indicators 1973 for list of sources Table 1 continued Available time series data Social variables Indicatora Yearsb State U.S. Source Student's general NA ability Student's sense NA of fate control Student's atti- High school 1901- X HEW, National Center for Educational tudes toward graduation 1975 Statistics in The Condition of Edu- education and rate biennial cation 1977, Vol. 3, Part 1 motivation for achievement School reten- 1924-32 X HEW, National Center for Educational tion rate, 5th to Statistics in Digest of Education grade to high 1967-75 Statistics (1976 edition) school gradua- tion Teacher expecta- NA tions Teacher behavior NA in classroom Administrative NA leadership style aIndicators presented are examples of time series data that might be used to operationally define the accompanying variable; a number of other indicators representing data aggregated to local, state, or national levels (depending upon purpose) could be added and/or substituted. bYears presented are those found in the course of this study and do not imply that data are in any way limited to these years. cNA: Time series indicator not available for variable. attempted. Furthermore, it is necessary to remember that since the population base increased over the decades covered by the data, a stable rate or percentage still represents larger absolute numbers of the phenomenon. The indicators selected are aggregated to either the state or national level; the appropriate level of aggregation would, of course, depend upon the specific planning activity. These indicators could be disaggregated by race, age, region, or sex (where appro- priate) for comparative analysis, and indeed this feature is a necessary characteristic in many of the definitions of social indicators (e.g., see definition by Land, 1971, pre- sented earlier in this chapter). CHAPTER III RATIONALE FOR EXTRAPOLATIVE METHODS SELECTED FOR COMPARISON One of the purposes of this study was to compare three purely extrapolative methods which could be used with social indicator data of the type described in Chapter II to fore- cast future values of those indicators. In order to select methods which were appropriate for this purpose, both the general forecasting literature and forecasting applications of extrapolative techniques in specific areas were reviewed. Detailed descriptions of each technique as well as statisti- cal assumptions, sensitivity, and evaluative criteria were derived primarily from the literature in economic statistics and regression analysis. The following sections provide (a) an overview of the extrapolative methods used in forecasting, (b) evaluation of the applicability of these methods for the purpose of this study, (c) a description of the three methods selected for comparison, including equations, parameters to be estimated, assumptions, and criteria to be used in the comparison of the three methods. Overview of Extrapolative Forecasting Methods An extrapolative forecasting method is a procedure for (a) identifying an underlying historical trend or cycle in time series data, and (b) estimating future states of a variable based on current and historical observations/mea- sures of that variable (Harrison, 1976). Extrapolation pro- vides a "surprise free" projection of the future, but not necessarily a future which is a bigger and better (or worse) version of the present. Martino (1976) noted that some extrapolation methods allow the forecaster to identify policy variables which are subject to manipulation and which allow the decision-maker to alter the future away from today's pattern of events. (p. 4) In the social realm extrapolation of trends may at least allow the planner or policy maker to make enlightened de- cisions to prepare for the future. Economic and Business Forecasting Because of the impetus in the 1930's and 1940's to describe and forecast the economic condition, many extra- polative methods were developed with economic applications in mind. Greenwald (1963, p. 187) classified methods for determining economic trends into (a) non-mathematical methods such as freehand curve fitting, first-order differences, semi- averages, selected points, and weighted and unweighted moving averages; and (b) mathematical methods such as least squares, moments, maximum likelihood, and others. In general, only the mathematical methods, which include a widely diverse array of complex curve-fitting techniques, seem to be relied upon for forecasting purposes while the non-mathematical methods are used for preliminary analysis of the shape of the time series data. (For descriptions of these methods, see Greenwald, 1963; Mayes & Mayes, 1976; Mendenhall & Reinmuth, 1971; Neiswanger, 1956; Tuttle, 1957.) Approaches to governmental/national economic forecasting (e.g.,Theil, 1966) often reach a relatively high level of mathematical and theoretical sophistication. This appears to be the result of decades of development, of applying method in light of theory, and developing both in turn. It is also the result of substantial investment of financial and manpower resources by both government and industry. The value of extrapolative forecasting to individual decision-makers in business has become apparent (Makridakis, Hodgsdon, & Wheelwright, 1974). Indeed, companies of all sizes are compelled to make forecasts for a number of varia- bles which affect them. Makridakis et al. (1974) have noted, however, that as with the development of most management science techniques, the application of these [extrapolative forecasting] methods has lagged behind their theoretical formu- lation and verification. (p. 153) Thus, the authors observed that while the need for forecasting methods is recognized by managers in business, few are famil- iar with the numerous techniques available and their charac- teristics in order that the one most appropriate for a given situation be selected. To help meet this need, Makridakis et al. have developed an interactive forecasting system (called Interactive Forecasting [SIBYL/RUNNER]) which allows a number of factors to be considered in the selection of a forecasting technique for a given set of data. Although the system has been well tested in teaching situations, it has not had extensive application in actual business settings. Quantitative techniques available in the Interactive Forecas- ting (SIBYL/RUNNER) system fall under the general headings of smoothing, decomposition, control, regression, and other techniques. The techniques considered under those headings are clearly explained in a subsequent work of two of the authors (Wheelwright & Makridakis, 1977). Technological Forecasting Martino (I973h; 1976) described the extrapolative methods most commonly used in technological forecasting in relation to the shape of their fitted curves: (a) growth curve, an S-shaped curve, which requires the setting of an upper limit; (b) trend curve, an exponential function which takes the form of a straight line when logarithmic transformation of the data is undertaken. Martino (1973b) illustrated the use of the growth curve with data on lowest temperature achieved in the laboratory by artificial means and the trend curve with data on productivity in the aircraft industry. It should be noted that both the growth curve and the trend curve applied to technological change by Martino (1973b) are highly versatile approaches with applications in a number of disciplines. Both methods are derived from the least squares formula for a straight line. The growth curve is a modified exponential, that is, it represents a variable which changes at a changing rate; the trend curve is a geometric straight line which represents a variable which changes at a constant rate (Neiswanger, 1956). Educational Forecasting Uses of extrapolative methods in education have generally been limited to projections of expenditures, school enroll- ments, and the number of instructional staff, high school graduates, and earned degrees. While many states and school districts have developed their own models, especially for projections of enrollments, the National Center for Educa- tion Statistics (U.S. Department of Health, Education, & Welfare, 1977c) in developing projections of education statis- tics to 1985-86 relied on regression methods wherever a trend could be established. Specifically, either arithmetic straight lines or logistic growth curves, depending upon the nature of data, were fitted by the method of least squares. The following was noted, however: For both the straight line and logistic growth curve, the fitted curve often lies considerably above or below the last ob- served point, resulting in an unusual rise or drop from the last actual observa- tion. To avoid this and give face validity to the projections, the fitted curve was used only to establish the last point, and a new curve was drawn through the last observed ratio and the end point on the fitted curve. (U.S. Department of Health, Education, & Welfare, 1977c, p. 92) Brown (1974) summarized the use of trend analysis methods in education and noted their potential applications in educa- tional administration. The four extrapolative methods that he critiqued were (a) arithmetic straight line extrapolation, (b) time series analysis (really a simplified version of the Box-Jenkins technique), (c) the S-shaped growth curve, and (d) cohort analysis (actually the trend curve described by Martino in the previous section). The examples selected by Brown do not reveal the versatility of the methods illustra- ted; he did, however, provide a comprehensive review of literature describing applications in other fields. A number of methodological concerns raised by Brown were considered in this study. In a critique of selected futures prediction techniques that might be employed by educational planners, Folk (1976) observed that exponential trend line and arithmetic straight line projections appear to be the most commonly used extra- polative techniques. This author provided a number of useful measures for evaluating statistically derived regression lines. The educational applications just described are basically attempts to project inputs such as money, pupils, or teachers to the educational system or outputs (graduates, degrees earned) of that system. No attempt to extrapolate the future status of variables which affect these student-related inputs or outputs was discovered in the literature search. Extrapolative Methods in Other Areas Several areas have developed highly specialized extrapo- lative methods in making forecasts of the future. Popula- tion, employment, and unemployment projections, for example, are usually based on fairly complex models which incorporate a number of factors. These particular applications are not reviewed here due to their highly specialized purposes and functions. Applicability of Reviewed Extrapolative Methods for Study In evaluating the applicability of the previously re- viewed extrapolative methods for projected future states of the time series indicators selected for use in this study, several points needed to be considered. Chief among these were (a) the underlying pattern of the data that can be recognized and (b) the type or class of model desired (from Wheelwright & Makridakis, 1977). Both of these will be briefly considered in relation to this study. The Pattern of the Data From graphical representations of each indicator, the data for each appeared to be characterized by a trend which either increased or decreased with time. Some also appeared to contain cyclical patterns and random fluctuations. It seemed as if major trends might follow the form of a straight line or curve with one or two bends. The Class of Model Wheelwright and Makridakis (1977) distinguished four classes or categories of models: 1. The time series model "always assumes that some pattern of combination of patterns is recurring over time" (p. 22). 2. The causal model assumes "that the value of a cer- tain variable is a function of several other variables" (p. 23). 3. The statistical model comprises a number of fore- casting techniques; it uses the language and procedures of statistical analysis to identify patterns in the variables being forecast and in making statements about the reliability of these forecasts. (p. 23) 4. The nonstatistical model includes "all models that do not follow the general rules of statistical analysis and probability" (p. 24). Of course, some techniques can be classified into more than one of the four types of models. It appeared that the statistical model, with its well-defined properties, and replicable procedures, would be an appropriate starting point for predicting the long-term trends in the selected time series data. The review of the literature revealed several techniques denoted by the form of their curves which are sensitive to long-term trends in the data and which are classified under the statistical model: (a) the arithmetic straight line, (b) the S-shaped growth or logistic curve, (c) the trend or exponential curve, (d) the polynomial curve. All of these techniques are regression techniques solved by least squares procedures. Techniques (b) through (d) require data trans- formations to satisfy the basic linear model used in regres- sion. The growth or logistic curve was eliminated from com- parison because this technique necessitates the setting of limits which might bias the results of the study due to its ex post facto nature. The remaining three techniques were considered to be appropriate for use in the comparison phase of this study. Description of Methods to be Compared Since the three techniques selected for comparison are intrinsically linear in their parameters (Draper & Smith, 1966), the general linear model denoted by the simple or bivariate regression equation is presented first. Addition- ally, estimation of the parameters of the equation by least squares procedures, the assumptions of the model, and criteria for evaluation and comparison of the three methods are dis- cussed. Each technique is then described in relation to the general linear model. The General Linear M'odel In the comparison of methods using selected time series indicators, time in years is considered the independent variable and the indicator is considered the dependent or response variable. Thus, if time is denoted by X, and the indicator is denoted by Y, a functional relationship in the form Y = f(X) might be stated. However, since most social relationships are stochastic (probabilistic) rather than deterministic in nature, a more appropriate form might be Y = f(X) + e, where e represents error, a measure of the unknown factors. When the relationship between the two variables, time and the indicator (Y) is assumed to be linear (that is, repre- sented by a straight line), the equation becomes Y = 60 + ix; and because many social relationships are stochastic for particular values of the variables, this equation is actually Y = 30 + 81X + E. Since the population parameters So and 6i are not known unless all possible occurrences of X and Y are known, the available data are used to provide estimates b0 and b, of So and Sj as in the following regression equation, Y = b0 + bjX + e (where Y denotes predicted values of Y). The constant b0 (the intercept) and the regression coefficient b, (the slope of the regression line) can be determined by ordinary least squares procedure, "so called because it estimates. in such a way that the sum of squared residuals, 2 Eei is as small as possible" (Mayes & Mayes, 1976, p. 112). (For detailed treatment of simple regression and least squares estimation of So and 61, see, for example: Draper & Smith, 1966; Kerlinger & Pedhazur, 1973; Mayes & Mayes, 1976; Men- denhall, Ott, & Larson, 1974; Mendenhall & Reinmuth, 1971; Runyon & laber, 1967.) The Assumntions of the Linear Model Draper and Smith (1966) noted that In many aspects of statistics it is necessary to assume a mathematical model to make progress. It might be well to emphasize that what we are usually doing is to consider or tentatively entertain our model. (p. 8) Thus, when the general linear model is employed as it is in this study, it becomes necessary to examine the assumptions upon which the model is based and to judge whether the model is in fact appropriate for the data. Assumptions for the general linear model include the following: 1. The regression equation Y = b0 + b1X + e is a better predictor of Y than Y = Y (bi/ 0). 2. The regression equation accounts for a significant portion of the variation in Y, that is, the relationship between X and Y described by the equation is not the result of chance. 3. The error term c has a mean value equal to zero and variance equal to 02; it is an independent random variable which is normally distributed. If the first two assumptions are not met, then the model is not a good predictor for that data. If the third assump- tion is not met, then it is not appropriate to interpret the results statistically, that is, in terms of the probability distribution of the random error e. It is possible to test Assumption 1 and Assumption 2 by the F statistic. Assumption 3 is best evaluated by plotting the residuals and examining the pattern of the deviations from the regression line (Anscombe, 1973; Anscombe & Tukey, 1963; Draper & Smith, 1966). Independence of the errors (Assumption 3[e]) may be tested by the Durbin-Watson test for serial correlation (Durbin & Watson, 1950; Durbin & Watson, 1951; Mayes & Mayes, 1976; Wheelwright & Makridakis, 1977). Criteria for Comparison of Methods The following questions were derived from the literature to guide the comparison of methods: 1. Do the data satisfy the assumptions of the model? (See previous section.) 2. How well does the regression line fit the data from which it was derived (the two-thirds of the data points used to generate the prediction equation)? Tufte (1974, pp. 69-70) listed four measures of quality of fit: a. the N residuals: Y. Y. 1 1 b. the residual variation: s2 (Y i ^(Y ) 2 y-x N k* -1 (or the square root of the residual variation, Sy.x, called the unbiased standard error of estimate). *k refers to the number of X terms in the regression equation. c. the ratio of explained to total variation: r2 = i E(y i y)2 d. the standard error of the estimate of the slope: Sbi = Syx Thus, for each set of data, the methods are compared according to these four measures. The observed and pre- dicted values of Y are also reported in tabular form; both observed and predicted values are plotted for visual com- parison as recommended by Anscombe (1973). 3. How well does the extrapolated line fit the data (the one-third of the data points that were not used to generate the prediction equation)? The residual variation around the extrapolated line, which is an indicator of the accuracy of the forecasting technique, may be expressed by its square root, the standard error for the extrapolated values. As in (2), the observed and extrapolated values are reported in tabular form; both observed and extrapolated values are plotted for visual comparison. Neiswanger (1956, p. 534) cautioned against accepting only mathematical tests of "goodness of fit" as proof that the mathematical expression is appropriate for the trend in the data. Other considerations such as the "reasonableness of the extrapolated values which the trend may yield" (p. 534) and "the extent to which this statistical manifestation of growth is supported by other evidence" (p. 534) should be kept in mind. Thus, the calculation of a trend is more than a mathematical analysis in curve fitting; it is essentially a problem of analysis of the phenomena represented by the data (Neiswanger, 1956). It should be also noted that while the standard error of estimate gives an overall measure of error around the regres- sion line, it may not be appropriate for computing confidence intervals for a specific forecast value. The reason for this is that the further an X is from X, the larger is the error that may be expected when predicting Y from the regression line. Draper and Smith (1966) noted: We might expect to make our "best" pre- dictions in the "middle" of our observed range of X and would expect our predic- tions to be less good away from the "middle." (p. 22) Therefore, the confidence limits for the true value of Y for a given X are two curved lines about the regression line. The limits change as the position of X changes. Hence the following equation was provided by Wheelwright and Makridakis (1977, p. 82) for computing the standard error of forecast (SEf): SEf = (Y i 31 + + )2 j N J(Xi _ for a specific forecast value. Method 1: Simple Linear Regression The equation Y = f(X) describes a natural functional relationship between X and Y. If this functional relationship can be expressed by a straight line on arithmetic paper, the linear, first-order regression equation Y = b0 + biX + 6 may be appropriate. The natural linear function is used when an absolute amount of change in Y per unit of X is hypothe- sized. Method 2: Log-linear Regression Occasionally when time series data are plotted on an arithmetic scale the scatter of points fall more in a curve than in a straight line with the curve rising or decreasing more rapidly as X increases. These same data when plotted on a semilogarithmic scale will produce a straight line. The relationship between X and Y may then be described by log Y = f(X) or Y = abx, the exponential form of the logarithmic relationship between X and Y. The exponential function is used when there is thought to be a constant rate of change in Y per unit absolute change in X. Thus, for each year (X), Y changes by a constant per- centage (rather than by an absolute amount as in Method 1). It is possible to fit the exponential function to the general linear model by transforming the values of Y to log Y. Thus Y = abx becomes log Y = a + bX log Y = log a + X log b or log Y = log b0 + X log bi + 8. As in the case of the natural number straight line, the method of least squares is used to estimate the parameters necessary for computing the logarithmic (or geometric) straight line. Tuttle (1957, p. 431) noted, therefore, that AA the log Y's are fitted to the log Y's, not the Y to the Y's, by the least squares criterion. Thus, Tuttle (1957, p. 432) recommended that the standard error of estimate be computed from the antilogs of the log Y values. If Sy.x was computed as the root mean square of the unexplained variation, "it would be in terms of the deviations of the logarithms of the YcS[Y's] from the logarithms of the Y's" (Tuttle, 1957, p. 432). The Sy x would not be comparable to those obtained from untransformed data as in Method 1. Similarly, Seidman (1976) has observed that in comparing linear and log-linear models, R2 may not be a sufficient criterion of choice. This is because the R2 represents "the proportion of variance of the logarithm of Y explained by the regression: log Y, not Y, is the dependent variable" (Seidman, 1976, p. 463). Therefore, Seidman recommended using the antilogs of the predicted values of log Y "in a regression explaining variability in Y" (p. 463). This R2 may then be used for comparison purposes. The examples given by Seidman (1976) were based on logarithmic transformations of both dependent and independent variables, but the same ob- servation may be made when only the dependent variable is transformed. Seidman's reservation about R2 has been con- sidered in this study. If an exponential curve appears to fit the data, it is often desirable to find the annual rate of change c. This can be derived from the regression coefficient bl, according to the following equation: log bi = (1 + c) change = antilog b1 1. The result should then be expressed as a percent (Mayes & Mayes, 1976, p. 94; Nie et al., 1975, p. 370). The common or Briggs logarithm, used in the Y trans- formation in this study, is the power to which 10 must be raised to equal the number (see Neiswanger, 1956, p. 210; Tufte, 1974, p. 108). Natural logs or logs to the base 2 could also have been used to obtain the same results (Snedecor, 1956, pp. 450-451). Method 3: Polynomial Regression In Method 1, the equation which expresses a straight line relationship between X and Y is Y = bo + biX + which is a linear (in the b's) first-order (in X) regression equation. When this functional relationship between X and Y can be expressed as a solid, or unbroken curve on arithmetic paper, the linear, second-order (or quadratic) regression equation Y = b0 + biX + b2X2 + 8 may be appropriate. When the relationship can be expressed as a curved line with two bends on arithmetic paper, the linear, third-order (or cubic) regression equation ^Y = b0 + biX + b2X2 + b3X3 + @ may be used. According to Kerlinger and Pedhazur (1973, p. 209), the highest order a polynomial equation may take is equal to N 1, where N is the number of distinct values in the independent variable. However, since one of the goals of scientific research is parsimony, our interest is not in the predictive power of the highest degree polynomial equation possible, but rather in the highest degree polynomial equation necessary to describe a set of data. (Kerlinger & Pedhazur, 1973, p. 209) Another reason for a parsimonious approach to polynomial curve fitting is that for each order added to the equation, a degree of freedom is lost. This is especially important when the number of observations are small as they are in this study (observations range from 8 to 20 in each of the eight sets of data). Also, higher order polynomial curves may possess statistical significance but be devoid of practical significance. Accordingly, only the quadratic and cubic forms of the polynomial regression equation are considered. In the polynomial regression the independent variable, X (time), is treated as a categorical variable and is raised to a certain power. In the quadratic equation, each value of X is squared to create a new vector of the squared X's, X2. Similarly, in the cubic equation, each value of X is cubed to create an additional vector of the cubed X's, X3. Thus, the resulting equation can be solved by a stepwise multiple regression procedure, in which at each step of the analysis, the R2 is tested to see if the higher-degree poly- nomial accounts for a significant proportion of the variance. While a least squares solution is used in this study, the values of the unknowns may also be found by orthogonal poly- nomials (see Draper & Smith, 1966, pp. 150-155; Greenwald, 1963, pp. 204-209; Kerlinger & Pedhazur, 1973, pp. 214-216). Neiswanger (1956, pp. 529-532) noted that the second- degree and third-degree parabolas provide greater flexi- bility in fitting a line to a set of data for the parabolas allow a trend to change direction. Whether or not the flexibility of the parabolic function enhances the predic- tability of extrapolated Y values, however, is not certain and is examined in this study. CHAPTER IV COMPARISON OF EXTRAPOLATIVE METHODS USING SELECTED TIME SERIES INDICATORS In Chapter II a rationale for the selection of social variables operationally defined as time series indicators was provided. The following eight time series indicators were selected for use in the method comparison phase of this study: 1. Median family income in the United States expressed as 1971 constant dollars. 2. Number of families in the United States headed by women expressed as a percentage of total families. 3. Number of wives in the labor force expressed as a percentage of total wives in the United States. 4. Number of marriages in Florida expressed as rate per 1,000 population in Florida. 5. Number of divorces in Florida expressed as rate per 1,000 population in Florida. 6. Number of resident live births in Florida expressed as rate per 1,000 population in Florida. 7. Number of 3 to 5 year olds enrolled in nursery school and kindergarten expressed as percentage of total children 3 to 5 years old in the United States. 8. Number of children involved in divorce or annulment expressed as rate per 1,000 children under 18 years old in the United States. A rationale for the three extrapolative methods selected for comparison in this study was presented in Chapter III. The three methods are simple linear regression, log-linear regression, and polynomial regression (specifically the quadratic and cubic forms). The following questions derived from the literature were proposed in Chapter III to guide the comparison of methods: 1. Do the data satisfy the assumptions of the general linear model? 2. How well does the regression line fit the data from which it was derived (the two-thirds of the data points used to generate the prediction equation)? 3. [low well does the extrapolated line fit the data (the one-third of the data points that were not used to generate the prediction equation)? To answer these questions in terms of each method and to facilitate comparison among the three methods, the results obtained from applying each of the methods to each of the eight indicator data sets are presented in the following manner: 1. The fit of the regression line to the observed data is indicated by r2 and the unbiased standard error of estimate S..x* For the simple linear and log-linear regres- sion methods the amount of variance accounted for by the regression line is tested by the F statistic (F value is the same as that obtained by dividing bi by SE ). For the quad- ratic and cubic forms of polynomial regression, both the r2 including all orders entered to that step (r2 or r2 y.12 ory.123) and the increase in r2 attributable to the last order entered in the regression (r (2.) or ry(3.12)) are tested with the F statistic.* (Of course, dividing the partial regression coefficients b2 in the quadratic form and b3 in the cubic form by their respective standard errors will also yield the same F value for the increase in r2.) 2. The fit of the extrapolated line to the data is indicated numerically by the standard error for the extra- polated values, Sext(y-x)* This measure reflects the average deviation of the extrapolated values from the ob- served values of Yi; thus, -N (f (Y -) Sext(y.x) =extrapolated values) (Note that this equation is not the "unbiased" form used in computing S .) y.x *Actually the increase in r2 is tested according to the following ratio: F = (r2 with kthorder term) (r2 without k th-order term). (1 r2 with kth-order term) / (N k 1) Total r2 is tested according to the following ratio: F =SS regression/k F S residual/(N -_1 3. All observed and predicted values of Y are reported in tabular and graphic form. 4. The residuals around the regression line were ex- amined for serial correlation by the Durbin-Watson d statistic, which is noted only when serial correlation is confirmed or questionable. Additionally, the standardized residuals were plotted against the sequence of cases and also against standardized Y values. Such visual inspection of the data is discussed as necessary to support the interpretation of re- sults in Chapter V. Presentation of Results Indicator 1 The mean and standard deviation for the Y values used to generate the regression equations are 6674 and 987, re- spectively. The following regression equations were used to derive Y: Linear Y = -3933.29 + 192.87 X Quadratic Y = 106.37 + 44.79 X + 1.35 X2 A Cubic Y = 135025.33 + (-7385.17)X + 137.08 X2 + (-.82)X3 Log-linear log Y = 3.12346 + .01266 X. The goodness of fit of the regression lines derived from these equations is indicated by r2, r2 change, and S in Table 2. y.x An ANOVA summary table is presented in Table 18 in the Appen- dix. The overall F's for all methods are significant (p<.01); the increases in r2 due to the higher order polynomials are not significant, however. Table 2 Indicator 1: Summary Statisticsfor Prediction Equations by Method r2 r2 change F df S y-x Linear .97440 571.04** 1,15 163.02 Quadratica .97531 276.47** 2,14 165.75 .00090 .51 1,14 Cubica .98137 228.25** 3,13 149.40 .00606 .42 1,13 Log-linearb log Y .97021 488.54** 1,15 log.01157 antilog Y .98726 164.92 Note. Indicator 1 is median family income expressed in 1971 constant dollars. aBoth quadratic and cubic forms of the polynomial regres- sion are presented. bBoth r2 and S have been recomputed using antilogs of y x 2 2 the log Y; much of the difference between r2(log Y) and r2 (antilog Y) may be due to rounding. **P<.01 The average errors for the extrapolated Y values (Sext(y.x)) according to method employed are (a) linear, 669; (b) quadratic, 487; (c) cubic, 1981; and (d) log-linear, 283. Observed Y's and predicted values of Y for both the original regression and the extrapolated lines are presented in Table 3. These data are graphically presented in Figure 2. Thus, there is very little difference in the total r2 for the methods; the quadratic and cubic forms added little to the r2 already provided by the linear component. The S y.x for the cubic form is smaller (149.40) than for the other methods. When the lines are extrapolated beyond the original values, however, the cubic form is clearly the "worst" fit with a Sext(y.x) of 1981 and the log-linear method the "best" with a Sext(y.x) of 283. Whether the exponential curve would continue to be a superior predictor is a matter of conjecture. Indicator 2 The mean and standard deviation for the Y values used to generate the regression equations are 10.4 and .77, re- spectively. The following regression equations were used to derive Y: Linear Y = 9.87860 + .02789 X Quadratic Y = 11.08794 + (-.19035)X + .00626 X2 Cubic Y = 11.45737 + (-.35750)X + .01954 X2 + (-.0027)X3 Log-linear log Y = .99387 + .00118 X. The goodness of fit of the regression lines derived from these equations is indicated by r2, r2 change, and S in y.x Table 3 Indicator 1: Observed Y's and Predicted Y's by Method Predicted Y's by Method Year Observed Y's Linear Quadratica Cubica Log-linear Original regressionb 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 5,4S3 5,367 5,278 5,594 5,783 5,939 6,433 6,288 6,693 7,122 7,138 7,126 7,524 7,688 7,765 7,975 8,267 5,131c 5,324 5,517 5,710 5,903 6,096 6,289 6,481 6,674 6,867 7,060 7,253 7,446 7,639 7,831 8,024 8,217 5 185 5,358 5 533 5 711 5,892 6,076 6,262 6 450 6,642 6,836 7,033 7 233 7,435 7,640 7 848 8 058 8 271 5,323 5,392 5,499 5 637 5 803 5,992 6,197 6,416 6,642 6,871 7 097 7 317 7,524 7, 714 7,882 8,023 8 133 5,231 5,386 5,545 5,709 5,878 6,052 6,231 6,415 6,605 6,800 7,002 7,209 7,422 7,642 7,868 8,100 8,340 Extrapolationd 1964 1965 1966 1967 1968 1969 1970 1971 8,579 8,932 9,360 9,683 10,049 10,423 10,289 10,285 8,410 8,603 8,796 8,989 9,182 9,374 9,567 9,760 8,487 8,705 8,926 9,150 9,377 9,606 9,838 10,072 8,205 8,235 8,220 8,152 8,028 7,843 7,591 7,268 8,584 8,838 9,100 9,369 9,646 9,931 10,225 10,527 Note. Indicator I is median family 1971 constant dollars. income expressed in aBoth quadratic and cubic forms of polynomial regression are presented. bThe regression line is derived from 2/3 of the known data points. Cpredicted Y's in terms of 1971 constant dollars are rounded to number of places in original data. dValues are extrapolated beyond the data points used to generate the regression equation. 11,000- 10,000- 9,000- 8,000- 7,000- 6,000- 5,000- .....Observed Y's -- Cubic Linear Ouadratic Log-linear ! 1947 6 60 I 70 I 75 Figure 2. Indicator 1: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extrapolation). I- 0,-4 U -4 4-J cz V) 0 U ct5 *H ,-4 4,000 I Table 4. An ANOVA summary table is presented in Table 19 in the Appendix. The overall F's for the quadratic and cubic forms of the polynomial regression are significant (p<.05); however, only the increase in r2 due to the quadratic is significant (p<.01). The average errors for the extrapolated Y values (Sext(y.x)) according to the method employed are (a) linear, 1.5; (b) quadratic, .35; (c) cubic, 1.1; and (d) log-linear, 1.5. Observed Y's and predicted values of Y for both the original regression and the extrapolated lines are presented by method in Table 5. These data are graphically presented in Figure 3. Thus, the cubic form of the polynomial accounts for the most variance in Y (89%) and has the smallest S (.33). y.x The quadratic form, accounting for 81% of the variance in Y, has a Sy. of .39; visual inspection of the second-degree curve reveals that this curve may, in fact, more closely fit observed values for the latter portion of the regression line than the cubic form. The linear and log-linear methods provide no better estimate of Y than does Y; indeed, the standard error of estimate approximates the standard devia- tion of the observed Y's. When the lines are extended beyond the original values, the quadratic provides the superior fit (Sext(y.x) = .35); the fit of the cubic, linear, and log-linear methods to the observed values is poor with residuals becoming larger for successive years. Table 4 Indicator 2: Summary Statistics for Prediction Equations by Method r2 r2 change F df S y.x Linear .16405 1.18 1,6 .76 Quadratica .81417 10.95* 2,5 .39 .65012 17.49** 1,5 Cubica .89255 11.08* 3,4 .33 .07838 2.92 1,4 Log-linearb log Y .16905 1.22 1,6 log.03175 antilog Y .16307 .78 Note. Indicator 2 is number of families in the United States headed by women expressed as a percentage of total families. aBoth quadratic and cubic forms of the polynomial regres- sion are presented. bBoth r' and S have been recomputed using antilogs of the log Y; much of the difference between r2(log Y) and r2 (antilog Y) may be due to rounding. V.05 *:*E<.O1 Table 5 Indicator 2: Observed Y's and Predicted Y's by Method Predicted Y's by Method Year(X) Observed Y's Linear Quadratica Cubica Log-linear Original regressionb 1940(l) 11.2 9.9c 10.9 11.1 9.9 1947(8) 9.5 10.1 10.0 9.7 10.1 1950(11) 9.4 10.2 9.7 9.5 10.2 1955(16) 10.1 10.3 9.6 9.6 10.3 1960(21) 10.0 10.5 9.9 10.1 10.4 1965(26) 10.5 10.6 10.4 10.6 10.6 1970(31) 10.9 10.7 11.2 11.1 10.7 1971(32) 11.5 10.8 11.4 11.2 10.8 Extrapolationd 1972(33) 11.6 10.8 11.6 11.2 10.8 1973(34) 12.1 10.8 11.9 11.3 10.8 1974(35) 12.4 10.9 12.1 11.3 10.8 1975(36) 13.0 10.9 12.4 11.3 10.9 Note. Indicator 2 is number of families in the United States headed by women expressed as a percentage of total families. a Both quadratic and cubic forms of polynomial regression are presented. bThe regression line is derived from 2/3 of the known data points. C Predicted Y's are rounded to number of places in original data. dValues are extrapolated beyond the data points used to generate the regression equation. Observed Y's Cubic Linear Quadratic Log-linear 12.0- 11.0- 10.0- 9.0- 80 - *1- 1 1940 Figure 3. I I I I - 547 '50 155 '60 Indicator 2: Observed Y's and predicted Y's separates values of original regression from I 1 '65 '70 by method (vertical extrapolation). * .. 