—— m *‘V,_ , 2‘“. * REGIONAL VARIATIONS IN THE STRUCTURING OF EDUCATIONAL AND MIGRATIONAL PLANS OF RURAL YOUTH Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY THOMAS A. LYSON 1976 I . IIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 3 1293 10490 3541 This is to certify that the thesis entitled REGIONAL VARIATIONS IN THE STRUCTURING OF EDUCATIONAL AND MIGRATIONAL PLANS 0F RURAL YOUTH presented by THOMAS A . LYSON has been accepted towards fulfillment of the requirements for Ph . D. degree in mm. 947mg k ~M’V‘D/Wficz’fi Major professor Date May 17, 1976 0-7639 ." gr“ 9. . i— I. J ’Nr' , ‘3‘ I. a, ;_v VV.‘ : 57‘ "9 ‘ r - ' ’3‘ \c 9? . . II A M" " s. - .l‘ x n! ~ “I l} g j] V ‘ 5:} 1' . A i , fl bk ; Ty/i “ 411* , ggcg; Til ,.-. ’j .2 L4 ’ I A T/ ‘ 7%‘7 u, H". 1’ I. 275 - f . . ‘ ‘// I \ I ' J ‘ ' . . ’3 . .3. -: .' - "I, W' fhm‘ {f- 64 -- - ""x - o ‘. _.~ , \ I (/ ‘ o . o . .‘xr / ; ,' ‘ 7N}?! ‘ 2 «001' .. . MAR 0 5 2002 .. MI». . - ,_ m JUN,242000 :2: 7' Nov 2 4 2008 ABSTRACT REGIONAL VARIATIONS IN THE STRUCTURING OF EDUCATIONAL AND MIGRATIONAL PLANS OF RURAL YOUTH By Thomas A. Lyson This study explores the influence of regional socioeconomic circumstances on the manner by which young people are sorted-out for career roles in the larger society. Our thesis holds that the nature and range of occupational opportunities in a region (i.e., the socioeconomic character of a region) affects the patterning of individual career development and, therefore, should be taken into account in the specification and elaboration of relevant theoretical formulations and exploratory hypotheses. Regional "opportunity structures," for example, as unique configurations of structural circumstances, may combine with or reinforce the effects of such career determinants as social class background, achieved scholastic record in school and sex. Region is viewed as a unique causal universe in which sets of variables may assume distinctive patterns of interrelationship. Data for this research were obtained through self administered questionaires from 2987 high school seniors in five areas of Kentucky, West Virginia, and Michigan. In this research, educational plans (plan to attend college) and propensity to migrate (a move out of the home region) are viewed as dependent variables. These two variables, representing individual components of the broader term "career plans," are treated both as independent indicants and as forming an interactive dependent variable Thomas A. Lyson typology. Social class origin and grade rank are introduced indiv- idually and jointly as "predictors" of career plans. They are viewed as independent and control variables in the statistical design. Rural/town setting (local community context) is employed as a contextual control and, where appropriate, introduced into the elabor- ation and prediction procedures. Regional context is treated as a supra-control variable and is not introduced directly into the statistical procedures. Finally, sex is also viewed as a control variable. Career planning models are positted for both boys and girls. To empirically test the effects of variations in regional and community context on the career planning model, percentage differences observed in contingency tables are employed. A method of non-linear causal analysis, based on Yule's Q, is used to ascertain the independent and interrelated effects, for boys and girls, of social class origin, local residence place, and grade rank on career plans. Empirical findings from this research clearly support the guiding hypothesis. We found that the relative importance and predictive power of social class and/or grades and/or local residence place in a career planning model is contingent upon the distribution of opportunities and the balance between educational and work opportunities within a region. Although correlation coefficients often mask important relative inter-regional percentage differences, in general, regional effects are noted in zero order as well as multiple control conditions. REGIONAL VARIATIONS IN THE STRUCTURING OF EDUCATIONAL AND MIGRATIONAL PLANS OF RURAL YOUTH By A <‘- l ’3‘" Thomas A: Lyson A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Sociology 1976 ACKNOWLEDGMENTS I wish to thank Professor Harry K. Schwarzweller, whose patience, guidance and humor made writing this dissertation a very enjoyable and rewarding experience. I wish also to thank the other members of my committee, Professor J. Allan Beegle, Professor Jay R. Artis and Professor Denton Morrison for their advice and criticism of the final draft of this dissertation. Special thanks goes to Loretta Carrillo, Lucy Hernandez, Joe Marshman, Brian McAteer, and Judy Roach for helping relieve the tedium that is inherent in a project of this sort. Finally, I wish to extend a very special thanks to Helen Lyson, my mother, without whose encouragement and support this dissertation may never have come about. ii TABLE OF CONTENTS Chapter Eage- LIST OF TABLES .................................................. vi I. THE RESEARCH PROBLEM ...................................... I INTRODUCTION ............................................ l LITERATURE OVERVIEW ..................................... 6 Social Class .......................................... 6 Grade rank ............................................ 7 Sex ................................................... 8 Regional/Community Context ............................ 