13.0- - -7 -_--7 I 75 line VQ v Indicator 3 The mean and standard deviation for the Y values used to generate the regression equations are 32.0 and 3.8, respec- tively. The following regression equations were used to de- rive Y: Linear Y = 23.27325 + .74603 X Quadratic Y = 23.38774 + .71893 X + .00126 X2 Cubic Y = 22.60710 + 1.18842 X + (-.05693)X2 + .00194 X3 Log-linear log Y = 1.37971 + .01047 X. The goodness of fit of the regression lines derived from these equations is indicated by r', r2 change, and S in Table 6. An ANOVA summary table is presented in Table 20 in the Appen- dix. The overall F's for all methods are significant (p<.01); however, the increases in r2 due to the higher order polyno- mials are not significant. The average errors for the extrapolated Y values (Sext(y.x)) according to the method employed are (a) linear, 1.4; (b) quadratic, 1.2; (c) cubic, 2.7; and (d) log-linear, .6. Observed Y's and predicted values of Y for both the original regression and the extrapolated lines are presented by method in Table 7. These data are graphically repre- sented in Figure 4. Thus, there is very little difference in the total r2 for the methods; the quadratic and cubic forms added an in- significant amount to the r2 already provided by the linear component. The Sy-x for the cubic form is only slightly better (.43) than for the other methods (.49-.54). Table 6 Indicator 3: Summary Statistics for Prediction Equations by Method r2 r2 change F df S y.x Linear .98451 826.05** 1,13 .49 Quadratica .98460 383.49** 2,12 .50 .00009 .07 1,12 Cubica .98953 346.51** 3,11 .43 .00493 5.18 1,11 Log-linearb log Y .98085 665.69** 1,13 log.00760 antilog Y .99999 .54 Note. Indicator 3 is the number of wives in the labor force expressed as a percentage of total wives in the United States. a Both quadratic and cubic forms of the polynomial regres- sion are presented. bBoth^ r2 and Syx have been recomputed using antilogs of the log Y; much of the difference between r2(log Y) and r2 (anitlog Y) may be due to rounding. :k*p<.O1. Table 7 Indicator 3: Observed Y's and Predicted Y's by Method Predicted Y's by Method Year(X) Observed Y's Linear Quadratica Cubica Log-linear Original regressionb 1950(1) 23.8 24.0c 24.1 23.7 24.6 1955(6) 27.7 27.7 27.8 28.1 27.7 1956(7) 29.0 28.5 28.5 28.8 28.4 1957(8) 29.6 29.2 29.2 29.5 29.1 1958(9) 30.2 30.0 30.0 30.1 29.8 1959(10) 30.9 30.7 30.7 30.7 30.5 1960(11) 30.5 31.5 31.4 31.4 31.3 1961(12) 32.7 32.2 32.2 32.0 32.0 1962(13) 32.7 33.0 33.0 32.7 32.8 1963(14) 33.7 33.7 33.7 33.4 33.6 1964(15) 34.4 34.5 34.5 34.2 34.4 1965(16) 34.7 35.2 35.2 35.0 35.3 1966(17) 35.4 36.0 36.0 35.9 36.1 1967(18) 36.8 36.7 36.7 36.9 37.0 1968(19) 38.3 37.4 37.5 38.0 37.9 Extrapolationd 1969(20) 39.6 38.2 38.3 39.1 38.8 1970(21) 40.8 38.9 39.0 40.4 39.8 1971(22) 40.8 39.7 39.8 41.9 40.7 1972(23) 41.5 40.4 40.6 43.4 41.7 1973(24) 42.2 41.2 41.4 45.2 42.8 1974(25) 43.0 41.9 42.1 47.0 43.8 1975(26) 44.4 42.7 42.9 49.1 44.9 Note. Indicator 3 is the number of wives in the labor force expressed as a percentage of total wives in the United States. a Both quadratic and cubic forms of polynomial regression are presented. bThe regression line is derived from 2/3 of the known data points. cPredicted Y's are rounded to number of places in original data. dValues are extrapolated beyond the data points used to generate the regression equation. 72 47.0 - 4.0.. nhserved Y's Cubic Linear 45.0- Quadratic Lou-linear . 43.0 - 41.0 U o 390- 0 ~237.0- S35.0-/ 6 33.0- 4-- U 3 29.0-- 27.0- 25.0 2 3. 0 1950 '5 5 '60 '65 '70 '75 Figure 4. Indicator 3: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extrapolation). When the lines are extrapolated beyond the original values, however, the cubic form is clearly the "worst" fit with a Sext(y.x) of 2.7 while the log-linear form is clearly the "best" fit with a Sext(y.x) of .6. Indicator 4 The mean and standard deviation for the Y values used to generate the regression equations are 9.8 and 2.6, re- spectively. The following regression equations were used to derive Y: Linear Y 13.76707 + (-.13216)X Quadratic Y = 14.16426 + (-.19720)X + .00148 X2 Cubic Y = 10.90554 + 1.15549 X + (-.07961)X2 + .00125 X3 Log-linear log Y = 1.12873 + (-.00496)X. The goodness of fit of the regression lines derived from these equations is indicated by r2, r2 change, and S in Table 8. y.x An ANOVA summary table is presented in Table 21 in the Appen- dix. The overall F statistic for the cubic form of the poly- nomial regression is significant (P<.01); the increase in r2 due to the third degree polynomial is also significant (p<.01). The F statistic for both the linear and log-linear methods is significant (p<.05). The average errors for the extrapolated Y values (Sext(y.x)) according to method employed are (a) linear, 2.9; (b) quadratic, 2.5; (c) cubic, 4.1; and (d) log-linear, 2.7. Observed Y's and predicted values of Y for both the original regression and the extrapolated lines are presented by method in Table 9. These data are graphically represented in Figure 5. Table 8 Indicator 4: Summary Statistics for Prediction Equations by Method r2 r2 change F df S y.x Linear .42085 7.27* 1,10 2.07 Quadratica .42592 3.34 2,9 2.17 .00507 .08 1,9 Cubica .90338 24.93** 3,8 .94 .47746 39.53** 1,8 Log-linearb log Y .42298 7.33* 1,10 log.07715 antilog Y .36243/.42922c 2.06 Note. Indicator 4 is number of marriages in Florida expressed as rate per 1,000 population in Florida. a Both quadratic and cubic forms of the polynomial regres- sion are presented. bBoth r2 and Syx have been recomputed using antilogs of 2 (l g A the log Y; much of the difference between r (log Y) and r2 (antilog Y) may be due to rounding. c2 Two methods of computing r' using antilogs of Y yielded different results. 0< **p<.050 Table 9 Indicator 4: Observed Y's and Predicted Y's by Method Predicted Y's by Method Year(X) Observed Y's Linear Quadratica Cubica Log-linear Original regressionb 1930(1) 11.6 13.6c 14.0 12.0 13.3 1940(11) 17.1 12.3 12.2 15.7 11.9 1950(21) 9.8 11.0 10.7 11.7 10.6 1960(31) 7.9 9.7 9.5 7.6 9.4 1963(34) 7.7 9.3 9.2 7.5 9.1 1964(35) 7.7 9.1 9.1 7.6 9.0 1965(36) 8.3 9.0 9.0 7.9 8.9 1966(37) 8.5 8.9 8.9 8.2 8.8 1967(38) 9.0 8.7 8.8 8.7 8.7 1968(39) 9.6 8.6 8.7 9.3 8.6 1969(40) 9.8 8.5 8.6 10.0 8.5 1970(41) 10.1 8.3 8.6 10.9 8.4 Extrapolationd 1971(42) 10.5 8.2 8.5 11.6 8.3 1972(43) 11.0 8.1 8.4 12.8 8.2 1973(44) 11.4 8.0 8.4 14.1 8.1 1974(45) 11.0 7.8 8.3 15.6 8.0 1975(46) 10.1 7.7 8.2 17.3 8.0 Note. Indicator pressed as rate per 4 is number of marriages in Florida ex- 1,000 population. a Both quadratic and cubic forms of polynomial regression are presented. poi bThe regression line is derived from 2/3 of the known data nts. c Predicted Y's are rounded to number of places in original data. dValues are extrapolated beyond the data points used to generate the regression equation. Observed Y's Cubic . -- Linear Quadratic Log-linear 0 2 14.0- o I.P 0 . CD 11.0 S12.0- 10.0 ' N 100- "" / N 9.0- 8.0 -* 7.0 f I i '1____I ! 1930 40 50 '60 '63 '65 '70 '75 Figure S. Indicator 4: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extrapolation). 18.0 17.0 16.0 15.0 Thus, it would appear that for the original regression the cubic form of the polynomial best rits the observed Y's. This method accounts for 90% of the variance in Y with less than half of the average error of the other methods. When the lines are extrapolated beyond the original values, however, the cubic form has the largest average error (S ext(y.x) = 4.1). The quadratic form of the polyno- mial is, in fact, the best predictor (Sext(y.x) = 2.5) of the methods compared. Actually, for this set of data the mean (9.8) of the observed values of Y used in the original re- gression would have been the best predictor of the future values of Y. Indicator 5 The mean and standard deviation for the Y values used to generate the regression equations are 4.6 and 1.0, respec- tively. The following regression equations were used to derive Y: Linear Y = 4.10332 + .01692 X Quadratic Y = 3.16091 + .17126 X + (-.00350)X2 Cubic Y = 1.59416 + .82162 X + (-.04249)X2 + .00060 X3 Log-linear log Y = .56596 + .00288 X. The goodness of fit of the regression lines derived from these equations is indicated by r2, r2 change, and S in y.x Table 10. An ANOVA summary table is presented in Table 22 in the Appendix. Only the overall F for the cubic form of the polynomial regression is significant (p<.05); the F Table 10 Indicator 5: Summary Statistics for Prediction Equations by Method r2 r2 change F df S y.x Linear .04360 .46 1,10 1.06 Quadratica .22401 1.30 2,9 1.00 .18041 2.09 1,9 Cubica .92134 31.23** 3,8 .34 .69733 70.92** 1,8 Log-linearb log Y .12007 A 1.36 1,10 log.10388 antilog Y .12779 1.07 Note. Indicator 5 is number of dissolutions of marriage in Florida expressed as rate per 1,000 population. a Both quadratic and cubic forms of the polynomial regres- sion are presented. bBoth r2 and S have been recomputed using antilogs of the log Y; much of the difference between r2(log Y) and r2 (antilog Y) may be due to rounding. **0<.O1 value for the increase in r2 for the cubic component is also significant (p<.01). The average errors for the extrapolated Y values (Sext(yx)) according to the method employed are (a) linear, 2.2; (b) quadratic, 3.1; (c) cubic, .5; and (d) log-linear, 2.1. Observed Y's and predicted values of Y for both the original regression and the extrapolated lines are presented by method in Table 11. These data are graphically repre- sented in Figure 6. For this set of data, the cubic form of the polynomial regression is a superior predictor of the observed Y's. This method accounts for 92% of the variance in Y with a S of y. x .34; the Sext(y.x) is .5, considerably less than the other three methods with average error ranging from 2.1 to 3.1. The quadratic form is definitely the least appropriate method for this set of data since the curve bends in an opposite direction to the observed Y values (see Figure 6). Indicator 6 The mean for the Y values used to generate the regres- sion equations is 16.9; the standard deviation, 1.3. The following regression equations were used to derive Y: Linear Y = 18.56785 + (-.36786)X Quadratic Y = 21.30892 + (-2.01250)X + (.18274)X2 Cubic Y = 22.82142 + (-3.58611)X + .59524 X2 + (-.03056)X3 Log-linear log Y = 1.26761 + (-.00900)X. The goodness of fit of the regression lines derived from Table 11 Indicator 5: Observed Y's and Predicted Y's by Method Predicted Y's by Method Year(X) Observed Y's Linear Quadratica Cubica Log-linear Original regression b 1930(1) 2.5 4.1c 3.3 2.4 3.7 1940(11) 5.8 4.3 4.6 6.3 4.0 1950(21) 6.4 4.5 5.2 5.7 4.2 1960(31) 3.9 4.6 5.1 4.2 4.5 1963(34) 4.1 4.7 4.9 4.1 4.6 1964(35) 4.1 4.7 4.9 4.2 4.6 1965(36) 4.2 4.7 4.8 4.3 4.7 1966(37) 4.2 4.7 4.7 4.4 4.7 1967(38) 4.6 4.7 4.6 4.6 4.7 1968(39) 4.9 4.8 4.5 4.8 4.8 1969(40) 5.2 4.8 4.4 5.1 4.8 1970(41) 5.5 4.8 4.2 5.4 4.8 Extrapolationd 1971(42) 6.1 4.8 4.2 5.6 4.9 1972(43) 6.9 4.8 4.1 6.1 4.9 1973(44) 7.1 4.8 3.9 6.6 4.9 1974(45) 7.2 4.9 3.8 7.2 5.0 1975(46) 7.5 4.9 3.6 7.9 5.0 Note. Indicator 5 is number of dissolutions of marriage in Florida expressed as rate per 1,000 population. aBoth quadratic and cubic forms of polynomial regression are presented. bThe regression line is derived from 2/3 of the known data points. CPredicted Y's are rounded to number of places in original data. d Values are extrapolated beyond the data points used to generate the regression equation. 8.0-- Observed Y's 1-C ubic 0 Linear 4-- Quadratic M -Log-linear 5.0 0 S5.01 4-I 2.0 I 1930 '40 50 '60 '63 '65 70 75 Figure 6. Indicator 5: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extrapolation). these equations is indicated by r2, r2 change, and S in y.x Table 12. An ANOVA summary table is presented in Table 23 in the Appendix. The overall F's for the quadratic and cubic forms of polynomial regression are significant (p<.01); only the increase in r2 due to the quadratic component is signifi- cant (p<.01), however. The average errors for the extrapolated Y values (Sext(yx)) according to the method employed are (a) linear, 1.2; (b) quadratic, 7.8; (c) cubic, 1.5; and (d) log-linear, 1.4. Observed Y's and predicted values of Y for both the original regression and the extrapolated lines are presented by method in Table 13. These data are graphically repre- sented in Figure 7. Thus, while the quadratic and cubic forms of polynomial regression best fit the observed Y's for the original regres- sion, they do not continue to be superior predictors. In fact, the quadratic form has a Sext(y.x) of 7.8 while the Sext(y.x) for the other three methods ranges from 1.2 to 1.5. No method is clearly the best predictor of Y when values are extrapolated beyond the original regression. It should be noted that the Durbin-Watson d for the linear and log-linear methods approaches the lower limits of d and the possibility of serial correlation of the resid- uals cannot be overlooked. Because of the small number of observations involved in this data set (N = 8), interpreta- tion of the Durbin-Watson d is more suggestive than con- clusive. Table 12 Indicator 6: Summary Statistics for Prediction Equations by Method r2 r change F df S y.x Linear .46628 5.24 1,6 1.04 Quadratica .92655 31.54** 2,5 .42 .46026 31.33"* 1,5 Cubic a .97205 46.37** 3,4 .29 .04550 6.51 1,4 Log-linearb log Y .45957 ^ 5.10 1,6 log.02582 antilog Y .41427 1.03 Note. Indicator 6 is number of resident live births in Florida expressed as rate per 1,000 population. a Both quadratic and cubic forms of the polynomial regres- sion are presented. bBoth r2 and S have been recomputed using antilogs of ^ y.x^ the log Y; much of the difference between r2(log Y) and r2 (antilog Y) may be due to rounding. **<.01 Table 13 Indicator 6: Observed Y's and Predicted Y's by Method Predicted Y's by Method Year(X) Observed Y's Linear Quadratica Cubica Log-linear Original regression 1964(l) 19.7 18.2c 19.5 19.8 18.1 1965(2) 17.9 17.8 18.0 17.8 17.8 1966(3) 16.8 17.5 16.9 16.6 17.4 1967(4) 15.9 17.1 16.2 16.0 17.0 1968(5) 15.7 16.7 15.8 16.0 16.7 1969(6) 16.1 16.4 15.8 16.1 16.4 1970(7) 16.8 16.0 16.1 16.4 16.0 1971(8) 16.4 15.6 16.9 16.6 15.7 Extrapolationd 1972(9) 14.8 15.3 18.0 16.5 15.4 1973(10) 13.7 14.9 19.5 15.9 15.1 1974(11) 13.4 14.5 22.1 14.7 14.7 1975(12) 12.5 14.1 23.5 12.7 14.4 Note_. Indicator 6 is number of resident live births in Florida expressed as rate per 1,000 population. aBoth quadratic and cubic forms of polynomial regression are presented. bThe regression line is derived from 2/3 of the known data points. CPredicted Y's are rounded to number of places in original data. dValues are extrapolated beyond the data points used to generate the regression equation. 25 24 0 *.so Observed Y Cubic Linear Is 23-- 'uaciratic ---Locg-linear 22- 4-' ~21 S20 -C 4- l19 -J > 16- 15- 14 - 13- 12- 10u- 1964 '65 '66 '67 '68 '69 '70 '71 '72 '73 '74 175 Figure 7.. Indicator 6: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extrapolation). Indicator 7 The mean and standard deviation for the Y values used to generate the regression equations are 32.2 and 4.8, re- spectively. The following regression equations were used to derive Y: Linear Y = 23.42857 + 1.95476 X Quadratic Y = 23.58928 + 1.85833 X + .01071 X2 Cubic Y = 23.26430 + 2.19645 X + (-.07792)X2 + .00657 X3 Log-linear log Y = 1.38414 + .02662 X. The goodness of fit of the regression lines derived from these equations is indicated by r2, r2 change, and S in Table 14. y-x An ANOVA summary table is presented in Table 24 in the Appen- dix. The overall F's for all methods are significant (p<.01); however, the increases in r2 due to the higher order polyno- mials are not significant. The average errors for the extrapolated Y values (Sext(y.x)) according to the method employed are (a) linear, 1.4; (b) quadratic, 1.3; (c) cubic, 1.8; and (d) log-linear, 2.4. Observed and predicted values of Y for both the original regression and the extrapolated lines are presented by method in Table 15. These data are graphically represented in Figure 8. There is very little difference in the predictive value of the methods for the original regression. Each method accounts for 99% of the variance in Y; the range of the S Y.x for all methods is from .34 to .43. When the lines are extrapolated beyond the original Table 14 Indicator 7: Summary Statistics for Prediction Equations by Method r2 r2 change F df S y.x Linear .99560 1358.03** 1,6 .34 Quadratica .99572 581.74** 2,5 .37 .00012 .14 1,5 Cubica .99588 322.27** 3,4 .41 .00016 .15 1,4 Log-linearb log Y .99333 ^ 893.91** 1,6 log.00577 antilog Y .99999 .43 Note. Indicator 7 is number of 3 to 5 year olds enrolled in nursery school and kindergarten expressed as percentage of total children 3 to 5 years old in the United States. a Both quadratic and cubic forms of the polynomial regres- sion are presented. bBoth r2 and Syx have been recomputed using antilogs of the log Y; much of the difference between r2(log Y) and r2 (antilog Y) may be due to rounding. **D<. 01 |

Full Text |

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n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nV (FRORJ\ RI (GXFDWLRQ 0RGHO 2SHUDWLRQDO 'HILQLWLRQ RI 9DULDEOHV DV 7LPH 6HULHV ,QGLFDWRUV &+$37(5 ,,, 5$7,21$/( )25 (;75$32/$7,9( 0(7+2'6 6(/(&7(' )25 &203$5,621 ,9 PAGE 5 7$%/( 2) &217(176 FRQWLQXHGf 3DJH 2YHUYLHZ RI ([WUDSRODWLYH )RUHFDVWLQJ 0HWKRGV (FRQRPLF DQG %XVLQHVV )RUHFDVWLQJ 7HFKQRORJLFDO )RUHFDVWLQJ (GXFDWLRQDO )RUHFDVWLQJ ([WUDSRODWLYH 0HWKRGV LQ 2WKHU $UHDV $SSOLFDELOLW\ RI 5HYLHZHG ([WUDSRODWLYH 0HWKRGV IRU 6WXG\ 7KH 3DWWHUQ RI WKH 'DWD 7KH &ODVV RI 0RGHO 'HVFULSWLRQ RI 0HWKRGV WR EH &RPSDUHG 7KH *HQHUDO /LQHDU 0RGHO 7KH $VVXPSWLRQV RI WKH /LQHDU 0RGHO &ULWHULD IRU &RPSDULVRQ RI 0HWKRGV 0HWKRG 6LPSOH /LQHDU 5HJUHVVLRQ 0HWKRG /RJOLQHDU 5HJUHVVLRQ 0HWKRG 3RO\QRPLDO 5HJUHVVLRQ &+$37(5 ,9 &203$5,621 2) (;75$32/$7,9( 0(7+2'6 86,1* 6(/(&7(' 7,0( 6(5,(6 ,1',&$7256 3UHVHQWDWLRQ RI 5HVXOWV ,QGLFDWRU ,QGLFDWRU } ,QGLFDWRU ,QGLFDWRU ,QGLFDWRU ,QGLFDWRU ,QGLFDWRU ,QGLFDWRU &+$37(5 9 ',6&866,21 7KH 9DULDEOHV 6HOHFWLRQ %URQIHQEUHQQHUnV (FRORJ\ RI (GXFDWLRQ 0RGHO 2SHUDWLRQDO 'HILQLWLRQ RI 9DULDEOHV 7KH ([WUDSRODWLYH 0HWKRGV 6WDWLVWLFDO &RQVLGHUDWLRQV 3UDFWLFDO &RQVLGHUDWLRQV &+$37(5 9, 6800$5< &21&/86,216 $1' ,03/,&$7,216 2) 678'< 6XPPDU\ 7KH 9DULDEOHV 7KH 0HWKRGV 5HVXOWV &RQFOXVLRQV 6XJJHVWLRQV IRU )XWXUH 5HVHDUFK ,PSOLFDWLRQV IRU 3ODQQHUV DQG 3ROLF\ 0DNHUV Y PAGE 6 7$%/( 2) &217(176 FRQWLQXHGf 3DJH $33(1',; 5()(5(1&( 127(6 5()(5(1&(6 %,2*5$3+,&$/ 6.(7&+ PAGE 7 /,67 2) 7$%/(6 3DJH 7DEOH 7LPH 6HULHV ,QGLFDWRUV RI 6RFLDO 9DULDEOHV $IIHFWLQJ 2XWFRPHV RI (GXFDWLRQ 7DEOH ,QGLFDWRU 6XPPDU\ 6WDWLVWLFV IRU 3UHGLFWLRQ (TXDWLRQV E\ 0HWKRG 7DEOH ,QGLFDWRU 2EVHUYHG PAGE 8 /,67 2) 7$%/(6 FRQWLQXHGf 3DJH 7DEOH ,QGLFDWRU 6XPPDU\ 6WDWLVWLFV IRU 3UHGLFWLRQ (TXDWLRQV E\ 0HWKRG 7DEOH ,QGLFDWRU 2EVHUYHG PAGE 9 /,67 2) ),*85(6 3DJH )LJXUH %URQIHQEUHQQHUnV HFRORJLFDO VWUXFWXUH RI WKH HGXFDWLRQDO HQYLURQPHQW %DVHG XSRQ %URQIHQEUHQQHUnV >@ GHVFULSWLRQ SS f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n DEOHV RSHUDWLRQDOO\ GHILQHG DV WLPH VHULHV LQGLFDWRUV FRXSOHG ZLWK DQ HPSLULFDOO\ VRXQG EDVLV IRU IRUHFDVWLQJ IXWXUH WUHQGV LQ VXFK LQGLFDWRUV $V HYLGHQFH RI WKH QHHG IRU GHYHORSLQJ VXFK D VRFLDO IRUHFDVWLQJ IUDPHZRUN IRU HGXFDWLRQ VWDWHV LQFOXGLQJ )ORULGD KDYH SURYLGHG JUDQWV IRU WKDW SXUSRVH 7KLV VWXG\ ZDV RQH DVSHFW RI VXFK D JUDQW 7KH SUREOHP LQ WKLV VWXG\ ZDV Df WR VHOHFW XVLQJ %URQIHQEUHQQHUnV HFRORJ\ RI HGXFDWLRQ PRGHO DQG RSHUDWLRQn DOO\ GHILQH DW OHDVW YDULDEOHV WKDW UHVHDUFK KDV VKRZQ WR EH UHODWHG WR WKH RXWFRPHV RI HGXFDWLRQ Ef WR XVH WKHVH YDULDEOHV RSHUDWLRQDOO\ GHILQHG DV WLPH VHULHV LQGLFDWRUV LQ WKH FRPSDULVRQ RI WKUHH SXUHO\ H[WUDSRODWLYH IRUHFDVWLQJ [ PAGE 11 PHWKRGV DQG Ff WR GHULYH LPSOLFDWLRQV IRU WKH XVH RI DQ HFRORJLFDO PRGHO VXFK DV %URQIHQEUHQQHUnV WLPH VHULHV LQGLFDWRUV DQG VHOHFWHG H[WUDSRODWLYH PHWKRGV IRU HGXFDn WLRQDO SODQQLQJ 7KH VWXG\ ZDV FRQGXFWHG LQ WKH IROORZLQJ SKDVHV 8VLQJ %URQIHQEUHQQHUnV HFRORJ\ RI HGXFDWLRQ PRGHO YDULDEOHV WKDW UHVHDUFK KDV VKRZQ WR EH UHODWHG WR WKH RXWFRPHV RI HGXFDWLRQ ZHUH VHOHFWHG DQG ZKHUH SRVVLEOH ZHUH RSHUDWLRQDOO\ GHILQHG DV VWDWH DQGRU QDWLRQDO WLPH VHULHV LQGLFDWRUV 'DWD ZHUH FROOHFWHG IRU WKHVH LQGLFDn WRUV HLJKW ZKLFK PHW WKH FULWHULD HVWDEOLVKHG LQ WKLV VWXG\ ZHUH XVHG LQ WKH FRPSDULVRQ RI H[WUDSRODWLYH WHFKQLTXHV 7KUHH SXUHO\ H[WUDSRODWLYH WHFKQLTXHV GHULYHG IURP WKH JHQHUDO OLQHDU PRGHO ZHUH FRPSDUHG DFFRUGLQJ WR VWDWLVn WLFDO FULWHULD DQG SUDFWLFDO FRQVLGHUDWLRQV GHULYHG IURP WKH OLWHUDWXUH LQ VWDWLVWLFV HFRQRPLFV WLPH VHULHV DQDO\VLV DQG IRUHFDVWLQJ PHWKRGRORJ\ 7KH PHWKRGV ZHUH Df OLQHDU UHJUHVVLRQ Ef FXUYLOLQHDU UHJUHVVLRQ TXDGn UDWLF DQG FXELF IRUPVf DQG Ff ORJOLQHDU UHJUHVVLRQ GHn SHQGHQW YDULDEOH XQGHUJRHV ORJDULWKPLF WUDQVIRUPDWLRQf (DFK PHWKRG ZDV DSSOLHG WR HDFK WLPH VHULHV LQGLFDWRU 7LPH LQ \HDUV ZDV XVHG DV WKH LQGHSHQGHQW YDULDEOH WKH DQQXDO PHDVXUH RI WKH LQGLFDWRU ZDV WUHDWHG DV WKH GHSHQGHQW YDULn DEOH (DFK GDWD VHW ZDV GLYLGHG LQWR WKLUGV WZRWKLUGV RI WKH GDWD SRLQWV ZHUH XVHG WR HVWDEOLVK WKH SUHGLFWLRQ HTXDWLRQ 7KLV HTXDWLRQ ZDV XVHG WR SUHGLFW WKH UHPDLQLQJ WKLUG RI WKH GDWD SRLQWV 3UHGLFWHG YDOXHV ZHUH FRPSDUHG ZLWK DFWXDO YDOXHV [ L PAGE 12 ,PSOLFDWLRQV IRU WKH XVH LQ HGXFDWLRQDO SODQQLQJ RI DQ HFRORJLFDO PRGHO VXFK DV %URQIHQEUHQQHUnV WLPH VHULHV LQGLFDWRUV DQG VHOHFWHG H[WUDSRODWLYH WHFKQLTXHV ZHUH GLVn FXVVHG 5HVXOWV RI WKH PHWKRG FRPSDULVRQ ZHUH Df QR PHWKRG ZDV D VXSHULRU SUHGLFWRU IRU DOO LQGLFDWRUV Ef HDFK PHWKRG ZDV D VXSHULRU SUHGLFWRU IRU DW OHDVW RQH LQGLFDWRU DQG Ff WKH VXPPDU\ VWDWLVWLFV IRU WKH RULJLQDO UHJUHVVLRQ ZHUH QRW FRQVLVWHQWO\ UHODWHG WR WKH DFFXUDF\ RI WKH H[WUDn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f .HSSHO LQ 7KRPDV /DUVRQ f DFNQRZOHGJHG RQH RI WKH UHDVRQV IRU WKLV FRQWLQXLQJ LQWHUHVW LQ VRFLHWDO WUHQGV E\ HGXFDWLRQDO SODQQHUV DQG SROLF\PDNHUV WKH LPSHWXV IRU FKDQJH LQ HGXFDWLRQDO LQVWLWXWLRQV IURP WKH SUHVFKRRO WKURXJK WKH XQLYHUVLW\ LV PRUH OLNHO\ WR GHULYH IURP FKDQJHV LQ WKH ZLGHU VRFLHW\ WKDQ IURP IRUFHV ZLWKLQ WKH LQVWLWXWLRQV )RUHZRUGf $GGLWLRQDOO\ .HSSHO QRWHG WKDW HGXFDWLRQDO SROLF\ PXVW EH IRUPHG LQ FRQFHUW ZLWK RWKHU DVSHFWV RI SXEOLF SROLF\ DQG SURJUDP GHYHORSPHQW )RUHZRUGf %URQIHQEUHQQHU f SURSRVHG DQ HFRORJLFDO VWUXFWXUH RI WKH HGXFDWLRQDO HQYLURQPHQW ZKLFK PXVW EH WDNHQ LQWR DFFRXQW LI DQ\ SURJUHVV LQ WKH VFLHQWLILF VWXG\ RI HGXFDn WLRQDO V\VWHPV DQG SURFHVVHV S f LV WR EH PDGH %URQIHQn EUHQQHU VWDWHG PAGE 15 :KHWKHU DQG KRZ SHRSOH OHDUQ LV D IXQFWLRQ RI VHWV RI IRUFHV RU V\VWHPV DW WZR OHYHOV D 7KH ILUVW FRPSULVHV WKH UHODWLRQV EHWZHHQ FKDUDFWHULVWLFV RI OHDUQHUV DQG WKHLU VXUn URXQGLQJV LQ ZKLFK WKH\ OLYH RXW WKHLU OLYHV HJ KRPH VFKRRO SHHU JURXS ZRUN SODFH QHLJKERUKRRG FRPPXQLW\f E 7KH VHFRQG HQFRPSDVVHV WKH UHODWLRQV DQG LQWHUFRQQHFWLRQV WKDW H[LVW EHWZHHQ WKHVH HQYLURQPHQWV S f %XLOGLQJ DQG HPSOR\LQJ RQ /HZLQnV WKHRU\ RI WRSRORJLFDO WHUULWRULHV D WHUPLQRORJ\ DGDSWHG IURP %ULP f %URQIHQ EUHQQHU IXUWKHU HODERUDWHG WKDW WKH FRQVWUXFW HQYLURQPHQW FDQ EH FRQFHLYHG WRSRORJLFDOO\ DV D QHVWHG DUUDQJHPHQW RI VWUXFWXUHV HDFK FRQWDLQHG ZLWKLQ WKH QH[W S f f $ PLFURV\VWHP LV DQ LPPHGLDWH VHWWLQJ FRQWDLQLQJ WKH OHDUQHU f 7KH PHVR V\VWHP FRPSULVHV WKH LQWHU UH OD)7RQV7LLSVDPRQJ WKH PDMRU VHWWLQJV FRQWDLQLQJ WKH OHDUQHU DW D SDUWLFXODU SRLQW LQ KLV RU KHU OLIH D V\VWHP RI PLFURV\VWHPV f 7KH H[R V\VWHP LV DQ H[WHQVLRQ RI WKH PHVRV\VWHP HPEUDFLQJ WKH FRQFUHWH VRFLDO VWUXFWXUHV ERWK IRUPDO DQG LQIRUPDO WKDW LPSLQJH XSRQ RU HQFRPSDVV WKH LPPHGLDWH VHWWLQJV FRQWDLQLQJ WKH OHDUQHU DQG WKHUHn E\ LQIOXHQFH DQG HYHQ GHWHUPLQH RU GHOLPLW ZKDW JRHV RQ WKHUH 7KHVH VWUXFWXUHV LQn FOXGH WKH PDMRU LQVWLWXWLRQV RI VRFLHW\ ERWK GHOLEHUDWHO\ VWUXFWXUHG DQG VSRQWDQHn RXVO\ HYROYLQJ DV WKH\ RSHUDWH DW WKH ORFDO FRPPXQLW\ OHYHO f 0DFUR V\VWHPV DUH WKH RYHUDUFKLQJ LQVWLn WXWLRQV RI WKH FXOWXUH RU VXEFXOWXUH VXFK DV WKH HFRQRPLF VRFLDO HGXFDWLRQDO OHJDO DQG SROLWLFDO V\VWHPV RI ZKLFK ORFDO PLFUR PHVR DQG H[RV\VWHPV DUH WKH FRQFUHWH PDQLIHVWDWLRQV SS f 6HH )LJXUH IRU D UHSUHVHQWDWLRQ RI WKHVH LGHDVf PAGE 16 )LJXUH %URQIHQEUHQQHUnV HFRORJLFDO VWUXFWXUH RI WKH HGXFDWLRQDO HQYLURQPHQW %DVHG XSRQ %URQIHQ EUHQQHUnV >@ GHVFULSWLRQ SS f PAGE 17 7KH )XWXUHVr 3HUVSHFWLYH LQ (GXFDWLRQDO 3ODQQLQJ :KLOH %URQIHQEUHQQHU SURSRVHG KLV HFRORJLFDO VWUXFWXUH SULPDULO\ DV D IUDPHZRUN IRU OHDUQLQJ UHVHDUFK HIIRUWV WKDW LV IRU H[DPLQLQJ UHODWLRQVKLSV DPRQJ YDULDEOHV DVVRFLDWHG ZLWK OHDUQLQJ RWKHUV +DUPDQ :HEVWHU f KDYH GLVn FXVVHG WKH VRFLHWDO FRQWH[W RI HGXFDWLRQ DV D IUDPHZRUN IRU IXWXUHRULHQWHG HGXFDWLRQDO SODQQLQJ ,QGHHG WKLV HPSKDVLV RQ IXWXUH DZDUHQHVV KDV HYROYHG LQWR D VLJQLILFDQW PRYHPHQW ZLWKLQ HGXFDWLRQ UHIHUUHG WR DV HGXFDWLRQDO IXWXULVP +HQFOH\ PAGE 18 :HEVWHU DOVR QRWHG WKH IXWXUHV SHUVSHFWLYH LPSOLHV WKDW ZH QRW MXVW DWWHQG WR DOWHUQDWLYHV LQ DQG IRU HGXFDn WLRQ EXW DOVR FRQVLGHU WKH VRFLHWDO FRQWH[W LQ PRUH FRPSUHKHQVLYH IDVKLRQ WKDQ LV XVXDO LQ HGXFDWLRQDO SODQQLQJ S f ,Q RUGHU WR DVVLVW GHFLVLRQ PDNHUV LQ WKH VHOHFWLRQ RI DOWHUQDWLYHV ZKLFK KDYH SRVLWLYH IXWXUH FRQVHTXHQFHV IRU VRFLHW\ HGXFDWLRQDO SODQQHUV DW ERWK QDWLRQDO DQG VWDWH OHYHOV PXVW WDNH LQWR DFFRXQW WKRVH VRFLHWDO IRUFHV ZKLFK DIIHFW QRW RQO\ WKH RXWFRPHV RI HGXFDWLRQ EXW DOVR WKH SXUn SRVHV RI HGXFDWLRQ DQG WKH KXPDQ DQG PDWHULDO UHVRXUFHV DYDLODEOH WR WKH HGXFDWLRQDO SURFHVV 7R GR WKLV KRZHYHU WKH SODQQHU PXVW GHOLQHDWH WKH VRFLHWDO IDFWRUV RU YDULDEOHV WR EH LQFOXGHG LQ WKH SODQQLQJ SURFHVV DQG GHYHORS D VRXQG UDWLRQDOH EDVHG RQ UHVHDUFK DQG WKHRU\ IRU VXFK LQFOXVLRQ 7KHQ WUHQGVSDVW SUHVHQW DQG IXWXUHLQ WKHVH YDULDEOHV PD\ EH H[DPLQHG LQ RUGHU WR GHULYH LPSOLFDWLRQV IRU HGXFDn WLRQDO SODQQLQJ DQG SROLF\ )RUHFDVWLQJ 7UHQGV LQ 6RFLDO 9DULDEOHV $YDLODEOH WR WKH HGXFDWLRQDO SODQQHU LQ WKLV XQGHUWDNLQJ DUH D QXPEHU RI SUHGLFWLYH DQG KHXULVWLF GHYLFHV WR H[SORUH DOWHUQDWLYH IXWXUHV ZKLFK KDYH EHHQ GHYHORSHG E\ JRYHUQPHQW LQGXVWU\ QRQSURILW RUJDQL]DWLRQV DQG IXWXUHV FRQVXOWLQJ JURXSV 7KHVH IRUHFDVWLQJ WHFKQLTXHV FDQ EH FDWHJRUL]HG LQWR H[SORUDWRU\ IRUHFDVWLQJ PHWKRGV DQG QRUPDWLYH IRUHFDVWLQJ PHWKRGV ([SORUDWRU\ IRUHFDVWLQJ PHWKRGV VWDUW IURP WKH SUHVHQW VLWXDWLRQ DQG LWV SUHFHGLQJ KLVWRU\ DQG DWWHPSW WR SURMHFW IXWXUH PAGE 19 GHYHORSPHQWV 1RUPDWLYH IRUHFDVWV RQ WKH FRQWUDU\ VWDUW ZLWK VRPH GHVLUHG RU SRVn WXODWHG IXWXUH VLWXDWLRQ DQG ZRUN EDFNn ZDUGV WR GHULYH IHDVLEOH URXWHV IRU WKH WUDQVLWLRQ IURP WKH SUHVHQW WR WKH GHVLUHG IXWXUH 0DUWLQR S f ([SORUDWRU\ IRUHFDVWLQJ PHWKRGV DOO RI ZKLFK DUH EDVHG XSRQ H[WUDSRODWLRQ RI VRPH NLQG LQFOXGH Df SXUHO\ H[WUDn SRODWLYH PHWKRGV Ef H[SODQDWRU\ PHWKRGV DQG Ff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n WLRQ RI IXWXUH VRFLHWDO WUHQGV FDQ SURYLGH WKH LPSHWXV WR H[DPLQH SUHVHQW SROLF\ DQG WR DQDO\]H WKH FRQVHTXHQFHV RI FRQWHPSODWHG FKDQJHV 7KLV DSSURDFK QHHG QRW EH RQO\ SUHn YHQWLYH IRUHFDVWLQJ LQ WKH VHQVH XVHG E\ =LHJOHU f RI SUHYHQWLQJ XQGHVLUDEOH IRUHFDVWV ,W PD\ DOVR EH H[WHQGHG WR H[DPLQH DOO FRQVHTXHQFHV RI DFWLRQ RU LQWHUYHQWLRQ LQn WHQGHG RU QRW 3XUHO\ H[WUDSRODWLYH PHWKRGV ZKHQ FRPELQHG ZLWK DX[LOLDU\ PHWKRGV VXFK DV WUHQGLPSDFW DQDO\VLV FURVVn LPSDFW PDWULFHV RU VFHQDULRV FDQ SURYLGH D YHKLFOH IRU H[n SORULQJ WKH UHODWLRQVKLSV DPRQJ LGHQWLILHG IXWXUH SDWWHUQV LQ VRFLHW\ PAGE 20 :KLOH WKH XVH RI SXUHO\ H[WUDSRODWLYH PHWKRGV ZLWK WLPH VHULHV GDWD LV IDLUO\ ZHOO GHILQHG LQ WHFKQRORJLFDO DQG HFRQRPLF DUHDV WKHLU DSSOLFDWLRQ WR VRFLDO IRUHFDVWLQJ KDV QRW EHHQ WKH IRFXV RI VLJQLILFDQW GHILQLWLYH VWXG\ ,QGHHG +DUULVRQ f HPSKDVL]HG WKH QHHG IRU VXFK UHVHDUFK VSHFLn ILFDOO\ WKH FRQVLGHUDWLRQ RI HDFK PHWKRG LQ WHUPV RI VRPH DVSHFW RI WKH VRFLDO SURFHVV LW ZRXOG OLNHO\ EH DSSOLHG WR S f )RU DV +DUULVRQ H[SODLQHG ZKLOH VRPH SUREOHPV LQ UHJUHVVLRQ DQG WLPH VHULHV DQDO\VLV ZKLFK UHPDLQ XQUHVROYHG DUH FXUUHQWO\ WKH FRQFHUQ RI VWDWLVWLFLDQV DQG PDWKHPDWLFLDQV LW DSSHDUV WKDW UHVROXWLRQ PLJKW EHVW OLH LQ WHUPV RI LQYHVn WLJDWLRQ LQ FRQFUHWH DSSOLFDWLRQ FDVHV S f ,Q VRFLDO IRUHFDVWLQJ WKHUH LV D JUHDW QHHG LQ DOPRVW DOO WKH NQRZQ H[WUDSRODWLYH PHWKRGV IRU DQ H[SOLFLW VWDWHPHQW RI WKH DOJRULWKPLF WKHRUHWLFDO DQG HPSLULFDO ZHDNQHVVHV RU VHQVLWLYLWLHV RI VXFK SURFHGXUHV 6XFK D GLVFXVVLRQ DV QRWHG ZRXOG EH PRUH PHDQn LQJIXO LI FDUULHG RQ LQ WKH FRQWH[W RI DQ DQDO\VLV RI VRPH VSHFLILF DVSHFW RU DVSHFWV RI VRFLDO SURFHVV +DUULVRQ S f 2QO\ WKURXJK HPSLULFDO VWXG\ RI WKH SHUIRUPDQFH RI YDULRXV H[WUDSRODWLYH PHWKRGV DSSOLHG WR SDUWLFXODU VRFLDO SKHQRPHQD ZLOO D EDVLV IRU VHOHFWLRQ RI DSSURSULDWH DQG DFFXUDWH WHFKn QLTXHV EH IRUPXODWHG 7KH 1HHG IRU 5HVHDUFK 6LQFH WKHUH DUH QR ZLGHO\DFFHSWHG SODQQLQJ PRGHOV LQn FRUSRUDWLQJ TXDQWLWDWLYH GDWD RQ VRFLDO YDULDEOHV WKH HGXn FDWLRQDO SODQQHU ZKR ZDQWV WR XWLOL]H VXFK LQIRUPDWLRQ LV FRQIURQWHG ZLWK D QXPEHU RI TXHVWLRQV UHODWHG WR Df WKH PAGE 21 LGHQWLILFDWLRQ RI VRFLDO YDULDEOHV WR EH LQFOXGHG Ef WKH RSHUDWLRQDO GHILQLWLRQ RI VRFLDO YDULDEOHV LQ WHUPV RI WLPH VHULHV LQGLFDWRUV Ff WKH VHOHFWLRQ RI D SXUHO\ H[WUDSRODWLYH WHFKQLTXH ZKLFK ZLOO \LHOG WKH PRVW DFFXUDWH IRUHFDVW IRU D VSHFLILF LQGLFDWRU DQG Gf WKH XWLOL]DWLRQ RI WKHVH IRUHFDVWV LQ WKH SODQQLQJ SURFHVV $QVZHUV UHTXLUH IXWXUHV UHVHDUFK ZKLFK LV GHULYHG IURP D FRQFHSWXDOO\ VRXQG IUDPHZRUN DQG LV SXUVXHG ZLWK PHWKRGRORJLFDO YLJRU $V HYLGHQFH RI WKH LPSRUn WDQFH RI VXFK LQYHVWLJDWLRQ WR WKH HGXFDWLRQDO SODQQHU WKH 6WDWH RI )ORULGD WKURXJK WKH 2IILFH RI 6WUDWHJ\ 3ODQQLQJ LQ WKH 'HSDUWPHQW RI (GXFDWLRQ IXQGHG LQ D VRFLDO IRUHn FDVWLQJ SURMHFW 67$5 3URMHFW 1R 5f DW WKH 8QLYHUVLW\ RI )ORULGD IRU WKH VHFRQG \HDU 7KH VWXG\ GHVFULEHG KHUHLQ ZDV SDUW RI WKDW HIIRUW WR IRUHFDVW VRFLDO WUHQGV DIIHFWLQJ HGXFDWLRQ LQ )ORULGD 7R VXPPDUL]H (GXFDWLRQDO SODQQHUV DQG SROLF\ PDNHUV QHHG DGHTXDWH LQIRUPDWLRQ WR PDNH DSSURSULDWH GHFLVLRQV DERXW WKH UROH DQG IXQFWLRQ RI HGXFDWLRQ LQ FUHDWLQJ LPSURYHG TXDOLW\ RI OLIH IRU FLWL]HQV RI WKH IXWXUH 7KH 6WDWH RI )ORULGD LQ IXQGLQJ 67$5 3URMHFW 1R 5 RI ZKLFK WKLV VWXG\ LV D SDUW DFNQRZOHGJHG WKDW QHHG 3DUW RI WKLV LQn IRUPDWLRQ PD\ EH SURYLGHG WKURXJK WKH XVH RI FRQFHSWXDOO\ VRXQG VRFLDO DQG HGXFDWLRQDO YDULDEOHV RSHUDWLRQDOO\ GHILQHG DV WLPH VHULHV LQGLFDWRUV FRXSOHG ZLWK DQ HPSLULFDOO\ VRXQG EDVLV IRU IRUHFDVWLQJ IXWXUH VWDWHV RI VXFK LQGLFDWRUV PAGE 22 7KH 3UREOHP 7KH SUREOHP LQ WKLV VWXG\ ZDV Df WR VHOHFW XVLQJ %URQIHQEUHQQHUnV HFRORJ\ RI HGXFDWLRQ PRGHO DQG RSHUDWLRQDOO\ GHILQH DW OHDVW YDULDEOHV WKDW UHVHDUFK KDV VKRZQ WR EH UHODWHG WR WKH RXWFRPHV RI HGXFDWLRQ Ef WR XVH WKHVH YDULDEOHV RSHUDWLRQDOO\ GHILQHG DV WLPH VHULHV LQGLFDWRUV LQ WKH FRPSDULVRQ RI WKUHH SXUHO\ H[WUDSRODWLYH IRUHFDVWLQJ PHWKRGV DQG Ff WR GHULYH LPSOLFDWLRQV IRU WKH XVH RI DQ HFRn ORJLFDO PRGHO VXFK DV %URQIHQEUHQQHUnV WLPH VHULHV LQGLFDWRUV DQG VHOHFWHG H[WUDSRODWLYH PHWKRGV IRU HGXFDWLRQDO SODQQLQJ 'HOLPLWDWLRQV DQG /LPLWDWLRQV 7KH %URQIHQEUHQQHU HFRORJ\ RI HGXFDWLRQ PRGHO ZDV XVHG SULPDULO\ DV D IUDPHZRUN IRU WKH VHOHFWLRQ RI VRFLDO DQG HGXFDWLRQDO YDULDEOHV DQG ZDV QRW HYDOXDWHG LWVHOI LQ WKLV VWXG\ 7HQ YDULDEOHV HJ VRFLRHFRQRPLF VWDWXV RI IDPLO\ SHHU JURXS FKDUDFWHULVWLFVf ZHUH VHOHFWHG WR EH RSHUDWLRQDOO\ GHILQHG ZKHUH SRVVLEOH LQ WHUPV RI QDWLRQDO DQGRU VWDWH OHYHO WLPH VHULHV LQGLFDWRUV HJ PHGLDQ IDPLO\ LQFRPH MXYHQLOH FULPH UDWHVf 2I WKHVH LGHQWLILHG LQGLFDWRUV HLJKW ZKLFK PHW WKH IROORZLQJ FULWHULD ZHUH XVHG LQ WKH FRPSDULVRQ RI H[WUDSRODWLYH WHFKQLTXHV Df WKH LQGLFDWRU ZDV UHDGLO\ DYDLODEOH Ef WKH GDWD ZHUH DYDLODEOH IRU D \HDU RU JUHDWHU WLPH VSDQ DQG Ff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n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f WKHUH ZDV QR DWWHPSW WR SUHVHQW WKH FRPSDULVRQ RI WKHVH WHFKQLTXHV LQ WKH GHWDLO GHVLUHG E\ WKHVH GLVFLSOLQHV 5DWKHU WKH FRPSDULn VRQ ZDV PDGH LQ VXFK D ZD\ DV WR EH PRVW UHOHYDQW WR WKH SODQQHU LQ HGXFDWLRQ 7KHUH ZDV QR DWWHPSW WR ZULWH RU DGDSW FRPSXWHU SURJUDPV IRU YDULRXV WHFKQLTXHV ,QVWHDG DQ HIIRUW ZDV PDGH WR LGHQWLI\ DQG XWLOL]H FRPSXWHU SURJUDPV DQG VWDWLVWLFDO SDFNn DJHV ZKLFK KDG DOUHDG\ EHHQ DGDSWHG IRU XVH DW WKH 1RUWK (DVW 5HJLRQDO 'DWD &HQWHUn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n ODWHG E\ PHDQV DV YDULHG DV PXOWLSOH UHJUHVVLRQ DQDO\VLV WLPH VHULHV DQDO\n VLV HQYHORSH FXUYH ILWWLQJ WKUHHPRGH IDFWRU DQDO\VLV FRUUHODWLRQDO DQDO\VLV DYHUDJHV RU DQ\ RWKHU PHWKRG WKDW WDNHV FXUUHQW DQG KLVWRULFDO GDWD DV WKH SULQn FLSDO EDVLV IRU HVWLPDWLQJ IXWXUH VWDWHV LQ D JLYHQ YDULDEOH +DUULVRQ S f ,QGLFDWRU HGXFDWLRQDO (GXFDWLRQDO LQGLFDWRUV DUH VWDWLVWLFV WKDW HQDEOH LQWHUHVWHG SXEOLFV WR NQRZ WKH VWDWXV RI HGXFDWLRQ DW D SDUWLFXODU PRPHQW LQ WLPH ZLWK UHVSHFW WR VRPH VHOHFWHG YDULDEOHV WR PDNH FRPSDULVRQV LQ WKDW VWDWXV RYHU WLPH DQG WR SURMHFW IXWXUH VWDWXV ,QGLFDWRUV DUH WLPHVHULHV VWDWLVWLFV WKDW SHUPLW D VWXG\ RI WUHQGV DQG FKDQJH LQ HGXFDWLRQ *RROHU S ,' ,QGLFDWRU VRFLDO 7KH RSHUDWLRQDO GHILQLWLRQ RU SDUW RI WKH RSHUDWLRQDO GHILQLWLRQ RI DQ\ RQH RI WKH FRQFHSWV PAGE 25 FHQWUDO WR WKH JHQHUDWLRQ RI DQ LQIRUPDWLRQ V\VWHP GHVFULSn WLYH RI WKH VRFLDO V\VWHP &DUOLVOH S f WLPH VHULHV WKDW DOORZ FRPSDULVRQV RYHU DQ H[WHQGHG SHULRG ZKLFK SHUPLW RQH WR JUDVS ORQJWHUP WUHQGV DV ZHOO DV XQXVXDOO\ VKDUS IOXFWXDWLRQV 6KHOGRQ i )UHHPDQ S f D VWDWLVWLF RI GLUHFW QRUPDWLYH LQWHUHVW ZKLFK IDFLOLWDWHV FRQFLVH FRPSUHKHQVLYH DQG EDODQFHG MXGJPHQWV DERXW WKH FRQn GLWLRQ RI PDMRU DVSHFWV RI D VRFLHW\ 86 'HSDUWPHQW RI +HDOWK (GXFDWLRQ I :HOIDUH S f 2XWFRPHV RI HGXFDWLRQ 7KRVH PHDVXUHV RI SHUIRUPDQFH VXFK DV DFKLHYHPHQW WHVW VFRUHV RU XWLOL]DWLRQ VXFK DV HPSOR\PHQW UDWHV ZKLFK DSSHDU WR EH WKH UHVXOW RI SDUWLFLn SDWLRQ LQ WKH IRUPDO HGXFDWLRQDO SURFHVV 5HJUHVVLRQ OLQHDU 0RVW FRPPRQ W\SH RI UHJUHVVLRQ LQ ZKLFK WKH REMHFWLYH LV WR ORFDWH WKH EHVWILWWLQJ VWUDLJKW OLQH WKURXJK D VFDWWHUJUDP EDVHG RQ LQWHUYDOOHYHO YDULDEOHV 1LH +XOO -HQNLQV 6WHLQEUHQQHU %HQW S f 5HJUHVVLRQ ORJOLQHDU $V XVHG LQ WKLV VWXG\ D OHDVW VTXDUHV UHJUHVVLRQ PHWKRG LQ ZKLFK D JHRPHWULF VWUDLJKW OLQH LV ORFDWHG WKURXJK D VFDWWHUJUDP SORWWHG RQ VHPLORJDULWKPLF SDSHU DOVR FDOOHG H[SRQHQWLDO FXUYH RU WUHQG FXUYH 5HJUHVVLRQ SRO\QRPLDO RU FXUYLOLQHDU 5HJUHVVLRQ PHWKRG IRU ILWWLQJ D FXUYH WR D VHW RI GDWD XVLQJ WKH FULn WHULRQ RI OHDVW VTXDUHV GLVWDQFHV 1LH HW DO S f 7LPH VHULHV $ VHW RI REVHUYDWLRQV JHQHUDWHG VHTXHQn WLDOO\ LQ WLPH %R[ I -HQNLQV S f PAGE 26 3URFHGXUHV 7KH VWXG\ SURFHHGHG LQ WKH IROORZLQJ SKDVHV Df XVLQJ %URQIHQEUHQQHUnV HFRORJ\ RI HGXFDWLRQ PRGHO YDULDEOHV WKDW UHVHDUFK KDV VKRZQ WR EH UHODWHG WR WKH RXWFRPHV RI HGXFDWLRQ ZHUH VHOHFWHG DQG ZKHUH SRVVLEOH ZHUH RSHUDWLRQDOn O\ GHILQHG DV WLPH VHULHV LQGLFDWRUV Ef GDWD ZHUH FROOHFWHG IRU WKHVH WLPH VHULHV LQGLFDWRUV HLJKW RI ZKLFK ZHUH XVHG LQ WKH FRPSDULVRQ RI WKH VHOHFWHG H[WUDSRODWLYH WHFKQLTXHV Ff XVLQJ WKH VHOHFWHG WLPH VHULHV LQGLFDWRUV WKUHH SXUHO\ H[WUDSRODWLYH WHFKQLTXHV ZHUH FRPSDUHG DFFRUGLQJ WR VWDWLVn WLFDO FULWHULD DQG SUDFWLFDO FRQVLGHUDWLRQV GHULYHG IURP WKH OLWHUDWXUH DQG Gf LPSOLFDWLRQV IRU WKH XVH LQ HGXFDn WLRQDO SODQQLQJ RI DQ HFRORJLFDO PRGHO VXFK DV %URQIHQ EUHQQHUnV WLPH VHULHV VRFLDO DQG HGXFDWLRQDO LQGLFDWRUV DQG VHOHFWHG H[WUDSRODWLYH WHFKQLTXHV ZHUH GHULYHG 7KH 6HOHFWLRQ DQG 2SHUDWLRQDO 'HILQLWLRQ RI 9DULDEOHV 7KH ZRUN E\ &ROOD]R /HZLV DQG 7KRPDV f FRPSOHWHG GXULQJ WKH ILUVW \HDU RI 67$5 3URMHFW 1R 5 RQ IRUHn FDVWLQJ VHOHFWHG HGXFDWLRQDO RXWFRPHV IURP VRFLDO YDULDEOHV ZDV XWLOL]HG 6LQFH WKH YDULDEOHV VHOHFWHG E\ WKHVH LQYHVn WLJDWRUV ZHUH GHULYHG IURP D UHYLHZ RI WKH UHVHDUFK OLWHUDn WXUH DQG ZHUH DFNQRZOHGJHG WR EH DSSURSULDWH IRU WKH VWDWHG VRFLDO IRUHFDVWLQJ SXUSRVHV E\ D SDQHO RI H[SHUWV LQ YDULRXV GLVFLSOLQHV WKH\ DSSHDUHG WR IXOILOO WKH UHTXLUHPHQWV RI WKLV VWXG\ $GGLWLRQDOO\ HDFK RI WKH YDULDEOHV VHOHFWHG PAGE 27 IRU XVH ZDV GHVFULEHG DQG FODVVLILHG DFFRUGLQJ WR %URQIHQ EUHQQHUnV HFRORJ\ RI HGXFDWLRQ PRGHO )RU HDFK YDULDEOH DQ DWWHPSW ZDV PDGH WR LGHQWLI\ RQH RU PRUH W\SHV RI WLPH VHULHV LQGLFDWRUV ZKLFK PLJKW ORJLFDOO\ UHSUHVHQW WKH YDULDEOH )RU VRPH YDULDEOHV VHYHUDO LQGLFDn WRUV ZHUH LGHQWLILHG ZKLOH IRU RWKHUV QR LQGLFDWRU FRXOG ORJLFDOO\ EH LGHQWLILHG RU QR WLPH VHULHV GDWD ZHUH DYDLOn DEOH IRU WKH LQGLFDWRU DW WKH WLPH RI WKH VWXG\ 7KLV SKDVH RI WKH VWXG\ LV H[SODLQHG IXUWKHU LQ &KDSWHU ,, &ROOHFWLRQ RI 7LPH 6HULHV ,QGLFDWRU 'DWD 6RXUFHV RI QHHGHG WLPH VHULHV GDWD DW ERWK WKH QDWLRQDO DQG VWDWH OHYHO ZHUH LGHQWLILHG LQ VHYHUDO ZD\V 7KH H[SDQn GLQJ OLWHUDWXUH RQ VRFLDO WUHQGV HJ86 'HSDUWPHQW RI +HDOWK (GXFDWLRQ i :HOIDUH f DQG VSHFLILFDOO\ WKH OLWHUDWXUH RQ WKHVH VRFLDO WUHQGV RSHUDWLRQDOL]HG DV VRFLDO LQGLFDWRUV HJ ([HFXWLYH 2IILFH RI WKH 3UHVLGHQW 2IILFH RI 0DQDJHPHQW %XGJHW f ZDV UHYLHZHG )XUWKHUPRUH H[DPLQDWLRQ RI LQLWLDO HIIRUWV LQ XVLQJ WLPH VHULHV LQGLFDn WRUV UHODWHG WR HGXFDWLRQ E\ WKH 2IILFH RI 7HFKQRORJ\ $VVHVVn PHQW IRU WKH 8QLWHG 6WDWHV &RQJUHVV &RDWHV 1RWH f DQG VHYHUDO VWDWH GHSDUWPHQWV RI HGXFDWLRQ HJ 2UHJRQ 3HQQn V\OYDQLD )ORULGDf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f LQLWLDO LGHQWLILFDWLRQ DQG WHVWLQJ RI PHWKRGV XVLQJ GDWD VLPLODU LQ IRUP WR VHOHFWHG LQGLFDWRUV Ef UHFRQn VLGHUDWLRQ DQG WHVWLQJ RI DGGLWLRQDO DYDLODEOH PHWKRGV Ff VHOHFWLRQ RI WKUHH PHWKRGV WR EH XVHG IRU FRPSDUDWLYH H[WUDn SRODWLRQV Gf GHULYDWLRQ RI VSHFLILF FULWHULD DQG SUDFWLFDO FRQVLGHUDWLRQV IURP WKH OLWHUDWXUH Hf DSSOLFDWLRQ RI WKUHH PHWKRGV WR HDFK GDWD VHW If H[WUDSRODWLRQ RI LGHQWLILHG WUHQG LQWR IXWXUH XVLQJ HTXDWLRQ JHQHUDWHG LQ Hf DQG Jf FRPSDULVRQ RI DFWXDO YHUVXV SUHGLFWHG YDOXHV RI LQGLFDWRUV )URP D SUHOLPLQDU\ UHYLHZ RI WKH OLWHUDWXUH LQ VWDWLVn WLFV HFRQRPLFV WLPH VHULHV DQDO\VLV DQG IRUHFDVWLQJ PHWKRGRORJ\ WKH IROORZLQJ IRXU PHWKRGV ZHUH WHQWDWLYHO\ LGHQWLILHG IRU FRPSDULVRQ Df OLQHDU UHJUHVVLRQ FRPSXWHU SURJUDP E\ 1LH HW DO f Ef FXUYLOLQHDU RU SRO\QRPLDO UHJUHVVLRQ FRPSXWHU SURJUDP E\ 1LH HW DO f Ff %R[ -HQNLQV WLPH VHULHV DQDO\VLV FRPSXWHU SURJUDP E\ &RRSHU 1RWH f DQG Gf ),7 FXUYHILWWLQJ ZLWK ZHLJKWHG GDWD FRPn SXWHU SURJUDP E\ 6WRYHU 1RWH f $Q LQLWLDO DQDO\VLV RI WKH PHWKRGV XVLQJ WULDO VHWV RI GDWD FRPELQHG ZLWK D YLVXDO DQDO\VLV RI WKH JHQHUDO IRUP RI PAGE 29 WKH GDWD WR EH XVHG UHYHDOHG WKDW WZR RI WKH PHWKRGV XQGHU FRQVLGHUDWLRQ ZHUH LQDSSURSULDWH 7KH %R[-HQNLQV SURFHGXUH ZKLOH DQ H[WUHPHO\ SRZHUIXO WRRO IRU WLPH VHULHV DQDO\VLV RI GDWD ZKLFK DUH FKDUDFWHUL]HG E\ VHDVRQDO RU F\FOLF YDULDWLRQ XVXDOO\ UHVXOWLQJ LQ DXWRFRUUHODWLRQ RI REVHUYDWLRQV DQG UHVLGXDOVf GLG QRW VHHP VXLWDEOH IRU WKH VRFLDO LQGLFDWRU GDWD FROOHFWHG 6KRXOG VXEVHTXHQW WHVWV UHYHDO DXWRFRUUHODn WLRQ DQG KHQFH D YLRODWLRQ RI WKH DVVXPSWLRQV RI WKH OLQHDU PRGHO %R[-HQNLQV FRXOG WKHQ EH DSSURSULDWHO\ HPSOR\HGf 7KH ),7 FXUYHILWWLQJ SURFHGXUH XWLOL]LQJ D ZHLJKWHG GDWD SULQFLSOH ZDV UHMHFWHG EHFDXVH WKH FRPSXWHU SURJUDP UHTXLUHG H[WHQVLYH PRGLILFDWLRQ WR \LHOG QHFHVVDU\ FRPSDUDWLYH VWDWLVn WLFV DQG UHOLDEOH RXWSXW 7KHRUHWLFDO MXVWLILFDWLRQ IRU WKH ZHLJKWLQJ IRUPXOD DQG GDWD WUDQVIRUPDWLRQV HPSOR\HG ZDV XQDYDLODEOH 7KXV WZR RI WKH IRXU PHWKRGV WHQWDWLYHO\ FRQVLGHUHG ZHUH UHMHFWHG 6LQFH WKH FRPSDULVRQ SKDVH ZDV WR LQYROYH WKUHH PHWKRGV WKH OLWHUDWXUH ZDV DJDLQ VHDUFKHG IRU RWKHU DSSURSULDWH PHWKRGV 7KH PRVW SURPLVLQJ RI WKHVH ZDV D FXUYH ILWWLQJ WHFKQLTXH ZKLFK XWLOL]HV DQ H[SRQHQWLDO IXQFWLRQ WR GHVFULEH D FRQVWDQW JURZWK UDWH 7KLV PHWKRG FDOOHG ORJ OLQHDU UHJUHVVLRQ LQ WKLV VWXG\ FDQ EH GHVFULEHG LQ WHUPV RI WKH JHQHUDO OLQHDU PRGHO DQG VROYHG E\ OHDVW VTXDUHV SURn FHGXUHV ZKHQ WKH GHSHQGHQW YDULDEOH XQGHUJRHV ORJDULWKPLF WUDQVIRUPDWLRQ 6LQFH VRFLDO SKHQRPHQD VRPHWLPHV H[KLELW ZKDW DSSHDUV WR EH D FRQVWDQW JURZWK UDWH ORJOLQHDU UHJUHVn VLRQ VHHPHG WR EH DQ DSSURSULDWH PHWKRG WR LQFOXGH LQ WKLV VWXG\ PAGE 30 7KH WKUHH PHWKRGV ILQDOO\ VHOHFWHG IRU FRPSDULVRQ ZHUH Df OLQHDU UHJUHVVLRQ ZLWKRXW GDWD WUDQVIRUPDWLRQf Ef FXUYLOLQHDU RU SRO\QRPLDO UHJUHVVLRQ DQG Ff ORJOLQHDU UHJUHVVLRQ 7KH PDWKHPDWLFDO SURSHUWLHV RI HDFK DUH SUHn VHQWHG LQ &KDSWHU ,,, $OO WKUHH DSSURDFKHV WR WUHQG H[WUDn SRODWLRQ ZHUH H[HFXWHG E\ XVLQJ YDULDWLRQV RI 6366 VXESURn JUDPV 6&$77(5*5$0 DQG 5(*5(66,21 DQG WKDW V\VWHPnV GDWD WUDQVIRUPDWLRQ FDSDELOLWLHV 1LH HW DO f (DFK RI WKH WKUHH PHWKRGV ZDV DSSOLHG WR HDFK RI WKH HLJKW VHOHFWHG WLPH VHULHV LQGLFDWRUV 7LPH LQ \HDUV ZDV XVHG DV WKH LQGHSHQGHQW YDULDEOH WKH DQQXDO PHDVXUH RU LQGH[ RI WKH LQGLFDWRU ZDV WUHDWHG DV WKH GHSHQGHQW RU UHVSRQVH YDULDEOH (DFK GDWD VHW ZDV GLYLGHG LQWR WKLUGV WZRWKLUGV RI WKH GDWD SRLQWV ZHUH XVHG WR HVWDEOLVK WKH SUHGLFWLRQ HTXDWLRQ 7KLV HTXDWLRQ ZDV WKHQ XVHG WR SUHGLFW WKH UHPDLQLQJ WKLUG RI WKH GDWD SRLQWV 3UHGLFWHG YDOXHV ZHUH WKHQ FRPSDUHG ZLWK DFWXDO YDOXHV 7KXV LQ WKLV SKDVH RI WKH VWXG\ WKUHH SUHGLFWLRQ HTXDWLRQV RQH IRU HDFK PHWKRGf ZHUH JHQHUDWHG IRU HDFK RI WKH HLJKW WLPH VHULHV LQGLFDWRUV 6WDWLVWLFDO FULWHULD GHn ULYHG IURP WKH OLWHUDWXUH ZHUH XVHG WR HYDOXDWH WKH JRRGn QHVV RI ILW RI WKH UHJUHVVLRQ OLQH GHULYHG IURP WKH SUHn GLFWLRQ HTXDWLRQ WR WKH GDWD 7KH GLVWULEXWLRQ RI HUURU UHVLGXDOVf DERXW WKH UHJUHVVLRQ OLQH ZDV DOVR H[DPLQHG WR GHWHUPLQH LI WKH GDWD VDWLVILHG WKH DVVXPSWLRQV RI WKH VWDn WLVWLFDO PRGHO 5HVXOWV RI WKH PHWKRG FRPSDULVRQ SKDVH DUH UHSRUWHG LQ &KDSWHU ,9 PAGE 31 'HYHORSPHQW RI ,PSOLFDWLRQV IRU (GXFDWLRQDO 3ODQQLQJ ,Q &KDSWHU 9 PHWKRGRORJLFDO VWUDWHJLHV LQYROYHG LQ WKH VHOHFWLRQ DQG RSHUDWLRQDO GHILQLWLRQ RI YDULDEOHV DUH DQDO\]HG LQ WHUPV RI YLDELOLW\ IRU IXWXUH XVH 5HVXOWV RI WKH WHFKQLTXH FRPSDULVRQ SKDVH DUH DQDO\]HG DFFRUGLQJ WR WKH VWDWLVWLFDO FULWHULD DQG SUDFWLFDO FRQVLGHUDWLRQV GHULYHG IURP WKH OLWHUDWXUH LQ IRUHFDVWLQJ PHWKRGRORJ\ DQG VWDWLVn WLFV ,Q &KDSWHU 9, D VXPPDU\ RI WKH VWXG\ DQG FRQFOXVLRQV ZDUUDQWHG E\ WKH UHVXOWV RI WKH VWXG\ DUH SUHVHQWHG )XWXUH GLUHFWLRQV IRU UHVHDUFK VXJJHVWHG E\ WKH UHVXOWV RI WKLV VWXG\ DUH GLVFXVVHG $GGLWLRQDOO\ LPSOLFDWLRQV IRU WKH XVH LQ HGXFDWLRQDO SODQQLQJ RI DQ HFRORJLFDO PRGHO VXFK DV %URQIHQEUHQQHUnV WLPH VHULHV VRFLDO DQG HGXFDWLRQDO LQn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n FXVVHG LQ UHODWLRQ WR WKHLU KLVWRULFDO GHYHORSPHQW GHILQLn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n SUHKHQVLYH ZRUN GRFXPHQWLQJ VRFLDO FKDQJH LQ WKH 8QLWHG 6WDWHV 3UHVLGHQWfV 5HVHDUFK &RPPLWWHH RQ 6RFLDO 7UHQGV f 7KH GHYHORSPHQW RI LQGLFDWRUV RU PHDVXUHV RI VRFLDO FKDQJH KRZHYHU GLG QRW UHFHLYH WKH VXVWDLQHG JRYHUQPHQWDO VXSSRUW WKDW ZDV SURYLGHG IRU LQGLFDWRUV RI WKH HFRQRPLF SURFHVV 7KXV ZKLOH WKH GHYHORSPHQW RI HFRQRPLF VWDWLVWLFV GXULQJ WKH nV DQG nV SURYLGHG D VROLG EDVLV IRU HFRQRPLF DQDO\VLV DQG HFRQRPLF UHSRUWLQJ ZKLFK HYHQWXDOO\ UHVXOWHG LQ WKH HVWDEOLVKPHQW RI WKH &RXQFLO RI (FRQRPLF $GYLVRUV DQG WKH (FRQRPLF 5HSRUW 86 'HSDUWPHQW RI +HDOWK (GXFDWLRQ :HOIDUH S Yf FRPSDUDEOH GHYHORSPHQW RI VRFLDO LQn GLFDWRUV ZDV QRW XQGHUWDNHQ ,Q WKH nV D UHQHZHG LQWHUHVW LQ VWDWLVWLFV GHVFULELQJ WKH VRFLDO FRQGLWLRQ EHFDPH DSSDUHQW ,PSHWXV IRU WKH GHn YHORSPHQW RI VRFLDO LQGLFDWRUV ZDV SURYLGHG E\ VRFLDO VFLHQWLVWV LQ YDULRXV GLVFLSOLQHV JRYHUQPHQW SROLF\ PDNHUV DQG EXVLQHVV OHDGHUV LQ WKH SULYDWH VHFWRU %URRNV S f :KLOH WKLV HDUO\ HIIRUW ZDV QRW ZHOO GHILQHG DV WR PHPEHUVKLS RUJDQL]DWLRQ RU REMHFWLYHV WKH SDUWLFLSDQWV LQ PAGE 34 WKH VRFLDO LQGLFDWRU PRYHPHQW VHQVHG JUHDW QHHGV DQG RSSRUn WXQLWLHV IRU FKDQJH >DQG@ FHOHEUDWHG VKDUHG EXW QHFHVVDULO\ DPELJXRXV V\PEROV 6KHOGRQ 3DUNH S f 7KH IROORZLQJ H[DPSOHV ZHUH GUDZQ IURP WKH PDQ\ PDQLn IHVWDWLRQV RI LQWHUHVW LQ WKH GHYHORSPHQW RI VRFLDO LQGLFDn WRUV GXULQJ WKH SHULRG IURP WKURXJK RYHU LWHPV ZHUH OLVWHG LQ D ELEOLRJUDSK\ LVVXHG LQ ODWH E\ :LOFR[ %URRNV %HDO .ORQJLDQf 7KH 5XVVHOO 6DJH )RXQGDWLRQ FRPPLVVLRQHG LQ DQG SXEOLVKHG LQ DQ LQGHSHQGHQW VWXG\ ,QGLFDWRUV RI 6RFLDO &KDQJH &RQFHSWV DQG 0HDVXUHPHQWV RQ D QXPEHU RI DVSHFWV RI VWUXFWXUDO FKDQJH LQ VRFLHW\ 6KHOGRQ 0RRUH f $ VWXG\ RQ ZD\V WR PHDVXUH WKH LPSDFW RI PDVVLYH VFLHQWLILF DQG WHFKQRORJLFDO FKDQJH RQ VRFLHW\ %DXHU f ZDV SUHSDUHG E\ WKH $PHULFDQ $FDGHP\ RI $UWV DQG 6FLHQFHV IRU WKH 1DWLRQDO $HURQDXWLFV DQG 6SDFH $GPLQLVWUDWLRQ 7KLV ZRUN 6RFLDO ,QGLFDWRUV ZDV DQ RYHUYLHZ RI WKH WDVN RI GHYHORSLQJ LQGLFDWRUV DV SDUW RI D IHHGEDFN PHFKDQLVP GRFXn PHQWLQJ VRFLDO FKDQJH 3UHVLGHQW -RKQVRQ LQ 0DUFK RI GLUHFWHG WKH 6HFUHWDU\ RI +HDOWK (GXFDWLRQ DQG :HOIDUH WR GHYHORS WKH QHFHVVDU\ VRFLDO VWDWLVWLFV DQG LQGLFDWRUV WR VXSSOHPHQW WKRVH SUHSDUHG E\ WKH %XUHDX RI /DERU 6WDWLVWLFV DQG WKH &RXQFLO RI (FRQRPLF $GYLVRUVf 86 'HSDUWPHQW RI +HDOWK (GXFDWLRQ i :HOIDUH S LLLf 7KH UHVXOW RI WKLV GLUHFWLYH 7RZDUG D 6RFLDO 5HSRUW LVVXHG LQ ZDV PAGE 35 FRQVLGHUHG D SUHOLPLQDU\ VWHS WRZDUG WKH HYROXWLRQ RI D UHJXODU V\VWHP RI VRFLDO UHSRUWLQJ 86 'HSDUWPHQW RI +HDOWK (GXFDWLRQ i :HOIDUH S LLLf 7KH 6RFLDO 6FLHQFH 5HVHDUFK &RXQFLO HVWDEOLVKHG LQ WKH &HQWHU IRU &RRUGLQDWLRQ RI 5HVHDUFK RQ 6RFLDO ,QGLFDWRUV ZKRVH REMHFWLYH LV WR HQKDQFH WKH FRQWULEXWLRQ RI VRFLDO VFLHQFH UHVHDUFK WR WKH GHYHORSPHQW RI D EURDG UDQJH RI LQGLFDWRUV RI VRFLDO FKDQJH :RUOG )XWXUH 6RFLHW\ S f 7KH DSSHDUDQFH RI WKH ODQGPDUN JRYHUQPHQW SXEOLFDn WLRQ 6RFLDO ,QGLFDWRUV ([HFXWLYH 2IILFH RI WKH 3UHVLn GHQW 2IILFH RI 0DQDJHPHQW %XGJHW fZDV KHUDOGHG DV D VLJQLILFDQW DWWHPSW WR SURYLGH D FROOHFWLRQ RI VRFLDO VWDn WLVWLFV GHVFULELQJ TXDOLW\ RI OLIH LQ WKH 8QLWHG 6WDWHV 7KLV ZRUN ZKLFK ZDV VFKHGXOHG WR EH XSGDWHG HYHU\ WKUHH \HDUV ZDV D FRPSLODWLRQ RI VWDWLVWLFV RQ HLJKW PDMRU DUHDV RI VRFLDO LQWHUHVW KHDOWK SXEOLF VDIHW\ HGXFDWLRQ HPn SOR\PHQW LQFRPH KRXVLQJ OHLVXUH DQG UHFUHDWLRQ DQG SRSXn ODWLRQ ,W PD\ EH QRWHG WKDW LW LV RIWHQ LPSRVVLEOH WR VWULFWO\ FDWHJRUL]H ZKDW LV VRFLDO DQG ZKDW LV HFRQRPLF DV DOPRVW DOO DVSHFWV RI OLIH DUH WKH UHVXOW RI LQWHUDFWLRQ EHn WZHHQ VRFLDO DQG HFRQRPLF IRUFHVf 7KH LGHD RI V\VWHPDWLF FROOHFWLRQ DQG XVH RI VRFLDO LQGLFDWRUV KDV QRW DOZD\V PHW D IDYRUDEOH UHFHSWLRQ %H]ROG 1RWH f DQG 6KRVWDN 1RWH f GRFXPHQWHG WKH XQVXFFHVVIXO HIIRUWV EHJLQQLQJ LQ E\ :DOWHU ) 0RQGDOH WR HDUQ FRQJUHVVLRQDO DSSURYDO RI D IDUUHDFKLQJ SODQ IRU QHZ PAGE 36 JRYHUQPHQW XVH RI WKH DSSOLHG VRFLDO VFLHQFHV 0RQGDOHnV EOXHSULQW IRU EHWWHU FROOHFWLRQ DQG XVH RI VRFLDO LQWHOOLn JHQFH LQYROYHG WZR VWDWXWRULO\PDQGDWHG DGGLWLRQV WR WKH ([HFXWLYH 2IILFH RI WKH 3UHVLGHQW Df D &RXQFLO RI 6RFLDO $GYLVRUV &6$f FRPSDUDEOH WR WKH &RXQFLO RI (FRQRPLF $GYLVRUV HVWDEOLVKHG LQ DQG Ef DQ DQQXDO 6RFLDO 5HSRUW RI WKH 3UHVLGHQW SUHSDUHG E\ WKH &6$ WR SDUDOOHO WKH DQQXDO (FRQRPLF 5HSRUW WR WKH 3UHVLGHQW :KLOH DVSHFWV RI 0RQGDOHnV SODQ PD\ EH VDWLVILHG E\ VXFK HIIRUWV DV 6RFLDO ,QGLFDWRUV DQG VHYHUDO FRQJUHVVLRQDO SURYLVLRQV ZKLFK UHTXLUHG WKH GHYHORSn PHQW DQG DSSOLFDWLRQ RI VRFLDO VFLHQFH WHFKQLTXHV WR WKH VWXG\ RI SUHVHQW DQG IXWXUH QDWLRQDO SUREOHPV 6KRVWDN 1RWH f WKH FRPSUHKHQVLYH QDWXUH RI WKH 0RQGDOH SODQ LV DEVHQW 7KH IXWXUH RI VXFK JRYHUQPHQWDO HIIRUWV DW VRFLDO DFFRXQWLQJ ZDV LQ XQFHUWDLQ ,QWHUHVW LQ VRFLDO LQGLFDWRUV KDV QRW EHHQ FRQILQHG WR WKH 8QLWHG 6WDWHV -RKQVRQ 6KHOGRQ i 3DUNH f ,Q IRU H[DPSOH WKH 2UJDQL]DWLRQ IRU (FRQRPLF &RRSHUDn WLRQ DQG 'HYHORSPHQW 2(&'f WR ZKLFK WKH 8QLWHG 6WDWHV DOVR EHORQJV LVVXHG D OLVW RI VRFLDO FRQFHUQV VKDUHG E\ PDQ\ PHPEHU FRXQWULHV 7KH LGHQWLILFDWLRQ RI FRQFHUQV ZDV D ILUVW VWHS LQ WKH GHYHORSPHQW RI D VHW RI VRFLDO LQGLFDWRUV GHVLJQHG H[SOLFLWO\ WR UHYHDO ZLWK YDOLGLW\ WKH OHYHO RI ZHOOEHLQJ IRU HDFK VRFLDO FRQFHUQ LQ WKH OLVW DQG WR PRQLWRU FKDQJHV LQ WKRVH OHYHOV RYHU WLPH 2(&' S f $GGLWLRQDOO\ LQWHUQDWLRQDO RUJDQL]DWLRQV VXFK DV WKH &RQIHUHQFH RI (XURSHDQ 6WDWLVWLFLDQV WKH 8QLWHG 1DWLRQV PAGE 37 5HVHDUFK ,QVWLWXWH IRU 6RFLDO 'HYHORSPHQW DQG WKH 8QLWHG 1DWLRQV (GXFDWLRQDO 6FLHQWLILFDQG &XOWXUDO 2UJDQL]DWLRQ KDYH EHHQ DFWLYHO\ FRQFHUQHG ZLWK VRFLDO LQGLFDWRUV 6KHOGRQ 3DUNH f (IIRUWV WR GHYHORS VRFLDO LQGLFDWRUV KDYH EHHQ LQLWLDWHG LQ FRXQWULHV VXFK DV )UDQFH *UHDW %ULWDLQ :HVW *HUPDQ\ &DQDGD -DSDQ 1RUZD\ 6ZHGHQ DQG 'HQPDUN %URRNV -RKQVRQ f 'HILQLWLRQ DQG 8VH RI 6RFLDO ,QGLFDWRUV 7KH VRFLDO LQGLFDWRU PRYHPHQW KDV EHHQ FKDUDFWHUL]HG E\ DPELJXLW\ RI GHILQLWLRQ DQG SXUSRVH GXH LQ SDUW WR WKH KHWHURJHQHRXV QDWXUH RI SDUWLFLSDQWV ZLWK WKHLU RZQ EDFNn JURXQGV VNLOOV DQG LQWHUHVWV DQG DOVR WR WKH QHFHVVDU\ VWDJHV RI HYROXWLRQ WKDW VXFK D PRYHPHQW H[SHULHQFHV 7KHVH SUREOHPV LQ GHILQLWLRQ DQG SXUSRVH RI VRFLDO LQGLFDWRUV KDYH EHHQ GLVFXVVHG E\ D QXPEHU RI FULWLFV /DQG /LWWOH 3OHVVDV i )HLQ 6KHOGRQ )UHHPDQ 6KHOGRQ /DQG f $WWHPSWV KDYH EHHQ PDGH WR UHVROYH D QXPEHU RI WKHVH SUREOHPV /DQG f IRU H[DPSOH SURSRVHG WKH IROORZLQJ VRFLDO VFLHQFHRULHQWHG GHILQLWLRQ RI VRFLDO LQGLFDWRUV VRFLDO LQGLFDWRUV UHIHU WR VRFLDO VWD WLVWLFV WKDW 7f DUH FRPSRQHQWV LQ D VRFLDO V\VWHP PRGHO LQFOXGLQJ VRFLRSV\ FKRORJLFDO HFRQRPLF GHPRJUDSKLF DQG HFRORJLFDOf RU RI VRPH SDUWLFXODU VHJPHQW RU SURFHVV WKHUHRI f FDQ EH FROOHFWHG DQG DQDO\]HG DW YDULRXV WLPHV DQG DFFXPXn ODWHG LQWR D WLPH VHULHV DQG f FDQ EH DJJUHJDWHG RU GLVDJJUHJDWHG WR OHYHOV DSSURSULDWH WR WKH VSHFLILFDWLRQV RI WKH PRGHO 7KH LPSRUWDQW SRLQW LV WKDW WKH FULWHULRQ IRU FODVVLI\LQJ D VRFLDO PAGE 38 VWDWLVWLF DV D VRFLDO LQGLFDWRU LV LWV LQIRUPDWLYH YDOXH ZKLFK GHULYHV IURP LWV HPSLULFDOO\ YHULILHG QH[XV LQ D FRQn FHSWXDOL]DWLRQ RI D VRFLDO SURFHVV S f 3DUW RI WKH FRQIXVLRQ RYHU GHILQLWLRQ LV WKH UHVXOW RI GLVDJUHHPHQW RYHU SXUSRVHV RU XVHV RI VRFLDO LQGLFDWRUV 7KHVH SXUSRVHV RU XVHV KDYH EHHQ FRQVLGHUHG XQGHU D QXPEHU RI RYHUODSSLQJ VRPHWLPHV V\QRQRPRXV KHDGLQJV Df GHVFULS WLRQV UHSRUWLQJ Ef SROLF\ SODQQLQJ Ff VRFLDO DFFRXQWLQJ Gf SURJUDP HYDOXDWLRQ Hf VRFLDO PRGHOLQJ If VRFLDO IRUH FDVWLQJ DQG Jf VRFLDO HQJLQHHULQJ :KLOH WKH XOWLPDWH REn MHFWLYH RI JXLGLQJ VRFLDO SROLF\ LV UDUHO\ GLVSXWHG WKH IRUP RI WKLV JXLGDQFH LV VWLOO GHEDWHG 6RFLDO VFLHQWLVWV DUH PRUH OLNHO\ WR EH FRQFHUQHG ZLWK WKH DQDO\VLV DQG SUHn GLFWLRQ RI VRFLDO FKDQJH ZKLOH SXEOLF DGPLQLVWUDWRUV DQG OHJLVODWRUV DUH RIWHQ PRUH FRQFHUQHG ZLWK XVHV RI LQGLFDWRUV UHODWHG WR SXEOLF SURJUDP HYDOXDWLRQ DQG DJHQF\ JRDO VHWWLQJ 6KHOGRQ DQG 3DUNH f LQ DFNQRZOHGJLQJ WKHVH FRQFHUQV VDLG ,W LV DSSDUHQW WKDW PDQ\ GLIIHUHQW W\SHV RI ZRUN JR RQ XQGHU WKH UXEULF RI VRFLDO LQGLFDWRUV :KDW LV LPSRUn WDQW LV WKDW WKH ILHOG EH VHHQ DV DQ DUHQD IRU ORQJWHUP GHYHORSPHQW DV DQ HIIRUW RI VRFLDO VFLHQWLVWV WR SXVK IRUHZDUG GHYHORSPHQWV LQ FRQFHSWV DQG LQ PHWKRGRORJ\ WKDW SURPLVH SD\RIIV WR ERWK VFLHQFH DQG SXEOLF SROLF\ S f 7R XQGHUVFRUH WKLV SRLQW 6KHOGRQ DQG 3DUNH f VHOHFWHG DQ REVHUYDWLRQ E\ 'XQFDQ 7KH YDOXH RI LPSURYHG PHDVXUHV RI VRFLDO FKDQJH LV QRW WKDW WKH\ QHFHVVDULO\ UHVROYH WKHRUHWLFDO LVVXHV PAGE 39 FRQFHUQLQJ VRFLDO G\QDPLFV RU VHWWOH SUDJPDWLF LVVXHV RI VRFLDO SROLF\ EXW WKDW WKH\ PD\ SHUPLW WKRVH LVVXHV WR EH DUJXHG PRUH SURGXFWLYHO\ S f 'DWD %DVH IRU 6RFLDO ,QGLFDWRUV 9DULRXV HIIRUWV KDYH EHHQ XQGHUWDNHQ WR LPSURYH WKH GDWD EDVH IRU VRFLDO LQGLFDWRUV $PRQJ WKH HIIRUWV LQ WKH HDUO\ nV ZHUH EDVLF VXUYH\V RQ FULPH DQG HGXFDWLRQ DV ZHOO DV UHSOLFDWLRQV RI SUHYLRXV VRFLDO VFLHQFH VWXGLHV DQG VXUYH\V 6KHOGRQ i 3DUNH f 0RVW VRFLDO VWDWLVWLFV DYDLODEOH RULPDULO\ IURP JRYHUQn PHQW VRXUFHV DUH REMHFWLYH LQ QDWXUH WKDW LV WKH\ PHDVXUH WKH IUHTXHQF\ RI RFFXUUHQFH RI DQ DWWULEXWH RU FRPPRGLW\ LQ WKH SRSXODWLRQ 1XPEHUV RI ELUWKV GHDWKV PDUULDJHV \HDUV RI VFKRROLQJ DQG SHUFHQW RI RFFXSLHG KRXVLQJ ZLWK WHOHYLVLRQ VHWV FRXOG WKXV EH FRQVLGHUHG REMHFWLYH PHDVXUHV 6RPH ZRXOG GLVDJUHH KRZHYHU ZLWK WKH REMHFWLYLW\ RI WKHVH PHDVXUHV VHH $QGUHZV i ,9LWKH\ S f 6HYHUDO UHVHDUFKHUV $QGUHZV :LWKH\ &DPSEHOO &RQYHUVH i 5RJHUV f KDYH DWWHPSWHG WR PHDVXUH SHRSOHnV SHUFHSWLRQV RI WKHLU ZHOOEHLQJ WKHLU TXDOLW\ RI OLIH 6XFK PHDVXUHV FROOHFWHG RQ D UHJXODU EDVLV DUH H[SHFWHG WR EH YDOXDEOH VXSSOHPHQWV WR WKH XVXDO REMHFWLYH TXDOLW\ RI OLIH LQGLFDWRUV 6HH IRU H[DPSOHV RI WKH ODWWHU /LX 7KRPSVRQ Ef &UHDWLRQ RI D VRFLDO LQGLFDWRU GDWD EDVH LV QRW ZLWKRXW FRQFHSWXDO DQG PHWKRGRORJLFDO SUREOHPV 9DULRXV DVSHFWV RI PAGE 40 WKH VRFLDO PHDVXUHPHQW SUREOHP KDYH EHHQ DFNQRZOHGJHG LQ WKH OLWHUDWXUH VHH IRU H[DPSOH GH 1HXIYLOOH SS (W]LRQL /HKPDQ 6RFLDO 0HDVXUHPHQW f :KLOH GHn WDLOHG GLVFXVVLRQ RI PHDVXUHPHQW G\VIXQFWLRQ LQ WKH WHUPLn QRORJ\ RI (W]LRQL i /HKPDQ f LV EH\RQG WKH VFRSH RI WKLV VWXG\ WKH IROORZLQJ REVHUYDWLRQ PLJKW EH NHSW LQ PLQG ,QFUHDVHG LQYHVWPHQW LQWHOOHFWXDO DV ZHOO DV ILQDQFLDO QR GRXEW FDQ JR D ORQJ ZD\ WR LQFUHDVH WKH HIILFDF\ RI VRFLDO PHDVXUHPHQWV DQG WR UHGXFH PXFK RI WKH OLNHOLKRRG RI G\VIXQFWLRQV %XW LQ WKH ILQDO DQDO\VLV WKHVH SUREOHPV FDQ QHYHU EH HOLPLQDWHG HQn WLUHO\ +HUH WKH FOLHQW RI V\VWHPDWLF PHDVXUHPHQW DQG DFFRXQWLQJ VKRXOG EH DOHUWHG WR WKH OLPLWDWLRQV RI VRFLDO LQGLFDWRUV ERWK WR PDNH KLV XVH RI WKHP PRUH VRSKLVWLn FDWHG DQG WR SUHYHQW KLP IURP XOWLPDWHO\ UHMHFWLQJ WKH LGHD RI VRFLDO DFFRXQWLQJ ZKHQ KH HQFRXQWHUV LWV OLPLWDWLRQV (W]LRQL i /HKPDQ S f (GXFDWLRQDO ,PSOLFDWLRQV (GXFDWLRQDO LQGLFDWRUV D VXEVHW RI VRFLDO LQGLFDWRUV KDYH WUDGLWLRQDOO\ EHHQ PHDVXUHV RI WKH HGXFDWLRQDO V\VWHPnV LQSXWV DQG RXWSXWV VWDWHG LQ VXFK WHUPV DV QXPEHUV RI WHDn FKHUV SHU SXSLO H[SHQGLWXUHV DQG DFKLHYHPHQW WHVW VFRUHV 7KHUH KDYH EHHQ DWWHPSWV KRZHYHU WR EURDGHQ WKLV EDVH RI HGXFDWLRQDO VWDWLVWLFV WR LQFOXGH ERWK REMHFWLYH DQG VXEn MHFWLYH LQGLFDWRUV XQGHU WKH FDWHJRULHV RI DFFHVV DVSLUDWLRQV DFKLHYHPHQW LPSDFW DQG UHVRXUFHV &RROHU S f 7KHUH KDYH DOVR EHHQ DWWHPSWV WR OLQN LQGLFDWRUV RI VRFLDO SURFHVVHV HJ GLYRUFH UDWHV YRWLQJ UDWHVf WR HGXFDWLRQDO JRDOV DQG WKXV WR HVWDEOLVK DFFRXQWDELOLW\ PHDVXUHV DOEHLW UHPRWH H[WHUQDO WR WKH HGXFDWLRQDO V\VWHP &OHPPHU )DLUEDQNV PAGE 41 +DOO ,PSDUD 1HOVRQ &ROOD]R /HZLV i 7KRPDV 1RWH *UDG\ f 7KH XVH DQG DEXVH RI LQGLFDWRUV LQ DQ HGXn FDWLRQDO VHWWLQJ KRZHYHU UHPDLQHG LQ D PDWWHU RI GHEDWH ,PSDUD 1RWH f DQG FDXWLRXV RSWLPLVP +DOO 1RWH f +RSHn IXOO\ LQYHVWLJDWLRQV RI WKH SUREOHP VXFK DV WKDW GHVFULEHG LQ WKLV VWXG\ ZLOO SURYLGH VRPH JXLGDQFH DV WR WKH PRVW SURPLVLQJ DSSOLFDWLRQV RI VRFLDO LQGLFDWRUV WR HGXFDWLRQ 6HOHFWLRQ RI 9DULDEOHV7LPH 6HULHV ,QGLFDWRUV 7KH 9DULDEOHV ,Q WKH ILUVW \HDU 6HSW -XQH f RI )ORULGD 'HSDUWPHQW RI (GXFDWLRQ 67$5 3URMHFW 5 RQ VRFLDO IRUHn FDVWLQJ IRU HGXFDWLRQDO SODQQLQJ WUHQGV LQ ILYH LQGLFDWRUV RI HGXFDWLRQDO RXWFRPHV ZHUH IRUHFDVW ,Q RUGHU WR GR WKLV LW ZDV QHFHVVDU\ WR LGHQWLI\ YDULDEOHV WKDW LQIOXHQFH WKH RXWFRPHV RI HGXFDWLRQ 7KURXJK D UHYLHZ RI WKH UHVHDUFK DQG WKHRUHWLFDO OLWHUDWXUH D QXPEHU RI VRFLDO YDULDEOHV ZHUH LGHQWLILHG 7KLV OLVW ZDV UHILQHG E\ DQ LQWHUGLVFLSOLQDU\ SDQHO RI H[SHUWV DW WKH 8QLYHUVLW\ RI )ORULGD WR WKH IROORZLQJ YDULDEOHV Df VRFLRHFRQRPLF VWDWXV Ef IDPLO\ H[SHFWDWLRQV DWWLWXGHV DQG DVSLUDWLRQV Ff VWXGHQWnV VHOI FRQFHSW Gf VWXGHQWnV JHQHUDO DELOLW\ Hf VWXGHQWnV VHQVH RI IDWH FRQWURO If VWXGHQWnV DWWLWXGHV DQG PRWLYDWLRQ Jf SHHU JURXS FKDUDFWHULVWLFV Kf WHDFKHU H[SHFWDWLRQV Lf WHDFKHU EHKDYLRU LQ WKH FODVVURRP DQG Mf DGPLQLVWUDWLYH OHDGHUVKLS VW\OH &ROOD]R HW DO f VDLG WKDW RQO\ WKH YDULDEOHV Df DQG Gf UHFHLYHG VWURQJ VXSSRUW IURP UHVHDUFK PAGE 42 D QXPEHU RI WKH RWKHU YDULDEOHV ZKLOH LGHQWLILHG DV LPSRUn WDQW LQ WKH WKHRUHWLFDO OLWHUDWXUH OLDG LQFRQFOXVLYH VXSSRUW IURP UHVHDUFK S f 6HH &ROOD]R /HZLV 7KRPDV 1RWH IRU D UHYLHZ RI WKH UHVHDUFK OLWHUDWXUH RQ YDULDEOHV DIIHFWLQJ HGXFDWLRQDO RXWFRPHVf 7KH SDQHO RI H[SHUWV ZDV IXUWKHU XWLOL]HG WR IRUHFDVW WKH IXWXUH WUHQGV RI WKHVH YDULDEOHV DQG WKHLU HIIHFW RQ VSHFLILHG SHUIRUPDQFH DQG XWLOL]DWLRQ PHDVXUHV RI WKH RXWn FRPHV RI HGXFDWLRQ &URVVLPSDFW DQDO\VLV D FRPSXWHU DVVLVWn HG PRGLILFDWLRQ RI WKH 'HOSKL IRUHFDVWLQJ WHFKQLTXH ZDV WKHQ XVHG E\ WKH SDQHO WR JHQHUDWH WKH IXWXUH WUHQGV LQ WKH ILYH RXWFRPH LQGLFDWRUV 7KH IUDPHZRUN IRU ORRNLQJ DW WKH IXWXUH HVWDEOLVKHG GXULQJ WKHVH ILUVW \HDU SURMHFW DFWLYLWLHV LV XWLOL]HG LQ WKH SUHVHQW VWXG\ 3UHYLRXV IRUHFDVWLQJ DFWLYLWLHV ZHUH EDVHG SULPDULO\ RQ WKH VXEMHFWLYH MXGJPHQW RI SDQHO SDUWLFLSDQWV ,Q WKLV VWXG\ KRZHYHU WKH IHDVLELOLW\ RI XVLQJ WLPH VHULHV GDWD ZKHUH DYDLODEOH DV WKH EDVLV IRU IRUHFDVWLQJ IXWXUH WUHQGV LQ WKH YDULDEOHV DIIHFWLQJ HGXFDWLRQDO RXWFRPHV LV H[DPLQHG ,Q DGGLWLRQ WKH XVH RI D PRGHO FRQWDLQLQJ WKH VHOHFWHG YDULDEOHV LV FRQVLGHUHG %URQIHQEUHQQHUV (FRORJ\ RI (GXFDWLRQ 0RGHO ,Q WKH SUHYLRXV VHFWLRQ WKH YDULDEOHV DIIHFWLQJ HGXFDWLRQDO RXWFRPHV ZKLFK ZHUH GHULYHG IURP WKH UHVHDUFK OLWHUDWXUH ZHUH SUHVHQWHG +RZ FDQ WKHVH YDULDEOHV EH SXW LQWR SHUVSHFWLYH DV VRFLDO IRUFHV LQIOXHQFLQJ ZKDW WKH VWXn GHQW OHDUQV" PAGE 43 7KH %URQIHQEUHQQHU f PRGHO ZKLFK ZDV SUHVHQWHG LQ &KDSWHU SS f LV D PXOWLGLPHQVLRQDO HFRORJLFDO VWUXFn WXUH RI WKH HGXFDWLRQDO HQYLURQPHQW $W WKH FHQWHU RI WKH LQWHUDFWLQJ PHVR H[R DQG PDFURV\VWHPV LV WKH PLFURV\VWHP WKH LPPHGLDWH VHWWLQJ FRQWDLQLQJ WKH OHDUQHU %URQIHQEUHQQHU S f 7KH PHVRV\VWHP LV DFWXDOO\ D V\VWHP RI PLFURn V\VWHPV WKDW LV LW FRPSULVHV WKH LQWHUUHODWLRQVKLSV DPRQJ WKH PDMRU VHWWLQJV FRQWDLQLQJ WKH OHDUQHU DW D SDUWLFXODU SRLQW LQ KLV RU KHU OLIH %URQIHQEUHQQHU S f 6RPH RI WKH VRFLDO YDULDEOHV WKDW ZHUH LGHQWLILHG SUHYLRXVO\ FRXOG EH FRQVLGHUHG DV SDUW RI WKH PHVRV\VWHP 7KH KRPH IRU H[n DPSOH LV UHSUHVHQWHG E\ VRFLRHFRQRPLF VWDWXV DQG IDPLO\ H[SHFWDWLRQV DWWLWXGHV DQG DVSLUDWLRQV WKH SHHU JURXS E\ SHHU JURXS FKDUDFWHULVWLFV DQG WKH VFKRRO E\ WHDFKHU H[SHFn WDWLRQV WHDFKHU EHKDYLRU LQ WKH FODVVURRP DQG DGPLQLVWUDn WLYH OHDGHUVKLS VW\OH 7KH RWKHU YDULDEOHV VWXGHQWnV VHOI FRQFHSW VWXGHQWnV JHQHUDO DELOLW\ VWXGHQWnV VHQVH RI IDWH FRQWURO DQG VWXGHQWnV DWWLWXGHV DQG PRWLYDWLRQ DUH DOO GLn UHFWO\ UHODWHG WR WKH OHDUQHU %URQIHQEUHQQHU f SURSRVHG WKDW OHDUQLQJ LV D IXQFn WLRQ RI Df WKH G\QDPLF UHODWLRQVKLS EHWZHHQ FKDUDFWHULVWLFV RI WKH OHDUQHUV DQG WKHLU YDULRXV VXUURXQGLQJV PHVRV\VWHPf DQG Ef WKH LQWHUDFWLRQ EHWZHHQ WKHVH YDULRXV HQYLURQPHQWV HJ KRPH VFKRRO SHHU JURXSf 7KH %URQIHQEUHQQHU HFRORJ\ RI HGXFDWLRQ PRGHO WKXV DSSHDUV WR SURYLGH WKH QHFHVVDU\ IUDPHZRUN WR VXSSRUW XVH RI WKH SUHVHQWO\ LGHQWLILHG YDULDEOHV DQG WR JHQHUDWH GLUHFWLRQV IRU IXWXUH IRUHFDVWLQJ UHVHDUFK PAGE 44 2SHUDWLRQDO 'HILQLWLRQ RI 9DULDEOHV DV 7LPH 6HULHV ,QGLFDWRUV ,Q SUHYLRXV VHFWLRQV YDULDEOHV DIIHFWLQJ HGXFDWLRQDO RXWFRPHV ZHUH SUHVHQWHG DQG WKHQ FODVVLILHG DFFRUGLQJ WR WKH %URQIHQEUHQQHU HFRORJ\ RI HGXFDWLRQ PRGHO ,Q RUGHU WR LGHQn WLI\ WUHQGV LQ WKHVH YDULDEOHV DQG WR H[WUDSRODWH WKHVH WUHQGV LQWR WKH IXWXUH LW ZDV QHFHVVDU\ WR RSHUDWLRQDOO\ GHILQH WKHVH YDULDEOHV DV WLPH VHULHV PHDVXUHV RU LQGLFDWRUV 6LQFH VRPH RI WKHVH YDULDEOHV ZHUH H[SUHVVHG LQ JHQHUDO WHUPV LW VHHPHG QHFHVVDU\ WR WU\ WR UHSUHVHQW HDFK E\ D QXPEHU RI PHDVXUHV DQG WKXV DYRLG IUDFWLRQDO PHDVXUHPHQW ZKLFK LV RIWHQ D FRQFHUQ ZKHQ RSHUDWLRQDOO\ GHILQLQJ D VRFLDO FRQFHSW (W]LRQL /HKPDQ f 6HYHUDO SUREOHPV EHFDPH DSSDUHQW LQ RSHUDWLRQDOL]LQJ WKH YDULDEOHV $ QXPEHU RI LQGLFDWRUV ZHUH LGHQWLILHG IRU WKH YDULDEOHV Df VRFLRHFRQRPLF VWDWXV Ef IDPLO\ H[SHFWDWLRQV DWWLWXGHV DQG DVSLUDWLRQV DQG Ff SHHU JURXS FKDUDFWHULVWLFV )RU VRPH LQGLFDWRUV KRZHYHU GDWD ZHUH QRW FROOHFWHG DQQXDOO\ IRU RWKHUV PHDVXUHV ZHUH QRW FRPSDUDEOH RYHU WLPH GXH WR D GLIIHUHQW EDVLV IRU PHDVXUHPHQW )RU WKH YDULDEOHV UHODWHG WR WKH VFKRRO DQG VWXGHQW FKDUDFWHULVWLFV H[FHSW VWXGHQW DWWLWXGHV DQG PRWLYDWLRQf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n OLVKHG IRU WKLV VWXG\ VHH S f ZHUH VHOHFWHG IRU XVH ZLWK WKH WKUHH H[WUDSRODWLYH PHWKRGV GHVFULEHG LQ &KDSWHU ,,, 0HGLDQ IDPLO\ LQFRPH LQ WKH 8QLWHG 6WDWHV H[n SUHVVHG LQ FRQVWDQW GROODUV 1XPEHU RI IDPLOLHV LQ WKH 8QLWHG 6WDWHV KHDGHG E\ ZRPHQ H[SUHVVHG DV D SHUFHQWDJH RI WRWDO IDPLOLHV 1XPEHU RI ZLYHV LQ WKH ODERU IRUFH H[SUHVVHG DV D SHUFHQWDJH RI WRWDO ZLYHV LQ WKH 8QLWHG 6WDWHV 1XPEHU RI PDUULDJHV LQ )ORULGD H[SUHVVHG DV UDWH SHU SRSXODWLRQ LQ )ORULGD 1XPEHU RI GLVVROXWLRQV RI PDUULDJH LQ )ORULGD H[n SUHVVHG DV UDWH SHU SRSXODWLRQ LQ )ORULGD 1XPEHU RI UHVLGHQW OLYH ELUWKV LQ )ORULGD H[n SUHVVHG DV UDWH SHU SRSXODWLRQ LQ )ORULGD 1XPEHU RI WR \HDU ROGV HQUROOHG LQ QXUVHU\ VFKRRO DQG NLQGHUJDUWHQ H[SUHVVHG DV SHUFHQWDJH RI WRWDO FKLOGUHQ WR \HDUV ROG LQ WKH 8QLWHG 6WDWHV 1XPEHU RI FKLOGUHQ LQYROYHG LQ GLYRUFH RU DQQXOPHQW H[SUHVVHG DV UDWH SHU FKLOGUHQ XQGHU \HDUV ROG LQ WKH 8QLWHG 6WDWHV :KLOH UDWHV RU SHUFHQWDJHV DUH XVHG IRU IRUHFDVWLQJ SXUSRVHV WKH PDJQLWXGH RI WKH DFWXDO QXPEHUV VKRXOG EH NHSW LQ PLQG EHIRUH LQWHUSUHWDWLRQ RI DQ LGHQWLILHG WUHQG LV PAGE 46 7DEOH 7LPH 6HULHV ,QGLFDWRUV RI 6RFLDO 9DULDEOHV $IIHFWLQJ 2XWFRPHV RI (GXFDWLRQ $YDLODEOH WLPH VHULHV GDWD 6RFLDO YDULDEOHV ,QGLFDWRU PAGE 47 7DEOH FRQWLQXHG $YDLODEOH WLPH VHULHV GDWD 6RFLDO YDULDEOHV ,QGLFDWRU PAGE 48 7DEOH FRQWLQXHG 6RFLDO YDULDEOHV ,QGLFDWRU D $YDLODEOH WLPH VHULHV GDWD PAGE 49 7DEOH FRQWLQXHG R 6RFLDO YDULDEOHV ,QGLFDWRU PAGE 50 DWWHPSWHG )XUWKHUPRUH LW LV QHFHVVDU\ WR UHPHPEHU WKDW VLQFH WKH SRSXODWLRQ EDVH LQFUHDVHG RYHU WKH GHFDGHV FRYHUHG E\ WKH GDWD D VWDEOH UDWH RU SHUFHQWDJH VWLOO UHSUHVHQWV ODUJHU DEVROXWH QXPEHUV RI WKH SKHQRPHQRQ 7KH LQGLFDWRUV VHOHFWHG DUH DJJUHJDWHG WR HLWKHU WKH VWDWH RU QDWLRQDO OHYHO WKH DSSURSULDWH OHYHO RI DJJUHJDWLRQ ZRXOG RI FRXUVH GHSHQG XSRQ WKH VSHFLILF SODQQLQJ DFWLYLW\ 7KHVH LQGLFDWRUV FRXOG EH GLVDJJUHJDWHG E\ UDFH DJH UHJLRQ RU VH[ ZKHUH DSSURn SULDWHf IRU FRPSDUDWLYH DQDO\VLV DQG LQGHHG WKLV IHDWXUH LV D QHFHVVDU\ FKDUDFWHULVWLF LQ PDQ\ RI WKH GHILQLWLRQV RI VRFLDO LQGLFDWRUV HJ VHH GHILQLWLRQ E\ /DQG SUHn VHQWHG HDUOLHU LQ WKLV FKDSWHUf PAGE 51 &+$37(5 ,,, 5$7,21$/( )25 (;75$32/$7,9( 0(7+2'6 6(/(&7(' )25 &203$5,621 2QH RI WKH SXUSRVHV RI WKLV VWXG\ ZDV WR FRPSDUH WKUHH SXUHO\ H[WUDSRODWLYH PHWKRGV ZKLFK FRXOG EH XVHG ZLWK VRFLDO LQGLFDWRU GDWD RI WKH W\SH GHVFULEHG LQ &KDSWHU ,, WR IRUHn FDVW IXWXUH YDOXHV RI WKRVH LQGLFDWRUV ,Q RUGHU WR VHOHFW PHWKRGV ZKLFK ZHUH DSSURSULDWH IRU WKLV SXUSRVH ERWK WKH JHQHUDO IRUHFDVWLQJ OLWHUDWXUH DQG IRUHFDVWLQJ DSSOLFDWLRQV RI H[WUDSRODWLYH WHFKQLTXHV LQ VSHFLILF DUHDV ZHUH UHYLHZHG 'HWDLOHG GHVFULSWLRQV RI HDFK WHFKQLTXH DV ZHOO DV VWDWLVWLn FDO DVVXPSWLRQV VHQVLWLYLW\ DQG HYDOXDWLYH FULWHULD ZHUH GHULYHG SULPDULO\ IURP WKH OLWHUDWXUH LQ HFRQRPLF VWDWLVWLFV DQG UHJUHVVLRQ DQDO\VLV 7KH IROORZLQJ VHFWLRQV SURYLGH Df DQ RYHUYLHZ RI WKH H[WUDSRODWLYH PHWKRGV XVHG LQ IRUHFDVWLQJ Ef HYDOXDWLRQ RI WKH DSSOLFDELOLW\ RI WKHVH PHWKRGV IRU WKH SXUSRVH RI WKLV VWXG\ Ff D GHVFULSWLRQ RI WKH WKUHH PHWKRGV VHOHFWHG IRU FRPSDULVRQ LQFOXGLQJ HTXDWLRQV SDUDPHWHUV WR EH HVWLPDWHG DVVXPSWLRQV DQG FULWHULD WR EH XVHG LQ WKH FRPSDULVRQ RI WKH WKUHH PHWKRGV 2YHUYLHZ RI ([WUDSRODWLYH )RUHFDVWLQJ 0HWKRGV $Q H[WUDSRODWLYH IRUHFDVWLQJ PHWKRG LV D SURFHGXUH IRU Df LGHQWLI\LQJ DQ XQGHUO\LQJ KLVWRULFDO WUHQG RU F\FOH LQ PAGE 52 WLPH VHULHV GDWD DQG Ef HVWLPDWLQJ IXWXUH VWDWHV RI D YDULDEOH EDVHG RQ FXUUHQW DQG KLVWRULFDO REVHUYDWLRQVPHDn VXUHV RI WKDW YDULDEOH +DUULVRQ f ([WUDSRODWLRQ SURn YLGHV D VXUSULVH IUHH SURMHFWLRQ RI WKH IXWXUH EXW QRW QHFHVVDULO\ D IXWXUH ZKLFK LV D ELJJHU DQG EHWWHU RU ZRUVHf YHUVLRQ RI WKH SUHVHQW 0DUWLQR f QRWHG WKDW VRPH H[WUDSRODWLRQ PHWKRGV DOORZ WKH IRUHFDVWHU WR LGHQWLI\ SROLF\ YDULDEOHV ZKLFK DUH VXEMHFW WR PDQLSXODWLRQ DQG ZKLFK DOORZ WKH GHFLVLRQPDNHU WR DOWHU WKH IXWXUH DZD\ IURP WRGD\nV SDWWHUQ RI HYHQWV S f ,Q WKH VRFLDO UHDOP H[WUDSRODWLRQ RI WUHQGV PD\ DW OHDVW DOORZ WKH SODQQHU RU SROLF\ PDNHU WR PDNH HQOLJKWHQHG GHn FLVLRQV WR SUHSDUH IRU WKH IXWXUH (FRQRPLF DQG %XVLQHVV )RUHFDVWLQJ %HFDXVH RI WKH LPSHWXV LQ WKH nV DQG nV WR GHVFULEH DQG IRUHFDVW WKH HFRQRPLF FRQGLWLRQ PDQ\ H[WUDn SRODWLYH PHWKRGV ZHUH GHYHORSHG ZLWK HFRQRPLF DSSOLFDWLRQV LQ PLQG *UHHQZDOG S f FODVVLILHG PHWKRGV IRU GHWHUPLQLQJ HFRQRPLF WUHQGV LQWR Df QRQPDWKHPDWLFDO PHWKRGV VXFK DV IUHHKDQG FXUYH ILWWLQJ ILUVWRUGHU GLIIHUHQFHV VHPLn DYHUDJHV VHOHFWHG SRLQWV DQG ZHLJKWHG DQG XQZHLJKWHG PRYLQJ DYHUDJHV DQG Ef PDWKHPDWLFDO PHWKRGV VXFK DV OHDVW VTXDUHV PRPHQWV PD[LPXP OLNHOLKRRG DQG RWKHUV ,Q JHQHUDO RQO\ WKH PDWKHPDWLFDO PHWKRGV ZKLFK LQFOXGH D ZLGHO\ GLYHUVH DUUD\ RI FRPSOH[ FXUYHILWWLQJ WHFKQLTXHV VHHP WR EH UHOLHG XSRQ IRU IRUHFDVWLQJ SXUSRVHV ZKLOH WKH QRQPDWKHPDWLFDO PHWKRGV DUH XVHG IRU SUHOLPLQDU\ DQDO\VLV RI WKH VKDSH RI WKH PAGE 53 WLPH VHULHV GDWD )RU GHVFULSWLRQV RI WKHVH PHWKRGV VHH *UHHQZDOG 0D\HV i 0D\HV 0HQGHQKDOO 5HLQPXWK 1HLVZDQJHU 7XWWOH f $SSURDFKHV WR JRYHUQPHQWDOQDWLRQDO HFRQRPLF IRUHFDVWLQJ HJ7KHLO f RIWHQ UHDFK D UHODWLYHO\ KLJK OHYHO RI PDWKHPDWLFDO DQG WKHRUHWLFDO VRSKLVWLFDWLRQ 7KLV DSSHDUV WR EH WKH UHVXOW RI GHFDGHV RI GHYHORSPHQW RI DSSO\LQJ PHWKRG LQ OLJKW RI WKHRU\ DQG GHYHORSLQJ ERWK LQ WXUQ ,W LV DOVR WKH UHVXOW RI VXEVWDQWLDO LQYHVWPHQW RI ILQDQFLDO DQG PDQSRZHU UHVRXUFHV E\ ERWK JRYHUQPHQW DQG LQGXVWU\ 7KH YDOXH RI H[WUDSRODWLYH IRUHFDVWLQJ WR LQGLYLGXDO GHFLVLRQPDNHUV LQ EXVLQHVV KDV EHFRPH DSSDUHQW 0DNULGDNLV +RGJVGRQ :KHHOZULJKW f ,QGHHG FRPSDQLHV RI DOO VL]HV DUH FRPSHOOHG WR PDNH IRUHFDVWV IRU D QXPEHU RI YDULDn EOHV ZKLFK DIIHFW WKHP 0DNULGDNLV HW DO f KDYH QRWHG KRZHYHU WKDW DV ZLWK WKH GHYHORSPHQW RI PRVW PDQDJHPHQW VFLHQFH WHFKQLTXHV WKH DSSOLFDWLRQ RI WKHVH >H[WUDSRODWLYH IRUHFDVWLQJ@ PHWKRGV KDV ODJJHG EHKLQG WKHLU WKHRUHWLFDO IRUPXn ODWLRQ DQG YHULILFDWLRQ S f 7KXV WKH DXWKRUV REVHUYHG WKDW ZKLOH WKH QHHG IRU IRUHFDVWLQJ PHWKRGV LV UHFRJQL]HG E\ PDQDJHUV LQ EXVLQHVV IHZ DUH IDPLOn LDU ZLWK WKH QXPHURXV WHFKQLTXHV DYDLODEOH DQG WKHLU FKDUDFn WHULVWLFV LQ RUGHU WKDW WKH RQH PRVW DSSURSULDWH IRU D JLYHQ VLWXDWLRQ EH VHOHFWHG 7R KHOS PHHW WKLV QHHG 0DNULGDNLV HW DO KDYH GHYHORSHG DQ LQWHUDFWLYH IRUHFDVWLQJ V\VWHP FDOOHG ,QWHUDFWLYH )RUHFDVWLQJ >6,% PAGE 54 IRUHFDVWLQJ WHFKQLTXH IRU D JLYHQ VHW RI GDWD $OWKRXJK WKH V\VWHP KDV EHHQ ZHOO WHVWHG LQ WHDFKLQJ VLWXDWLRQV LW KDV QRW KDG H[WHQVLYH DSSOLFDWLRQ LQ DFWXDO EXVLQHVV VHWWLQJV 4XDQWLWDWLYH WHFKQLTXHV DYDLODEOH LQ WKH ,QWHUDFWLYH )RUHFDVn WLQJ 6,% PAGE 55 JHRPHWULF VWUDLJKW OLQH ZKLFK UHSUHVHQWV D YDULDEOH ZKLFK FKDQJHV DW D FRQVWDQW UDWH 1HLVZDQJHU f (GXFDWLRQDO )RUHFDVWLQJ 8VHV RI H[WUDSRODWLYH PHWKRGV LQ HGXFDWLRQ KDYH JHQHUDOO\ EHHQ OLPLWHG WR SURMHFWLRQV RI H[SHQGLWXUHV VFKRRO HQUROOn PHQWV DQG WKH QXPEHU RI LQVWUXFWLRQDO VWDII KLJK VFKRRO JUDGXDWHV DQG HDUQHG GHJUHHV :KLOH PDQ\ VWDWHV DQG VFKRRO GLVWULFWV KDYH GHYHORSHG WKHLU RZQ PRGHOV HVSHFLDOO\ IRU SURMHFWLRQV RI HQUROOPHQWV WKH 1DWLRQDO &HQWHU IRU (GXFDn WLRQ 6WDWLVWLFV 86 'HSDUWPHQW RI +HDOWK (GXFDWLRQ :HOIDUH Ff LQ GHYHORSLQJ SURMHFWLRQV RI HGXFDWLRQ VWDWLVn WLFV WR UHOLHG RQ UHJUHVVLRQ PHWKRGV ZKHUHYHU D WUHQG FRXOG EH HVWDEOLVKHG 6SHFLILFDOO\ HLWKHU DULWKPHWLF VWUDLJKW OLQHV RU ORJLVWLF JURZWK FXUYHV GHSHQGLQJ XSRQ WKH QDWXUH RI GDWD ZHUH ILWWHG E\ WKH PHWKRG RI OHDVW VTXDUHV 7KH IROORZLQJ ZDV QRWHG KRZHYHU )RU ERWK WKH VWUDLJKW OLQH DQG ORJLVWLF JURZWK FXUYH WKH ILWWHG FXUYH RIWHQ OLHV FRQVLGHUDEO\ DERYH RU EHORZ WKH ODVW REn VHUYHG SRLQW UHVXOWLQJ LQ DQ XQXVXDO ULVH RU GURS IURP WKH ODVW DFWXDO REVHUYDn WLRQ 7R DYRLG WKLV DQG JLYH IDFH YDOLGLW\ WR WKH SURMHFWLRQV WKH ILWWHG FXUYH ZDV XVHG RQO\ WR HVWDEOLVK WKH ODVW SRLQW DQG D QHZ FXUYH ZDV GUDZQ WKURXJK WKH ODVW REVHUYHG UDWLR DQG WKH HQG SRLQW RQ WKH ILWWHG FXUYH 86 'HSDUWPHQW RI +HDOWK (GXFDWLRQ :HOIDUH F S f %URZQ f VXPPDUL]HG WKH XVH RI WUHQG DQDO\VLV PHWKRGV LQ HGXFDWLRQ DQG QRWHG WKHLU SRWHQWLDO DSSOLFDWLRQV LQ HGXFDn WLRQDO DGPLQLVWUDWLRQ 7KH IRXU H[WUDSRODWLYH PHWKRGV WKDW KH FULWLTXHG ZHUH Df DULWKPHWLF VWUDLJKW OLQH H[WUDSRODWLRQ PAGE 56 Ef WLPH VHULHV DQDO\VLV UHDOO\ D VLPSOLILHG YHUVLRQ RI WKH %R[-HQNLQV WHFKQLTXHf Ff WKH 6VKDSHG JURZWK FXUYH DQG Gf FRKRUW DQDO\VLV DFWXDOO\ WKH WUHQG FXUYH GHVFULEHG E\ 0DUWLQR LQ WKH SUHYLRXV VHFWLRQf 7KH H[DPSOHV VHOHFWHG E\ %URZQ GR QRW UHYHDO WKH YHUVDWLOLW\ RI WKH PHWKRGV LOOXVWUDn WHG KH GLG KRZHYHU SURYLGH D FRPSUHKHQVLYH UHYLHZ RI OLWHUDWXUH GHVFULELQJ DSSOLFDWLRQV LQ RWKHU ILHOGV $ QXPEHU RI PHWKRGRORJLFDO FRQFHUQV UDLVHG E\ %URZQ ZHUH FRQVLGHUHG LQ WKLV VWXG\ ,Q D FULWLTXH RI VHOHFWHG IXWXUHV SUHGLFWLRQ WHFKQLTXHV WKDW PLJKW EH HPSOR\HG E\ HGXFDWLRQDO SODQQHUV )RON f REVHUYHG WKDW H[SRQHQWLDO WUHQG OLQH DQG DULWKPHWLF VWUDLJKW OLQH SURMHFWLRQV DSSHDU WR EH WKH PRVW FRPPRQO\ XVHG H[WUDn SRODWLYH WHFKQLTXHV 7KLV DXWKRU SURYLGHG D QXPEHU RI XVHIXO PHDVXUHV IRU HYDOXDWLQJ VWDWLVWLFDOO\ GHULYHG UHJUHVVLRQ OLQHV 7KH HGXFDWLRQDO DSSOLFDWLRQV MXVW GHVFULEHG DUH EDVLFDOO\ DWWHPSWV WR SURMHFW LQSXWV VXFK DV PRQH\ SXSLOV RU WHDFKHUV WR WKH HGXFDWLRQDO V\VWHP RU RXWSXWV JUDGXDWHV GHJUHHV HDUQHGf RI WKDW V\VWHP 1R DWWHPSW WR H[WUDSRODWH WKH IXWXUH VWDWXV RI YDULDEOHV ZKLFK DIIHFW WKHVH VWXGHQWUHODWHG LQSXWV RU RXWSXWV ZDV GLVFRYHUHG LQ WKH OLWHUDWXUH VHDUFK ([WUDSRODWLYH 0HWKRGV LQ 2WKHU $UHDV 6HYHUDO DUHDV KDYH GHYHORSHG KLJKO\ VSHFLDOL]HG H[WUDSRn ODWLYH PHWKRGV LQ PDNLQJ IRUHFDVWV RI WKH IXWXUH 3RSXODn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f WKH XQGHUO\LQJ SDWWHUQ RI WKH GDWD WKDW FDQ EH UHFRJQL]HG DQG Ef WKH W\SH RU FODVV RI PRGHO GHVLUHG IURP :KHHOZULJKW i 0DNULGDNLV f %RWK RI WKHVH ZLOO EH EULHIO\ FRQVLGHUHG LQ UHODWLRQ WR WKLV VWXG\ 7KH 3DWWHUQ RI WKH 'DWD )URP JUDSKLFDO UHSUHVHQWDWLRQV RI HDFK LQGLFDWRU WKH GDWD IRU HDFK DSSHDUHG WR EH FKDUDFWHUL]HG E\ D WUHQG ZKLFK HLWKHU LQFUHDVHG RU GHFUHDVHG ZLWK WLPH 6RPH DOVR DSSHDUHG WR FRQWDLQ F\FOLFDO SDWWHUQV DQG UDQGRP IOXFWXDWLRQV ,W VHHPHG DV LI PDMRU WUHQGV PLJKW IROORZ WKH IRUP RI D VWUDLJKW OLQH RU FXUYH ZLWK RQH RU WZR EHQGV 7KH &ODVV RI 0RGHO :KHHOZULJKW DQG 0DNULGDNLV f GLVWLQJXLVKHG IRXU FODVVHV RU FDWHJRULHV RI PRGHOV 7KH WLPH VHULHV PRGHO DOZD\V DVVXPHV WKDW VRPH SDWWHUQ RI FRPELQDWLRQ RI SDWWHUQV LV UHFXUULQJ RYHU WLPH S f PAGE 58 7KH FDXVDO PRGHO DVVXPHV WKDW WKH YDOXH RI D FHUn WDLQ YDULDEOH LV D IXQFWLRQ RI VHYHUDO RWKHU YDULDEOHV S f 7KH VWDWLVWLFDO PRGHO FRPSULVHV D QXPEHU RI IRUHn FDVWLQJ WHFKQLTXHV LW XVHV WKH ODQJXDJH DQG SURFHGXUHV RI VWDWLVWLFDO DQDO\VLV WR LGHQWLI\ SDWWHUQV LQ WKH YDULDEOHV EHLQJ IRUHFDVW DQG LQ PDNLQJ VWDWHPHQWV DERXW WKH UHOLDELOLW\ RI WKHVH IRUHFDVWV S f 7KH QRQVWDWLVWLFDO PRGHO LQFOXGHV DOO PRGHOV WKDW GR QRW IROORZ WKH JHQHUDO UXOHV RI VWDWLVWLFDO DQDO\VLV DQG SUREDELOLW\ S f 2I FRXUVH VRPH WHFKQLTXHV FDQ EH FODVVLILHG LQWR PRUH WKDQ RQH RI WKH IRXU W\SHV RI PRGHOV ,W DSSHDUHG WKDW WKH VWDWLVWLFDO PRGHO ZLWK LWV ZHOOGHILQHG SURSHUWLHV DQG UHSOLFDEOH SURFHGXUHV ZRXOG EH DQ DSSURSULDWH VWDUWLQJ SRLQW IRU SUHGLFWLQJ WKH ORQJWHUP WUHQGV LQ WKH VHOHFWHG WLPH VHULHV GDWD 7KH UHYLHZ RI WKH OLWHUDWXUH UHYHDOHG VHYHUDO WHFKQLTXHV GHQRWHG E\ WKH IRUP RI WKHLU FXUYHV ZKLFK DUH VHQVLWLYH WR ORQJWHUP WUHQGV LQ WKH GDWD DQG ZKLFK DUH FODVVLILHG XQGHU WKH VWDWLVWLFDO PRGHO Df WKH DULWKPHWLF VWUDLJKW OLQH Ef WKH 6VKDSHG JURZWK RU ORJLVWLF FXUYH Ff WKH WUHQG RU H[SRQHQWLDO FXUYH Gf WKH SRO\QRPLDO FXUYH $OO RI WKHVH WHFKQLTXHV DUH UHJUHVVLRQ WHFKQLTXHV VROYHG E\ OHDVW VTXDUHV SURFHGXUHV 7HFKQLTXHV Ef WKURXJK Gf UHTXLUH GDWD WUDQVn IRUPDWLRQV WR VDWLVI\ WKH EDVLF OLQHDU PRGHO XVHG LQ UHJUHVn VLRQ 7KH JURZWK RU ORJLVWLF FXUYH ZDV HOLPLQDWHG IURP FRPn SDULVRQ EHFDXVH WKLV WHFKQLTXH QHFHVVLWDWHV WKH VHWWLQJ RI PAGE 59 OLPLWV ZKLFK PLJKW ELDV WKH UHVXOWV RI WKH VWXG\ GXH WR LWV H[ SRVW IDFWR QDWXUH 7KH UHPDLQLQJ WKUHH WHFKQLTXHV ZHUH FRQVLGHUHG WR EH DSSURSULDWH IRU XVH LQ WKH FRPSDULVRQ SKDVH RI WKLV VWXG\ 'HVFULSWLRQ RI 0HWKRGV WR EH &RPSDUHG 6LQFH WKH WKUHH WHFKQLTXHV VHOHFWHG IRU FRPSDULVRQ DUH LQWULQVLFDOO\ OLQHDU LQ WKHLU SDUDPHWHUV 'UDSHU i 6PLWK f WKH JHQHUDO OLQHDU PRGHO GHQRWHG E\ WKH VLPSOH RU ELYDULDWH UHJUHVVLRQ HTXDWLRQ LV SUHVHQWHG ILUVW $GGLWLRQn DOO\ HVWLPDWLRQ RI WKH SDUDPHWHUV RI WKH HTXDWLRQ E\ OHDVW VTXDUHV SURFHGXUHV WKH DVVXPSWLRQV RI WKH PRGHO DQG FULWHULD IRU HYDOXDWLRQ DQG FRPSDULVRQ RI WKH WKUHH PHWKRGV DUH GLVn FXVVHG (DFK WHFKQLTXH LV WKHQ GHVFULEHG LQ UHODWLRQ WR WKH JHQHUDO OLQHDU PRGHO 7KH *HQHUDO /LQHDU 0RGHO ,Q WKH FRPSDULVRQ RI PHWKRGV XVLQJ VHOHFWHG WLPH VHULHV LQGLFDWRUV WLPH LQ \HDUV LV FRQVLGHUHG WKH LQGHSHQGHQW YDULDEOH DQG WKH LQGLFDWRU LV FRQVLGHUHG WKH GHSHQGHQW RU UHVSRQVH YDULDEOH 7KXV LI WLPH LV GHQRWHG E\ ; DQG WKH LQGLFDWRU LV GHQRWHG E\ < D IXQFWLRQDO UHODWLRQVKLS LQ WKH IRUP < I;f PLJKW EH VWDWHG +RZHYHU VLQFH PRVW VRFLDO UHODWLRQVKLSV DUH VWRFKDVWLF SUREDELOLVWLFf UDWKHU WKDQ GHWHUPLQLVWLF LQ QDWXUH D PRUH DSSURSULDWH IRUP PLJKW EH PAGE 60 < I;f H ZKHUH H UHSUHVHQWV HUURU D PHDVXUH RI WKH XQNQRZQ IDFWRUV :KHQ WKH UHODWLRQVKLS EHWZHHQ WKH WZR YDULDEOHV WLPH DQG WKH LQGLFDWRU PAGE 61 7KH $VVXPSWLRQV RI WKH /LQHDU 0RGHO 'UDSHU DQG 6PLWK f QRWHG WKDW ,Q PDQ\ DVSHFWV RI VWDWLVWLFV LW LV QHFHVVDU\ WR DVVXPH D PDWKHPDWLFDO PRGHO WR PDNH SURJUHVV ,W PLJKW EH ZHOO WR HPSKDVL]H WKDW ZKDW ZH DUH XVXDOO\ GRLQJ LV WR FRQVLGHU RU WHQWDWLYHO\ HQWHUWDLQ RXU PRGHO S nf 7KXV ZKHQ WKH JHQHUDO OLQHDU PRGHO LV HPSOR\HG DV LW LV LQ WKLV VWXG\ LW EHFRPHV QHFHVVDU\ WR H[DPLQH WKH DVVXPSWLRQV XSRQ ZKLFK WKH PRGHO LV EDVHG DQG WR MXGJH ZKHWKHU WKH PRGHO LV LQ IDFW DSSURSULDWH IRU WKH GDWD $VVXPSWLRQV IRU WKH JHQHUDO OLQHDU PRGHO LQFOXGH WKH IROORZLQJ 7KH UHJUHVVLRQ HTXDWLRQ $ < E R EL; i LV D EHWWHU SUHGLFWRU RI < WKDQ < < &ELA f 7KH UHJUHVVLRQ HTXDWLRQ DFFRXQWV IRU D VLJQLILFDQW SRUWLRQ RI WKH YDULDWLRQ LQ < WKDW LV WKH UHODWLRQVKLS EHWZHHQ ; DQG < GHVFULEHG E\ WKH HTXDWLRQ LV QRW WKH UHVXOW RI FKDQFH 7KH HUURU WHUP H KDV D PHDQ YDOXH HTXDO WR ]HUR DQG YDULDQFH HTXDO WR D LW LV DQ LQGHSHQGHQW UDQGRP YDULDEOH ZKLFK LV QRUPDOO\ GLVWULEXWHG ,I WKH ILUVW WZR DVVXPSWLRQV DUH QRW PHW WKHQ WKH PRGHO LV QRW D JRRG SUHGLFWRU IRU WKDW GDWD ,I WKH WKLUG DVVXPSn WLRQ LV QRW PHW WKHQ LW LV QRW DSSURSULDWH WR LQWHUSUHW WKH UHVXOWV VWDWLVWLFDOO\ WKDW LV LQ WHUPV RI WKH SUREDELOLW\ PAGE 62 GLVWULEXWLRQ RI WKH UDQGRP HUURU H ,W LV SRVVLEOH WR WHVW $VVXPSWLRQ DQG $VVXPSWLRQ E\ WKH ) VWDWLVWLF $VVXPSWLRQ LV EHVW HYDOXDWHG E\ SORWWLQJ WKH UHVLGXDOV DQG H[DPLQLQJ WKH SDWWHUQ RI WKH GHYLDWLRQV IURP WKH UHJUHVVLRQ OLQH $QVFRPEH $QVFRPEH i 7XNH\ 'UDSHU i 6PLWK f ,QGHSHQGHQFH RI WKH HUURUV $VVXPSWLRQ >H@f PD\ EH WHVWHG E\ WKH 'XUELQ:DWVRQ WHVW IRU VHULDO FRUUHODWLRQ 'XUELQ i :DWVRQ 'XUELQ :DWVRQ 0D\HV 0D\HV :KHHOZULJKW 0DNULGDNLV f &ULWHULD IRU &RPSDULVRQ RI 0HWKRGV 7KH IROORZLQJ TXHVWLRQV ZHUH GHULYHG IURP WKH OLWHUDWXUH WR JXLGH WKH FRPSDULVRQ RI PHWKRGV 'R WKH GDWD VDWLVI\ WKH DVVXPSWLRQV RI WKH PRGHO" 6HH SUHYLRXV VHFWLRQf +RZ ZHOO GRHV WKH UHJUHVVLRQ OLQH ILW WKH GDWD IURP ZKLFK LW ZDV GHULYHG WKH WZRWKLUGV RI WKH GDWD SRLQWV XVHG WR JHQHUDWH WKH SUHGLFWLRQ HTXDWLRQf" 7XIWH SS f OLVWHG IRXU PHDVXUHV RI TXDOLW\ RI ILW $ D WKH 1 UHVLGXDOV < < L L E WKH UHVLGXDO YDULDWLRQ $ V \ f [ WT <3 1 Nr RU WKH VTXDUH URRW RI WKH UHVLGXDO YDULDWLRQ 6 FDOOHG \ f [ WKH XQELDVHG VWDQGDUG HUURU RI HVWLPDWHf rN UHIHUV WR WKH QXPEHU RI ; WHUPV LQ WKH UHJUHVVLRQ HTXDWLRQ PAGE 63 F WKH UDWLR RI H[SODLQHG WR WRWDO YDULDWLRQ U (< PAGE 64 DQG WKH H[WHQW WR ZKLFK WKLV VWDWLVWLFDO PDQLIHVWDWLRQ RI JURZWK LV VXSSRUWHG E\ RWKHU HYLGHQFH S f VKRXOG EH NHSW LQ PLQG 7KXV WKH FDOFXODWLRQ RI D WUHQG LV PRUH WKDQ D PDWKHPDWLFDO DQDO\VLV LQ FXUYH ILWWLQJ LW LV HVVHQWLDOO\ D SUREOHP RI DQDO\VLV RI WKH SKHQRPHQD UHSUHVHQWHG E\ WKH GDWD 1HLVZDQJHU f ,W VKRXOG EH DOVR QRWHG WKDW ZKLOH WKH VWDQGDUG HUURU RI HVWLPDWH JLYHV DQ RYHUDOO PHDVXUH RI HUURU DURXQG WKH UHJUHVn VLRQ OLQH LW PD\ QRW EH DSSURSULDWH IRU FRPSXWLQJ FRQILGHQFH LQWHUYDOV IRU D VSHFLILF IRUHFDVW YDOXH 7KH UHDVRQ IRU WKLV LV WKDW WKH IXUWKHU DQ ; LV IURP ; WKH ODUJHU LV WKH HUURU WKDW PD\ EH H[SHFWHG ZKHQ SUHGLFWLQJ < IURP WKH UHJUHVVLRQ OLQH 'UDSHU DQG 6PLWK f QRWHG :H PLJKW H[SHFW WR PDNH RXU EHVW SUHn GLFWLRQV LQ WKH PLGGOHf RI RXU REVHUYHG UDQJH RI ; DQG ZRXOG H[SHFW RXU SUHGLFn WLRQV WR EH OHVV JRRG DZD\ IURP WKH PLGGOH S f 7KHUHIRUH WKH FRQILGHQFH OLPLWV IRU WKH WUXH YDOXH RI < IRU D JLYHQ ; DUH WZR FXUYHG OLQHV DERXW WKH UHJUHVVLRQ OLQH 7KH OLPLWV FKDQJH DV WKH SRVLWLRQ RI ; FKDQJHV +HQFH WKH IROORZLQJ HTXDWLRQ ZDV SURYLGHG E\ :KHHOZULJKW DQG 0DNULGDNLV S f IRU FRPSXWLQJ WKH VWDQGDUG HUURU RI IRUHFDVW 6(If a V f n n ,, =< PAGE 65 0HWKRG 6LPSOH /LQHDU 5HJUHVVLRQ 7KH HTXDWLRQ < e;f GHVFULEHV D QDWXUDO IXQFWLRQDO UHODWLRQVKLS EHWZHHQ ; DQG < ,I WKLV IXQFWLRQDO UHODWLRQVKLS FDQ EH H[SUHVVHG E\ D VWUDLJKW OLQH RQ DULWKPHWLF SDSHU WKH OLQHDU ILUVWRUGHU UHJUHVVLRQ HTXDWLRQ < ER EL;i PD\ EH DSSURSULDWH 7KH QDWXUDO OLQHDU IXQFWLRQ LV XVHG ZKHQ DQ DEVROXWH DPRXQW RI FKDQJH LQ < SHU XQLW RI ; LV K\SRWKHn VL]HG 0HWKRG /RJOLQHDU 5HJUHVVLRQ 2FFDVLRQDOO\ ZKHQ WLPH VHULHV GDWD DUH SORWWHG RQ DQ DULWKPHWLF VFDOH WKH VFDWWHU RI SRLQWV IDOO PRUH LQ D FXUYH WKDQ LQ D VWUDLJKW OLQH ZLWK WKH FXUYH ULVLQJ RU GHFUHDVLQJ PRUH UDSLGO\ DV ; LQFUHDVHV 7KHVH VDPH GDWD ZKHQ SORWWHG RQ D VHPLORJDULWKPLF VFDOH ZLOO SURGXFH D VWUDLJKW OLQH 7KH UHODWLRQVKLS EHWZHHQ ; DQG < PD\ WKHQ EH GHVFULEHG E\ ORJ < I;f RU < DE[ WKH H[SRQHQWLDO IRUP RI WKH ORJDULWKPLF UHODWLRQVKLS EHWZHHQ ; DQG < 7KH H[SRQHQWLDO IXQFWLRQ LV XVHG ZKHQ WKHUH LV WKRXJKW WR EH D FRQVWDQW UDWH RI FKDQJH LQ < SHU XQLW DEVROXWH FKDQJH LQ ; 7KXV IRU HDFK \HDU ;f < FKDQJHV E\ D FRQVWDQW SHUn FHQWDJH UDWKHU WKDQ E\ DQ DEVROXWH DPRXQW DV LQ 0HWKRG