9 REGION AND REGIONALISM .................................. ll Anthropology: The Search for Cultural Areas .......... l2 Physical and Cultural Areas in Geography .............. 13 The Politcal Science Perspective ...................... l4 Regional Economics .................................... l4 Regional Science: A Synthesis of Perspectives ........ l5 REGIONAL APPROACH IN SOCIOLOGY .......................... 15 The Problem of Delineating Homogeneous Regions ........ l8 Intra-Regional Analysis ............................... 20 Comparative Study of Regional Social Systems .......... 24 II. RESEARCH METHODS AND PROCEDURES ........................... 30 DESCRIPTION OF REGIONAL SETTINGS ........................ 3O Appalachian Non-coal Area ............................. 32 Appalachian Coal Area ................................. 36 m Chapter ' Page The Commercial Farming Region of Central Kentucky ..... 37 The Rural Industrialized Western Kentucky Region ...... 38 The Copper/Pulp Area of Upper Michigan: Ontonagon County .............................................. 39 STUDY POPULATIONS ....................................... 4O STUDY VARIABLES ......................................... 42 Career Plans .......................................... 42 Social Class Origin ................................... 43 Grade Rank ............................................ 44 Community Context/Rural-Town Setting .................. 44 STATISTICAL PROCEDURES AND ANALYTICAL STRATEGY .......... 45 III. ANALYSIS OF FINDINGS ...................................... 48 BASIC INTERRELATIONSHIPS ................................ 48 College Plan and Migrational Plan ..................... 48 Father's Occupational Status and Career Plans ......... 52 Grade Rank and Educational Plan ....................... 56 Local Residence Setting and Career Plans .............. 59 THREE AND FOUR VARIABLE ANALYSIS BY REGION AND SEX ...... 64 Model l: Father's Occupational Status -- Grade Rank -- College Plan ........................................ 64 Model 2: Father's Occupational Status -- Residence Place -- College Plan ............................... 70 Model 3: Father's Occupational Status -- Residence Place -- Migrational Plan ........................... 76 Model 4: Father's Occupational Status -- Grade Rank -- Residence Place -- College Plan ..................... 81 An Intra-Appalachian Note ............................. 88 IV. SUMMARY AND CONCLUSIONS ................................... 93 SUMMARY OF FINDINGS ..................................... 93 iv Chapter Page GENERAL CONCLUSIONS ..................................... 97 LIMITATIONS OF STUDY AND SUGGESTIONS FOR FUTURE RESEARCH 105 THEORETICAL IMPLICATIONS ................................ 107 PRACTICAL IMPLICATIONS .................................. 109 BIBLIOGRAPHY .................................................... 113 APPENDIX ........................................................ 119 Table Table Table Table Table Table Table Table Table Table Table 10. 11. LIST OF TABLES Selected social and economic characteristics of the study regions. Percent planning college and percent planning to migrate in each study region. Regional study populations compared. Percent going on to college and percent planning to migrate; individually and combined by sex and region of residence. college and percent planning to occupational status, sex, and Percent going on to migrate by father's region of residence. Percent going on to college by grade rank, sex, and region of residence. Percent going on to college and percent planning to migrate by local residence place, sex, and region of residence. Three variable non-linear causal analysis: correlations (Yule's Q) among father's occupational status, grade rank, and college plan by sex and region of residence. Percent going on to college by father's occupational status, grade rank, sex, and region of residence. Three variable non-linear causal analysis: correlations (Yule's Q) among father's occupational status, local residence place, and college plan by sex and region of residence. Percent going on to college by father's occupational status, local residence place, sex, and region of residence. vi 49,50 53,54 58 60,61 66,67 68,69 71,72 73 Table Table Table Table Table Table 12. 13. 14. 15. 16. 17. Page Three variable non-linear causal analysis: correlations (Yule's O) among father's occupational status, local residence place, and migrational plan by sex and region of residence. 77,78 Percent planning to migrate by father's occupational status, local residence place, sex, and region of residence. 79 Four variable non-linear causal analysis: correlations (Yule's Q) among father's occupational status, grade rank, local residence place, and educational plan by sex and region of residence. 83-85 Percent going on to college by father's occupational status, local residence place, grade rank, sex, and region of residence. 86,87 "Lorenz Curve" -- educational "chances" for boys and girls by region and certain specified control conditions. 100 "Lorenz Curve" -- migration "chances" for boys and girls by region and certain specified control conditions. 