f PAGE 66 ,W LV SRVVLEOH WR ILW WKH H[SRQHQWLDO IXQFWLRQ WR WKH JHQHUDO OLQHDU PRGHO E\ WUDQVIRUPLQJ WKH YDOXHV RI < WR ORJ < 7KXV < DE EHFRPHV ORJ < D E; ORJ < ORJ D ; ORJ E RU $ ORJ < ORJ E ; ORJ EL H $V LQ WKH FDVH RI WKH QDWXUDO QXPEHU VWUDLJKW OLQH WKH PHWKRG RI OHDVW VTXDUHV LV XVHG WR HVWLPDWH WKH SDUDPHWHUV QHFHVVDU\ IRU FRPSXWLQJ WKH ORJDULWKPLF RU JHRPHWULFf VWUDLJKW OLQH 7XWWOH S f QRWHG WKHUHIRUH WKDW ? ? WKH ORJ PAGE 67 XVLQJ WKH DQWLORJV RI WKH SUHGLFWHG YDOXHV RI ORJ < LQ D UHJUHVVLRQ H[SODLQLQJ YDULDELOLW\ LQ < S f 7KLV 5 PD\ WKHQ EH XVHG IRU FRPSDULVRQ SXUSRVHV 7KH H[DPSOHV JLYHQ E\ 6HLGPDQ f ZHUH EDVHG RQ ORJDULWKPLF WUDQVIRUPDWLRQV RI ERWK GHSHQGHQW DQG LQGHSHQGHQW YDULDEOHV EXW WKH VDPH REn VHUYDWLRQ PD\ EH PDGH ZKHQ RQO\ WKH GHSHQGHQW YDULDEOH LV WUDQVIRUPHG 6HLGPDQnV UHVHUYDWLRQ DERXW 5 KDV EHHQ FRQn VLGHUHG LQ WKLV VWXG\ ,I DQ H[SRQHQWLDO FXUYH DSSHDUV WR ILW WKH GDWD LW LV RIWHQ GHVLUDEOH WR ILQG WKH DQQXDO UDWH RI FKDQJH F 7KLV FDQ EH GHULYHG IURP WKH UHJUHVVLRQ FRHIILFLHQW EL DFFRUGLQJ WR WKH IROORZLQJ HTXDWLRQ ORJ EL Ff FKDQJH DQWLORJ EL 7KH UHVXOW VKRXOG WKHQ EH H[SUHVVHG DV D SHUFHQW 0D\HV 0D\HV S 1LH HW DO S f 7KH FRPPRQ RU %ULJJV ORJDULWKP XVHG LQ WKH < WUDQVn IRUPDWLRQ LQ WKLV VWXG\ LV WKH SRZHU WR ZKLFK PXVW EH UDLVHG WR HTXDO WKH QXPEHU VHH 1HLVZDQJHU S 7XIWH S f 1DWXUDO ORJV RU ORJV WR WKH EDVH FRXOG DOVR KDYH EHHQ XVHG WR REWDLQ WKH VDPH UHVXOWV 6QHGHFRU SS f 0HWKRG 3RO\QRPLDO 5HJUHVVLRQ ,Q 0HWKRG WKH HTXDWLRQ ZKLFK H[SUHVVHV D VWUDLJKW OLQH UHODWLRQVKLS EHWZHHQ ; DQG < LV $ < E R E L ; PAGE 68 ZKLFK LV D OLQHDU LQ WKH EnVf ILUVWRUGHU LQ ;f UHJUHVVLRQ HTXDWLRQ :KHQ WKLV IXQFWLRQDO UHODWLRQVKLS EHWZHHQ ; DQG < FDQ EH H[SUHVVHG DV D VROLG RU XQEURNHQ FXUYH RQ DULWKPHWLF SDSHU WKH OLQHDU VHFRQGRUGHU RU TXDGUDWLFf UHJUHVVLRQ HTXDWLRQ $ < E E; E; PD\ EH DSSURSULDWH :KHQ WKH UHODWLRQVKLS FDQ EH H[SUHVVHG DV D FXUYHG OLQH ZLWK WZR EHQGV RQ DULWKPHWLF SDSHU WKH OLQHDU WKLUGRUGHU RU FXELFf UHJUHVVLRQ HTXDWLRQ $ < ER EL; E; ED; i PD\ EH XVHG $FFRUGLQJ WR .HUOLQJHU DQG 3HGKD]XU S f WKH KLJKHVW RUGHU D SRO\QRPLDO HTXDWLRQ PD\ WDNH LV HTXDO WR 1 ZKHUH 1 LV WKH QXPEHU RI GLVWLQFW YDOXHV LQ WKH LQGHSHQGHQW YDULDEOH +RZHYHU VLQFH RQH RI WKH JRDOV RI VFLHQWLILF UHVHDUFK LV SDUVLPRQ\ RXU LQWHUHVW LV QRW LQ WKH SUHGLFWLYH SRZHU RI WKH KLJKHVW GHJUHH SRO\QRPLDO HTXDWLRQ SRVVLEOH EXW UDWKHU LQ WKH KLJKHVW GHJUHH SRO\QRPLDO HTXDWLRQ QHFHVVDU\ WR GHVFULEH D VHW RI GDWD .HUOLQJHU i 3HGKD]XU S f $QRWKHU UHDVRQ IRU D SDUVLPRQLRXV DSSURDFK WR SRO\QRPLDO FXUYH ILWWLQJ LV WKDW IRU HDFK RUGHU DGGHG WR WKH HTXDWLRQ D GHJUHH RI IUHHGRP LV ORVW 7KLV LV HVSHFLDOO\ LPSRUWDQW ZKHQ WKH QXPEHU RI REVHUYDWLRQV DUH VPDOO DV WKH\ DUH LQ WKLV VWXG\ REVHUYDWLRQV UDQJH IURP WR LQ HDFK RI WKH HLJKW VHWV RI GDWDf $OVR KLJKHU RUGHU SRO\QRPLDO FXUYHV PD\ SRVVHVV VWDWLVWLFDO VLJQLILFDQFH EXW EH GHYRLG RI SUDFWLFDO PAGE 69 VLJQLILFDQFH $FFRUGLQJO\ RQO\ WKH TXDGUDWLF DQG FXELF IRUPV RI WKH SRO\QRPLDO UHJUHVVLRQ HTXDWLRQ DUH FRQVLGHUHG ,Q WKH SRO\QRPLDO UHJUHVVLRQ WKH LQGHSHQGHQW YDULDEOH ; WLPHf LV WUHDWHG DV D FDWHJRULFDO YDULDEOH DQG LV UDLVHG WR D FHUWDLQ SRZHU ,Q WKH TXDGUDWLF HTXDWLRQ HDFK YDOXH RI ; LV VTXDUHG WR FUHDWH D QHZ YHFWRU RI WKH VTXDUHG ;nV ; 6LPLODUO\ LQ WKH FXELF HTXDWLRQ HDFK YDOXH RI ; LV FXEHG WR FUHDWH DQ DGGLWLRQDO YHFWRU RI WKH FXEHG ;nV ; 7KXV WKH UHVXOWLQJ HTXDWLRQ FDQ EH VROYHG E\ D VWHSZLVH PXOWLSOH UHJUHVVLRQ SURFHGXUH LQ ZKLFK DW HDFK VWHS RI WKH DQDO\VLV WKH 5 LV WHVWHG WR VHH LI WKH KLJKHUGHJUHH SRO\n QRPLDO DFFRXQWV IRU D VLJQLILFDQW SURSRUWLRQ RI WKH YDULDQFH :KLOH D OHDVW VTXDUHV VROXWLRQ LV XVHG LQ WKLV VWXG\ WKH YDOXHV RI WKH XQNQRZQV PD\ DOVR EH IRXQG E\ RUWKRJRQDO SRO\n QRPLDOV VHH 'UDSHU i 6PLWK SS *UHHQZDOG SS .HUOLQJHU i 3HGKD]XU SS f 1HLVZDQJHU SS f QRWHG WKDW WKH VHFRQG GHJUHH DQG WKLUGGHJUHH SDUDERODV SURYLGH JUHDWHU IOH[Ln ELOLW\ LQ ILWWLQJ D OLQH WR D VHW RI GDWD IRU WKH SDUDERODV DOORZ D WUHQG WR FKDQJH GLUHFWLRQ :KHWKHU RU QRW WKH IOH[LELOLW\ RI WKH SDUDEROLF IXQFWLRQ HQKDQFHV WKH SUHGLFn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f 7KH IROORZLQJ TXHVWLRQV GHULYHG IURP WKH OLWHUDWXUH ZHUH SURSRVHG LQ &KDSWHU ,,, WR JXLGH WKH FRPSDULVRQ RI PHWKRGV 'R WKH GDWD VDWLVI\ WKH DVVXPSWLRQV RI WKH JHQHUDO OLQHDU PRGHO" +RZ ZHOO GRHV WKH UHJUHVVLRQ OLQH ILW WKH GDWD IURP ZKLFK LW ZDV GHULYHG WKH WZRWKLUGV RI WKH GDWD SRLQWV XVHG WR JHQHUDWH WKH SUHGLFWLRQ HTXDWLRQf" +RZ ZHOO GRHV WKH H[WUDSRODWHG OLQH ILW WKH GDWD WKH RQHWKLUG RI WKH GDWD SRLQWV WKDW ZHUH QRW XVHG WR JHQHUDWH WKH SUHGLFWLRQ HTXDWLRQf" 7R DQVZHU WKHVH TXHVWLRQV LQ WHUPV RI HDFK PHWKRG DQG WR IDFLOLWDWH FRPSDULVRQ DPRQJ WKH WKUHH PHWKRGV WKH UHVXOWV REWDLQHG IURP DSSO\LQJ HDFK RI WKH PHWKRGV WR HDFK RI WKH HLJKW LQGLFDWRU GDWD VHWV DUH SUHVHQWHG LQ WKH IROORZLQJ PDQQHU 7KH ILW RI WKH UHJUHVVLRQ OLQH WR WKH REVHUYHG GDWD LV LQGLFDWHG E\ U DQG WKH XQELDVHG VWDQGDUG HUURU RI HVWLPDWH 6 )RU WKH VLPSOH OLQHDU DQG ORJOLQHDU UHJUHV f VLRQ PHWKRGV WKH DPRXQW RI YDULDQFH DFFRXQWHG IRU E\ WKH UHJUHVVLRQ OLQH LV WHVWHG E\ WKH ) VWDWLVWLF ) YDOXH LV WKH PAGE 72 VDPH DV WKDW REWDLQHG E\ GLYLGLQJ EL E\ 6(A f )RU WKH TXDGn UDWLF DQG FXELF IRUPV RI SRO\QRPLDO UHJUHVVLRQ ERWK WKH U LQFOXGLQJ DOO RUGHUV HQWHUHG WR WKDW VWHS U A RU U\ r DQG WKH LQFUHDVH LQ U DWWULEXWDEOH WR WKH ODVW RUGHU HQWHUHG LQ WKH UHJUHVVLRQ UUR RU UU DUH WHVWHG ZLWK WKH E \f \fn ) VWDWLVWLFr 2I FRXUVH GLYLGLQJ WKH SDUWLDO UHJUHVVLRQ FRHIILFLHQWV E LQ WKH TXDGUDWLF IRUP DQG E LQ WKH FXELF IRUP E\ WKHLU UHVSHFWLYH VWDQGDUG HUURUV ZLOO DOVR \LHOG WKH VDPH ) YDOXH IRU WKH LQFUHDVH LQ Uf 7KH ILW RI WKH H[WUDSRODWHG OLQH WR WKH GDWD LV LQGLFDWHG QXPHULFDOO\ E\ WKH VWDQGDUG HUURU IRU WKH H[WUDn SRODWHG YDOXHV 6 7KLV PHDVXUH UHIOHFWV WKH A f H[W\[f DYHUDJH GHYLDWLRQ RI WKH H[WUDSRODWHG YDOXHV IURP WKH REn VHUYHG YDOXHV RI < WKXV f f nAH[W \ f [f 1RWH WKDW WKLV HTXDWLRQ FRPSXWLQJ 6 [f Q RI LV QRW =< PAGE 73 $OO REVHUYHG DQG SUHGLFWHG YDOXHV RI < DUH UHSRUWHG LQ WDEXODU DQG JUDSKLF IRUP 7KH UHVLGXDOV DURXQG WKH UHJUHVVLRQ OLQH ZHUH H[n DPLQHG IRU VHULDO FRUUHODWLRQ E\ WKH 'XUELQ:DWVRQ G VWDWLVWLF ZKLFK LV QRWHG RQO\ ZKHQ VHULDO FRUUHODWLRQ LV FRQILUPHG RU TXHVWLRQDEOH $GGLWLRQDOO\ WKH VWDQGDUGL]HG UHVLGXDOV ZHUH SORWWHG DJDLQVW WKH VHTXHQFH RI FDVHV DQG DOVR DJDLQVW $ VWDQGDUGL]HG < YDOXHV 6XFK YLVXDO LQVSHFWLRQ RI WKH GDWD LV GLVFXVVHG DV QHFHVVDU\ WR VXSSRUW WKH LQWHUSUHWDWLRQ RI UHn VXOWV LQ &KDSWHU 9 3UHVHQWDWLRQ RI 5HVXOWV ,QGLFDWRU 7KH PHDQ DQG VWDQGDUG WR JHQHUDWH WKH UHJUHVVLRQ VSHFWLYHO\ 7KH IROORZLQJ $ GHULYH < $ /LQHDU < GHYLDWLRQ HTXDWLRQV UHJUHVVLRQ IRU WKH < YDO DUH DQG HTXDWLRQV ZH XHV XVHG UHn UH XVHG WR ; 4XDGUDWLF < ; ; &XELF < f; ; f; /RJOLQHDU ORJ < ; 7KH JRRGQHVV RI ILW RI WKH UHJUHVVLRQ OLQHV GHULYHG IURP WKHVH HTXDWLRQV LV LQGLFDWHG E\ U U FKDQJH DQG 6 [ LQ 7DEOH $Q $129$ VXPPDU\ WDEOH LV SUHVHQWHG LQ 7DEOH LQ WKH $SSHQn GL[ 7KH RYHUDOO )nV IRU DOO PHWKRGV DUH VLJQLILFDQW M!f WKH LQFUHDVHV LQ U GXH WR WKH KLJKHU RUGHU SRO\QRPLDOV DUH QRW VLJQLILFDQW KRZHYHU PAGE 74 7DEOH ,QGLFDWRU 6XPPDU\ 6WDWLVWLFV IRU 3UHGLFWLRQ (TXDWLRQV E\ 0HWKRG U U FKDQJH ) GI 6 \ f [ /LQHDU rr 4XDGUDWLF rr &XELF rr ORJ < DQWLORJ < U L f E /RJOLQHDU rr ORJ FRQVWDQW GROODUV F%RWK TXDGUDWLF DQG FXELF IRUPV RI WKH SRO\QRPLDO UHJUHVn VLRQ DUH SUHVHQWHG A%RWK U DQG 6 KDYH EHHQ UHFRPSXWHG XVLQJ DQWLORJV RI D \ f ; WKH ORJ < PXFK RI WKH GLIIHUHQFH EHWZHHQ UORJ PAGE 75 7KH DYHUDJH HUURUV IRU WKH H[WUDSRODWHG < YDOXHV 6 2 DFFRUGLQJ WR PHWKRG HPSOR\HG DUH Df OLQHDU H[W L\f[ M Ef TXDGUDWLF Ff FXELF DQG Gf ORJOLQHDU 2EVHUYHG PAGE 76 7DEOH ,QGLFDWRU 2EVHUYHG PAGE 77 0HGLDQ IDPLO\ LQFRPH FRQVWDQW GROODUVf )LJXUH ,QGLFDWRU 2EVHUYHG PAGE 78 7DEOH $Q $129$ VXPPDU\ WDEOH LV SUHVHQWHG LQ 7DEOH LQ WKH $SSHQGL[ 7KH RYHUDOO )nV IRU WKH TXDGUDWLF DQG FXELF IRUPV RI WKH SRO\QRPLDO UHJUHVVLRQ DUH VLJQLILFDQW Sf KRZHYHU RQO\ WKH LQFUHDVH LQ U GXH WR WKH TXDGUDWLF LV VLJQLILFDQW MNf 7KH DYHUDJH HUURUV IRU WKH H[WUDSRODWHG < YDOXHV 6H[ W & \ ;ff DFFRUGLQi W WALH PHWKRG HPSOR\HG DUH Df OLQHDU Ef TXDGUDWLF Ff FXELF DQG Gf ORJOLQHDU 2EVHUYHG PAGE 79 7DEOH ,QGLFDWRU 6XPPDU\ 6WDWLVWLFV IRU 3UHGLFWLRQ (TXDWLRQV E\ 0HWKRG U U FKDQJH ) GI 6 \ f [ /LQHDU R 4XDGUDWLF r rr &XELF r ORJ < DQWLORJ < /RJ LQHDUN ORJ 1RWH ,QGLFDWRU LV QXPEHU RI IDPLOLHV LQ WKH 8QLWHG 6WDWHV KHDGHG E\ ZRPHQ H[SUHVVHG DV D SHUFHQWDJH RI WRWDO IDPLOLHV %RWK TXDGUDWLF DQG FXELF IRUPV RI WKH SRO\QRPLDO UHJUHVn VLRQ DUH SUHVHQWHG A%RWK U DQG 6 KDYH EHHQ UHFRPSXWHG XVLQJ DQWLORJV RI D \n [ WKH ORJ < PXFK RI WKH GLIIHUHQFH EHWZHHQ UORJ PAGE 80 7DEOH ,QGLFDWRU 2EVHUYHG PAGE 81 3HUFHQW RI IDPLOLHV KHDGHG E\ ZRPHQ PAGE 82 ,QGLFDWRU 7KH PHDQ DQG VWDQGDUG GHYLDWLRQ IRU WKH < YDOXHV XVHG WR JHQHUDWH WKH UHJUHVVLRQ HTXDWLRQV DUH DQG UHVSHFn WLYHO\ 7KH IROORZLQJ UHJUHVVLRQ HTXDWLRQV ZHUH XVHG WR GH $ ULYH < /LQHDU < ; $ 4XDGUDWLF < ; ; $ &XELF < ; f; ; $ /RJOLQHDU ORJ < ; 7KH JRRGQHVV RI ILW RI WKH UHJUHVVLRQ OLQHV GHULYHG IURP WKHVH HTXDWLRQV LV LQGLFDWHG E\ U U FKDQJH DQG 6Y LQ 7DEOH $Q $129$ VXPPDU\ WDEOH LV SUHVHQWHG LQ 7DEOH LQ WKH $SSHQn GL[ 7KH RYHUDOO )nV IRU DOO PHWKRGV DUH VLJQLILFDQW Sf KRZHYHU WKH LQFUHDVHV LQ U GXH WR WKH KLJKHU RUGHU SRO\QRn PLDOV DUH QRW VLJQLILFDQW 7KH DYHUDJH HUURUV IRU WKH H[WUDSRODWHG < YDOXHV 6H[W\[ff DFFRUGLQJ WR WKH PHWKRG HPSOR\HG DUH Df OLQHDU Ef TXDGUDWLF Ff FXELF DQG Gf ORJOLQHDU 2EVHUYHG PAGE 83 7DEOH ,QGLFDWRU 6XPPDU\ 6WDWLVWLFV IRU 3UHGLFWLRQ (TXDWLRQV E\ 0HWKRG U U FKDQJH ) GI 9} /LQHDU rr 4XDGUDWLF rr &XELF rr ORJ < $ DQWLORJ < W L f E /RJ LQHDU rr ORJ 1RWH ,QGLFDWRU 7V WKH QXPEHU RI ZLYHV LQ WKH ODERU IRUFH H[SUHVVHG DV D SHUFHQWDJH RI WRWDO ZLYHV LQ WKH 8QLWHG 6WDWHV %RWK TXDGUDWLF DQG FXELF IRUPV RI WKH SRO\QRPLDO UHJUHVn VLRQ DUH SUHVHQWHG E %RWK U DQG 6 KDYH EHHQ UHFRPSXWHG XVLQJ DQWLORJV RI Q \ f [ A WKH ORJ < PXFK RI WKH GLIIHUHQFH EHWZHHQ UORJ PAGE 84 7DEOH ,QGLFDWRU 2EVHUYHG PAGE 85 3HUFHQW RI ZLYHV LQ ODERU IRUFH n n n n n )LJXUH ,QGLFDWRU 2EVHUYHG PAGE 86 :KHQ WKH OLQHV DUH H[WUDSRODWHG EH\RQG WKH RULJLQDO YDOXHV KRZHYHU WKH FXELF IRUP LV FOHDUO\ WKH ZRUVW ILW ZLWK D 6 A RI ZKLOH WKH ORJOLQHDU IRUP LV FOHDUO\ H[W\[f E WKH EHVW ILW ZLWK D 6 F RI H[W\f[f ,QGLFDWRU 7KH PHDQ DQG VWDQGDUG GHYLDWLRQ IRU WKH < YDOXHV XVHG WR JHQHUDWH WKH UHJUHVVLRQ HTXDWLRQV DUH DQG UHn VSHFWLYHO\ 7KH IROORZLQJ UHJUHVVLRQ HTXDWLRQV ZHUH XVHG WR 9 GHULYH < /LQHDU < 4XDGUDWLF < f; ; Y &XELF < ; f; ; $ /RJOLQHDU ORJ < f; 7KH JRRGQHVV RI ILW RI WKH UHJUHVVLRQ OLQHV GHULYHG IURP WKHVH HTXDWLRQV LV LQGLFDWHG EY U U FKDQJH DQG 6 LQ 7DEOH $Q $129$ VXPPDU\ WDEOH LV SUHVHQWHG LQ 7DEOH LQ WKH $SSHQn GL[ 7KH RYHUDOO ) VWDWLVWLF IRU WKH FXELF IRUP RI WKH SRO\n QRPLDO UHJUHVVLRQ LV VLJQLILFDQW Sf WKH LQFUHDVH LQ U GXH WR WKH WKLUG GHJUHH SRO\QRPLDO LV DOVR VLJQLILFDQW Sf 7KH ) VWDWLVWLF IRU ERWK WKH OLQHDU DQG ORJOLQHDU PHWKRGV LV VLJQLILFDQW Sf 7KH DYHUDJH HUURUV IRU WKH H[WUDSRODWHG < YDOXHV 6H[W\[ff DFFRUGLQJ WR PHWKRG HPSOR\HG DUH Df OLQHDU Ef TXDGUDWLF Ff FXELF DQG Gf ORJOLQHDU 2EVHUYHG PAGE 87 7DEOH ,QGLFDWRU 6XPPDU\ 6WDWLVWLFV IRU 3UHGLFWLRQ (TXDWLRQV E\ 0HWKRG U U FKDQJH ) GI 69; /LQHDU r 4XDGUDWLF &XELF rr rr ORJ < DQWLORJ $ < r /RJOLQHDUA r ORJ H[SUHVVHG DV UDWH SHU SRSXODWLRQ LQ )ORULGD %RWK TXDGUDWLF DQG FXELF IRUPV RI WKH SRO\QRPLDO UHJUHVn VLRQ DUH SUHVHQWHG E %RWK U DQG 6 KDYH EHHQ UHFRPSXWHG XVLQJ DQWLORJV RI WKH ORJ < PXFK RI WKH GLIIHUHQFH EHWZHHQ UORJ PAGE 88 7DEOH ,QGLFDWRU 2EVHUYHG PAGE 89 0DUULDJHV SRSXODWLRQ 2EVHUYHG PAGE 90 7KXV LW ZRXOG DSSHDU WKDW IRU WKH RULJLQDO UHJUHVVLRQ WKH FXELF IRUP RI WKH SRO\QRPLDO EHVW ILWV WKH REVHUYHG PAGE 91 7DEOH ,QGLFDWRU 6XPPDU\ 6WDWLVWLFV IRU 3UHGLFWLRQ (TXDWLRQV E\ 0HWKRG U U FKDQJH ) GI 6 \[ /LQHDU 4XDGUDWLF &XELFD rr rr /RJ LQHDUA ORJ < DQWLORJ < ORJ 1RWH ,QGLFDWRU LV QXPEHU RI GLVVROXWLRQV RI PDUULDJH LQ )ORULGD H[SUHVVHG DV UDWH SHU SRSXODWLRQ %RWK TXDGUDWLF DQG FXELF IRUPV RI WKH SRO\QRPLDO UHJUHVn VLRQ DUH SUHVHQWHG A%RWK U DQG 6 KDYH EHHQ UHFRPSXWHG XVLQJ DQWLORJV RI D \ f D D WKH ORJ < PXFK RI WKH GLIIHUHQFH EHWZHHQ UORJ PAGE 92 YDOXH IRU WKH LQFUHDVH LQ U IRU WKH FXELF FRPSRQHQW LV DOVR VLJQLILFDQW Sf 7KH DYHUDJH HUURUV IRU WKH H[WUDSRODWHG < YDOXHV 6H[W\[ff DFFRUGLQJ WR WKH PHWKRG HPSOR\HG DUH Df OLQHDU Ef TXDGUDWLF Ff FXELF DQG Gf ORJOLQHDU 2EVHUYHG PAGE 93 7DEOH ,QGLFDWRU 2EVHUYHG PAGE 94 'LVVROXWLRQV RI PDUULDJH SRSXODWLRQ )LJXUH f n L 6 n n n n ,QGLFDWRU 2EVHUYHG PAGE 95 ? WKHVH HTXDWLRQV LV LQGLFDWHG E\ U U FKDQJH DQG 6 LQ \ f [ 7DEOH $Q $129$ VXPPDU\ WDEOH LV SUHVHQWHG LQ 7DEOH LQ WKH $SSHQGL[ 7KH RYHUDOO )nV IRU WKH TXDGUDWLF DQG FXELF IRUPV RI SRO\QRPLDO UHJUHVVLRQ DUH VLJQLILFDQW Sf RQO\ WKH LQFUHDVH LQ U GXH WR WKH TXDGUDWLF FRPSRQHQW LV VLJQLILn FDQW Sf KRZHYHU 7KH DYHUDJH HUURUV IRU WKH H[WUDSRODWHG < YDOXHV 6H[W\ DFFRUGLQJ WR WKH PHWKRG HPSOR\HG DUH Df OLQHDU Ef TXDGUDWLF Ff FXELF DQG Gf ORJOLQHDU 2EVHUYHG PAGE 96 7DEOH ,QGLFDWRU 6XPPDU\ 6WDWLVWLFV IRU 3UHGLFWLRQ (TXDWLRQV E\ 0HWKRG U U FKDQJH ) GI 6 \ f [ /LQHDU R 4XDGUDWLF rr rr &XELFD rr W L f E /RJOLQHDU ORJ < f DQWLORJ < ORJ 1RWH ,QGLFDWRU LV QXPEHU RI UHVLGHQW OLYH ELUWKV 7Q )ORULGD H[SUHVVHG DV UDWH SHU SRSXODWLRQ f%RWK TXDGUDWLF DQG FXELF IRUPV RI WKH SRO\QRPLDO UHJUHVn VLRQ DUH SUHVHQWHG A%RWK U DQG 6 KDYH EHHQ UHFRPSXWHG XVLQJ DQWLORJV RI Q \ Â‘ [ Q WKH ORJ < PXFK RI WKH GLIIHUHQFH EHWZHHQ UORJ PAGE 97 7DEOH ,QGLFDWRU 2EVHUYHG PAGE 98 5HVLGHQW OLYH ELUWKV SRSXODWLRQ )LJXUH ,QGLFDWRU 2EVHUYHG PAGE 99 ,QGLFDWRU 7KH PHDQ DQG VWDQGDUG GHYLDWLRQ IRU WKH < YDOXHV XVHG WR JHQHUDWH WKH UHJUHVVLRQ HTXDWLRQV DUH DQG UHn VSHFWLYHO\ 7KH IROORZLQJ UHJUHVVLRQ HTXDWLRQV ZHUH XVHG WR $ GHULYH < $ /LQHDU < ; 4XDGUDWLF < ; ; &XELF < ; f; ; $ /RJOLQHDU ORJ < ; 7KH JRRGQHVV RI ILW RI WKH UHJUHVVLRQ OLQHV GHULYHG IURP WKHVH HTXDWLRQV LV LQGLFDWHG E\ U U FKDQJH DQG 6 LQ 7DEOH \ f [ $Q $129$ VXPPDU\ WDEOH LV SUHVHQWHG LQ 7DEOH LQ WKH $SSHQn GL[ 7KH RYHUDOO )nV IRU DOO PHWKRGV DUH VLJQLILFDQW ef KRZHYHU WKH LQFUHDVHV LQ U GXH WR WKH KLJKHU RUGHU SRO\QRn PLDOV DUH QRW VLJQLILFDQW 7KH DYHUDJH HUURUV IRU WKH H[WUDSRODWHG < YDOXHV 6H[W \ [Mf DFFRUGLQJ WR WKH PHWKRG HPSOR\HG DUH Df OLQHDU Ef TXDGUDWLF Ff FXELF DQG Gf ORJOLQHDU 2EVHUYHG DQG SUHGLFWHG YDOXHV RI < IRU ERWK WKH RULJLQDO UHJUHVVLRQ DQG WKH H[WUDSRODWHG OLQHV DUH SUHVHQWHG E\ PHWKRG LQ 7DEOH 7KHVH GDWD DUH JUDSKLFDOO\ UHSUHVHQWHG LQ )LJXUH 7KHUH LV YHU\ OLWWOH GLIIHUHQFH LQ WKH SUHGLFWLYH YDOXH RI WKH PHWKRGV IRU WKH RULJLQDO UHJUHVVLRQ (DFK PHWKRG DFFRXQWV IRU b RI WKH YDULDQFH LQ < WKH UDQJH RI WKH 6\ IRU DOO PHWKRGV LV IURP WR :KHQ WKH OLQHV DUH H[WUDSRODWHG EH\RQG WKH RULJLQDO PAGE 100 7DEOH ,QGLFDWRU 6XPPDU\ 6WDWLVWLFV IRU 3UHGLFWLRQ (TXDWLRQV E\ 0HWKRG U U FKDQJH ) GI 6 \ f [ /LQHDU rr T 4XDGUDWLF rr &XELF rr /RJ LQHDU' ORJ < f DQWLORJ < rr ORJ 1RWH ,QGLFDWRU LV QXPEHU RI WR \HDU ROGV HQUROOHG LQ QXUVHU\ VFKRRO DQG NLQGHUJDUWHQ H[SUHVVHG DV SHUFHQWDJH RI WRWDO FKLOGUHQ WR \HDUV ROG LQ WKH 8QLWHG 6WDWHV T %RWK TXDGUDWLF DQG FXELF IRUPV RI WKH SRO\QRPLDO UHJUHVn VLRQ DUH SUHVHQWHG A%RWK U DQG 6 KDYH EHHQ UHFRPSXWHG XVLQJ DQWLORJV RI WKH ORJ < PXFK RI WKH GLIIHUHQFH EHWZHHQ UORJ PAGE 101 7DEOH ,QGLFDWRU 2EVHUYHG PAGE 102 3HUFHQW RI WR \HDU ROGV LQ QXUVHU\ VFKRRO DQG NLQGHUJDUWHQ )LJXUH ,QGLFDWRU 2EVHUYHG PAGE 103 YDOXHV WKH TXDGUDWLF DQG OLQHDU PHWKRGV DUH VRPHZKDW EHWWHU SUHGLFWRUV ZLWK 6 RI DQG UHVSHFWLYHO\ 6KRXOG WKH OLQHV EH H[WUDSRODWHG IXUWKHU LQWR WKH IXWXUH KRZHYHU WKH FXELF DQG ORJOLQHDU PHWKRGV PLJKW LPSURYH LQ SUHGLFWLYH DFFXUDF\ VHH )LJXUH f ,QGLFDWRU 7KH PHDQ DQG VWDQGDUG GHYLDWLRQ IRU WKH < YDOXHV XVHG WR JHQHUDWH WKH UHJUHVVLRQ HTXDWLRQV DUH DQG UHn VSHFWLYHO\ 7KH IROORZLQJ UHJUHVVLRQ HTXDWLRQV ZHUH XVHG WR GHULYH < /LQHDU < ; 2XDGUDWLF < ; ; &XELF < f; ; & f; /RJOLQHDU ORJ < ; 7KH JRRGQHVV RI ILW RI WKH UHJUHVVLRQ OLQHV GHULYHG IURP WKHVH HTXDWLRQV LV LQGLFDWHG E\ U U FKDQJH DQG 6 LQ \ f [ 7DEOH $Q $129$ VXPPDU\ WDEOH LV SUHVHQWHG LQ 7DEOH LQ WKH $SSHQGL[ 7KH RYHUDOO )nV IRU DOO PHWKRGV DUH VLJQLILn FDQW Sf KRZHYHU RQO\ WKH LQFUHDVH LQ U GXH WR WKH TXDGUDWLF FRPSRQHQW LV VLJQLILFDQW Sf 7KH DYHUDJH HUURUV IRU WKH H[WUDSRODWHG < YDOXHV 6H[W\ [ff DFFRUGLQJ WR WKH PHWKRG HPSOR\HG DUH Df OLQHDU Ef TXDGUDWLF Ff FXELF DQG Gf ORJOLQHDU 2EVHUYHG DQG SUHGLFWHG YDOXHV RI < IRU ERWK WKH RULJLQDO UHJUHVVLRQ DQG WKH H[WUDSRODWHG OLQHV DUH SUHVHQWHG E\ PHWKRG LQ 7DEOH 7KHVH GDWD DUH JUDSKLFDOO\ UHSUHn VHQWHG LQ )LJXUH PAGE 104 7DEOH ,QGLFDWRU 6XPPDU\ 6WDWLVWLFV IRU 3UHGLFWLRQ (TXDWLRQV E\ 0HWKRG U U FKDQJH ) GI 6 \ f [ /LQHDU rr D 4XDGUDWLF rr rr &XELF rr /RJ LQHDUr ORJ < f DQWLORJ < rr ORJ GLYRUFH RU DQQXOPHQW H[SUHVVHG DV UDWH SHU FKLOGUHQ XQGHU ,6 \HDUV ROG LQ WKH 8QLWHG 6WDWHV %RWK TXDGUDWLF DQG FXELF IRUPV RI WKH SRO\QRPLDO UHJUHVn VLRQ DUH SUHVHQWHG A%RWK U DQG 6 KDYH EHHQ UHFRPSXWHG XVLQJ DQWLORJV RI WKH ORJ < PXFK RI WKH GLIIHUHQFH EHWZHHQ UORJ PAGE 105 7DEOH ,QGLFDWRU 2EVHUYHG PAGE 106 &KLOGUHQ LQYROYHG LQ GLYRUFH FKLOGUHQ XQGHU +fÂ§O n n n n n )LJXUH ,QGLFDWRU 2EVHUYHG PAGE 107 $OO WKH PHWKRGV DUH DGHTXDWH SUHGLFWRUV RI < IRU WKH UDQJH RI WKH RULJLQDO UHJUHVVLRQ 7KH TXDGUDWLF DQG FXELF IRUPV RI WKH SRO\QRPLDO UHJUHVVLRQ DUH RQO\ VOLJKWO\ EHWWHU WKDQ WKH OLQHDU DQG ORJOLQHDU PHWKRGV KRZHYHU DFFRXQWLQJ IRU DSSUR[LPDWHO\ b PRUH YDULDQFH LQ < ZLWK D 6AU [ RI DQG UHVSHFWLYHO\ FRPSDUHG ZLWK DQG IRU WKH ODWWHU PHWKRGVf :KHQ WKH OLQHV DUH H[WUDSRODWHG EH\RQG WKH RULJLQDO YDOXHV WKH TXDGUDWLF IRUP FOHDUO\ EHFRPHV WKH VXSHULRU SUHGLFWRU ZLWK D 6 RI 7KH 6 IRU WKH U H[W\[f H[W\[f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n FOXVLYHO\ UHMHFWHG PAGE 108 &+$37(5 9 ',6&866,21 7ZR REMHFWLYHV RI WKLV VWXG\ ZHUH Df WR VHOHFW XVLQJ %URQIHQEUHQQHUnV HFRORJ\ RI HGXFDWLRQ PRGHO DQG RSHUDWLRQDOO\ GHILQH DW OHDVW YDULDEOHV WKDW UHVHDUFK KDV VKRZQ WR EH UHODWHG WR WKH RXWFRPHV RI HGXFDWLRQ Ef WR XVH WKHVH YDULDEOHV RSHUDWLRQDOO\ GHILQHG DV WLPH VHULHV LQGLFDWRUV LQ WKH FRPSDULVRQ RI WKUHH SXUHO\ H[WUDSRODWLYH IRUHFDVWLQJ PHWKRGV ,Q WKLV FKDSWHU PHWKRGRORJLFDO VWUDWHJLHV LQYROYHG LQ WKH VHOHFWLRQ DQG RSHUDWLRQDO GHILQLWLRQ RI VRFLDO YDUn LDEOHV DV ZHOO DV WKH XVH RI WKH %URQIHQEUHQQHU PRGHO DUH GLVFXVVHG LQ WHUPV RI WKHLU YLDELOLW\ IRU IXWXUH XVH 5Hn VXOWV RI WKH PHWKRG FRPSDULVRQ SKDVH DUH GLVFXVVHG DFFRUGLQJ WR VWDWLVWLFDO DQG SUDFWLFDO FRQVLGHUDWLRQV GHULYHG IURP WKH OLWHUDWXUH 7KH 9DULDEOHV 6HOHFWLRQ 7KH YDULDEOHV VHOHFWHG QHHG WR EH UHFRQVLGHUHG VLQFH GHQWnV VHQVH RI IDWH FRQWURO URRPf DUH UDWKHU REVFXUH ERWK IRU XVH LQ WKLV VWXG\ PD\ VHYHUDO RI WKHP HJ VWX WHDFKHU EHKDYLRU LQ WKH FODVV LQ WKHLU FRQVWLWXWLYH DQG LQ PAGE 109 WKHLU RSHUDWLRQDO GHILQLWLRQV $GGLWLRQDOO\ VHYHUDO RI WKH YDULDEOHV HJ DGPLQLVWUDWLYH OHDGHUVKLS VW\OHf DUH RQO\ SHULSKHUDOO\ UHODWHG WR HGXFDWLRQDO RXWFRPHV 7KHUH DSSHDUV WR EH D QHHG WR GHYHORS LQVWHDG D QXPEHU RI GHPRJUDSKLF YDULDEOHV UHODWHG WR WKH IDPLO\ 3UHVHQWO\ WKHUH DUH YHU\ IHZ HIIRUWV WR GHILQH V\VWHPDWLFDOO\ DQDO\]H DQG IRUHFDVW WKH GHPRJUDSKLF FRQWH[W RI HGXFDWLRQ &RDWHV 1RWH 0RUULVRQ f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nV (FRORJ\ RI (GXFDWLRQ 0RGHO %URQIHQEUHQQHUn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f DQG 6LVPRQGR f 7KH %URQIHQEUHQQHU PRGHO PD\ KHOS WR ILOO WKLV QHHG IRU HGXFDn WLRQDO SODQQHUV 2SHUDWLRQDO 'HILQLWLRQ RI 9DULDEOHV 7KH RSHUDWLRQDO GHILQLWLRQ RI VRFLDO YDULDEOHV LQ WHUPV RI WLPH VHULHV LQGLFDWRUV LV D FKDOOHQJLQJ DQG RIWHQ IUXVn WUDWLQJ WDVN 6HYHUDO RI WKH SUREOHPV DFNQRZOHGJHG E\ )HUULVV f DQG RWKHUV LQYROYH WKH QDWXUH DQG DYDLODELOLW\ RI WKH GDWD 6RPH GDWD DUH WKRXJKW WR EH RI SRRU TXDOLW\ ZLWK WKH EDVLV IRU FROOHFWLRQ FKDQJLQJ IUHTXHQWO\ RYHU WLPH 2WKHU GDWD DUH FROOHFWHG RQO\ IRU FHUWDLQ \HDUV RU DUH PDGH DYDLOn DEOH RQO\ IRU FHUWDLQ \HDUV 7KHUH DUH QXPHURXV VRXUFHV RI GDWD KRZHYHU LW UHTXLUHV FDUHIXO UHYLHZ DQG FRQVLGHUDWLRQ RI DYDLODEOH GDWD WR ILQG WKH PRVW UHOLDEOH PHDVXUHV IRU XVH LQ DQDO\VLV DQG IRUHFDVWLQJ 2WKHU SUREOHPV DULVH IURP WKH SURFHVV RI VRFLDO PHDVXUHn PHQW LWVHOI 6RFLDO LQYHVWLJDWRUV QHHG QRW OHW WKHVH FRQFHUQV GHWHU HIIRUWV WR GHYHORS VRFLDO LQGLFDWRUV UDWKHU WKHVH FRQFHUQV DV GHVFULEHG E\ (W]LRQL DQG /HKPDQ f PLJKW EH XVHG WR JXLGH IXWXUH HIIRUWV WR RSHUDWLRQDOO\ GHILQH VRFLDO YDULDEOHV 3HUKDSV WKH PRVW LPSRUWDQW IXQFWLRQV RI WLPH VHULHV LQGLFDWRUV DUH Df WKH FRPSDULVRQ RI WUHQGV LQ DQ LQGLFDWRU PAGE 111 RYHU WLPH DQG Ef WKH FRPSDULVRQ RI GLIIHUHQW JURXSV LQ UHn ODWLRQ WR D VSHFLILF LQGLFDWRU 7KXV IRU DQ LQGLFDWRU VXFK DV ELUWKV WR XQZHG PRWKHUV WKH DJJUHJDWHG GDWD PD\ EH H[DPLQHG IRU D JHQHUDO VRFLHWDO WUHQG WKH JHQHUDO WUHQG PD\ EH GLVDJJUHJDWHG DFFRUGLQJ WR VWDWH RU FRXQW\ DQG WKHQ DFFRUGLQJ WR UDFH DQG DJH RI PRWKHU ,W LV RIWHQ RQO\ ZKHQ WKH GDWD DUH GLVDJJUHJDWHG IRU H[DPSOH ELUWKV WR XQZHG PRWKHUV DJH f WKDW WKH\ EHFRPH PRVW PHDQLQJIXO IRU WKH SDUWLFXODU SODQQLQJ DFWLYLW\ $ JUDSKLF SUHVHQWDWLRQ RI WKH GLVDJJUHJDWHG GDWD LV D YDOXDEOH SDUW RI DQ\ DWWHPSW WR XQGHUn VWDQG WKH XQGHUO\LQJ VRFLDO SURFHVV UHSUHVHQWHG E\ WKH WLPH VHULHV LQGLFDWRU 7KH ([WUDSRODWLYH 0HWKRGV 6WDWLVWLFDO &RQVLGHUDWLRQV 7KH XVH RI WKH JHQHUDO OLQHDU PRGHO ZDV DSSURSULDWH LQ WKLV VWXG\ :KLOH YLRODWLRQV RI WKH DVVXPSWLRQV GLG RFFXU RFFDVLRQDOO\ LQ WKH OLQHDU RU ORJOLQHDU PHWKRGV WKH\ FRXOG EH FRUUHFWHG E\ WKH DGGLWLRQ RI WKH TXDGUDWLF DQGRU FXELF WHUPV WR WKH UHJUHVVLRQ HTXDWLRQ ,Q DOO GDWD VHWV RQH RU PRUH RI WKH PHWKRGV DFFRXQWHG IRU VLJQLILFDQW DPRXQWV RI WKH YDULDQFH LQ < ,Q WZR RI WKH GDWD VHWV KRZHYHU WKHUH UHPDLQHG TXHVWLRQDEOH SDWWHUQV LQ WKH UHVLGXDOV WKDW ZHUH QRW FRUUHFWHG E\ DQ\ RI WKH PHWKRGV HPSOR\HG &RLQFLGHQWDOO\ WKH H[WUDn SRODWHG OLQHV RI WKHVH LQGLFDWRUV QXPEHUV DQG f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n PLQDWLRQ U DQG WKH XQELDVHG VWDQGDUG HUURU RI HVWLPDWH 6 \ f [ ,Q WKLV VWXG\ KRZHYHU U DQG 6 ZHUH QRW DOZD\V JRRG LQ \ r [ GLFDWRUV RI WKH DFFXUDF\ RI H[WUDSRODWHG YDOXHV WR WKH REVHUYHG GDWD )RU H[DPSOH ZKLOH WKH FXELF IRUP RI WKH SRO\QRPLDO EHVW ILW WKH REVHUYHG PAGE 113 ,I WKH VXPPDU\ VWDWLVWLFV RI U DQG 6AU [ DUH QRW DOZD\V XVHIXO LQ FKRRVLQJ WKH EHVW PHWKRG IRU WKH H[WUDSRODWLRQ SKDVH WKHQ RQ ZKDW EDVLV VKRXOG WKH PHWKRG EH FKRVHQ" 3HUKDSV WKLV TXHVWLRQ FDQ RQO\ EH DQVZHUHG E\ D FRPELQDWLRQ RI VWUDWHJLHV Df DSSOLFDWLRQ RI VHYHUDO GLIIHUHQW PHWKRGV WR WKH GDWD Ef IRU HDFK PHWKRG H[WUDSRODWLRQ RI WUHQG IRU VSHFLILHG QXPEHU RI \HDUV LQWR WKH IXWXUH Ff JUDSKLF UHSUHVHQWDWLRQ RI RULJLn QDO GDWD DQG SUHGLFWHG GDWD IRU HDFK PHWKRG Gf YLVXDO H[n DPLQDWLRQ RI GDWD WR GHWHUPLQH PRVW OLNHO\ GLUHFWLRQ IRU REn VHUYHG WUHQG WR WDNH EDVHG RQ NQRZOHGJH RI WKH VRFLDO SKHQRPn HQRQ EHLQJ VWXGLHG DQG Hf FRQVWUXFWLRQ RI XSSHU DQG ORZHU OLPLWV IRU WKH H[WUDSRODWHG WUHQG FRQVLGHUHG WR EH WKH PRVW DSSURSULDWH 7KHUHIRUH ZKLOH WKH PDWKHPDWLFDO H[WUDSRODWLRQ RI WUHQGV LV D XVHIXO DSSURDFK WR IRUHFDVWLQJ DSSURSULDWH DSSOLFDWLRQ UHPDLQV DQ DUW ,QGHHG IRUHFDVWLQJ DFFXUDF\ LV GHSHQGHQW XSRQ WKH MXGJPHQW RI WKH LQYHVWLJDWRU DQG KLV RU KHU NQRZOHGJH RI WKH VRFLDO SKHQRPHQRQ /HVW WKH SUREOHPV SUHVHQWHG RYHUVKDGRZ WKH SRVLWLYH DWWULEXWHV RI PDWKHPDWLFDO H[WUDSRODWLRQ LW VKRXOG EH QRWHG WKDW OLQHDU UHJUHVVLRQ LV D SRZHUIXO WRRO ZKLFK FDQ EH HPSOR\HG ERWK LQ SUHOLPLQDU\ DQDO\VLV RI WKH GDWD DQG LQ WKH SURMHFWLRQ RI IXWXUH WUHQGV (YHQ LI WKH DVVXPSWLRQV DUH YLRODWHG WKH JHQHUDO OLQHDU PRGHO PD\ VWLOO EH HPSOR\HG IRU DQDO\VLV RI WUHQGV YLRODWLRQ RI WKH DVVXPSWLRQ KRZHYHU SUHFOXGHV PDNLQJ DQ\ VWDWHPHQW RI D SUREDELOLVWLF QDWXUH +D\V S f PAGE 114 3UDFWLFDO &RQVLGHUDWLRQV (YDOXDWLRQ RI D IRUHFDVW PXVW EH EDVHG RQ PRUH WKDQ VWDWLVWLFDO FULWHULD 2WKHU FRQVLGHUDWLRQV VXFK DV SXUSRVH RI WKH IRUHFDVW DQG WKH XVHUnV QHHG IRU DFFXUDF\ DUH LQYROYHG VHH +DUULVRQ 0DUWLQR Df 7KXV ILQDO HYDOXDn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n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nV WLPH VHULHV LQn GLFDWRUV DQG VHOHFWHG H[WUDSRODWLYH PHWKRGV DUH GLVFXVVHG 6XPPDU\ (GXFDWLRQDO SODQQHUV DQG SROLF\ PDNHUV QHHG DGHTXDWH LQIRUPDWLRQ DERXW WKH VRFLHWDO FRQWH[W RI HGXFDWLRQ LQ RUGHU WR PDNH DSSURSULDWH GHFLVLRQV DERXW WKH IXWXUH UROH DQG IXQFWLRQ RI HGXFDWLRQ *LYHQ WKLV QHHG WKH SUREOHP LQ WKLV VWXG\ ZDV Df WR VHOHFW XVLQJ %URQIHQEUHQQHUnV HFRORJ\ RI HGXFDWLRQ PRGHO DQG RSHUDWLRQDOO\ GHILQH DW OHDVW YDULDEOHV WKDW UHVHDUFK KDV VKRZQ WR EH UHODWHG WR WKH RXWn FRPHV RI HGXFDWLRQ Ef WR XVH WKHVH YDULDEOHV RSHUDWLRQDOO\ GHILQHG DV WLPH VHULHV LQGLFDWRUV LQ WKH FRPSDULVRQ RI WKUHH SXUHO\ H[WUDSRODWLYH IRUHFDVWLQJ PHWKRGV DQG Ff WR GHULYH LPSOLFDWLRQV IRU WKH XVH RI DQ HFRORJLFDO PRGHO VXFK DV %URQIHQEUHQQHUnV WLPH VHULHV LQGLFDWRUV DQG VHOHFWHG H[WUDSRODWLYH PHWKRGV IRU HGXFDWLRQDO SODQQLQJ PAGE 116 7KH 9DULDEOHV 7HQ YDULDEOHV UHODWHG WR WKH RXWFRPHV RI HGXFDWLRQ ZHUH VHOHFWHG WKHVH YDULDEOHV ZHUH WKHQ FODVVLILHG DFFRUGLQJ WR %URQIHQEUHQQHUfV HFRORJ\ RI HGXFDWLRQ PRGHO 7KH YDULDEOHV ZHUH RSHUDWLRQDOO\ GHILQHG ZKHQ SRVVLEOH DV WLPH VHULHV LQn GLFDWRUV 'DWD ZHUH FROOHFWHG IRU WKHVH LQGLFDWRUV DQG HLJKW LQGLFDWRUV ZKLFK PHW WKH FULWHULD HVWDEOLVKHG LQ WKH VWXG\ ZHUH VHOHFWHG IRU XVH LQ WKH PHWKRG FRPSDULVRQ SKDVH 7KH 0HWKRGV 7KH OLWHUDWXUH LQ IRUHFDVWLQJ PHWKRGRORJ\ HFRQRPLFV VWDWLVWLFV DQG WLPH VHULHV DQDO\VLV ZDV UHYLHZHG LQ RUGHU WR GHWHUPLQH WKH PRVW DSSURSULDWH H[WUDSRODWLYH PHWKRGV WR XVH ZLWK WKH VHOHFWHG WLPH VHULHV LQGLFDWRUV 7KUHH VWDWLVn WLFDO PHWKRGV GHULYHG IURP WKH JHQHUDO OLQHDU PRGHO ZHUH VHOHFWHG IRU FRPSDULVRQ 7KHVH PHWKRGV LQFOXGHG VLPSOH OLQHDU UHJUHVVLRQ VHFRQG DQG WKLUG GHJUHH SRO\QRPLDO UHn JUHVVLRQ DQG ORJOLQHDU UHJUHVVLRQ LQ ZKLFK WKH GHSHQGHQW YDULDEOH XVXDOO\ D UDWH RU SHUFHQWDJH XQGHUJRHV ORJDULWKPLF WUDQVIRUPDWLRQf 7LPH LQ \HDUV ZDV XVHG DV WKH LQGHSHQGHQW YDULDEOH (DFK PHWKRG ZDV DSSOLHG WR HDFK WLPH VHULHV LQGLFDWRU (DFK GDWD VHW LYDV GLYLGHG LQWR WKLUGV WZRWKLUGV RI WKH GDWD SRLQWV ZHUH XVHG WR HVWDEOLVK WKH SUHGLFWLRQ HTXDWLRQ 7KLV HTXDWLRQ ZDV XVHG WR SUHGLFW WKH UHPDLQLQJ WKLUG RI WKH GDWD SRLQWV 7KH PHWKRGV ZHUH WKHQ FRPSDUHG DFFRUGLQJ WR Df WKH ILW RI WKH SUHGLFWHG YDOXHV WR WKH REVHUYHG YDOXHV PAGE 117 IRU WKH RULJLQDO UHJUHVVLRQ DQG Ef WKH ILW RI WKH H[WUDn SRODWHG YDOXHV WR REVHUYHG YDOXHV %RWK VXPPDU\ VWDWLVWLFV DQG YLVXDO LQVSHFWLRQ RI WKH GDWD ZHUH XWLOL]HG 5HVXOWV 5HVXOWV RI WKH PHWKRG FRPSDULVRQ ZHUH Df QR RQH PHWKRG ZDV D VXSHULRU SUHGLFWRU IRU DOO LQGLFDWRUV Ef HDFK PHWKRG ZDV D VXSHULRU SUHGLFWRU RI DW OHDVW RQH LQGLFDWRU DQG Ff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n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f WKH %URQIHQEUHQQHU PRGHO Ef VHOHFWHG VRFLDO YDULDEOHV RSHUDWLRQDOL]HG DV WLPH VHULHV LQGLFDWRUV DQG Ff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n VLW\ RI SHUVSHFWLYHV DQG ZLWK D YDULHW\ RI VNLOOV ,Q &KDSWHU 9 S f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f QRWHG WKDW WKH SUREOHPV RI VRFLDO IRUHFDVWLQJ DUH WKH SUREOHPV RI DOO VRFLDO VFLHQFH LQTXLU\ UHODWHG WR VRFLDO SURFHVVHV DQG WKH PRGHV IRU UHVROYLQJ WKHVH SUREOHPV DUH LQ PDQ\ ZD\V WKH VDPH S f 7KXV WKHRU\EDVHG VRFLDO IRUHFDVWLQJ XWLOL]LQJ WLPH VHULHV LQGLFDWRUV KDV FRQVLGHUDEOH SRWHQWLDO DSSOLFDWLRQ LQ HGXFDn WLRQDO SODQQLQJ RQJRLQJ HPSLULFDO VWXGLHV DUH QHHGHG KRZn HYHU WR DFWXDOL]H WKLV SRWHQWLDO ,PSOLFDWLRQV IRU 3ODQQHUV DQG 3ROLF\ 0DNHUV 7KH %URQIHQEUHQQHU HFRORJ\ RI HGXFDWLRQ PRGHO VRFLDO LQGLFDWRUV DQG H[WUDSRODWLYH PHWKRGV FDQ EH XVHIXO WRROV IRU HGXFDWLRQDO SODQQHUV DQG SROLF\ PDNHUV DW WKH QDWLRQDO VWDWH DQG ORFDO OHYHOV -RKQVRQ f KDV QRWHG )URP WKH YLHZSRLQW RI IXWXUHV UHVHDUFK VRFLDO LQGLFDWRUV PD\ EH UHJDUGHG DV D PHDQV IRU GHYHORSLQJ PRUH DGHTXDWH DQVZHUV WR WKH DJHROG TXHVWLRQV PAGE 120 :KDW DUH ZH ZKDW DUH ZH LQ WKH SURFHVV RI EHFRPLQJ DQG GR ZH OLNH ZKDW ZH VHH" S f &HUWDLQO\ HGXFDWLRQDO SODQQHUV QHHG WR PRQLWRU VRFLHWDO SURFHVVHV DQG LQWHUSUHW FKDQJH LQ UHODWLRQ WR SDVW SUHVHQW DQG IXWXUH HGXFDWLRQDO SXUSRVHV DQG RXWFRPHV 3ROLF\ PDNHUV QHHG WR EH LQIRUPHG RI WKH SUREDEOH RU SRVVLEOH FRQVHTXHQFHV RI D FRQWLQXDWLRQ RI REVHUYHG WUHQGV DQG RI SRVVLEOH UHVSRQVLYHQHVV RI WKHVH WUHQGV WR GLIIHUHQW VHWV RI SRVWXODWHG FRQGLn WLRQV -RKQVRQ S f 7KXV VRFLDO IRUHFDVWLQJ VWUDWHJLHV VXFK DV WKRVH GHVFULEHG LQ WKLV VWXG\ VKRXOG EH D SDUW RI RQJRLQJ HGXFDn WLRQDO SODQQLQJ HIIRUWV 'DWD RQ VRFLDO LQGLFDWRUV PLJKW EH LQFRUSRUDWHG LQWR H[LVWLQJ PDQDJHPHQW LQIRUPDWLRQ V\VWHPV :KLOH LQLWLDWLRQ RI VXFK D GDWD EDQN ZRXOG UHTXLUH D VXEn VWDQWLDO LQYHVWPHQW RI UHVRXUFHV PDLQWHQDQFH ZRXOG EH PLQLPDO )RUHFDVWV FRXOG WKHQ EH LVVXHG DQG UHYLVHG DW UHJXODU LQWHUYDOV &HUWDLQO\ WKH XVH RI VRFLDO IRUHFDVWLQJ VWUDWHJLHV FDQ DVVLVW HGXFDWLRQDO SODQQHUV DQG SROLF\ PDNHUV LQ GHYHORSLQJ DQ XQGHUVWDQGLQJ RI WKH SDVW SHUn VSHFWLYH LQ DQDO\]LQJ WKH SUHVHQW DQG KRSHIXOO\ YLVLRQ LQ SODQQLQJ IRU WKH IXWXUH PAGE 121 ;,1GG9 PAGE 122 7DEOH ,QGLFDWRU $129$ 6XPPDU\ 7DEOHV E\ 0HWKRG 6RXUFH RI 9DULDWLRQ 66 GI 06 ) /LQHDU 5HJUHVVLRQ 5HVLGXDO rr 4XDGUDWLF 5HJUHVVLRQ 5HVLGXDO rr &XELF 5HJUHVVLRQ 5HVLGXDO rr 5HJUHVVLRQ 5HVLGXDO /RJOLQHDU rr 1RWH ,QGLFDWRU LV PHGLDQ IDPLO\ FRQVWDQW GROODUV LQFRPH H[SUHVVHG LQ rref PAGE 123 7DEOH ,QGLFDWRU $129$ 6XPPDU\ 7DEOHV E\ 0HWKRG 6RXUFH RI 9DULDWLRQ 66 GI 06 ) /LQHDU 5HJUHVVLRQ 5HVLGXDO 4XDGUDWLF 5HJUHVVLRQ r 5HVLGXDO &XELF 5HJUHVVLRQ r 5HVLGXDO /RJOLQHDU 5HJUHVVLRQ 5HVLGXDO 1RWH ,QGLFDWRU LV QXPEHU RI IDPLOLHV LQ WKH 8QLWHG 6WDWHV KHDGHG E\ ZRPHQ H[SUHVVHG DV D SHUFHQWDJH RI WRWDO IDPLOLHV re PAGE 124 ,OO 7DEOH ,QGLFDWRU $129$ 6XPPDU\ 7DEOHV E\ 0HWKRG 6RXUFH RI 9DULDWLRQ 66 GI 06 ) 5HJUHVVLRQ /LQHDU rr 5HVLGXDO 5HJUHVVLRQ 4XDGUDWLF rr 5HVLGXDO 5HJUHVVLRQ &XELF rr 5HVLGXDO 5HJUHVVLRQ /RJ LQHDU rr 5HVLGXDO 1RWH ,QGLFDWRU LV QXPEHU RI ZLYHV LQ WKH ODERU IRUFH H[SUHVVHG DV D SHUFHQWDJH RI WRWDO ZLYHV LQ WKH 8QLWHG 6WDWHV rre PAGE 125 7DEOH ,QGLFDWRU $129$ 6XPPDU\ 7DEOHV E\ 0HWKRG 6RXUFH RI 9DULDWLRQ 66 GI 06 ) /LQHDU 5HJUHVVLRQ r 5HVLGXDO 4XDGUDWLF 5HJUHVVLRQ 5HVLGXDO &XELF 5HJUHVVLRQ rr 5HVLGXDO /RJ LQHDU 5HJUHVVLRQ r 5HVLGXDO 1RWH ,QGLFDWRU LV QXPEHU RI PDUULDJHV LQ )ORULGD H[n SUHVVHG DV UDWH SHU SRSXODWLRQ LQ )ORULGD rS rre PAGE 126 7DEOH ,QGLFDWRU $129$ 6XPPDU\ 7DEOHV E\ 0HWKRG 6RXUFH RI 9DULDWLRQ 66 GI 06 ) /LQHDU 5HJUHVVLRQ 5HVLGXDO 4XDGUDWLF 5HJUHVVLRQ 5HVLGXDO &XELF 5HJUHVVLRQ rr 5HVLGXDO /RJOLQHDU 5HJUHVVLRQ 5HVLGXDO 1RWH ,QGLFDWRU LV QXPEHU RI GLVVROXWLRQV RI PDUULDJH LQ )ORULGD H[SUHVVHG DV UDWH SHU SRSXODWLRQ rrS PAGE 127 7DEOH ,QGLFDWRU $129$ 6XPPDU\ 7DEOHV E\ 0HWKRG 6RXUFH RI 9DULDWLRQ 66 GI 06 ) /LQHDU 5HJUHVVLRQ 5HVLGXDO 4XDGUDWLF 5HJUHVVLRQ rr 5HVLGXDO &XELF 5HJUHVVLRQ rr 5HVLGXDO /RJ OLQHDU 5HJUHVVLRQ 5HVLGXDO 1RWH ,QGLFDWRU LV QXPEHU RI UHVLGHQW OLYH ELUWKV LQ )ORULGD H[SUHVVHG DV UDWH SHU SRSXODWLRQ rrS PAGE 128 7DEOH ,QGLFDWRU $129$ 6XPPDU\ 7DEOHV E\ 0HWKRG 6RXUFH RI 9DULDWLRQ 66 GI 06 ) /LQHDU 5HJUHVVLRQ rr 5HVLGXDO 4XDG UDWLF 5HJUHVVLRQ rr 5HVLGXDO &XELF 5HJUHVVLRQ rr 5HVLGXDO /RJ LQHDU 5HJUHVVLRQ rr 5HVLGXDO 1RWH ,QGLFDWRU LV QXPEHU RI WR \HDU ROGV HQUROOHG LQ QXUVHU\ VFKRRO DQG NLQGHUJDUWHQ H[SUHVVHG DV SHUFHQWDJH RI WRWDO FKLOGUHQ WR \HDUV ROG LQ WKH 8QLWHG 6WDWHV rrS PAGE 129 7DEOH ,QGLFDWRU $129$ 6XPPDU\ 7DEOHV E\ 0HWKRG 6RXUFH RI 9DULDWLRQ 66 GI 06 ) /LQHDU 5HJUHVVLRQ rr 5HVLGXDO 4XDGUDWLF 5HJUHVVLRQ rr 5HVLGXDO &XELF 5HJUHVVLRQ rr 5HVLGXDO /RJ LQHDU 5HJUHVVLRQ rr 5HVLGXDO 1RWH ,QGLFDWRU LV QXPEHU RI FKLOGUHQ LQYROYHG LQ GLYRUFH RU DQQXOPHQW H[SUHVVHG DV UDWH SHU FKLOGUHQ XQGHU \HDUV ROG LQ WKH 8QLWHG 6WDWHV rre PAGE 130 5()(5(1&( 127(6 &RDWHV ) 7UHQGV DQG LVVXHV LQIOXHQFLQJ HOHPHQWDU\ DQG VHFRQGDU\ HGXFDWLRQ :RUNLQJ SDSHU SUHSDUHG IRU WKH 86 &RQJUHVVf :DVKLQJWRQ '& 2IILFH RI 7HFKQRORJ\ $VVHVVPHQW -DQXDU\ &RRSHU 3 (FRQRPHWULF VRIWZDUH SDFNDJH &RPSXWHU SURJUDP PDQXDOff &KLFDJR 8QLYHUVLW\ RI &KLFDJR 6FKRRO RI %XVLQHVV 0D\ 6WRYHU 7UHQG LPSDFW DQDO\VLV SURJUDP PDQXDO *ODVWRQEXU\ &RQQ 7KH )XWXUHV *URXS 1RYHPEHU %H]ROG & 6WUDWHJLF SROLF\ DVVHVVPHQW $FWLYLWLHV DQG SURVSHFWV 3DSHU SUHSDUHG IRU WKH &RPPLVVLRQ RQ &ULWLFDO &KRLFHV IRU $PHULFDQVf &HQWHU IRU *RYHUQPHQWDO 5HVSRQn VLELOLW\ 6KRVWDN $ % 7KH 0RQGDOH EOXHSULQW WKH VWDWH DQG WKH DSSOLHG VRFLDO VFLHQFHV 3DSHU SUHVHQWHG DW WKH PHHWLQJ RI WKH (DVWHUQ 6RFLRORJLFDO 6RFLHW\ 1HZ PAGE 131 5()(5(1&(6 $GYLVRU\ &RPPLWWHH RQ &KLOG 'HYHORSPHQW 7RZDUG D QDWLRQDO SROLF\ IRU FKLOGUHQ DQG IDPLOLHV :DVKLQJWRQ '& 1DWLRQDO $FDGHP\ RI 6FLHQFHV 7 $OODUGW ( $ IUDPH RI UHIHUHQFH IRU VHOHFWLQJ VRFLDO LQGLn FDWRUV +HOVLQNL 6RFLHWDV 6FLHQWLDUXP )HQQLFD $QGUHZV ) 0 :LWKH\ 6 % 6RFLDO LQGLFDWRUV RI ZHOOn EHLQJ $PHULFDQV n SHUFHSWLRQV RI OLIH TXDOLW\ 1HZ PAGE 132 %URZQ (GXFDWLRQDO WUHQG DQDO\VLV PHWKRGV ,Q 6 +HQ FOF\ PAGE 133 *RROHU ' 7KH GHYHORSPHQW DQG XVH RI HGXFDWLRQDO LQGLFDn WRUV ,Q (GXFDWLRQDO LQGLFDWRUV 0RQLWRULQJ WKH VWDWH RI HGXFDWLRQ 3URFHHGLQJV RI WKH (76 ,QYLWDWLRQDO &RQIHUHQFHf 3ULQFHWRQ 1(GXFDWLRQDO 7HVWLQJ 6HUn YLFH *RUGRQ & ,9 (Gf 8VHV RI WKH VRFLRORJ\ RI HGXFDWLRQ 166( PAGE 134 0DNULGDNLV 6 ,ORGJVGRQ $ eM :KHHOZULJKW 6 & $Q LQWHUn DFWLYH IRUHFDVWLQJ V\VWHP 7KH $PHULFDQ 6WDWLVWLFLDQ 0DULHQ 0 =LHJOHU : / (GVf 7KH SRWHQWLDO RI HGXFDWLRQDO IXWXUHV %HOPRQW &DOLI :DGVZRUWK 3XEOLVKLQJ &RPSDQ\ ,QF 0DUWLQR 3 (YDOXDWLQJ IRUHFDVW YDOLGLW\ ,Q 5 %ULJKW I7 0 ( ) 6FKRHPDQ (GVf $ JXLGH WR SUDFWLFDO WHFK QRORJLFDO IRUHFDVWLQJ (QJOHZRRG &OLIIV 13UHQWLFH +DOO ,QF IM 0DUWLQR 3 7UHQG H[WUDSRODWLRQ ,Q 5 %ULJKW i 0 ( ) 6FKRHPDQ (GVf $ JXLGH WR SUDFWLFDO WHFKQRORJLFDO IRUHFDVWLQJ (QJOHZRRG &OLIIV 13UHQWLFH+DOO ,QF Ef 0DUWLQR 3 6XUYH\ RI IRUHFDVWLQJ PHWKRGV3DUW :RUOG )XWXUH 6RFLHW\ %XOOHWLQ 1RY'HF SS 0D\HV $ & 0D\HV ,QWURGXFWRU\ HFRQRPLF VWDWLVWLFV 1HZ PAGE 135 3XOOLDP I %RZPDQ 5 (GXFDWLRQDO IXWXULVP,Q SXUVXDQFH RI VXUYLYDO 1RUPDQ 2NOD 8QLYHUVLW\ R7 2NODKRPD 3UHVV 5XQ\RQ 5 3 i +DEHU $ )XQGDPHQWDOV RI EHKDYLRUDO VWDWLVWLFV 5HDGLQJ 0DVV $GGLVRQ:HVOH\ 3XEOLVKLQJ &R 6HLGPDQ 2Q FKRRVLQJ EHWZHHQ OLQHDU DQG ORJOLQHDU PRGHOV -RXUQDO RI 3ROLWLFV 6KDQH + 7KH HGXFDWLRQDO VLJQLILFDQFH RI WKH IXWXUH %ORRPLQJWRQ ,QG 3KL 'HOWD .DSSD ,QF 6KHOGRQ ( % i )UHHPDQ + ( 1RWHV RQ VRFLDO LQGLFDWRUV 3URPLVHV DQG SRWHQWLDO 3ROLF\ 6FLHQFHV 6KHOGRQ ( % /DQG & 6RFLDO UHSRUWLQJ IRU WKH nV $ UHYLHZ DQG SURJUDPPDWLF VWDWHPHQW 3ROLF\ 6FLHQFHV 6KHOGRQ ( % 0RRUH : ( (GVf ,QGLFDWRUV RI VRFLDO FKDQJH &RQFHSWV DQG PHDVXUHPHQWV 1HZ PAGE 136 7KRPSVRQ 5 % (Gf )ORULGD VWDWLVWLFDO DEVWUDFW *DLQHVYLOOH )OD 7KH 8QLYHUVLW\ 3UHVVHV RI )ORULGD Df 7KRPSVRQ 5 % 0HDVXULQJ WKH TXDOLW\ RI OLIH 3DUW %XVLQHVV DQG (FRQRPLF 'LPHQVLRQV f P 7KRPS V QQ 5 % 0HDVXULQJ WKH TXDOLW\ RI OLIH 3DUW ,, %XVLQHVV DQG (FRQRPLF 'LPHQVLRQV f 7XIWH ( 5 'DWD DQDO\VLV IRU SROLWLFV DQG SROLF\ (QJOHn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n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a *UDGXDWH 6FKRRO PAGE 139 81,9(56,7< 2) )/25,'$ THE PROJECTION OF SOCIAL TRENDS MS TNG TIME SERIES INDICATORS: METHODOLOGY AND APPLICATION IN EDUCATIONAL PLANNING By JANE COUNIHAN NELSON A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL 01 THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE RHOUIRFMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 197 7 Copyright 1977 by Jane Counihan Nelson ACKNOWLEDGMENTS I wish to express my sincere appreciation to my Supervisory Committee: to Dr. Michael Y. Nunnery, chairman, for his candor and understanding whether acting as counselor or critic; to Dr. Phillip A. Clark and Dr. Gordon D. Lawrence, for providing valuable suggestions and support. A special thank you goes to Dr. Arthur J. Lewis, director of the DOE social forecasting project, for his leadership, receptivity to new ideas, and especially for his confidence in me; and to my colleagues on the project, Dr. Robert S. Soar and Ms. Linda Troup, for their substantial contribution to the conceptual framework presented in this study. For carefully reviewing the statistical portions of this manuscript, I am indebted to my friend, Dr. Azza S. Guertin. I am especially grateful to my husband Edward for his love and encouragement and for sharing with me hi$f enthuÂ¬ siasm for scientific inquiry. in TABLE OF CONTENTS Page ACKNOWLEDGMENTS iii LIST OF TABLES vii LIST OF FIGURES ix ABSTRACT x CHAPTER I INTRODUCTION 1 Background and Significance of the Study 1 The Social Context of Education 1 The Futures Perspective in Educational Planning 4 Forecasting Trends in Social Variables 5 The Need for Research 7 The Problem 9 Delimitations and Limitations 9 Definition of Terms 11 Procedures 13 The Selection and Operational Definition of Variables 13 Collection of Time Series Indicator Data .... 14 Comparison of Extrapolative Methods Using Time Series Indicators 15 Development of Implications for Educational Planning 18 CHAPTER II RATIONALE FOR SELECTION OF VARIABLES/ TIME SERIES INDICATORS 19 The Social Indicator Movement 20 Historical Development 20 Definition and Use of Social Indicators 24 Data Base for Social Indicators 26 Educational Implications 27 Selection of Variables/Time Series Indicators ... 28 The Variables 28 Bronfenbrenner's Ecology of Education Model. . . 29 Operational Definition of Variables as Time Series Indicators 31 CHAPTER III RATIONALE FOR EXTRAPOLATIVE METHODS SELECTED FOR COMPARISON 38 IV TABLE OF CONTENTS (continued) Page Overview of Extrapolative Forecasting Methods ... 38 Economic and Business Forecasting 39 Technological Forecasting 41 Educational Forecasting 42 Extrapolative Methods in Other Areas 43 Applicability of Reviewed Extrapolative Methods for Study 44 The Pattern of the Data 44 The Class of Model 44 Description of Methods to be Compared 46 The General Linear Model 46 The Assumptions of the Linear Model 48 Criteria for Comparison of Methods 49 Method 1: Simple Linear Regression 52 Method 2: Log-linear Regression 52 Method 3: Polynomial Regression 54 CHAPTER IV COMPARISON OF EXTRAPOLATIVE METHODS USING SELECTED TIME SERIES INDICATORS 57 Presentation of Results 60 Indicator 1 60 Indicator 2 Â» 62 Indicator 3 69 Indicator 4 73 Indicator 5 77 Indicator 6 79 Indicator 7 86 Indicator 8 90 CHAPTER V DISCUSSION 95 The Variables 95 Selection 95 Bronfenbrenner's Ecology of Education Model. . . 96 Operational Definition of Variables 97 The Extrapolative Methods 98 Statistical Considerations 98 Practical Considerations 101 CHAPTER VI SUMMARY, CONCLUSIONS, AND IMPLICATIONS OF STUDY 102 Summary 102 The Variables 103 The Methods 103 Results 104 Conclusions 104 Suggestions for Future Research 105 Implications for Planners and Policy Makers .... 106 v TABLE OF CONTENTS (continued) Page APPENDIX 109 REFERENCE NOTES 117 REFERENCES 118 BIOGRAPHICAL SKETCH 124 LIST OF TABLES Page Table 1 Time Series Indicators of Social Variables Affecting Outcomes of Education 33 Table 2 Indicator 1: Summary Statistics for Prediction Equations by Method 61 Table 3 Indicator 1: Observed Y's and Predicted Y ' s by Method 63 Table 4 Indicator 2: Summary Statistics for Prediction Equations by Method 66 Table 5 Indicator 2: Observed Y's and Predicted Y's by Method 67 Table 6 Indicator 3: Summary Statistics for Prediction Equations by Method 70 Table 7 Indicator 3: Observed Y's and Predicted Y's by Method 71 Table 8 Indicator 4: Summary Statistics for Prediction Equations by Method 74 Table 9 Indicator 4: Observed Y's and Predicted Y's by Method 75 Table 10 Indicator 5: Summary Statistics for Prediction Equations by Method 78 Table 11 Indicator 5: Observed Y's and Predicted Y's by Method 80 Table 12 Indicator 6: Summary Statistics for Prediction Equations by Method 83 Table 13 Indicator 6: Observed Y's and Predicted Y's by Method 84 Table 14 Indicator 7: Summary Statistics for Prediction Equations by Method 87 Table 15 Indicator 7: Observed Y's and Predicted Y's by Method 88 v i i LIST OF TABLES (continued) Page Table 16 Indicator 8: Summary Statistics for Prediction Equations by Method 91 Table 17 Indicator 8: Observed Y's and Predicted Y's by Method 92 Table 18 Indicator 1: ANOVA Summary Tables by Method 109 Table 19 Indicator 2: ANOVA Summary Tables by Method 110 Table 20 Indicator 3: ANOVA Summary Tables by Method Ill Table 21 Indicator 4: ANOVA Summary Tables by Method 112 Table 22 Indicator 5: ANOVA Summary Tables by Method 113 Table 23 Indicator 6: ANOVA Summary Tables by Method 114 Table 24 Indicator 7: ANOVA Summary Tables by Method 115 Table 25 Indicator 8: ANOVA Summary Tables by Method 116 vii i LIST OF FIGURES Page Figure 1 Bronfenbrenner's ecological structure of the educational environment. (Based upon Bronfenbrenner's [1976] description, pp. 5-6) 3 Figure 2 Indicator 1: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extraÂ¬ polation) 64 Figure 3 Indicator 2: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extraÂ¬ polation) 68 Figure 4 Indicator 3: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extraÂ¬ polation) 72 Figure 5 Indicator 4: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extraÂ¬ polation) 76 Figure 6 Indicator 5: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extraÂ¬ polation) 81 Figure 7 Indicator 6: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extraÂ¬ polation) 85 Figure 8 Indicator 7: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extraÂ¬ polation) 89 Figure 9 Indicator 8: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extraÂ¬ polation) 93 IX Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE PROJECTION OF SOCIAL TRENDS USING TIME SERIES INDICATORS: METHODOLOGY AND APPLICATION IN EDUCATIONAL PLANNING By Jane Counihan Nelson December 1977 Chairman: Michael Y. Nunnery Major Department: Educational Administration Educational planners and policy makers need adequate information about the societal context of education to make appropriate decisions about the future role and function of education. Some of this information may be provided through the use of conceptually sound social and educational variÂ¬ ables operationally defined as time series indicators coupled with an empirically sound basis for forecasting future trends in such indicators. As evidence of the need for developing such a social forecasting framework for education, states including Florida have provided grants for that purpose. This study was one aspect of such a grant. The problem in this study was (a) to select, using Bronfenbrenner's ecology of education model, and operationÂ¬ ally define at least 10 variables that research has shown to be related to the outcomes of education; (b) to use these variables operationally defined as time series indicators in the comparison of three purely extrapolative forecasting x methods; and (c) to derive implications for the use of an ecological model such as Bronfenbrenner's, time series indicators, and selected extrapolative methods for educaÂ¬ tional planning. The study was conducted in the following phases: 1. Using Bronfenbrenner's ecology of education model, 10 variables that research has shown to be related to the outcomes of education were selected and where possible, were operationally defined as state and/or national time series indicators. Data were collected for these indicaÂ¬ tors; eight which met the criteria established in this study were used in the comparison of extrapolative techniques. 2. Three purely extrapolative techniques derived from the general linear model were compared according to statisÂ¬ tical criteria and practical considerations derived from the literature in statistics, economics, time series analysis, and forecasting methodology. The methods were (a) linear regression, (b) curvilinear regression (quadÂ¬ ratic and cubic forms), and (c) log-linear regression (deÂ¬ pendent variable undergoes logarithmic transformation). Each method was applied to each time series indicator. Time in years was used as the independent variable; the annual measure of the indicator was treated as the dependent variÂ¬ able. Each data set was divided into thirds; two-thirds of the data points were used to establish the prediction equation. This equation was used to predict the remaining third of the data points. Predicted values were compared with actual values. 3. Implications for the use in educational planning of an ecological model such as Bronfenbrenner's, time series indicators, and selected extrapolative techniques were disÂ¬ cussed . Results of the method comparison were (a) no method was a superior predictor for all indicators; (b) each method was a superior predictor for at least one indicator; and (c) the summary statistics for the original regression were not consistently related to the accuracy of the extraÂ¬ polated values. The following conclusions appear to be warranted by the results of this study: 1. The Bronfenbrenner model is a useful framework for considering the numerous factors impinging upon the learner. 2. Time series indicators provide a means to compare trends in an indicator over time or to compare different groups in relation to a specific indicator. 3. The general linear model is appropriate for the analysis and extrapolation of the selected time series indicators used in this study. 4. Each method is appropriate for use with some indicators but not with others. Measures of "best fit" such as r2 and the standard error of estimate are not reliable criteria for the selection of an extrapolative method. A combination of strategies such as graphic representation of original and predicted data, analysis of residuals, and XI 1 knowledge of the social phenomena being studied may provide guidance as to the most appropriate method for a particular indicator. Xll i CHAPTER I INTRODUCTION Background and Significance of the Study The Social Context of Education Educators have become increasingly cognizant of the myriad forces in society impinging upon various facets of the educational process. The influence of a number of these forces upon educational purposes, outcomes, and resources has been analyzed from several social science perspectives (Boocock, 1976; Henry, 1961; Gordon, 1974). Keppel (in Thomas 5 Larson, 1976) acknowledged one of the reasons for this continuing interest in societal trends by educational planners and policy-makers: the impetus for change in educational institutions, from the preschool through the university, is more likely to derive from changes in the wider society than from forces within the institutions. (Foreword) Additionally, Keppel noted that "educational policy must be formed in concert with other aspects of public policy and program development" (Foreword). Bronfenbrenner (1976) proposed an ecological structure of the educational environment which must be taken into account if "any progress in the scientific study of educaÂ¬ tional systems and processes" (p. 5) is to be made. BronfenÂ¬ brenner stated: 1 2 Whether and how people learn is a function of sets of forces, or systems, at two levels: a. The first comprises the relations between characteristics of learners and their surÂ¬ roundings in which they live out their lives (e.g., home, school, peer group, work place, neighborhood, community). b. The second encompasses the relations and interconnections that exist between these environments. (p. 5) Building and employing on Lewin's theory of topological territories a terminology adapted from Brim (1975), Bronfen- brenner further elaborated that the construct environment can be "conceived topologically as a nested arrangement of structures, each contained within the next" (p. 5). 1) A micro-system is an immediate setting containing the learner .... 2) The meso - system comprises the inter- re laFTonsTiips-among the major settings containing the learner at a particular point in his or her life ... a system of micro-systems. 3) The exo- system is an extension of the meso-system embracing the concrete social structures, both formal and informal, that impinge upon or encompass the immediate settings containing the learner and, thereÂ¬ by, influence and even determine or delimit what goes on there. These structures inÂ¬ clude the major institutions of society, both deliberately structured and spontaneÂ¬ ously evolving, as they operate at the local community level .... 4) Macro - systems are the overarching instiÂ¬ tutions of the culture or subculture, such as the economic, social, educational, legal and political systems, of which local micro-, meso-, and exo-systems are the concrete manifestations. (pp. 5-6) (See Figure 1 for a representation of these ideas.) .3 Figure 1. Bronfenbrenner's ecological structure of the educational environment. (Based upon Bronfen- brenner's [1976] description, pp. 5-6.) 4 The Futures* Perspective in Educational Planning While Bronfenbrenner proposed his ecological structure primarily as a framework for learning research efforts, that is, for examining relationships among variables associated with learning, others (Harman, 1976; Webster, 1976) have disÂ¬ cussed the societal context of education as a framework for future-oriented educational planning. Indeed, this emphasis on future awareness has evolved into a significant movement within education referred to as educational futurism (Hencley 5 Yates, 1974; Pulliam 6 Bowman, 1974), (Marien 5 Ziegler, 1972), or alternative (Webster, 1976). The primary purpose of future studies is "to help policy makers educational futures futures perspective futures research or choose wisely--in terms of their purposes and values--among alternative courses of action that are open to leadership at a given time" (Shane, 1973, p. 1). According to Webster (1976), this requires that we attend to alternatives--to alternative assumptions, ends and means. It requires us to examine alternative plausible futures that might be rendered more or less possible by our planning and action; to identify unÂ¬ intended as well as intended consequences for others of achieving the goals that seem desirable to us; to analyze alternative straÂ¬ tegies and tactics for achieving any desired future; and to anticipate the variety of potential consequences of our strategies, tactics, and short-run planning. Perhaps, most fundamentally it asks of us that we look hard at our basic premises about the nature of man and the world and consider implications and alternatives for the future. (p. 2) *"Futures" refers to the number of different possible views of what is ahead in subsequent time periods for society and, thus, for education. 5 Webster also noted: the futures perspective implies that we not just attend to alternatives in and for educaÂ¬ tion, but also consider the societal context in more comprehensive fashion than is usual in educational planning. (p. 2) In order to assist decision makers in the selection of alternatives which have positive future consequences for society, educational planners at both national and state levels must take into account those societal forces which affect not only the outcomes of education, but also the purÂ¬ poses of education, and the human and material resources available to the educational process. To do this, however, the planner must delineate the societal factors or variables to be included in the planning process and develop a sound rationale based on research and theory for such inclusion. Then trends--past, present, and future--in these variables may be examined in order to derive implications for educaÂ¬ tional planning and policy. Forecasting Trends in Social Variables Available to the educational planner in this undertaking are a number of predictive and heuristic devices to explore alternative futures which have been developed by government, industry, non-profit organizations, and futures consulting groups. These forecasting techniques can be categorized into exploratory forecasting methods and normative forecasting methods: Exploratory forecasting methods start from the present situation and its preceding history, and attempt to project future 6 developments. Normative forecasts, on the contrary, start with some desired or posÂ¬ tulated future situation, and work backÂ¬ wards to derive feasible routes for the transition from the present to the desired future. (Martino, 1976, p. 4) Exploratory forecasting methods, all of which are based upon extrapolation of some kind, include (a) purely extraÂ¬ polative methods, (b) explanatory methods, and (c) auxiliary methods. Since forecasting of social phenomena is still in a highly intuitive developmental phase, there is a growing interest in examining those exploratory methods considered to be purely extrapolative, which are based upon time series data representing social and educational variables. These time series data, often called time series indicators, are defined measurements made at specified intervals over a period of time. By extrapolating identified patterns in the time series data into the future, planners may compare present, past, and future states of that indicator. Thus, a projecÂ¬ tion of future societal trends can provide the impetus to examine present policy and to analyze the consequences of contemplated changes. This approach need not be only "preÂ¬ ventive" forecasting, in the sense used by Ziegler (1972) of preventing undesirable forecasts. It may also be extended to examine all consequences of action or intervention, inÂ¬ tended or not. Purely extrapolative methods, when combined with auxiliary methods such as trend-impact analysis, crossÂ¬ impact matrices, or scenarios, can provide a vehicle for exÂ¬ ploring the relationships among identified future patterns in society. 7 While the use of purely extrapolative methods with time series data is fairly well defined in technological and economic areas, their application to social forecasting has not been the focus of significant definitive study. Indeed Harrison (1976) emphasized the need for such research, speciÂ¬ fically the consideration of "each method in terms of some aspect of the social process it would likely be applied to" (p. 13). For, as Harrison explained, while some problems in regression and time series analysis which remain unresolved are currently the concern of statisticians and mathematicians, "it appears that resolution might best lie in terms of invesÂ¬ tigation in concrete application cases" (p. 14). In social forecasting there is a great need in almost all the known extrapolative methods for an explicit statement of the algorithmic, theoretical, and empirical weaknesses or sensitivities of such procedures. Such a discussion, as noted, would be more meanÂ¬ ingful if carried on in the context of an analysis of some specific aspect or aspects of social process. (Harrison, 1976, p. 17) Only through empirical study of the performance of various extrapolative methods applied to particular social phenomena will a basis for selection of appropriate and accurate techÂ¬ niques be formulated. The Need for Research Since there are no widely-accepted planning models inÂ¬ corporating quantitative data on social variables, the eduÂ¬ cational planner who wants to utilize such information is confronted with a number of questions related to (a) the 8 identification of social variables to be included, (b) the operational definition of social variables in terms of time series indicators, (c) the selection of a purely extrapolative technique which will yield the most accurate forecast for a specific indicator, and (d) the utilization of these forecasts in the planning process. Answers require futures research which is derived from a conceptually sound framework and is pursued with methodological vigor. As evidence of the imporÂ¬ tance of such investigation to the educational planner, the State of Florida through the Office of Strategy Planning in the Department of Education funded in 1976 a social foreÂ¬ casting project (STAR Project No. R5-175) at the University of Florida for the second year. The study described herein was part of that effort to forecast social trends affecting education in Florida. To summarize: Educational planners and policy makers need adequate information to make appropriate decisions about the role and function of education in creating improved quality of life for citizens of the future. The State of Florida, in funding STAR Project No. R5-175 of which this study is a part, acknowledged that need. Part of this inÂ¬ formation may be provided through the use of conceptually sound social and educational variables operationally defined as time series indicators coupled with an empirically sound basis for forecasting future states of such indicators. 9 The Problem The problem in this study was (a) to select, using Bronfenbrenner's ecology of education model, and operationally define at least 10 variables that research has shown to be related to the outcomes of education; (b) to use these variables operationally defined as time series indicators in the comparison of three purely extrapolative forecasting methods; and (c) to derive implications for the use of an ecoÂ¬ logical model such as Bronfenbrenner's, time series indicators, and selected extrapolative methods for educational planning. Delimitations and Limitations The Bronfenbrenner ecology of education model was used primarily as a framework for the selection of social and educational variables and was not evaluated itself in this study. Ten variables (e.g., socio-economic status of family, peer group characteristics) were selected to be operationally defined, where possible, in terms of national and/or state level time series indicators (e.g., median family income, juvenile crime rates). Of these identified indicators, eight which met the following criteria were used in the comparison of extrapolative techniques: (a) the indicator was readily available, (b) the data were available for a 10 year or greater time span, and (c) the indicator was a reasonably reliable and valid measure of one aspect of the social or educational variable that it represented. It should be noted that the selection of the eight indicators used in this study 10 was in many cases influenced more by data availability than the logic or appropriateness of the indicator to represent a specific social variable. Thus, the eight indicators are examples of the type of data that might be employed to operationally define the variables; utilization in a specific planning situation would require evaluation of the appropriÂ¬ ateness of the indicators presented in this study and the addition and/or substitution of other indicators. In this study only the variables related to the outcomes of education were used. As previously noted, this study was part of a larger social forecasting and educational planning effort which also included the status of education 1976-77, social trends affecting the purposes of education, and social trends affecting the resources for education. While the literature in mathematics, statistics, and economics was reviewed and considered in preparation for the selection and use of the three extrapolative techniques (linear, log-linear, and curvilinear regression), there was no attempt to present the comparison of these techniques in the detail desired by these disciplines. Rather the compariÂ¬ son was made in such a way as to be most relevant to the planner in education. There was no attempt to write or adapt computer programs for various techniques. Instead, an effort was made to identify and utilize computer programs and statistical packÂ¬ ages which had already been adapted for use at the North East Regional Data Center's computer facilities. 11 Additionally, the projection of specific trends per se was not of interest in this study. Rather the focus of this study was the development of the conceptual framework and methodology for such projection. Also, there has not been any attempt to forecast educational outcomes from the operationally defined social and educational variables. The present work may be considered an initial step in determining the feasibility of developing such a mathematical forecasting model. Definition of Terms Extrapolative forecasting. The procedure consists of identifying an underlying historical trend or cycle in social processes that can be extrapoÂ¬ lated by means as varied as multiple regression analysis, time series analyÂ¬ sis, envelope curve fitting, three-mode factor analysis, correlational analysis, averages, or any other method that takes current and historical data as the prinÂ¬ cipal basis for estimating future states in a given variable. (Harrison, 1976, p. 3) Indicator, educational. Educational indicators are statistics that enable interested publics to know the status of education at a particular moment in time with respect to some selected variables, to make comparisons in that status over time and to project future status. Indicators are time-series statistics that permit a study of trends and change in education. (Gooler, 1976, p. ID Indicator, social. "The operational definition or part of the operational definition of any one of the concepts 12 central to the generation of an information system descripÂ¬ tive of the social system" (Carlisle, 1972, p. 25); "time- series that allow comparisons over an extended period which permit one to grasp long-term trends as well as unusually sharp fluctuations" (Sheldon Â§ Freeman, 1970, p. 97); "a statistic of direct normative interest which facilitates concise, comprehensive and balanced judgments about the conÂ¬ dition of major aspects of a society" (U.S. Department of Health, Education, f7 Welfare, 1970, p. 97). Outcomes of education. Those measures of performance, such as achievement test scores, or utilization, such as employment rates, which appear to be the result of particiÂ¬ pation in the formal educational process. Regression, linear. Most common type of regression in which the objective is to locate the best-fitting straight line through a scattergram based on interval-level variables (Nie, Hull, Jenkins, Steinbrenner, 5 Bent, 1975, p. 278). Regression, log-linear. As used in this study, a least squares regression method in which a geometric straight line is located through a scattergram plotted on semi-logarithmic paper; also called exponential curve or trend curve. Regression, polynomial or curvilinear. Regression method for fitting a curve to a set of data using the criÂ¬ terion of least squares distances (Nie et al., 1975, p. 278). Time series. "A set of observations generated sequenÂ¬ tially in time" (Box f, Jenkins, 1970, p. 23). 13 Procedures The study proceeded in the following phases: (a) using Bronfenbrenner's ecology of education model, 10 variables that research has shown to be related to the outcomes of education were selected and, where possible, were operationalÂ¬ ly defined as time series indicators; (b) data were collected for these time series indicators, eight of which were used in the comparison of the selected extrapolative techniques; (c) using the selected time series indicators, three purely extrapolative techniques were compared according to statisÂ¬ tical criteria and practical considerations derived from the literature; and (d) implications for the use in educaÂ¬ tional planning of an ecological model such as Bronfen- brenner's, time series social and educational indicators, and selected extrapolative techniques were derived. The Selection and Operational Definition of Variables The work by Collazo, Lewis, and Thomas (1977), completed during the first year of STAR Project No. R5-175, on foreÂ¬ casting selected educational outcomes from social variables was utilized. Since the variables selected by these invesÂ¬ tigators were derived from a review of the research literaÂ¬ ture and were acknowledged to be appropriate for the stated social forecasting purposes by a panel of experts in various disciplines, they appeared to fulfill the requirements of this study. Additionally, each of the 10 variables selected 14 for use was described and classified according to Bronfen- brenner's ecology of education model. For each variable an attempt was made to identify one or more types of time series indicators which might logically represent the variable. For some variables several indicaÂ¬ tors were identified, while for others, no indicator could logically be identified or no time series data were availÂ¬ able for the indicator at the time of the study. This phase of the study is explained further in Chapter II. Collection of Time Series Indicator Data Sources of needed time series data at both the national and state level were identified in several ways. The expanÂ¬ ding literature on social trends (e.g.,U.S. Department of Health, Education, Â§ Welfare, 1970) and specifically the literature on these social trends operationalized as social indicators (e.g., Executive Office of the President, Office of Management 5 Budget, 1973) was reviewed. Furthermore, examination of initial efforts in using time series indicaÂ¬ tors related to education by the Office of Technology AssessÂ¬ ment for the United States Congress (Coates, Note 1) and several state departments of education (e.g., Oregon, PennÂ¬ sylvania, 5 Florida) yielded additional sources. Published sources of data such as U.S. Census Reports and Florida Statistical Abstracts were consulted. When data did not appear to be available in suitable form or for desired time periods, inquiries and requests were directed to appropriate sources. Any apparent limitations in the data such as known 15 measurement error due to sampling technique were noted. After data collection was completed, eight indicators which met the criteria outlined in a previous section were selected for inclusion in the next phase of the study. Comparison of Extrapolative Methods Using Time Series Indicators The following steps were involved in this phase of the study: (a) initial identification and testing of methods using data similar in form to selected indicators, (b) reconÂ¬ sideration and testing of additional available methods, (c) selection of three methods to be used for comparative extraÂ¬ polations, (d) derivation of specific criteria and practical considerations from the literature, (e) application of three methods to each data set, (f) extrapolation of identified trend into future using equation generated in (e), and (g) comparison of actual versus predicted values of indicators. From a preliminary review of the literature in statisÂ¬ tics, economics, time series analysis, and forecasting methodology, the following four methods were tentatively identified for comparison: (a) linear regression (computer program by Nie et al., 1975), (b) curvilinear or polynomial regression (computer program by Nie et al., 1975), (c) Box- Jenkins time series analysis (computer program by Cooper, Note 2), and (d) FIT curve-fitting with weighted data (comÂ¬ puter program by Stover, Note 3). An initial analysis of the methods using trial sets of data combined with a visual analysis of the general form of 16 the data to be used revealed that two of the methods under consideration were inappropriate. The Box-Jenkins procedure, while an extremely powerful tool for time series analysis of data which are characterized by seasonal or cyclic variation (usually resulting in autocorrelation of observations and residuals), did not seem suitable for the social indicator data collected. (Should subsequent tests reveal autocorrelaÂ¬ tion and hence a violation of the assumptions of the linear model, Box-Jenkins could then be appropriately employed.) The FIT curve-fitting procedure utilizing a weighted data principle was rejected because the computer program required extensive modification to yield necessary comparative statisÂ¬ tics and reliable output. Theoretical justification for the weighting formula and data transformations employed was unavailable. Thus, two of the four methods tentatively considered were rejected. Since the comparison phase was to involve three methods, the literature was again searched for other appropriate methods. The most promising of these was a curve fitting technique which utilizes an exponential function to describe a constant growth rate. This method, called log- linear regression in this study, can be described in terms of the general linear model and solved by least squares proÂ¬ cedures when the dependent variable undergoes logarithmic transformation. Since social phenomena sometimes exhibit what appears to be a constant growth rate, log-linear regresÂ¬ sion seemed to be an appropriate method to include in this study. 17 The three methods finally selected for comparison were (a) linear regression (without data transformation), (b) curvilinear or polynomial regression, and (c) log-linear regression. The mathematical properties of each are preÂ¬ sented in Chapter III. All three approaches to trend extraÂ¬ polation were executed by using variations of SPSS subproÂ¬ grams SCATTERGRAM and REGRESSION and that system's data transformation capabilities (Nie et al., 1975). Each of the three methods was applied to each of the eight selected time series indicators. Time in years was used as the independent variable; the annual measure or index of the indicator was treated as the dependent or response variable. Each data set was divided into thirds; two-thirds of the data points were used to establish the prediction equation. This equation was then used to predict the remaining third of the data points. Predicted values were then compared with actual values. Thus, in this phase of the study three prediction equations (one for each method) were generated for each of the eight time series indicators. Statistical criteria deÂ¬ rived from the literature were used to evaluate the "goodÂ¬ ness of fit" of the regression line derived from the preÂ¬ diction equation to the data. The distribution of error (residuals) about the regression line was also examined to determine if the data satisfied the assumptions of the staÂ¬ tistical model. Results of the method comparison phase are reported in Chapter IV. 18 Development of Implications for Educational Planning In Chapter V methodological strategies involved in the selection and operational definition of variables are analyzed in terms of viability for future use. Results of the technique comparison phase are analyzed according to the statistical criteria and practical considerations derived from the literature in forecasting methodology and statisÂ¬ tics. In Chapter VI a summary of the study and conclusions warranted by the results of the study are presented. Future directions for research suggested by the results of this study are discussed. Additionally, implications for the use in educational planning of an ecological model such as Bronfenbrenner's, time series social and educational inÂ¬ dicators, and selected extrapolative methods are discussed. CHAPTER II RATIONALE FOR SELECTION OF VARIABLES/ TIME SERIES INDICATORS In the previous chapter, the need for educational planners and policy makers to have an awareness of the societal context of education was emphasized. To this end the Bronfenbrenner ecology of education model was proposed as a framework for the selection of social variables which affect the outcomes of the educational process. The selected social variables may then be operationalized as time series indicators; trends in these indicators can be identified and extrapolated into the future. Such information might then be incorporated into a planning model in order to assist planners and policy makers in making informed decisions about the role and function of education in the future. In order to place the use of time series indicators described in this study into perspective, in the first section of the present chapter social indicators are disÂ¬ cussed in relation to their historical development, definiÂ¬ tion and use, and data base. Educational applications of indicators are briefly noted. In the second section the social variables selected for use in this study are presented in relation to the Bronfenbrenner model. These variables are then operationally defined as time series indicators, 19 20 and the eight indicators selected for use in the comparison of the three extrapolative methods are listed. The Social Indicator Movement Historical Development Interest in societal trends by policy planners is not of recent origin in the United States. Indeed, in 1933 a presidential task force reported on social trends in a comÂ¬ prehensive work documenting social change in the United States (Presidentâ€™s Research Committee on Social Trends, 1933). The development of indicators, or measures, of social change, however, did not receive the sustained governmental support that was provided for indicators of the economic process. Thus, while the development of economic statistics during the 1930's and 1940's provided "a solid basis for economic analysis and economic reporting which eventually resulted in the establishment of the Council of Economic Advisors and the Economic Report" (U.S. Department of Health, Education, 5 Welfare, 1970, p. v), comparable development of social inÂ¬ dicators was not undertaken. In the 1960's a renewed interest in statistics describing the social condition became apparent. Impetus for the deÂ¬ velopment of social indicators was provided by social scientists in various disciplines, government policy makers, and business leaders in the private sector (Brooks, 1972, p. 1). While this early effort was not well defined as to membership, organization, or objectives, the participants in 21 the social indicator movement "sensed great needs and opporÂ¬ tunities for change, [and] celebrated shared but necessarily ambiguous symbols" (Sheldon 5 Parke, 1975, p. 693). The following examples were drawn from the many maniÂ¬ festations of interest in the development of social indicaÂ¬ tors during the period from 1965 through 1975 (over 1000 items were listed in a bibliography issued in late 1972 by Wilcox, Brooks, Beal, Â§ Klongian): 1. The Russell Sage Foundation commissioned in 1965 and published in 1968 an independent study, Indicators of Social Change: Concepts and Measurements, on a number of aspects of structural change in society (Sheldon 5 Moore, 1968). 2. A study on ways to measure the impact of massive scientific and technological change on society (Bauer, 1966) was prepared by the American Academy of Arts and Sciences for the National Aeronautics and Space Administration. This work, Social Indicators, was an overview of the task of developing indicators as part of a feedback mechanism docuÂ¬ menting social change. 3. President Johnson in March of 1966 directed the Secretary of Health, Education and Welfare "to develop the necessary social statistics and indicators to supplement those prepared by the Bureau of Labor Statistics and the Council of Economic Advisors" (U.S. Department of Health, Education, Â§ Welfare, 1970, p. iii) . The result of this directive, Toward a Social Report issued in 1969, was 22 considered "a preliminary step toward the evolution of a regular system of social reporting" (U.S. Department of Health, Education, Â§ Welfare, 1970, p. iii). 4. The Social Science Research Council established in 1972 the Center for Coordination of Research on Social Indicators, whose objective is "to enhance the contribution of social science research to the development of a broad range of indicators of social change" (World Future Society, 1977, p. 97). 5. The appearance of the landmark government publicaÂ¬ tion, Social Indicators 1973 (Executive Office of the PresiÂ¬ dent: Office of Management $ Budget, 1973),was heralded as a significant attempt to provide a collection of social staÂ¬ tistics describing quality of life in the United States. This work, which was scheduled to be up-dated every three years, was a compilation of statistics on eight major areas of social interest: health, public safety, education, emÂ¬ ployment, income, housing, leisure and recreation, and popuÂ¬ lation. (It may be noted that it is often impossible to strictly categorize what is social and what is economic, as almost all aspects of life are the result of interaction beÂ¬ tween social and economic forces.) The idea of systematic collection and use of social indicators has not always met a favorable reception. Bezold (Note 4) and Shostak (Note 5) documented the unsuccessful efforts, beginning in 1967, by Walter F. Mondale to earn congressional approval of a far-reaching plan for new 23 government use of the applied social sciences. Mondale's blueprint for better collection and use of social intelliÂ¬ gence involved two statutorily-mandated additions to the Executive Office of the President: (a) a Council of Social Advisors (CSA) comparable to the Council of Economic Advisors established in 1946, and (b) an annual Social Report of the President prepared by the CSA to parallel the annual Economic Report to the President. While aspects of Mondale's plan may be satisfied by such efforts as Social Indicators 1973 and several congressional provisions which required the developÂ¬ ment and application of social science techniques to the study of present and future national problems (Shostak, Note 5), the comprehensive nature of the Mondale plan is absent. The future of such governmental efforts at social accounting was in 1977 uncertain. Interest in social indicators has not been confined to the United States (Johnson, 1975; Sheldon Â§ Parke, 1975). In 1973, for example, the Organization for Economic CooperaÂ¬ tion and Development (OECD) to which the United States also belongs issued a list of social concerns shared by many member countries. The identification of concerns was a first step in the development of a set of social indicators designed explicitly to reveal, with validity, the level of well-being for each social concern in the list and to monitor changes in those levels over time. (OECD, 1973, p. 4) Additionally, international organizations such as the Conference of European Statisticians, the United Nations 24 Research Institute for Social Development, and the United Nations Educational, Scientific,and Cultural Organization have been actively concerned with social indicators (Sheldon 5 Parke, 1975). Efforts to develop social indicators have been initiated in countries such as France, Great Britain, West Germany, Canada, Japan, Norway, Sweden, and Denmark (Brooks, 1972; Johnson, 1975). Definition and Use of Social Indicators The social indicator movement has been characterized by ambiguity of definition and purpose due, in part, to the heterogeneous nature of participants with their own backÂ¬ grounds, skills, and interests, and also, to the necessary stages of evolution that such a movement experiences. These problems in definition and purpose of social indicators have been discussed by a number of critics (Land, 1971; Little, 1975; Plessas Â§ Fein, 1972; Sheldon 5 Freeman, 1970; Sheldon 5 Land, 1972). Attempts have been made to resolve a number of these problems. Land (1971), for example, proposed the following social science-oriented definition of social indicators: social indicators refer to social sta- tistics that (T) are components in a social system model (including sociopsy- chological, economic, demographic, and ecological) or of some particular segment or process thereof, (2) can be collected and analyzed at various times and accumuÂ¬ lated into a time series, and (3) can be aggregated or disaggregated to levels appropriate to the specifications of the model. . . . The important point is that the criterion for classifying a social 25 statistic as a social indicator is its informative value which derives from its empirically verified nexus in a conÂ¬ ceptualization of a social process. (p. 323) Part of the confusion over definition is the result of disagreement over purposes, or uses, of social indicators These purposes, or uses, have been considered under a number of overlapping, sometimes synonomous, headings: (a) descrip tions reporting, (b) policy planning, (c) social accounting, (d) program evaluation, (e) social modeling, (f) social fore casting, and (g) social engineering. While the ultimate obÂ¬ jective of guiding social policy is rarely disputed, the form of this guidance is still debated. Social scientists are more likely to be concerned with the analysis and preÂ¬ diction of social change, while public administrators and legislators are often more concerned with uses of indicators related to public program evaluation and agency goal setting Sheldon and Parke (1975) in acknowledging these concerns, said: It is apparent that many different types of work go on under the rubric of social indicators. What is imporÂ¬ tant is that the field be seen as an arena for long-term development, as an effort of social scientists to push foreward developments in concepts and in methodology that promise payoffs to both science and public policy. (p. 698) To underscore this point, Sheldon and Parke (1975) selected an observation by Duncan: The value of improved measures of social change. . .is not that they necessarily resolve theoretical issues 26 concerning social dynamics or settle pragmatic issues of social policy, but that they may permit those issues to be argued more productively. (p. 698) Data Base for Social Indicators Various efforts have been undertaken to improve the data base for social indicators. Among the efforts in the early 1970's were basic surveys on crime and education as well as replications of previous social science studies and surveys (Sheldon Â§ Parke, 1975). Most social statistics, available orimarily from governÂ¬ ment sources, are objective in nature; that is, they measure the frequency of occurrence of an attribute or commodity in the population. Numbers of births, deaths, marriages, years of schooling, and percent of occupied housing with television sets could thus be considered objective measures. (Some would disagree, however, with the objectivity of these measures, see Andrews Â§ IVithey, 1976 , p. 5.) Several researchers (Andrews $ Withey, 1976; Campbell, Converse, Â§ Rogers, 1976) have attempted to measure people's perceptions of their well-being, their quality of life. Such measures collected on a regular basis are expected to be valuable supplements to the usual objective quality of life indicators. (See for examples of the latter: Liu, 1976; Thompson, 1976b,1977.) Creation of a social indicator data base is not without conceptual and methodological problems. Various aspects of 27 the social measurement problem have been acknowledged in the literature (see, for example, de Neufville, 1975, pp. 175-179; Etzioni 5 Lehman, 1969; Social Measurement, 1972). While deÂ¬ tailed discussion of measurement dysfunction (in the termiÂ¬ nology of Etzioni Â§ Lehman, 1969) is beyond the scope of this study, the following observation might be kept in mind: Increased investment, intellectual as well as financial, no doubt can go a long way to increase the efficacy of social measurements and to reduce much of the likelihood of dysfunctions. But, in the final analysis, these problems can never be eliminated enÂ¬ tirely. Here, the client of systematic measurement and accounting should be alerted to the limitations of social indicators, both to make his use of them more sophistiÂ¬ cated and to prevent him from ultimately rejecting the idea of social accounting when he encounters its limitations. (Etzioni Â§ Lehman, 1969, p. 62) Educational Implications Educational indicators, a subset of social indicators, have traditionally been measures of the educational system's inputs and outputs stated in such terms as numbers of teaÂ¬ chers, per pupil expenditures, and achievement test scores. There have been attempts, however, to broaden this base of educational statistics to include both objective and subÂ¬ jective indicators under the categories of access, aspirations, achievement, impact, and resources (Cooler, 1976, p. 15). There have also been attempts to link indicators of social processes (e.g., divorce rates, voting rates) to educational goals and thus to establish accountability measures, albeit remote, external to the educational system (Clemmer, Fairbanks, 28 Hall, Impara, 5 Nelson, 1974; Collazo, Lewis, Â§ Thomas, Note 6; Grady, 1974). The use and abuse of indicators in an eduÂ¬ cational setting, however, remained in 1977 a matter of debate (Impara, Note 7) and cautious optimism (Hall, Note 8). HopeÂ¬ fully, investigations of the problem, such as that described in this study, will provide some guidance as to the most promising applications of social indicators to education. Selection of Variables/Time Series Indicators The Variables In the first year (Sept. 1975-June 1976) of Florida Department of Education STAR Project R5-175 on social foreÂ¬ casting for educational planning, trends in five indicators of educational outcomes were forecast. In order to do this, it was necessary to identify variables that influence the outcomes of education. Through a review of the research and theoretical literature, a number of social variables were identified. This list was refined by an interdisciplinary panel of experts at the University of Florida to the following 10 variables: (a) socio-economic status; (b) family expectations, attitudes, and aspirations; (c) student's self- concept; (d) student's general ability; (e) student's sense of fate control; (f) student's attitudes and motivation; (g) peer group characteristics; (h) teacher expectations; (i) teacher behavior in the classroom; and (j) administrative leadership style. Collazo et al. (1977) said that only the variables (a) and (d) received strong support from research; 29 a number of the other variables, while "identified as imporÂ¬ tant in the theoretical literature. . .liad inconclusive support from research" (p. 298). (See Collazo, Lewis, 8 Thomas, Note 9, for a review of the research literature on variables affecting educational outcomes.) The panel of experts was further utilized to forecast the future trends of these variables and their effect on specified performance and utilization measures of the outÂ¬ comes of education. Cross-impact analysis, a computer assistÂ¬ ed modification of the Delphi forecasting technique, was then used by the panel to generate the future trends in the five outcome indicators. The framework for looking at the future established during these first year project activities is utilized in the present study. Previous forecasting activities were based primarily on the subjective judgment of panel participants. In this study, however, the feasibility of using time series data, where available, as the basis for forecasting future trends in the 10 variables affecting educational outcomes is examined. In addition, the use of a model containing the selected variables is considered. Bronfenbrenner1s Ecology of Education Model In the previous section, the 10 variables affecting educational outcomes which were derived from the research literature were presented. How can these variables be put into perspective as social forces influencing what the stuÂ¬ dent learns? 30 The Bronfenbrenner (1976) model which was presented in Chapter I (pp. 1-3) is a multi-dimensional ecological strucÂ¬ ture of the educational environment. At the center of the interacting meso-, exo- and macro-systems is the micro-system, "the immediate setting containing the learner" (Bronfenbrenner, 1976, p. 5). The meso-system is actually a system of microÂ¬ systems; that is, it "comprises the inter-relationships among the major settings containing the learner at a particular point in his or her life" (Bronfenbrenner, 1976, p. 5). Some of the social variables that were identified previously could be considered as part of the meso-system. The home, for exÂ¬ ample, is represented by socioeconomic status and family expectations, attitudes, and aspirations; the peer group by peer group characteristics; and the school by teacher expecÂ¬ tations, teacher behavior in the classroom, and administraÂ¬ tive leadership style. The other variables: student's self- concept, student's general ability, student's sense of fate control, and student's attitudes and motivation are all diÂ¬ rectly related to the learner. Bronfenbrenner (1976) proposed that learning is a funcÂ¬ tion of (a) the dynamic relationship between characteristics of the learners and their various surroundings (meso-system) and (b) the interaction between these various environments (e.g., home, school, peer group). The Bronfenbrenner ecology of education model thus appears to provide the necessary framework to support use of the presently identified variables and to generate directions for future forecasting research. 31 Operational Definition of Variables as Time Series Indicators In previous sections 10 variables affecting educational outcomes were presented and then classified according to the Bronfenbrenner ecology of education model. In order to idenÂ¬ tify trends in these variables and to extrapolate these trends into the future, it was necessary to operationally define these variables as time series measures, or indicators. Since some of these variables were expressed in general terms, it seemed necessary to try to represent each by a number of measures and thus avoid "fractional measurement" which is often a concern when operationally defining a social concept (Etzioni $ Lehman, 1969). Several problems became apparent in operationalizing the variables: 1. A number of indicators were identified for the variables (a) socioeconomic status; (b) family expectations, attitudes, and aspirations; and (c) peer group characteristics. For some indicators, however, data were not collected annually; for others, measures were not comparable over time due to a different basis for measurement. 2. For the variables related to the school and student characteristics (except student attitudes and motivation), no time series data which met the criteria for selection were available. 3.Operational definitions were in many cases influenced by the availability of indicators rather than the logic or appropriateness of the indicator to measure the social concept it represented. 32 The social variables, examples of indicators that might be used to operationally define these variables, and sources of the available time series data are presented in Table 1. The following eight indicators which met the criteria estabÂ¬ lished for this study (see p. 9) were selected for use with the three extrapolative methods described in Chapter III: 1. Median family income in the United States exÂ¬ pressed in 1971 constant dollars. 2. Number of families in the United States headed by women expressed as a percentage of total families. 3. Number of wives in the labor force expressed as a percentage of total wives in the United States. 4. Number of marriages in Florida expressed as rate per 1,000 population in Florida. 5. Number of dissolutions of marriage in Florida exÂ¬ pressed as rate per 1,000 population in Florida. 6. Number of resident live births in Florida exÂ¬ pressed as rate per 1,000 population in Florida. 7. Number of 3 to 5 year olds enrolled in nursery school and kindergarten expressed as percentage of total children 3 to 5 years old in the United States. 8. Number of children involved in divorce or annulment expressed as rate per 1,000 children under 18 years old in the United States. While rates or percentages are used for forecasting purposes, the magnitude of the actual numbers should be kept in mind before interpretation of an identified trend is Table 1 Time Series Indicators of Social Variables Affecting Outcomes of Education Available time series data K Social variables Indicator Years State U.S. Source Socioeconomic status of family Median family 1947- X U.S. Dept. income 1971 Census in Employment 1947- X U.S. Dept. rate-total 1976 Statistics labor force Dec. 1976 1947- X U.S. Dept. 1975 Statistics Unemployment Dec. 1976 rate 1950- Fla. Fla. Dept. 1975 Employment Statistics Husband-wife 1950, X U.S. Dept. families with 1955- Statistics two workers 1957 or more Female headed 1940, X U.S. Dept. families 47,50 55,60 65,70- 75 Statistics of Commerce, Bureau of the Social Indicators 1975 of Labor, Bureau of Labor in Employment and Earnings, of Labor, Bureau of Labor in Employment and Earnings, of Commerce, Division of Security Research and of Labor, Bureau of Labor in Special Report 189 of Labor, Bureau of Labor in Special Report 190 04 04 Table 1 - continued Available time series data Social variables Indicator3 Years*7 State U.S. Source Labor force participation rates of wives 1950 , 55-75 X U.S. Dept, of Labor, Bureau of Labor Statistics in Special Report 189 Children living in poverty 1959 , 62,65 68,71 75,76 X U.S. Dept, of Commerce, Bureau of the Census and 1976 Survey of Income and Education Family expectaÂ¬ tions , attitudes, aspirations Marriage rate 1930 , 40,50 60,63- 75 Fla. State of Florida, Dept, of Health and Rehabilitative Services, Division of Health in Florida Statistical Abstract 1976 (Thompson, 1976a) Divorce rate 1930, 40,50 60,63- 75 Fla. I ! Birth rate 1964- 1975 Fla. X t T U.S. Dept, of Commerce, Bureau of the Census in Florida Statistic Abstract 1976 Births to unwed mothers By age 10-18 1957- 1975 Fla. Public Health Statistics Section, Fla. Dept, of Health 8 Rehabilitative Services Table 1 continued Social variables Indicator a Available time series data Years State U.S. Source By race 1930 , 40,50, 56-75 Fla. Public Health Statistics Section, Fla Dept, of Health 5 Rehabilitative Services By age Â§ race 1956- 1975 Fla. Public Health Statistics Section, Fla Dept, of Health Â§ Rehabilitative Services Public attiÂ¬ tude toward education 1969- 1976 X Gallup Poll (published annually in Phi Delta Kappan, 1969-77 Peer group characteristics 3 to 5 year olds enrolled in nursery school and kindergarten 1964- 1975 X HEW, National Center for Educational Statistics in Advisory Committee on Child Development (1976) study Children under 18 involved in divorce 1953- 1975 X 1953-67: Ferriss (1970) 1968-75: Bane (1977) Suspected offenders for 1958- 1972 X See Social Indicators 1973 for list of sources four violent crimes, by age Student's self- NAC concept Table 1 - continued o Social variables Indicator Years b Available time series data State U.S. Source Student's general NA ability Student's sense NA of fate control Student's atti- High school 1901- X HEW, National Center for Educational tudes toward education and graduation rate 1975 biennial Statistics in The Condition of Edu- cation 1977, Vol. 3, Part 1 motivation for achievement School reten- 1924-32 X HEW, National Center for Educational tion rate, 5th grade to high school graduaÂ¬ tion to 1967-75 Statistics in Digest of Education Statistics (1976 edition) Teacher expecta- NA t ions Teacher behavior NA in classroom Administrative NA leadership style 3- Indicators presented are examples of time series data that might be used to operationally define the accompanying variable; a number of other indicators representing data aggregated to local, state, or national levels (depending upon purpose) could be added and/or substituted. bYears presented are those found in the course of this study and do not imply that data are in any way limited to these years. CNA: Time series indicator not available for variable. 37 attempted. Furthermore, it is necessary to remember that since the population base increased over the decades covered by the data, a stable rate or percentage still represents larger absolute numbers of the phenomenon. The indicators selected are aggregated to either the state or national level; the appropriate level of aggregation would, of course, depend upon the specific planning activity. These indicators could be disaggregated by race, age, region, or sex (where approÂ¬ priate) for comparative analysis, and indeed this feature is a necessary characteristic in many of the definitions of social indicators (e.g., see definition by Land, 1971, preÂ¬ sented earlier in this chapter). CHAPTER III RATIONALE FOR EXTRAPOLATIVE METHODS SELECTED FOR COMPARISON One of the purposes of this study was to compare three purely extrapolative methods which could be used with social indicator data of the type described in Chapter II to foreÂ¬ cast future values of those indicators. In order to select methods which were appropriate for this purpose, both the general forecasting literature and forecasting applications of extrapolative techniques in specific areas were reviewed. Detailed descriptions of each technique as well as statistiÂ¬ cal assumptions, sensitivity, and evaluative criteria were derived primarily from the literature in economic statistics and regression analysis. The following sections provide (a) an overview of the extrapolative methods used in forecasting, (b) evaluation of the applicability of these methods for the purpose of this study, (c) a description of the three methods selected for comparison, including equations, parameters to be estimated, assumptions, and criteria to be used in the comparison of the three methods. Overview of Extrapolative Forecasting Methods An extrapolative forecasting method is a procedure for (a) identifying an underlying historical trend or cycle in 38 39 time series data, and (b) estimating future states of a variable based on current and historical observations/meaÂ¬ sures of that variable (Harrison, 1976). Extrapolation proÂ¬ vides a "surprise free" projection of the future, but not necessarily a future which is a bigger and better (or worse) version of the present. Martino (1976) noted that some extrapolation methods allow the forecaster to identify policy variables which are subject to manipulation and which allow the decision-maker to alter the future away from today's pattern of events. (p. 4) In the social realm extrapolation of trends may at least allow the planner or policy maker to make enlightened deÂ¬ cisions to prepare for the future. Economic and Business Forecasting Because of the impetus in the 1930's and 1940's to describe and forecast the economic condition, many extraÂ¬ polative methods were developed with economic applications in mind. Greenwald (1963, p. 187) classified methods for determining economic trends into (a) non-mathematical methods such as freehand curve fitting, first-order differences, semiÂ¬ averages, selected points, and weighted and unweighted moving averages; and (b) mathematical methods such as least squares, moments, maximum likelihood, and others. In general, only the mathematical methods, which include a widely diverse array of complex curve-fitting techniques, seem to be relied upon for forecasting purposes while the non-mathematical methods are used for preliminary analysis of the shape of the 40 time series data. (For descriptions of these methods, see Greenwald, 1963; Mayes Â§ Mayes, 1976; Mendenhall $ Reinmuth, 1971; Neiswanger, 1956; Tuttle, 1957.) Approaches to governmental/national economic forecasting (e.g.,Theil, 1966) often reach a relatively high level of mathematical and theoretical sophistication. This appears to be the result of decades of development, of applying method in light of theory, and developing both in turn. It is also the result of substantial investment of financial and manpower resources by both government and industry. The value of extrapolative forecasting to individual decision-makers in business has become apparent (Makridakis, Hodgsdon, 5 Wheelwright, 1974). Indeed, companies of all sizes are compelled to make forecasts for a number of variaÂ¬ bles which affect them. Makridakis et al. (1974) have noted, however, that as with the development of most management science techniques, the application of these [extrapolative forecasting] methods has lagged behind their theoretical formuÂ¬ lation and verification. (p. 153) Thus, the authors observed that while the need for forecasting methods is recognized by managers in business, few are familÂ¬ iar with the numerous techniques available and their characÂ¬ teristics in order that the one most appropriate for a given situation be selected. To help meet this need, Makridakis et al. have developed an interactive forecasting system (called Interactive Forecasting [SIBYL/RUNNER]) which allows a number of factors to be considered in the selection of a 41 forecasting technique for a given set of data. Although the system has been well tested in teaching situations, it has not had extensive application in actual business settings. Quantitative techniques available in the Interactive ForecasÂ¬ ting (SIBYL/RUMNER) system fall under the general headings of smoothing, decomposition, control, regression, and other techniques. The techniques considered under those headings are clearly explained in a subsequent work of two of the authors (Wheelwright 5 Makridakis, 1977). Technological Forecasting Martino (1973b; 1976) described the extrapolative methods most commonly used in technological forecasting in relation to the shape of their fitted curves: (a) growth curve, an S-shaped curve, which requires the setting of an upper limit; (b) trend curve, an exponential function which takes the form of a straight line when logarithmic transformation of the data is undertaken. Martino (1973b) illustrated the use of the growth curve with data on lowest temperature achieved in the laboratory by artificial means and the trend curve with data on productivity in the aircraft industry. It should be noted that both the growth curve and the trend curve applied to technological change by Martino (1973b) are highly versatile approaches with applications in a number of disciplines. Both methods are derived from the least squares formula for a straight line. The growth curve is a modified exponential, that is, it represents a variable which changes at a changing rate; the trend curve is a 42 geometric straight line which represents a variable which changes at a constant rate (Neiswanger, 1956). Educational Forecasting Uses of extrapolative methods in education have generally- been limited to projections of expenditures, school enrollÂ¬ ments, and the number of instructional staff, high school graduates, and earned degrees. While many states and school districts have developed their own models, especially for projections of enrollments, the National Center for EducaÂ¬ tion Statistics (U.S. Department of Health, Education, 6 Welfare, 1 977c) in developing projections of education statisÂ¬ tics to 1985-86 relied on regression methods wherever a trend could be established. Specifically, either arithmetic straight lines or logistic growth curves, depending upon the nature of data, were fitted by the method of least squares. The following was noted, however: For both the straight line and logistic growth curve, the fitted curve often lies considerably above or below the last obÂ¬ served point, resulting in an unusual rise or drop from the last actual observaÂ¬ tion. To avoid this and give face validity to the projections, the fitted curve was used only to establish the last point, and a new curve was drawn through the last observed ratio and the end point on the fitted curve. (U.S. Department of Health, Education, 6 Welfare, 1977c, p. 92) Brown (1974) summarized the use of trend analysis methods in education and noted their potential applications in educaÂ¬ tional administration. The four extrapolative methods that he critiqued were (a) arithmetic straight line extrapolation, 43 (b) time series analysis (really a simplified version of the Box-Jenkins technique), (c) the S-shaped growth curve, and (d) cohort analysis (actually the trend curve described by Martino in the previous section). The examples selected by Brown do not reveal the versatility of the methods illustraÂ¬ ted; he did, however, provide a comprehensive review of literature describing applications in other fields. A number of methodological concerns raised by Brown were considered in this study. In a critique of selected futures prediction techniques that might be employed by educational planners, Folk (1976) observed that exponential trend line and arithmetic straight line projections appear to be the most commonly used extraÂ¬ polative techniques. This author provided a number of useful measures for evaluating statistically derived regression lines. The educational applications just described are basically attempts to project inputs such as money, pupils, or teachers to the educational system or outputs (graduates, degrees earned) of that system. No attempt to extrapolate the future status of variables which affect these student-related inputs or outputs was discovered in the literature search. Extrapolative Methods in Other Areas Several areas have developed highly specialized extrapoÂ¬ lative methods in making forecasts of the future. PopulaÂ¬ tion, employment, and unemployment projections, for example, 44 are usually based on fairly complex models which incorporate a number of factors. These particular applications are not reviewed here due to their highly specialized purposes and functions. Applicability of Reviewed Extrapolative Methods for Study In evaluating the applicability of the previously re- viewed extrapolative methods for projected future states of the time series indicators selected for use in this study, several points needed to be considered. Chief among these were (a) the underlying pattern of the data that can be recognized and (b) the type or class of model desired (from Wheelwright Â§ Makridakis, 1977). Both of these will be briefly considered in relation to this study. The Pattern of the Data From graphical representations of each indicator, the data for each appeared to be characterized by a trend which either increased or decreased with time. Some also appeared to contain cyclical patterns and random fluctuations. It seemed as if major trends might follow the form of a straight line or curve with one or two bends. The Class of Model Wheelwright and Makridakis (1977) distinguished four classes or categories of models: 1. The time series model "always assumes that some pattern of combination of patterns is recurring over time" (p. 22). 45 2. The causal model assumes "that the value of a cerÂ¬ tain variable is a function of several other variables" (p. 23). 3. The statistical model comprises a number of foreÂ¬ casting techniques; it uses the language and procedures of statistical analysis to identify patterns in the variables being forecast and in making statements about the reliability of these forecasts. (p. 23) 4. The nonstatistical model includes "all models that do not follow the general rules of statistical analysis and probability" (p. 24). Of course, some techniques can be classified into more than one of the four types of models. It appeared that the statistical model, with its well-defined properties, and replicable procedures, would be an appropriate starting point for predicting the long-term trends in the selected time series data. The review of the literature revealed several techniques denoted by the form of their curves which are sensitive to long-term trends in the data and which are classified under the statistical model: (a) the arithmetic straight line, (b) the S-shaped growth or logistic curve, (c) the trend or exponential curve, (d) the polynomial curve. All of these techniques are regression techniques solved by least squares procedures. Techniques (b) through (d) require data transÂ¬ formations to satisfy the basic linear model used in regresÂ¬ sion. The growth or logistic curve was eliminated from comÂ¬ parison because this technique necessitates the setting of 46 limits which might bias the results of the study due to its ex post facto nature. The remaining three techniques were considered to be appropriate for use in the comparison phase of this study. Description of Methods to be Compared Since the three techniques selected for comparison are intrinsically linear in their parameters (Draper Â§ Smith, 1966), the general linear model denoted by the simple or bivariate regression equation is presented first. AdditionÂ¬ ally, estimation of the parameters of the equation by least squares procedures, the assumptions of the model, and criteria for evaluation and comparison of the three methods are disÂ¬ cussed. Each technique is then described in relation to the general linear model. The General Linear Model In the comparison of methods using selected time series indicators, time in years is considered the independent variable and the indicator is considered the dependent or response variable. Thus, if time is denoted by X, and the indicator is denoted by Y, a functional relationship in the form Y = f(X) might be stated. However, since most social relationships are stochastic (probabilistic) rather than deterministic in nature, a more appropriate form might be 47 Y = f(X) + e, where e represents error, a measure of the unknown factors. When the relationship between the two variables, time and the indicator (Y) is assumed to be linear (that is, repreÂ¬ sented by a straight line), the equation becomes Y = Bo + 31X; and because many social relationships are stochastic for particular values of the variables, this equation is actually Y = Bo + 31X + e. Since the population parameters Bo and Bi are not known unless all possible occurrences of X and Y are known, the available data are used to provide estimates b0 and bi of Bo and Bi as in the following regression equation, Y=bo +biX+e A (where Y denotes predicted values of Y). The constant bo (the intercept) and the regression coefficient bi (the slope of the regression line) can be determined by ordinary least squares procedure, "so called because it estimates. . .in such a way that the sum of squared residuals, 2 Ee^ , is as small as possible" (Mayes Â§ Mayes, 1976, p. 112). (For detailed treatment of simple regression and least squares estimation of Bo and Bi, see, for example: Draper 5 Smith, 1966; Kerlinger Pedhazur, 1973; Mayes Â§ Mayes, 1976; MenÂ¬ denhall, Ott, 5 Larson, 1974; Mendenhall 5 Reinmuth, 1971; Runyon 5 Haber, 1967.) 48 The Assumptions of the Linear Model Draper and Smith (1966) noted that In many aspects of statistics it is necessary to assume a mathematical model to make progress. It might be well to emphasize that what we are usually doing is to consider or tentatively entertain our model. (p. '8) Thus, when the general linear model is employed as it is in this study, it becomes necessary to examine the assumptions upon which the model is based and to judge whether the model is in fact appropriate for the data. Assumptions for the general linear model include the following: 1. The regression equation A Y = b o + biX + e is a better predictor of Y than Y = Y (bi^ 0). 2. The regression equation accounts for a significant portion of the variation in Y, that is, the relationship between X and Y described by the equation is not the result of chance. 3. The error term e has a mean value equal to zero and variance equal to a2; it is an independent random variable which is normally distributed. If the first two assumptions are not met, then the model is not a good predictor for that data. If the third assumpÂ¬ tion is not met, then it is not appropriate to interpret the results statistically, that is, in terms of the probability 49 distribution of the random error e. It is possible to test Assumption 1 and Assumption 2 by the F statistic. Assumption 3 is best evaluated by plotting the residuals and examining the pattern of the deviations from the regression line (Anscombe, 1973; Anscombe Â§ Tukey, 1963; Draper 6 Smith, 1966). Independence of the errors (Assumption 3[e]) may be tested by the Durbin-Watson test for serial correlation (Durbin Â§ Watson, 1950; Durbin 6 Watson, 1951; Mayes Â§ Mayes, 1976; Wheelwright 6 Makridakis, 1977). Criteria for Comparison of Methods The following questions were derived from the literature to guide the comparison of methods: 1. Do the data satisfy the assumptions of the model? (See previous section.) 2. How well does the regression line fit the data from which it was derived (the two-thirds of the data points used to generate the prediction equation)? Tufte (1974, pp. 69-70) listed four measures of quality of fit: A a. the N residuals: Y. - Y. i i b. the residual variation: A 2 s2 y â€¢ x tq - YP N - k* -1 (or the square root of the residual variation, S , called y â€¢ x the unbiased standard error of estimate). *k refers to the number of X terms in the regression equation. 50 c. the ratio of explained to total variation: r2 = E(Y. - Y)2 2(Yi - 7) 2 d. the standard error of the estimate of the slope /z(xi - X)2 Thus, for each set of data, the methods are compared according to these four measures. The observed and preÂ¬ dicted values of Y are also reported in tabular form; both observed and predicted values are plotted for visual comÂ¬ parison as recommended by Anscombe (1973). 3. How well does the extrapolated line fit the data (the one-third of the data points that were not used to generate the prediction equation)? The residual variation around the extrapolated line, which is an indicator of the accuracy of the forecasting technique, may be expressed by its square root, the standard error for the extrapolated values. As in (2), the observed and extrapolated values are reported in tabular form; both observed and extrapolated values are plotted for visual comparison. Neiswanger (1956, p. 534) cautioned against accepting only mathematical tests of "goodness of fit" as proof that the mathematical expression is appropriate for the trend in the data. Other considerations such as the "reasonableness of the extrapolated values which the trend may yield" (p. 534) 51 and "the extent to which this statistical manifestation of growth is supported by other evidence" (p. 534) should be kept in mind. Thus, the calculation of a trend is more than a mathematical analysis in curve fitting; it is essentially a problem of analysis of the phenomena represented by the data (Neiswanger, 1956). It should be also noted that while the standard error of estimate gives an overall measure of error around the regresÂ¬ sion line, it may not be appropriate for computing confidence intervals for a specific forecast value. The reason for this is that the further an X is from X, the larger is the error that may be expected when predicting Y from the regression line. Draper and Smith (1966) noted: We might expect to make our "best" preÂ¬ dictions in the "middleâ€ of our observed range of X and would expect our predicÂ¬ tions to be less good away from the "middle." (p. 22) Therefore, the confidence limits for the true value of Y for a given X are two curved lines about the regression line. The limits change as the position of X changes. Hence the following equation was provided by Wheelwright and Makridakis (1977, p. 82) for computing the standard error of forecast (SEf): ~ - - - ' II Z(Y. - Y.)2 y l lJ ! + 1 + u N - k n uxi - X)2 _ for a specific forecast value. 52 Method 1: Simple Linear Regression The equation Y = Â£(X) describes a natural functional relationship between X and Y. If this functional relationship can be expressed by a straight line on arithmetic paper, the linear, first-order regression equation Y=bo +biX+Â§ may be appropriate. The natural linear function is used when an absolute amount of change in Y per unit of X is hypotheÂ¬ sized. Method 2: Log-linear Regression Occasionally when time series data are plotted on an arithmetic scale the scatter of points fall more in a curve than in a straight line with the curve rising or decreasing more rapidly as X increases. These same data when plotted on a semilogarithmic scale will produce a straight line. The relationship between X and Y may then be described by log Y = f(X) or Y = abx, the exponential form of the logarithmic relationship between X and Y. The exponential function is used when there is thought to be a constant rate of change in Y per unit absolute change in X. Thus, for each year (X), Y changes by a constant perÂ¬ centage (rather than by an absolute amount as in Method 1). 53 It is possible to fit the exponential function to the general linear model by transforming the values of Y to log Y. Thus Y = ab becomes log Y = a + bX log Y = log a + X log b or A log Y = log b0 + X log bi + e. As in the case of the natural number straight line, the method of least squares is used to estimate the parameters necessary for computing the logarithmic (or geometric) straight line. Tuttle (1957, p. 431) noted, therefore, that /\ /\ the log Y's are fitted to the log Y's, not the Y to the Y's, by the least squares criterion. Thus, Tuttle (1957, p. 432) recommended that the standard error of estimate be computed /\ from the antilogs of the log Y values. If S # was computed as the root mean square of the unexplained variation, "it would be in terms of the deviations of the logarithms of the A Ycs[Y's] from the logarithms of the Y's" (Tuttle, 1957, p. 432). The S x would not be comparable to those obtained from untransformed data as in Method 1. Similarly, Seidman (1976) has observed that in comparing linear and log-linear models, R2 may not be a sufficient criterion of choice. This is because the R2 represents "the proportion of variance of the logarithm of Y explained by the regression: log Y, not Y, is the dependent variable" (Seidman, 1976, p. 463). Therefore, Seidman recommended 54 using the antilogs of the predicted values of log Y "in a regression explaining variability in Y" (p. 463). This R2 may then be used for comparison purposes. The examples given by Seidman (1976) were based on logarithmic transformations of both dependent and independent variables, but the same obÂ¬ servation may be made when only the dependent variable is transformed. Seidman's reservation about R2 has been conÂ¬ sidered in this study. If an exponential curve appears to fit the data, it is often desirable to find the annual rate of change c. This can be derived from the regression coefficient bi, according to the following equation: log bi = (1 + c) change = antilog bi - 1. The result should then be expressed as a percent (Mayes 5 Mayes, 1976, p. 94; Nie et al., 1975, p. 370). The common or Briggs logarithm, used in the Y transÂ¬ formation in this study, is the power to which 10 must be raised to equal the number (see Neiswanger, 1956, p. 210; Tufte, 1974, p. 108). Natural logs or logs to the base 2 could also have been used to obtain the same results (Snedecor, 1956, pp. 450-451). Method 3: Polynomial Regression In Method 1, the equation which expresses a straight line relationship between X and Y is A Y = b o + b i X + Ã© 55 which is a linear (in the b's) first-order (in X) regression equation. When this functional relationship between X and Y can be expressed as a solid, or unbroken curve on arithmetic paper, the linear, second-order (or quadratic) regression equation A Y = b0 + b,X + b2X2 + Ã© may be appropriate. When the relationship can be expressed as a curved line with two bends on arithmetic paper, the linear, third-order (or cubic) regression equation A Y = bo + biX + b2X2 + baX3 + Â§ may be used. According to Kerlinger and Pedhazur (1973, p. 209), the highest order a polynomial equation may take is equal to N - 1, where N is the number of distinct values in the independent variable. However, since one of the goals of scientific research is parsimony, our interest is not in the predictive power of the highest degree polynomial equation possible, but rather in the highest degree polynomial equation necessary to describe a set of data. (Kerlinger Â§ Pedhazur, 1973, p. 209) Another reason for a parsimonious approach to polynomial curve fitting is that for each order added to the equation, a degree of freedom is lost. This is especially important when the number of observations are small as they are in this study (observations range from 8 to 20 in each of the eight sets of data). Also, higher order polynomial curves may possess statistical significance but be devoid of practical 56 significance. Accordingly, only the quadratic and cubic forms of the polynomial regression equation are considered. In the polynomial regression the independent variable, X (time), is treated as a categorical variable and is raised to a certain power. In the quadratic equation, each value of X is squared to create a new vector of the squared X's, X2. Similarly, in the cubic equation, each value of X is cubed to create an additional vector of the cubed X's, X3. Thus, the resulting equation can be solved by a stepwise multiple regression procedure, in which at each step of the analysis, the R2 is tested to see if the higher-degree polyÂ¬ nomial accounts for a significant proportion of the variance. While a least squares solution is used in this study, the values of the unknowns may also be found by orthogonal polyÂ¬ nomials (see Draper Â§ Smith, 1966, pp. 150-155; Greenwald, 1963, pp. 204-209; Kerlinger Â§ Pedhazur, 1973, pp. 214-216). Neiswanger (1956, pp. 529-532) noted that the second- degree and third-degree parabolas provide greater flexiÂ¬ bility in fitting a line to a set of data for the parabolas allow a trend to change direction. Whether or not the flexibility of the parabolic function enhances the predicÂ¬ tability of extrapolated Y values, however, is not certain and is examined in this study. CHAPTER IV COMPARISON OF EXTRAPOLATIVE METHODS USING SELECTED TIME SERIES INDICATORS In Chapter II a rationale for the selection of social variables operationally defined as time series indicators was provided. The following eight time series indicators were selected for use in the method comparison phase of this study: 1. Median family income in the United States expressed as 1971 constant dollars. 2. Number of families in the United States headed by women expressed as a percentage of total families. 3. Number of wives in the labor force expressed as a percentage of total wives in the United States. 4. Number of marriages in Florida expressed as rate per 1,000 population in Florida. 5. Number of divorces in Florida expressed as rate per 1,000 population in Florida. 6. Number of resident live births in Florida expressed as rate per 1,000 population in Florida. 7. Number of 3 to 5 year olds enrolled in nursery school and kindergarten expressed as percentage of total children 3 to 5 years old in the United States. 8. Number of children involved in divorce or annulment 57 58 expressed as rate per 1,000 children under 18 years old in the United States. A rationale for the three extrapolative methods selected for comparison in this study was presented in Chapter III. The three methods are simple linear regression, log-linear regression, and polynomial regression (specifically the quadratic and cubic forms). The following questions derived from the literature were proposed in Chapter III to guide the comparison of methods: 1. Do the data satisfy the assumptions of the general linear model? 2. How well does the regression line fit the data from which it was derived (the two-thirds of the data points used to generate the prediction equation)? 3. How well does the extrapolated line fit the data (the one-third of the data points that were not used to generate the prediction equation)? To answer these questions in terms of each method and to facilitate comparison among the three methods, the results obtained from applying each of the methods to each of the eight indicator data sets are presented in the following manner: 1. The fit of the regression line to the observed data is indicated by r2 and the unbiased standard error of estimate S . For the simple linear and log-linear regres- â€¢ sion methods the amount of variance accounted for by the regression line is tested by the F statistic (F value is the 59 same as that obtained by dividing bi by SE^ ). For the quadÂ¬ ratic and cubic forms of polynomial regression, both the r2 including all orders entered to that step (r2 ^ or ry 123-* and the increase in r2 attributable to the last order entered in the regression (r2ro .or r2r, -.-â€¢>) are tested with the b y(2.1) y(3.12)' F statistic.* (Of course, dividing the partial regression coefficients b2 in the quadratic form and b3 in the cubic form by their respective standard errors will also yield the same F value for the increase in r2.) 2. The fit of the extrapolated line to the data is indicated numerically by the standard error for the extraÂ¬ polated values, S . This measure reflects the ^ â€™ ext(yx) average deviation of the extrapolated values from the obÂ¬ served values of Y.; thus, 1 â€™ â€™ ^ext(yâ€¢x) (Note that this equation computing S x.) is Z(Y - Y)2 . (of extrapolated values) not the "unbiased" form used in *Actually the increase in r2 is tested according to the following ratio: P _ (r2 with k^-order term) - (r2 without k^^-order term). (1 - r2 with kt -order term) / (N - k - 1) Total r2 is tested according to the following ratio: p _ SS regression/k Sf) residual/(N - 1c - IT 60 3. All observed and predicted values of Y are reported in tabular and graphic form. 4. The residuals around the regression line were exÂ¬ amined for serial correlation by the Durbin-Watson d statistic, which is noted only when serial correlation is confirmed or questionable. Additionally, the standardized residuals were plotted against the sequence of cases and also against A standardized Y values. Such visual inspection of the data is discussed as necessary to support the interpretation of reÂ¬ sults in Chapter V. Presentation of Results Indicator 1 The mean and standard to generate the regression spectively. The following A derive Y: A Linear Y = -3933.29 + deviation equations regression for the Y val are 6674 and equations we ues used 987, reÂ¬ re used to 192.87 X Quadratic Y = 106.37 + 44.79 X + 1.35 X2 Cubic Y = 135025.33 + (-7385.17)X + 137.08 X2 + (-.82)X3 Log-linear log Y = 3.12346 + .01266 X. The goodness of fit of the regression lines derived from these equations is indicated by r2, r2 change, and S x in Table 2. An ANOVA summary table is presented in Table 18 in the AppenÂ¬ dix. The overall F's for all methods are significant (j><.01); the increases in r2 due to the higher order polynomials are not significant, however. 61 Table 2 Indicator 1: Summary Statistics for Prediction Equations by Method r2 r2 change F df S y â€¢ x . 97440 Linear 571.04** 1,15 163.02 .97531 .00090 Quadratic3 276.47** . 51 2.14 1.14 165.75 .98137 .00606 Cubic3 228.25** .42 3.13 1.13 149.40 log Y .97021 antilog Y .98726 r i â€¢ b Log-linear 488.54** 1,15 log.01157 164.92 1971 constant dollars. cBoth quadratic and cubic forms of the polynomial regresÂ¬ sion are presented. ^Both r2 and S have been recomputed using antilogs of /N Y * X y-V the log Y; much of the difference between r2(log Y) and r2 A (antilog Y) may be due to rounding. **Â£<.01 62 The average errors for the extrapolated Y values (S , ..) according to method employed are (a) linear, 669; ext iyâ€¢x j (b) quadratic, 487; (c) cubic, 1981; and (d) log-linear, 283. Observed Y's and predicted values of Y for both the original regression and the extrapolated lines are presented in Table 3. These data are graphically presented in Figure 2. Thus, there is very little difference in the total r2 for the methods; the quadratic and cubic forms added little to the r2 already provided by the linear component. The x for the cubic form is smaller (149.40) than for the other methods. When the lines are extrapolated beyond the original values, however, the cubic form is clearly the "worst" fit with a Sexj.Â£ of 1981 and the log-linear method the "best" with a Sext^v of 283. Whether the exponential curve would continue to be a superior predictor is a matter of conjecture. Indicator 2 The mean and standard deviation for the Y values used to generate the regression equations are 10.4 and .77, reÂ¬ spectively. The following regression equations were used to A derive Y: A Linear Y = 9.87860 + .02789 X A Quadratic Y = 11.08794 + C- -19035)X + .00626 X2 A Cubic Y = 11.45737 + (-.35750)X + .01954 X2 + (-.0027)X3 A Log-linear log Y = .99387 + .00118 X. The goodness of fit of the regression lines derived from these equations is indicated by r2, r2 change, and S in y â€¢ x 63 Table 3 Indicator 1: Observed Y's and Predicted Y's by Method Year Observed Y's Predicted Y1 1 s by Method Linear Quadratic Cubic3 Log-linear Original regression^ 1947 5,483 5 ,131C 5,185 5,323 5,231 1948 5,367 5,324 5,353 5,39 2 5,386 1949 5,278 5,517 5,533 5,499 5,545 1950 5,594 5,710 5,711 5,637 5,709 1951 5,783 5,903 5,892 5,803 5,878 1952 5,939 6,096 6,076 5,992 6,052 1953 6,433 6,289 6,262 6,197 6,231 1954 6,288 6,481 6,450 6,416 6,415 1955 6,693 6,674 6,642 6,642 6,605 1956 7,122 6,867 6,836 6,871 6,800 1957 7,138 7,060 7,033 7,097 7,002 1958 7,126 7,253 7,233 7,317 7,209 1959 7,524 7,446 7,435 7,524 7,422 1960 7,688 7,639 7,640 7,714 7,642 1961 7,765 7,831 7,848 7,882 7,868 1962 7,975 8,024 8,058 8,023 8,100 1963 8,267 8,217 8,271 8,133 8,340 Extrapolation^ 1964 8,579 8,410 8,487 8,205 8,584 1965 8,932 8,603 8,705 8,235 8,838 1966 9,360 8,796 8,926 8,220 9,100 1967 9,683 3,989 9,150 3,152 9,369 1968 10,049 9,182 9,377 8,028 9,646 1969 10,423 9,374 9,606 7,843 9,931 1970 10,289 9,567 9,838 7,591 10,225 1971 10,285 9,760 10,072 7,268 10,527 Note. Indicator I is median family income expressed in 1971 constant dollars. aBoth quadratic and cubic forms of polynomial regression are presented. ^The regression line is derived from 2/3 of the known data points. Predicted Y's in terms of 1971 constant dollars are rounded to number of places in original data. ^Values are extrapolated beyond the data points used to generate the regression equation. Median family income (1971 constant dollars) 64 Figure 2. Indicator 1: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extrapolation). 65 Table 4. An ANOVA summary table is presented in Table 19 in the Appendix. The overall F's for the quadratic and cubic forms of the polynomial regression are significant (p<.05); however, only the increase in r2 due to the quadratic is significant (jk.01). The average errors for the extrapolated Y values 1.5. Observed Y's and predicted values of Y for both the original regression and the extrapolated lines are presented by method in Table 5. These data are graphically presented in Figure 3. Thus, the cubic form of the polynomial accounts for the most variance in Y (89%) and has the smallest S (.33). The quadratic form, accounting for 81% of the variance in Y, has a S of .39; visual inspection of the second-degree y â€¢ x curve reveals that this curve may, in fact, more closely fit observed values for the latter portion of the regression line than the cubic form. The linear and log-linear methods A provide no better estimate of Y than does Y; indeed, the standard error of estimate approximates the standard deviaÂ¬ tion of the observed Y's. When the lines are extended beyond the original values, the quadratic provides the superior fit CSext^ = .35); the fit of the cubic, linear, and log-linear methods to the observed values is poor with residuals becoming larger for successive years. Table 4 Indicator 2: Summary Statistics for Prediction Equations by Method r2 r2 change F df S y â€¢ x .16405 Linear 1.18 1,6 .76 .81417 .65012 o Quadratic 10.95* 17.49** 2.5 1.5 . 39 .89255 .07838 Cubic3 11.08* 2.92 3.4 1.4 .33 log Y .16905 antilog Y .16307 Log -1ineark 1.22 1,6 log.03175 . 78 Note. Indicator 2 is number of families in the United States headed by women expressed as a percentage of total families. Both quadratic and cubic forms of the polynomial regresÂ¬ sion are presented. ^Both r2 and S have been recomputed using antilogs of a y' x ^ the log Y; much of the difference between r2(log Y) and r2 A (antilog Y) may be due to rounding. * Â£<.05 67 Table 5 Indicator 2: Observed Y's and Predicted Y's by Method Predicted Y's by Method Year(X) Observed Y's Linear Quadratic3 Cubic3 Log-linear Original regression^ 1940(1) 11.2 9.9C 10.9 11.1 9.9 1947(8) 9.5 10.1 10.0 9.7 10.1 1950(11) 9.4 10.2 9.7 9.5 10.2 1955(16) 10.1 10.3 9.6 9.6 10.3 1960(21) 10.0 10.5 9.9 10.1 10.4 1965(26) 10.5 10.6 10.4 10.6 10.6 1970(31) 10.9 10.7 11.2 11.1 10.7 1971(32) 11.5 10.8 11.4 11.2 10.8 Extrapolation^ 1972(33) 11.6 10.8 11.6 11.2 10.8 1973(34) 12.1 10.8 11.9 11.3 10.8 1974(35) 12.4 10.9 12.1 11.3 10.8 1975(36) 13.0 10.9 12.4 11.3 10.9 Note. Indicator 2 is number of families in the United States headed by women expressed as a percentage of total families. aBoth quadratic and cubic forms of polynomial regression are presented. ^The regression line is derived from 2/3 of the known data points. cPredicted Y's are rounded to number of places in original data. ^Values are extrapolated beyond the data points used to generate the regression equation. Percent of families headed by women 69 Indicator 3 The mean and standard deviation for the Y values used to generate the regression equations are 32.0 and 3.8, respecÂ¬ tively. The following regression equations were used to de- A rive Y: Linear Y = 23.27325 + .74603 X A Quadratic Y = 23.38774 + .71893 X + .00126 X2 A Cubic Y = 22.60710 + 1.18842 X + (-.05693)X2 + .00194 X3 A Log-linear log Y = 1.37971 + .01047 X. The goodness of fit of the regression lines derived from these equations is indicated by r2, r2 change, and Sv in Table 6. An ANOVA summary table is presented in Table 20 in the AppenÂ¬ dix. The overall F's for all methods are significant (p<.01); however, the increases in r2 due to the higher order polynoÂ¬ mials are not significant. The average errors for the extrapolated Y values (Sext(y.x)) according to the method employed are (a) linear, 1.4; (b) quadratic, 1.2; (c) cubic, 2.7; and (d) log-linear, .6. Observed Y's and predicted values of Y for both the original regression and the extrapolated lines are presented by method in Table 7. These data are graphically repreÂ¬ sented in Figure 4. Thus, there is very little difference in the total r2 for the methods; the quadratic and cubic forms added an inÂ¬ significant amount to the r2 already provided by the linear component. The for the cubic form is only slightly y* X so/ better (.43) than for the other methods (.49-.54). 70 Table 6 Indicator 3: Summary Statistics for Prediction Equations by Method r2 r2 change F df VÂ» Linear .98451 826.05** 1,13 .49 .98460 .00009 Quadratic3 383.49** .07 2,12 1,12 . 50 Cubic3 .98953 .00493 346.51** 5.18 3.11 1.11 .43 log Y .98085 A antilog Y .99999 t i â€¢ b Log -1inear 665.69** 1,13 log.00760 . 54 Note. Indicator 3 is the number of wives in the labor force expressed as a percentage of total wives in the United States. 3. Both quadratic and cubic forms of the polynomial regresÂ¬ sion are presented, b 2 Both r and S have been recomputed using antilogs of /n y â€¢ x ^ the log Y; much of the difference between r2(log Y) and r2 A (anitlog Y) may be due to rounding. **Â£<.01. 71 Table 7 Indicator 3: Observed Y's and Predicted Y's by Method Predicted Y 's by Method Year(X) Observed Y's Linear Quadratic" Cubic3 Log-linear Original b regression 1950(1) 23.8 24.0C 24.1 23.7 24.6 1955(6) 27.7 27 . 7 27.8 28.1 27.7 1956(7) 29.0 28.5 28.5 28.8 28.4 1957(8) 29.6 29.2 29.2 29.5 29.1 1958(9) 30.2 30.0 30.0 30.1 29.8 1959(10) 30.9 30.7 30.7 30.7 30.5 1960(11) 30.5 31.5 31.4 31.4 51.3 1961(12) 32.7 32.2 32.2 32.0 3 2. U 1962(13) 32.7 33.0 33.0 32.7 32.8 1963(14) 33.7 33.7 33.7 33.4 33.6 1964(15) 34.4 34.5 34.5 34.2 34.4 1965(16) 34.7 35.2 35.2 35.0 3 5.5 1966(17) 3 5.4 36.0 36.0 35.9 36.1 1967(18) 36.8 36.7 36.7 36.9 37.0 1968(19) 38.3 37.4 37.5 38.0 37.9 Extrapolation^ 1969(20) 59.6 38.2 38.3 39.1 38.8 1970(21) 40.8 38.9 39.0 40.4 39.8 1971(22) 40.8 39.7 39.8 41.9 40.7 1972(23) 41.5 40.4 40.6 43.4 41.7 1973(24) 42.2 41.2 41.4 45.2 42.8 1974(25) 43.0 41.9 42.1 47.0 43.8 1975(26) 44.4 42.7 42.9 49.1 44.9 Note. Indicator 3 is the number of wives in the labor force expressed as a percentage of total wives in the United States. aBoth quadratic and cubic forms of polynomial regression are presented. ^The regression line is derived from 2/3 of the known data points. cPredicted Y's are rounded to number of places in original data. ^Values are extrapolated beyond the data points used to generate the regression equation. Percent of wives in labor force 72 1950 '55 '60 '65 '70 '75 Figure 4. Indicator 3: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extrapolation). 73 When the lines are extrapolated beyond the original values, however, the cubic form is clearly the "worst" fit with a Sext^ of 2.7 while the log-linear form is clearly the "best" fit with a ^ of .6. Indicator 4 The mean and standard deviation for the Y values used to generate the regression equations are 9.8 and 2.6, reÂ¬ spectively. The following regression equations were used to /V derive Y: Linear Y = 13.76707 + C--13216)X Quadratic Y = 14.16426 + (-.19720)X + .00148 X2 /v Cubic Y =10.90554 + 1.15549 X + (-.07961)X2 + .00125 X3 A Log-linear log Y = 1.12873 + (-.00496)X. The goodness of fit of the regression lines derived from these equations is indicated bv r2, r2 change, and S in Table 8. y â€¢ x An ANOVA summary table is presented in Table 21 in the AppenÂ¬ dix. The overall F statistic for the cubic form of the polyÂ¬ nomial regression is significant (p<.01); the increase in r2 due to the third degree polynomial is also significant (p<.01) The F statistic for both the linear and log-linear methods is significant (p<.05). The average errors for the extrapolated Y values (Sext(y.x)) according to method employed are (a) linear, 2.9; (b) quadratic, 2.5; (c) cubic, 4.1; and (d) log-linear, 2.7. Observed Y's and predicted values of Y for both the original regression and the extrapolated lines are presented by method in Table 9. These data are graphically represented in Figure 5. 74 Table 8 Indicator 4: Summary Statistics for Prediction Equations by Method r2 r2 change F df SVX Linear .42085 7.27* 1,10 2.07 .42592 .00507 3. Quadratic 3.34 . 08 2.9 1.9 2.17 Cubic3 .90338 .47746 24.93** 39.53** 3.8 1.8 .94 log Y .42298 antilog .36243/ A Y .42922Â° Log-linear^ 7.33* 1,10 log.07715 2.06 expressed as rate per 1,000 population in Florida. Both quadratic and cubic forms of the polynomial regresÂ¬ sion are presented, b ? Both r and S have been recomputed using antilogs of ^ y â€¢ a /v the log Y; much of the difference between r2(log Y) and r2 A (antilog Y) may be due to rounding. C A Two methods of computing r2 using antilogs of Y yielded different results. * Â£<.05 **Â£<.01 75 Table 9 Indicator 4: Observed Y's and Predicted Y's by Method Year(X) Observed Y Predicted Y's by Method 's Linear Quadratic Cubic3 Log-linear Original regression 1930(1) 11.6 13.6C 14.0 12.0 13.3 1940(11) 17.1 12.3 12.2 15.7 11.9 1950(21) 9.8 11.0 10.7 11.7 10.6 1960(31) 7.9 9.7 9.5 7.6 9.4 1963(34) 7.7 9.3 9.2 7.5 9.1 1964(35) 7.7 9.1 9.1 7.6 9.0 1965(36) 8.3 9.0 9.0 7.9 8.9 1966(37) 8.5 8.9 8.9 8.2 8.8 1967(38) 9.0 8.7 8.8 8.7 8.7 1968(39) 9.6 8.6 8.7 9.3 8.6 1969(40) 9.8 8.5 8.6 10.0 8 . 5 1970(41) 10.1 8.3 8.6 10.9 8.4 Extrapolation^ 1971(42) 10.5 8.2 8.5 11.6 8.3 1972(43) 11.0 8.1 8.4 12.8 8.2 1973(44) 11.4 8.0 8.4 14.1 8.1 1974(45) 11.0 7.8 8.3 15.6 8.0 1975(46) 10.1 7.7 8.2 17.3 8.0 Note. Indicator 4 is number of marriages in Florida exÂ¬ pressed as rate per 1,000 population. aBoth quadratic and cubic forms of polynomial regression are presented. ^The regression line is derived from 2/3 of the known data points. cPredicted Y's are rounded to number of places in original data. ^Values are extrapolated beyond the data points used to generate the regression equation. Marriages/1,000 population 18.0- Observed Y' s Cubic Figure 5. Indicator 4: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extrapolation). 77 Thus, it would appear that for the original regression the cubic form of the polynomial best fits the observed Y's. This method accounts for 90% of the variance in Y with less than half of the average error of the other methods. When the lines are extrapolated beyond the original values, however, the cubic form has the largest average error (Sext^ = 4.1). The quadratic form of the polynoÂ¬ mial is, in fact, the best predictor (Sextj- xj = 2.5) of the methods compared. Actually, for this set of data the mean (9.8) of the observed values of Y used in the original reÂ¬ gression would have been the best predictor of the future values of Y. Indicator 5 The mean and standard deviation for the Y values used to generate the regression equations are 4.6 and 1.0, respecÂ¬ tively. The following regression equations were used to A derive Y: A Linear Y = 4.10332 + .01692 X /\ Quadratic Y = 3.16091 + .17126 X + (-.00350)X2 A Cubic Y = 1.59416 + .82162 X + (-.04249)X2 + .00060 X3 Log-linear log Y = .56596 + .00288 X. The goodness of fit of the regression lines derived from these equations is indicated by r2, r2 change, and S x in Table 10. An ANOVA summary table is presented in Table 22 in the Appendix. Only the overall F for the cubic form of the polynomial regression is significant (Â£<.05); the F 78 Table 10 Indicator 5: Summary Statistics for Prediction Equations by Method r2 r2 change F df V* Linear .04360 .46 1,10 1.06 Quadratic .22401 1.30 2,9 1.00 .18041 2.09 1,9 Cubica .92134 31.23** 3,8 . 34 .69733 70.92** 1,8 Log -1inear^ log Y .12007 . antilog Y 1.36 1,10 log.10388 .12779 1.07 Note. Indicator 5 is number of dissolutions of marriage in Florida expressed as rate per 1,000 population. Both quadratic and cubic forms of the polynomial regresÂ¬ sion are presented. ^Both r2 and S have been recomputed using antilogs of a y â€¢ a a the log Y; much of the difference between r2(log Y) and r2 A (antilog Y) may be due to rounding. **Â£<.01 79 value for the increase in r2 for the cubic component is also significant (p<.01). The average errors for the extrapolated Y values (Sext(y.x)) according to the method employed are (a) linear, 2.2; (b) quadratic, 3.1; (c) cubic, .5; and (d) log-linear, 2.1. Observed Y's and predicted values of Y for both the original regression and the extrapolated lines are presented by method in Table 11. These data are graphically repreÂ¬ sented in Figure 6. For this set of data, the cubic form of the polynomial regression is a superior predictor of the observed Y's. This method accounts for 92% of the variance in Y with a S of y â€¢ x .34; the Sext^ is .5, considerably less than the other three methods with average error ranging from 2.1 to 3.1. The quadratic form is definitely the least appropriate method for this set of data since the curve bends in an opposite direction to the observed Y values (see Figure 6). Indicator 6 The mean for the Y values used to generate the regresÂ¬ sion equations is 16.9; the standard deviation, 1.3. The A following regression equations Avere used to derive Y: A Linear Y = 18.56785 + (-.36786)X Quadratic Y = 21.30892 + C-2.01250)X + (.18274)X2 A Cubic Y = 22.82142 + (-3.58611)X + .59524 X2 + (-.03056)X3 Log-linear log Y = 1.26761 + (-.00900)X. The goodness of fit of the regression lines derived from 80 Table 11 Indicator 5: Observed Y's and Predicted Y's by Method Year(X) < Observed Y Predicted Y's by Method 's Linear g Quadratic Cubic3 Log-linear Original regression^ 1930(1) 2.5 4.1C 3.3 2.4 3.7 1940(11) 5.8 4.3 4.6 6.3 4.0 1950(21) 6.4 4.5 5.2 5.7 4.2 1960(31) 3.9 4.6 5.1 4.2 4.5 1963(34) 4.1 4.7 4.9 4.1 4.6 1964(35) 4.1 4.7 4.9 4.2 4.6 1965(36) 4.2 4.7 4.8 4.3 4.7 1966(37) 4.2 4.7 4.7 4.4 4.7 1967(38) 4.6 4.7 4.6 4.6 4.7 1968(39) 4.9 4.8 4.5 4.8 4.8 1969(40) 5 . 2 4.8 4.4 5.1 4.8 1970(41) 5.5 4.8 4.2 5.4 4.8 Extrapolation^ 1971(42) 6.1 4.8 4.2 5.6 4.9 1972(43) 6.9 4.8 4.1 6.1 4.9 1973(44) 7 . 1 4.8 3.9 6.6 4.9 1974(45) 7.2 4.9 3.8 7.2 5.0 1975(46) 7.5 4.9 3.6 7.9 5.0 Note. Indicator 5 is number of dissolutions of marriage in Florida expressed as rate per 1,000 population. aBoth quadratic and cubic forms of polynomial regression are presented. ^The regression line is derived from 2/3 of the known data points. cPredicted Y's are rounded to number of places in original data. ^Values are extrapolated beyond the data points used to generate the regression equation. Dissolutions of marriage/1,000 population 2.0 1930 Figure 6 â€˜40 '50 1 1â€”T~ '60 '63 '65 70 '75 Indicator 5: Observed Y's and predicted Y's by method (vertical separates values of original regression from extrapolation). 1 ine 82 \ these equations is indicated by r2 , r2 change, and S in y â€¢ x Table 12. An ANOVA summary table is presented in Table 23 in the Appendix. The overall F's for the quadratic and cubic forms of polynomial regression are significant (p<.01); only the increase in r2 due to the quadratic component is signifiÂ¬ cant (j) < .01), however. The average errors for the extrapolated Y values (Sext(y xj) according to the method employed are (a) linear, 1.2; (b) quadratic, 7.8; (c) cubic, 1.5; and (d) log-linear, 1.4. Observed Y's and predicted values of Y for both the original regression and the extrapolated lines are presented by method in Table 13. These data are graphically repreÂ¬ sented in Figure 7. Thus, while the quadratic and cubic forms of polynomial regression best fit the observed Y's for the original regresÂ¬ sion, they do not continue to be superior predictors. In fact, the quadratic form has a S ., ^ of 7.8 while the â€™ 1 ext(y.x) Â£>ext(y-x) for the other three methods ranges from 1.2 to 1.5. No method is clearly the best predictor of Y when values are extrapolated beyond the original regression. It should be noted that the Durbin-'.Vatson d for the linear and log-linear methods approaches the lower limits of d and the possibility of serial correlation of the residÂ¬ uals cannot be overlooked. Because of the small number of observations involved in this data set (N = 8), interpretaÂ¬ tion of the Durbin-Watson d is more suggestive than conÂ¬ clusive . 83 Table 12 Indicator 6: Summary Statistics for Prediction Equations by Method r2 r2 change F df S y-x Linear .46628 5.24 1,6 1.04 o Quadratic .92655 31.54** 2,5 .42 .46026 31.33** 1,5 Cubica .97205 46.37** 3,4 . 29 .04550 6.51 1,4 t i â€¢ b Log-linear log Y .45957 â€ž antilog Y 5.10 1,6 log.02 58 2 .41427 1.03 Note. Indicator 6 is number of resident live births Tn Florida expressed as rate per 1,000 population. â€˜Both quadratic and cubic forms of the polynomial regresÂ¬ sion are presented. ^Both r2 and S have been recomputed using antilogs of /n y â– x /n the log Y; much of the difference between r2(log Y) and r2 /s (antilog Y) may be due to rounding. **Â£<.01 84 Table 13 Indicator 6: Observed Y's and Predicted Y's by Method Year(X) Observed Y Predicted Y's by Method 's Linear g Quadratic 1 Cubic3 Log-linear Original regression*3 1964(1) 19.7 18.2C 19.5 19.8 18.1 1965(2) 17.9 17.8 18.0 17.8 17.8 1966(3) 16.8 17.5 16.9 16.6 17.4 1967(4) 15.9 17.1 16.2 16.0 17.0 1968(5) 15.7 16.7 15.8 16.0 16.7 1969(6) 16.1 16.4 15.8 16.1 16.4 1970(7) 16.8 16.0 16.1 16.4 16.0 1971(8) 16.4 15.6 16.9 16.6 15.7 Extrapolation*^ 1972(9) 14.8 15.3 18.0 16.5 15.4 1973(10) 13.7 14.9 19.5 15.9 15.1 1974(11) 13.4 14.5 22.1 14.7 14.7 1975(12) 12.5 14.1 23.5 12.7 14.4 Note ! Indicator 6 is number of resident live births in Florida expressed as rate per 1,000 population. aBoth quadratic and cubic forms of polynomial regression are presented. ^The regression line is derived from 2/5 of the known data points. cPredicted Y's are rounded to number of places in original data. ^Values are extrapolated beyond the data points used to generate the regression equation. Resident live births/1,000 population 85 Figure 7. Indicator 6: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extrapolation). 86 Indicator 7 The mean and standard deviation for the Y values used to generate the regression equations are 32.2 and 4.8, reÂ¬ spectively. The following regression equations were used to A derive Y: /\ Linear Y = 23.42857 + 1.95476 X Quadratic Y = 23.58928 + 1.85833 X + .01071 X2 Cubic Y = 23.26430 + 2.19645 X + (-.07792)X2 + . 00657 X3 Log-linear log Y = 1.38414 + .02662 X. The goodness of fit of the regression lines derived from these equations is indicated by r2, r2 change, and ^ in Table 14. An ANOVA summary table is presented in Table 24 in the AppenÂ¬ dix. The overall F's for all methods are significant (Â£<.01); however, the increases in r2 due to the higher order polynoÂ¬ mials are not significant. The average errors for the extrapolated Y values (S ext(y.x)) according to the method employed are (a) linear, 1.4; (b) quadratic, 1.3; (c) cubic, 1.8; and (d) log-linear, 2.4. Observed and predicted values of Y for both the original regression and the extrapolated lines are presented by method in Table 15. These data are graphically represented in Figure 8. There is very little difference in the predictive value of the methods for the original regression. Each method accounts for 99% of the variance in Y; the range of the SA for all methods is from .34 to .43. When the lines are extrapolated beyond the original v â€¢ x 87 Table 14 Indicator 7: Summary Statistics for Prediction Equations by Method r2 r2 change F df S y â€¢ x Linear . 99560 1358.03** 1,6 . 34 q Quadratic .99572 581.74** 2,5 . 37 .00012 . 14 1,5 Cubic8 .99588 322.27** 3,4 .41 .00016 .15 1,4 Log -1inearD log Y .99333 â€ž antilog Y 893.91** 1,6 log.00577 .99999 .43 Note. Indicator 7 is number of 3 to 5 year olds enrolled in nursery school and kindergarten expressed as percentage of total children 3 to 5 years old in the United States. q Both quadratic and cubic forms of the polynomial regresÂ¬ sion are presented. ^Both r2 and S have been recomputed using antilogs of the log Y; much of the difference between r2(log Y) and r2 A (antilog Y) may be due to rounding. **o<.01 88 Table 15 Indicator 7: Observed Y's and Predicted Y's by Method Predicted Y' 1 s by Method Year(X) Observed Y's Linear Quadratic Cubic3 Log-linear 1964(1) 25.5 _ . . . â€¢ b Original regression 25.4C 25.5 25.4 25.7 1965(2) 27.1 27.3 27.3 27.4 27.4 1966(3) 29.4 29.3 29.3 29.3 29.1 1967(4) 31.6 31.2 31.2 31.2 30.9 1968(5) 33.0 33.2 33.1 33.1 32.9 1969(6) 34.6 35.2 35.1 35.1 35.0 1970(7) 37.5 37.1 37.1 37.1 37.2 1971(8) 39.1 39.1 39.1 39.2 39. -5 1972(9) 41.6 Extrapolation^ 41.0 41.2 41.5 42.0 1973(10) 40.9 43.0 43.2 44.0 44.7 1974(11) 45.2 45.0 45.3 46.7 47.5 1975(12) 48.7 46.9 47.4 49.8 50.5 Note. Indicator 7 is number of 3 to 5 year olds enrolled in nursery school and kindergarten expressed as percentage of total children 3 to 5 years old in the United States. aBoth quadratic and cubic forms of polynomial regression are presented. ^The regression line is derived from 2/3 of the known data points. cPredicted Y's are rounded to number of places in original data. ^Values are extrapolated beyond the data points used to generate the regression equation. Percent of 3 to 5 year olds in nursery school and kindergarten 89 Figure 8. Indicator 7: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extrapolation). 90 values, the quadratic and linear methods are somewhat better predictors with S of 1.3 and 1.4, respectively. Should the lines be extrapolated further into the future, however, the cubic and log-linear methods might improve in predictive accuracy (see Figure 8). Indicator 8 The mean and standard deviation for the Y values used to generate the regression equations are 7.6 and 1.2, reÂ¬ spectively. The following regression equations were used /N to derive Y: Linear Y = 5.48381 + .26286 X Ouadratic Y = 6.22593 + .00093 X + .01637 X2 Cubic Y = 6.52952 + (-.19576)X + .04613 X2 + C - - 00124)X3 Log-linear log Y = .75640 + .01481 X. The goodness of fit of the regression lines derived from these equations is indicated by r2, r2 change, and S in y â€¢ x Table 16. An ANOVA summary table is presented in Table 25 in the Appendix. The overall F's for all methods are signifiÂ¬ cant (p<.01); however, only the increase in r2 due to the quadratic component is significant (p<.01). The average errors for the extrapolated Y values (Sext(y x)) according to the method employed are (a) linear, 4.3; (b) quadratic, 1.9; (c) cubic, 4.2; and (d) log-linear, 3.7. Observed and predicted values of Y for both the original regression and the extrapolated lines are presented by method in Table 17. These data are graphically repreÂ¬ sented in Figure 9. 91 Table 16 Indicator 8: Summary Statistics for Predici tion Equations by Method r2 r2 change 3 F df S y.x Linear .92137 152.33** 1,13 . 36 a Quadratic .97402 224.95** 2,12 .21 .05265 24.32** 1,12 Cubic3 . 97822 164.65** 3,11 . 20 .00420 2 .12 1,11 Log-1inear^ log Y .93096 175.29** 1,13 log . 01872 antilog Y .89333 .30 Note. Indicator 8 is number of children invo lved in divorce or annulment e xpressed as rate per 1,000 children under 18 years old in the United States. aBoth quadratic and cubic forms of the polynomial regres - sion are presented. ^Both r /\ the log Y; 2 and S have y.x much of the di been recomputed fference between using antilogs of r2(log Y) and r2 (antilog Y) may be due to rounding. **p<.01 92 Table 17 Indicator 8: Observed Y's and Predicted Y's by Method Year(X) Observed Predicted Y 's by Method Y's Linear Quadratic Cubic a Log-linear â€ž . . , . b Original regression 1953(1) 6.4 5. 7C 6.2 6.4 5.9 1954(2) 6.4 6.0 6.3 6.3 6.1 1955(3) 6.3 6.3 6.4 6.3 6.3 1956(4) 6.3 6.5 6.5 6.4 6.5 1957(5) 6.4 6.8 6.6 6.5 6.8 1958(6) 6.5 7.1 6.8 6.7 7.0 1959(7) 7.5 7.3 7.0 7.0 7 . 2 1960(8) 7.2 7.6 7.3 7.3 7. 5 1961(9) 7.8 7.8 7.6 7.6 7.8 1962(10) 7.9 8.1 7.9 7.9 8.0 1963(11) 8.2 8.4 8.2 8.3 8.3 1964(12) 8.7 8.6 8.6 8.7 8.6 1965(13) 8.9 8.9 9.0 9.1 8.9 1966(14) 9.4 9.2 9.4 9.4 9.2 1967(15) 9.9 9.4 9.9 9.8 9.5 Extrapolation 1970(18) 12.5 10.2 11.5 10.7 10.5 1971(19) 13.6 10.5 12.1 11.0 10.9 1972(20) 14.8 10.7 12.8 11.1 11.3 1973(21) 15.9 11.0 13.5 11.3 11.7 1974(22) 16.4 11.3 14.2 11.3 12.1 1975(23) 17.1 11.5 14.9 11.3 12.5 Note. Indicator 8 is number of children involved in divorce or annulment expressed as rate per 1,000 children under 18 years old in the United States. aBoth quadratic and cubic forms of polynomial regression are presented. bThe regression line is derived from 2/3 of the known data points. cPredicted Y's are rounded to number of places in original data. ^Values are extrapolated beyond the data points used to generate the regression equation. Children involved in divorce/1,000 children under 18 93 50 Hâ€”l 1 1 1 1 1953 '55 '60 '65 '70 '75 Figure 9. Indicator 3: Observed Y's and predicted Y's by method (vertical line separates values of original regression from extrapolation). 94 All the methods are adequate predictors of Y for the range of the original regression. The quadratic and cubic forms of the polynomial regression are only slightly better than the linear and log-linear methods, however, accounting for approximately 5% more variance in Y with a x of .21 and .20, respectively (compared with .36 and .30 for the latter methods). When the lines are extrapolated beyond the original values, the quadratic form clearly becomes the superior predictor with a S . , , of 1.9. The S ., . for the r ext(yx) ext(y-x) other methods ranges from 3.7 to 4.3. Based on examination of the plotted values in Figure 9, it would appear that the quadratic form might continue to have the highest predictive accuracy. Of course, if all the data were employed in the regression, new curves influenced by the later data points would be drawn and the predictive value of all methods would have to be reassessed. The Durbin-Watson d for the linear and log-linear methods approaches the lower limits of d. The possibility of serial correlation of the residuals thus cannot be conÂ¬ clusively rejected. CHAPTER V DISCUSSION Two objectives of this study were (a) to select, using Bronfenbrenner's ecology of education model, and operationally define at least 10 variables that research has shown to be related to the outcomes of education; (b) to use these variables operationally defined as time series indicators in the comparison of three purely extrapolative forecasting methods. In this chapter, methodological strategies involved in the selection and operational definition of social varÂ¬ iables as well as the use of the Bronfenbrenner model are discussed in terms of their viability for future use. ReÂ¬ sults of the method comparison phase are discussed according to statistical and practical considerations derived from the literature. The Variables Selection The 10 variables selected for use in this study may need to be reconsidered, since several of them (e.g., stuÂ¬ dent's sense of fate control, teacher behavior in the classÂ¬ room) are rather obscure both in their constitutive and in 95 96 their operational definitions. Additionally, several of the variables (e.g., administrative leadership style) are only peripherally related to educational outcomes. There appears to be a need to develop instead a number of demographic variables related to the family. Presently there are very few efforts to define, systematically analyze, and forecast the demographic context of education (Coates, Note 1; Morrison, 1976). Inclusion of social variables representing changing patterns in the family as to fertility, stability, employment, income, and so on may result in new perspectives on both educational outcomes and purposes. These variables may be examined individually and then in clusters to discover trends and interdependencies which may continue into the future. Of course, the selection of variables need not be confined to demographic variables; rather demographic variables may provide a starting place which has potential for development and expansion. Bronfenbrenner's Ecology of Education Model Bronfenbrenner's model of the ecology of education is a conceptually useful representation which accounts for the numerous factors which impinge upon the educational process. The model includes those areas which are considered under force analysis and contextual mapping, two strategies used in macro-level educational planning. (See Hencley and Yates, 1974, for a description of these techniques.) Furthermore, the model includes those variables which traditionally have 97 been the concern of educational researchers. The need for a theoretical scheme, or a model derived from theory, when selecting social indicators for study has been acknowledged by Allardt (1971) and Sismondo (1973). The Bronfenbrenner model may help to fill this need for educaÂ¬ tional planners. Operational Definition of Variables The operational definition of social variables in terms of time series indicators is a challenging and often frusÂ¬ trating task. Several of the problems acknowledged by Ferriss (1970) and others involve the nature and availability of the data. Some data are thought to be of poor quality with the basis for collection changing frequently over time. Other data are collected only for certain years or are made availÂ¬ able only for certain years. There are numerous sources of data; however, it requires careful review and consideration of available data to find the most reliable measures for use in analysis and forecasting. Other problems arise from the process of social measureÂ¬ ment itself. Social investigators need not let these concerns deter efforts to develop social indicators; rather these concerns (as described by Etzioni and Lehman, 1969) might be used to guide future efforts to operationally define social variables. Perhaps the most important functions of time series indicators are (a) the comparison of trends in an indicator 98 over time and (b) the comparison of different groups in reÂ¬ lation to a specific indicator. Thus, for an indicator such as births to unwed mothers, the aggregated data may be examined for a general societal trend; the general trend may be disaggregated according to state or county, and then according to race and age of mother. It is often only when the data are disaggregated (for example, births to unwed mothers age 10-18) that they become most meaningful for the particular planning activity. A graphic presentation of the disaggregated data is a valuable part of any attempt to underÂ¬ stand the underlying social process represented by the time series indicator. The Extrapolative Methods Statistical Considerations The use of the general linear model was appropriate in this study. While violations of the assumptions did occur occasionally in the linear or log-linear methods, they could be corrected by the addition of the quadratic and/or cubic terms to the regression equation. In all data sets one or more of the methods accounted for significant amounts of the variance in Y. In two of the data sets, however, there remained questionable patterns in the residuals that were not corrected by any of the methods employed. Coincidentally, the extraÂ¬ polated lines of these indicators (numbers 4 and 8) were extremely poor approximations of the observed data. The 99 Durbin-Watson d statistic for serial correlation of residuals was inconclusive for the simple linear and log-linear methods for indicators 6 and 8 but did not indicate this problem with the other methods in any of the other data sets. Thus, at least one of the three methods based on the general linear model was appropriate to use with the data sets in this study. The application of the general linear model to other data sets of time series indicators or even to these data sets when additional observations are included would require further investigation, however. In the general forecasting literature it is usually acknowledged that a method which "best fits" the observed data should be used for extrapolative purposes. This "best fit" is indicated by magnitude of the coefficient of deterÂ¬ mination r2 and the unbiased standard error of estimate S y â€¢ x In this study, however, r2 and S were not always good in- y * x dicators of the accuracy of extrapolated values to the observed data. For example, while the cubic form of the polynomial best fit the observed Y's for the original regression in Indicator 5, it was the worst fit when expolated into the future. In several other data sets (namely indicators 1 and 3) the r2 and error were similar for the simple linear and log-linear methods. When extrapolated into the future, howÂ¬ ever, the log-linear clearly was the most appropriate method. No one method was superior to the others overall; each was superior in the extrapolation phase for at least one data set. 100 If the summary statistics of r2 and S^r x are not always useful in choosing the best method for the extrapolation phase, then on what basis should the method be chosen? Perhaps this question can only be answered by a combination of strategies: (a) application of several different methods to the data; (b) for each method, extrapolation of trend for specified number of years into the future; (c) graphic representation of origiÂ¬ nal data and predicted data for each method; (d) visual exÂ¬ amination of data to determine most likely direction for obÂ¬ served trend to take based on knowledge of the social phenomÂ¬ enon being studied; and (e) construction of upper and lower limits for the extrapolated trend considered to be the most appropriate. Therefore, while the mathematical extrapolation of trends is a useful approach to forecasting, appropriate application remains an art. Indeed, forecasting accuracy is dependent upon the judgment of the investigator and his or her knowledge of the social phenomenon. Lest the problems presented overshadow the positive attributes of mathematical extrapolation, it should be noted that linear regression is a powerful tool which can be employed both in preliminary analysis of the data and in the projection of future trends. Even if the assumptions are violated, the general linear model may still be employed for analysis of trends; violation of the assumption, however, precludes making any statement of a probabilistic nature (Hays, 1973, p. 636). 101 Practical Considerations Evaluation of a forecast must be based on more than statistical criteria. Other considerations such as purpose of the forecast and the user's need for accuracy are involved (see Harrison, 1976; Martino, 1973a). Thus, final evaluaÂ¬ tion of a forecast must be made in relation to the specific situation and the underlying motivation for the forecasting activity. The results of the mathematical extrapolation are greatly influenced by the data points used in the original regression especially if the sequence of observations is short. Thus, one highly discrepant observation can change the slope of the line. A decision must be made as to whether all data available should be included. The effect of editing data is not always advantageous; in fact, it may obscure the natural periodicity of the data or the occurrence of unexÂ¬ pected events which may be critical to an understanding of the social process. Only experimentation and human judgment will resolve this particular issue. CHAPTER VI SUMMARY, CONCLUSIONS, AND IMPLICATIONS OF STUDY In this chapter the main aspects of the study are summarized. Conclusions derived from an analysis of the results of the study are presented. Future directions for research suggested by the results of the study are presented. Implications for the use in educational planning of an ecological model such as Bronfenbrenner's, time series inÂ¬ dicators, and selected extrapolative methods are discussed. Summary Educational planners and policy makers need adequate information about the societal context of education in order to make appropriate decisions about the future role and function of education. Given this need, the problem in this study was (a) to select, using Bronfenbrenner's ecology of education model, and operationally define at least 10 variables that research has shown to be related to the outÂ¬ comes of education; (b) to use these variables operationally defined as time series indicators in the comparison of three purely extrapolative forecasting methods; and (c) to derive implications for the use of an ecological model such as Bronfenbrenner's, time series indicators, and selected extrapolative methods for educational planning. 102 103 The Variables Ten variables related to the outcomes of education were selected; these variables were then classified according to Bronfenbrenner's ecology of education model. The variables were operationally defined when possible as time series inÂ¬ dicators. Data were collected for these indicators, and eight indicators which met the criteria established in the study were selected for use in the method comparison phase. The Methods The literature in forecasting methodology, economics, statistics, and time series analysis was reviewed in order to determine the most appropriate extrapolative methods to use with the selected time series indicators. Three statisÂ¬ tical methods derived from the general linear model were selected for comparison. These methods included simple linear regression, second and third degree polynomial reÂ¬ gression, and log-linear regression (in which the dependent variable, usually a rate or percentage, undergoes logarithmic transformation). Time in years was used as the independent variable. Each method was applied to each time series indicator. Each data set ivas divided into thirds; two-thirds of the data points were used to establish the prediction equation. This equation was used to predict the remaining third of the data points. The methods were then compared according to (a) the fit of the predicted values to the observed values 104 for the original regression and (b) the fit of the extraÂ¬ polated values to observed values. Both summary statistics and visual inspection of the data were utilized. Results Results of the method comparison were (a) no one method was a superior predictor for all indicators; (b) each method was a superior predictor of at least one indicator; and (c) the summary statistics for the original regression were not consistently related to the accuracy of the extrapolated values. Despite the limitations noted in this study, the Bronfenbrenner model, social variables operationalized as time series indicators, and selected extrapolative methods have significant potential for development and application in educational planning. Conclusions The following conclusions appear to be warranted by the results of this study: 1. The Bronfenbrenner model is a useful framework for considering the numerous factors impinging upon the learner. 2. Time series indicators provide a means to compare trends in an indicator over time or to compare different groups in relation to a specific indicator. 3. The general linear model is appropriate for the analysis and extrapolation of the selected time series inÂ¬ dicators used in the study. 4. Each of the three extrapolative methods is 105 appropriate for use with some indicators but not for others. Measures of "best fit" such as r2 and the standard error of estimate are not reliable criteria for the selection of an extrapolative method. A combination of strategies such as graphic representation of original and predicted data, analysis of residuals, and knowledge of the social phenomena being studied may provide guidance as to the most appropriate method for a particular indicator. Suggestions for Future Research The data from the present study clearly indicate the potential application of (a) the Bronfenbrenner model, (b) selected social variables operationalized as time series indicators, and (c) selected extrapolative techniques in educational planning. However, a number of issues raised in Chapter V remain to be clarified and resolved. Some of the issues such as selection and operational definition of variables are conceptual in nature, while others are methodological. It would be useful to replicate the procedures used in this study with other time series indicators or for other time periods of the indicators used in this study. For example, does it increase the accuracy of the extrapolation to edit the data? And if so, by how much? At what price? Many other questions remain; answers will require coordinated research efforts in a number of disciplines in order to approach the problems from a diverÂ¬ sity of perspectives and with a variety of skills. In Chapter V (p. 100) it was suggested that a 106 combination of strategies might be appropriately utilized to extrapolate trends in time series data. Demonstration of the application of such strategies to actual data would provide valuable assistance to educational planners. The possibility of multimethod forecasting should be explored; that is, extrapolative methods may be combined with an intuitive method such as trend impact analysis to look at the effect of unexpected events on projected trends. The probability of various trends occurring together may be evaluated in a cross impact matrix, or the logical consistency of various trends might be examined in scenarios. Harrison (1976) noted that the problems of social forecasting are the problems of all social science inquiry related to social processes, and the modes for resolving these problems are in many ways the same. (p. 80) Thus, theory-based social forecasting utilizing time series indicators has considerable potential application in educaÂ¬ tional planning; on-going empirical studies are needed, howÂ¬ ever, to actualize this potential. Implications for Planners and Policy Makers The Bronfenbrenner ecology of education model, social indicators, and extrapolative methods can be useful tools for educational planners and policy makers at the national, state, and local levels. Johnson (1975) has noted: From the viewpoint of futures research, social indicators may be regarded as a means for developing more adequate answers to the age-old questions: 107 "What are we, what are we in the process of becoming, and do we like what we see?" (p. 10) Certainly educational planners need to monitor societal processes and interpret change in relation to past, present, and future educational purposes and outcomes. Policy makers need to be informed of the probable or possible consequences of a continuation of observed trends and of possible responsiveness of these trends to different sets of postulated condiÂ¬ tions. (Johnson, 1975, p. 10) Thus, social forecasting strategies such as those described in this study should be a part of on-going educaÂ¬ tional planning efforts. Data on social indicators might be incorporated into existing management information systems. While initiation of such a data bank would require a subÂ¬ stantial investment of resources, maintenance would be minimal. Forecasts could then be issued and revised at regular intervals. Certainly the use of social forecasting strategies can assist educational planners and policy makers in developing an understanding of the past, perÂ¬ spective in analyzing the present, and hopefully vision in planning for the future. XIQNdddV Table 13 Indicator 1: ANOVA Summary Tables by Method Source of Variation SS df MS F Linear Regression Residual 15176369.41422 1 398650.11520 15 15176369.41422 26576.67435 571.04** Quadratic Regression Residual 15190416.01187 2 384603.51754 14 7595208.00593 27471.67982 276.47** Cubic Regression Residual 15284842.00799 3 290177.52142 13 5094947.33600 22321.34780 228.25** Regression Residual Log-linear .06541 1 .00201 15 .06541 . 00013 488.54** Note. Indicator 1 is median family 1971 constant dollars. income expressed in **Â£<â€¢01 109 110 Table 19 Indicator 2: ANOVA Summary Tables by Method Source of Variation SS df MS F Linear Regression .68388 1 .68388 1.18 Residual 3.48487 6 .58081 Quadratic Regression 3.39409 2 1.69704 10.95* Residual .77466 5 .15493 Cubic Regression 3.72084 3 1.24028 11.08* Residual .44791 4 .11198 Log-linear Regression . 00123 1 .00123 1.22 Residual .00605 6 .00101 Note. Indicator 2 is number of families in the United States headed by women expressed as a percentage of total families. *Â£<.05 Ill Table 20 Indicator 3: ANOVA Summary Tables by Method Source of Variation SS df MS F Regression 195.31547 Linear 1 195.31547 826.05** Residual 3.07378 13 .23644 Regression 195.33311 Quadratic 2 97.66655 383.49** Residual 3.05615 12 .25468 Regression 196.31194 Cubic 3 65.43731 346.51** Residual 2.07732 11 .18885 Regression Log -1inear .03848 1 . 03848 665.69** Residual .00075 13 .00006 Note. Indicator 3 is number of wives in the labor force expressed as a percentage of total wives in the United States. **Â£<.01 112 Table 21 Indicator 4: ANOVA Summary Tables by Method Source of Variation SS df MS F Linear Regression 30.99500 1 30.99500 7.27* Residual 42.65403 10 4.26540 Quadratic Regression 31.36855 2 15.68428 3.34 Residual 42.28048 9 4.69783 Cubic Regression 66.53315 3 22.17772 24.93** Residual 7.11589 8 .88949 Log -1inear Regression . 04364 1 .04364 7.33* Residual .05953 10 . 00595 Note. Indicator 4 is number of marriages in Florida exÂ¬ pressed as rate per 1,000 population in Florida. *p<.05 **Â£<.01 113 Table 22 Indicator 5: ANOVA Summary Tables by Method Source of Variation SS df MS F Linear Regression 0.50827 1 .50827 .46 Residual 11.14840 10 1.11484 Quadratic Regression 2.61121 2 1.30561 1.30 Residual 9.04545 9 1.00505 Cubic Regression 10.73976 3 3.57992 31.23** Residual .91691 8 .11461 Log-linear Regression .01472 1 . 01472 1.36 Residual .10791 10 . 01079 Note. Indicator 5 is number of dissolutions of marriage in Florida expressed as rate per 1,000 population. **p<.01 114 Table 23 Indicator 6: ANOVA Summary Tables by Method Source of Variation SS df MS F Linear Regression 5.68341 1 5.68341 5.24 Residual 6.50534 6 1.08422 Quadratic Regression 11.29346 2 5.64673 31.54** Residual .89529 5 .17906 Cubic Regression 11.84805 3 3.94935 46.37** Residual . 34070 4 .08518 Log- linear Regression .00340 1 .00340 5.10 Residual . 00400 6 .00067 Note. Indicator 6 is number of resident live births in Florida expressed as rate per 1,000 population. **p<.01 115 Table 24 Indicator 7: ANOVA Summary Tables by Method Source of Variation SS df MS F Linear Regression 160.48592 1 160.48592 1358.03** Residual .70905 6 .11818 Quad ratic Regression 160.50520 2 80.25260 581.74** Residual .68977 5 .13795 Cubic Regression 160.53081 3 53.51027 322.27** Residual .66416 4 .16604 Log- 1inear Regression .02975 1 .02975 893.91** Residual .00020 6 .00003 Note. Indicator 7 is number of 3 to 5 year olds enrolled in nursery school and kindergarten expressed as percentage of total children 3 to 5 years old in the United States. **p<.01 116 Table 25 Indicator 8: ANOVA Summary Tables by Method Source of Variation SS df MS F Linear Regression 19.34629 1 19.34629 152.33** Residual 1.65105 13 .12700 Quadratic Regression 20.45183 2 10.22592 224.95** Residual . 54550 12 .04546 Cubic Regression 20.53993 3 6.84664 164.65** Residual .45740 11 . 04158 Log -1inear Regression . 06146 1 . 06146 175.29** Residual .00456 13 . 00035 Note. Indicator 8 is number of children involved in divorce or annulment expressed as rate per 1,000 children under 18 years old in the United States. **Â£<.01 REFERENCE NOTES 1. Coates, J. F. Trends and issues influencing elementary and secondary education (Working paper prepared for the U.S. Congress). Washington, D.C.: Office of Technology Assessment, January 5, 1977. 2. Cooper, J. P. Econometric software package (Computer program manual)â€ Chicago: University of Chicago School of Business, May 1973. 3. Stover, J. Trend impact analysis program manual. Glastonbury, Conn.: The Futures Group, November 1975. 4. Bezold, C. Strategic policy assessment: Activities and prospects (Paper prepared for the Commission on Critical Choices for Americans). Center for Governmental ResponÂ¬ sibility, 1974. 5. Shostak, A. B. The Mondale blueprint, the state, and the applied social sciences. Paper presented at the meeting of the Eastern Sociological Society, New York, March 1977. 6. Collazo, A., Lewis, A., Â§ Thomas, W. Social indicators for assessing educational achievement of Florida citizens (Technical Report No. T~, DOE Project on Social Indicators) . Gainesville, Fla.: University of Florida, December 19, 1975. 7. Impara, J. C. Are indicators the yellow brick road? Paper presented at a joint session of the American Edu- cational Research Association and the National Council on Measurement in Education, New York, April 1977. 8. Hall, M. J. (Moderator and discussant). Social and eduÂ¬ cational indicators: Tool of the future or fallacy of the pasT? Remarks presented at a joint session of the American Educational Research Association and the National Council on Measurement in Education, New York, April 1977. 9. Collazo, A., Lewis, A., Â§ Thomas, W. Components of a model for forecasting future status for selected social indicators (Technical Report No. DOE Project on Social Indicators). Gainesville, Fla.: University of Florida, March 26, 1976. 117 REFERENCES Advisory Committee on Child Development. Toward a national policy for children and families. Washington, D.C.: National Academy of Sciences, T97 6. Allardt, E. A frame of reference for selecting social indiÂ¬ cators . Helsinki: Societas Scientiarum Fennica, 1971. Andrews, F. M., 8 Withey, S. B. Social indicators of wellÂ¬ being - -Americans ' perceptions of life quality. New York: Plenum Press, 1976. Anscombe, F. J. Graphs in statistical analysis. The American Statistician, 1973 , 21_, 17-21. Anscombe, F. J., Â§ Tukey, J. W. The examination and analysis of residuals. Technometrics, 1963, _5, 141-160. Bane, M. J. Children, divorce, and welfare. The Wilson Quarterly, 1977, 1(2), 72-75. Bauer, R. A. (Ed.). Social indicators. Cambridge, Mass.: M.I.T. Press, 1966. Boocock, S. S. Students, schools, and educational policy: A sociological view' New York: Aspen Institute for Humanistic Studies, Program on Education for a Changing Society, 1976. Box, G. E. R., ^ Jenkins, G. M. Time series analysis: ForeÂ¬ casting and control. San Francisco, Calif.: Holden-Day, Inc., 1970. Brim, 0. G. Macro-structural influences on child development and the need for childhood social indicators. American Journal of Orthopsychiatry, 1975 , 4_5, 516-524. Bronfenbrenner, U. The experimental ecology of education. Educational Researcher, 1976, ,5(9), 5-15. Brooks, R. M. Social planning and societal monitoring. In L. D. Wilcox, R. M. Brooks, G. M. Beal, $ G. E. Klongian, Social indicators and societal monitoring: An annotated biblio graphy' San Francisco: Jossey-Bass, 1972. 118 119 Brown, D. J. Educational trend analysis methods. In S. Hen- clcy 8 J. Yates (Eds.), Futurism in education. Berkeley, Calif.: McCutchan, 1974. Campbell, A., Converse, P. E., $ Rogers, W. L. The quality of American life--Perceptions, evaluations, and satisÂ¬ factions'^ New York: Russell Sage Foundation, 1976. Carlisle, E. The conceptual structure of social indicators. In A. Shonfield and S. Shaw (Eds.), Social indicators and social policy. London: Heinemann Educational Books, 1972 . Clemmer, R., Fairbanks, D., Hall, M., Impara, J., Â§ Nelson, C. Indicators and statewide assessment. Denver: Cooperative Accountability Project, March 1974. (ERIC Document ReÂ¬ production Service No. ED 091 869) Collazo, A., Lewis, A., 8 Thomas, W. Forecasting future trends in education. Educational Leadership, 1977, 34, 298, 300-302, 304. de Neufville, J. I. Social indicators and public policy. Amsterdam: Elsevier Scientific Publishing Co., 1975. Draper, N. R., $ Smith, H. Applied regression analysis. New York: Wiley, 1966. Durbin, J., f, Watson, G. S. Testing for serial correlation in least squares regression, I. Biometrica, 1950, 37_, 409-428. Durbin, J. , f, Watson, G. S. Testing for serial correlation in least squares regression, II. Biometrica, 1951, 38, 159- 178. Etzioni, A., 8 Lehman, E. W. Some dangers in "valid" social measurement. In B. Gross (Ed.), Social intelligence for America's future: Explorations in societal problems. Boston: Allyn and Bacon, Inc., 1969. Executive Office of the President: Office of Management 5 Budget. Social indicators, 1973. Washington, D.C.: U.S. Government Printing Office, 1973. Ferriss, A. L. Indicators of change in the American family. New York: Russell Sage Foundation, 1970. Folk, M. A critique of some futures prediction techniques and their implications for educational planners. Edu- cational Planning, 1976 , 2^(3), 35-52. 120 Gooler, D. D. The development and use of educational indicaÂ¬ tors. In Educational indicators: Monitoring the state of education (Proceedings of the 1975 ETS Invitational Conference). Princeton, N.J.: Educational Testing SerÂ¬ vice, 1976. Gordon, C. IV. (Ed.). Uses of the sociology of education (NSSE Yearbook). Chicago, Ill.: National Society for the Study of Education, 1974. Grady, M. J. Using educational indicators for program accountability. Denver: Cooperative Accountability Project, September 1974. (ERIC Document Reproduction Service No. ED 096 740) Greenwald, W. I. Statistics for economics. Columbus, Ohio: Charles E. Merrill Books, Inc., 1963. Harman, W. W. The societal context of U.S. public education in the next quarter century. Educational Planning, 1976, 2(3), 5-19. Harrison, D. P. Social forecasting methodology: Suggestions for research" New York: Russell Sage Foundation, 1976. Hays, W. L. Statistics for the social sciences. New York: Holt, Rinehart and Winston, Inc., 1973. Hencley, S., Yates, J. (Eds.). Futurism in education. Berkeley, Calif.: McCutchan, 1974. Henry, N. B. (Ed.). Social forces influencing American eduÂ¬ cation (NSSE Yearbook). Chicago, Ill.: National Society for the Study of Education, 1961. Johnson, D. F. National social indicator reports: CompariÂ¬ sons and prospects. World Future Society Bulletin, July-August 1975, pp. 8-15. Kerlinger, F. N., 5 Pedhazur, E. J. Multiple regression in behavioral research. New York: Holt, Rinehart and Winston , Inc., 1973. Land, K. C. On the definition of social indicators. American Sociologist, 1971, 6^, 322-325. Little, D. L. Social indicators and public policy: Some unanswered questions. Futures, February 1975, pp. 41- 51. Liu, B. C. Quality of life indicators in U.S. metropolitan areas--A statistical analysis. New York: Praeger Publishers, 1976. 121 Makridakis, S., Hodgsdon, A., Â§ Wheelwright, S. C. An interÂ¬ active forecasting system. The American Statistician, 1974, 28, 153-158. Marien, M., $ Ziegler, W. L. (Eds.). The potential of educational futures. Belmont, Calif.: Wadsworth Publishing Company, Inc., 1972. Martino, J. P. Evaluating forecast validity. In J. R. Bright fT M. E. F. Schoeman (Eds.), A guide to practical tech- nological forecasting. Englewood Cliffs, N.J.: Prentice- lla'll', Inc.', 1973. (IJ Martino, J. P. Trend extrapolation. In J. R. Bright 8 M. E. F. Schoeman (Eds.), A guide to practical technological forecasting. Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1973. (b) Martino, J. P. Survey of forecasting methods--Part I. World Future Society Bulletin, Nov.-Dec. 1976, pp. 3-10. Mayes, A. C., Â§ Mayes, D. G. Introductory economic statistics. New York: John Wiley and Sons, 1976." Mendenhall, W., Ott, L., 8 Larson, R. F. Statistics: A tool for the social sciences. Belmont, Calif.: Wadsworth Publishing Co., 1974. Mendenhall, W., 8 Reinmuth, J. E. Statistics for management and economics. Belmont, Calif.: Wadsworth Publishing Co., Inc., 1971. Morrison, P. A. The demographic context of educational policy planning. New York: Aspen Institute for Humanistic Studies, Program on Education for a Changing Society, 1976. Neiswanger, W. A. Elementary statistical methods applied to business and economics. New York: The Macmillan Co., 1956. Nie, N. H., Hull, C. H., Jenkins, J. G., Steinbrenner, K., Â§ Bent, D. H. SPSS: Statistical package for the social sciences. New York: McGraw-Hill, Inc., 1975. Organization for Economic Indicator Programme, most OECD countries. Co-operation and Development, Social List of social concerns common to Paris: OECD, 1973. Plessas, D. J., 5 Fein, R. An evaluation of social indicators. AIP Journal, Jan. 1972, pp. 43-51. President's Research Committee on Social Trends. Recent social trends in the United States. New York: McGraw-Hi11 Book Company, Inc. , 1933. 122 Pulliam, J. D. , f, Bowman, J. R. Educational futurism--In pursuance of survival. Norman, Okla.: University oT Oklahoma Press, 1974. Runyon, R. P., Â§ Haber, A. Fundamentals of behavioral statistics. Reading, Mass.: Addison-Wesley Publishing Co., 1967. Seidman, D. On choosing between linear and log-linear models. Journal of Politics, 1976 , 3J3, 461-466. Shane, H. G. The educational significance of the future. Bloomington, Ind.: Phi Delta Kappa, Inc., 1973. Sheldon, E. B., Â§ Freeman, H. E. Notes on social indicators: Promises and potential. Policy Sciences, 1970, 1, 97-111. Sheldon, E. B., 5 Land, K. C. Social reporting for the 1970's --A review and programmatic statement. Policy Sciences, 1972, 3, 137-151. Sheldon, E. B., 8 Moore, W. E. (Eds.). Indicators of social change: Concepts and measurements. New York: Russell Sage Foundation, 1968. Sheldon, E. B., 8 Parke, R. Social indicators. Science, 1975 , 188, 693-699. Sismondo, S. Social indicators for policy research and democratic action: A paradigm and some examples. Ottawa, Ontario: Department of Regional Economic Expansion, August 1973. (ERIC Document Reproduction Service No. ED 082 908) Snedecor, G. W. Statistical methods applied to experiments in agriculture and biology"! Ames, Iowa: Iowa State University Press, 1956. Social measurement: Points of view of sociologists, business- men, political scientists, government officials, economists, CPAs. New York: American Institute of Certified Public Accountants, Inc., 1972. Theil, II. Applied economic forecasting. Chicago: Rand McÂ¬ Nally and Co., 1966. Thomas, T. C., 8 Larson, M. A. Educational indicators and educational policy. New York: Aspen Institute for Humanistic Studies, Program on Education for a Changing Society, 1976. 123 Thompson, R. B. (Ed.). Florida statistical abstract 1976. Gainesville, Fla.: The University Presses of Florida, 1976. (a) Thompson, R. B. Measuring the quality of life, Part I. Business and Economic Dimensions, 1976, 12(6), 6-14. m Thomp s on, R. B. Measuring the quality of life, Part II. Business and Economic Dimensions, 1977 , 1/(1), 6-19. Tufte, E. R. Data analysis for politics and policy. EngleÂ¬ wood Cliffs, N.J.: Prentice-Hall, Inc., 1974/ Tuttle, A. M. Elementary business and economic statistics. New York: McGraw-Hill, 1957. U.S. Department of Health, Education, Â§ Welfare. Toward a social report. Ann Arbor, Mich.: The University of Michigan Press, 1970. U.S. Department of Health, Education, 5 Welfare, National Center for Education Statistics. The condition of education (1977 ed.) (Vol. 3, Part 1). Washington: U.S. Government Printing Office, 1977. (a) U.S. Department of Health, Education, Â§ Welfare, National Center for Education Statistics. Digest of education statistics (1976 ed.). Washington! U.S. Government Printing Office, 1977. (b) U.S. Department of Health, Education, Â§ Welfare, National Center for Education Statistics. Projections of education statistics to 1985-86. Washington: U.S. Government Printing Office, 1977. (c) Webster, M. M. Guest editorial. Educational Planning, 1976, 2(3), 1-4. Wheelwright, S. C., 5 Makridakis, S. Forecasting methods for management. New York: John Wiley and Sons, Inc., 1977. Wilcox, L. D., Brooks, R. M., Beal, G. M., Klongian, G. E. Social indicators and societal monitoring: An annotated bibliography. San Francisco: Jossey-Bass, 1972. World Future Society. The future: A guide to information sources. Washington, D.C.: World Future Society, 1977. Ziegler, W. L. The Marien W. L. tional futures Company, Inc., potential of educational futures. In M Ziegler (Eds.), The potential of educa- Belmont, CalifT! Wadsworth Publishing 1972. BIOGRAPHICAL SKETCH Jane Counihan Nelson was born in Forest Hills, New York,on December 16, 1941. She attended parochial schools in St. Petersburg, Florida,and graduated from St. Paul's High School in 1960. Her undergraduate work was completed at Loyola UniÂ¬ versity in New Orleans, the Junior College of Broward County, Florida, and Barry College in Miami, Florida. She graduated magna cum laude from Barry College in June 1965 with a major in English and minors in secondary education and journalism. She received her M.Ed. in special education from the UniÂ¬ versity of Washington, Seattle,in June 1969. She has been employed as a senior high school English teacher and federal project writer in the New Orleans Public Schools; also, as a special education teacher and administrator in the Alachua County, Florida Public Schools. She is married to Edward W. Nelson, Jr., M.D. Her scholastic and professional affiliations include Phi Theta Kappa, Delta Epsilon Sigma, Kappa Gamma Pi, and Lambda Sigma, honorary societies; Lambda Iota Tau, interÂ¬ national literary society; Phi Delta Kappa; Council for Exceptional Children; American Educational Research AssoÂ¬ ciation; and National Council for Measurement in Education. I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. I certify that I have read this study and that/ in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Phi flip A.yGlark â€˜ Associate Professor of Education I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. \ Associate Professor of Education This dissertation was submitted to the Graduate Faculty of the Department of Educational Administration in the College of Education and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December 1977 Dean", Graduate School UNIVERSITY OF FLORIDA 3 1262 08555 2833 xml version 1.0 encoding UTF-8 REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd INGEST IEID E5IIX6TT7_PQ52FX INGEST_TIME 2014-07-23T23:01:37Z PACKAGE AA00022214_00001 AGREEMENT_INFO ACCOUNT UF PROJECT UFDC FILES |