104 vii CHAPTER I THE RESEARCH PROBLEM INTRODUCTION This study explores the influence of regional socioeconomic circumstances on the manner by which young people are sorted-out for career roles in the larger society. Our thesis holds that the nature and range of occupational opportunities in a region (i.e., the socio- economic character of a region affects the patterning of individual career development and, therefore, should be taken into account in the specification and elaboration of relevant theoretical formulations and exploratory hypotheses. Regional "opportunity structures", for example, as unique configurations of structural circumstances, may combine with or reinforce the effects of such career determinants as social class background, achieved scholastic record in school and sex. The career path a young person pursues and the eventual status he or she attains are dependent, in part at least, on numerous earlier decisions. Perhaps the most critical of these decisions occurs at the point of planning for additional training after high school. If, for whatever reasons, a young person chooses not to pursue further formal education, he or she is, in effect, limiting his or her chances to eventually attain more prestigeful and more rewarding status roles in the world of work. All professional positions and many white collar jobs require that a person have had advanced training, usually at the college or university level. An additional factor to consider in the career development sequence is the availability and accessability of employment l 2 opportunities. To seek jobs that will match their level of skills, ambitions, abilities, and drives youth living in economically depressed or relatively isolated rural regions must consider out-migration, regardless of their educational credentials or feelings of attachment to family, home community, and/or region. Clearly, then, two of the more important career decisions facing a young person after high school concernihe pursuit of further academic training and whether or not to leave the immediate area in search of work. these two dimensions (as conceptualized here) combine to form an interactive typology that delimits the range and scope of a youth's life chances and eventual career options. In this research the interrelationship between educational and migrational plans will be viewed as the dependent variable cluster. Sociological attempts to explain educational and occupational career choosing patterns of rural youth have genrally centered on factors such as academic performance in school, 1.0., sex, and family background variables such as occupation of the father, education of the parents, etc. In most studies these variables consistently account for the majority of variance in educational and occupational plans. Increasingly, however, broader, macro-level social, economic and even demographic contexts have been elevated to the level of problematic or conditional variables in an attempt to better refine our predictions and explanations of career planning. Lack of positive reinforcements, favorable circumstnaces, and stimulating opportunities can serve to dampen a young person's career ambitions, his subsequent career mobility and consequently, may set him on a course towards a work career and life style that is characterized by social estrangement and lack of hope for the future. 3 A simple predictive model often employed to help explain an interpret the nature of educational tracking and the structuring of ambition in rural areas combine three of the above mentioned factors — social class, grades, and sex. It has been repeatedly shown that these factors tend to structure a youth's life chances and as such can be conceived as indicative or predictive of a young person's relative opportunities for upward career mobility. Thus, for example, an upper class boy whose grades in school are excellent would be more likely than an academically unqualified girl from working class origins to advance in the career hierarchy. However, as was also noted above, regional context plays an important part in the educational sorting process in rural areas. The fact that past studies in the United States using similar prob- lematic variables (e.g., social class, grades, sex) but carried out in different regional settings, have failed to produce consistently similar results suggest that universal generalizations, at least at the societal level (i.e., national) are lacking. Consequently the creation of valid and useful theories of educational sorting and the generalizability of sociological knowledge is dependent upon specifying those conditions which account for variations in the career planning model. In order to utilize "region" as an analytical tool, useful in generating significant insights and generalizable knowledge about the process of career development among rural youth, we must specify those regionally organized situational circumstances that may affect var- iables in the predictive model. A classification on the basis of regional "opportunity structures" is a useful concept here. As noted earlier, regional situations can be arranged in terms of the nature 4 and range of occupational choices and possibilities that a young person believes are available and accessable to him in his home region. The greater the number of job opportunities and the wider the range of potential job chocies, the better the chance a young person has to find suitable employment in the home region and, in general, the less likely he will be to migrate. If, however, opportunities for advance- ment are limited in a region and/or the range of occupational choices is limited, young people living in that region must decide whether or not to cope with the situation as it exists and perhaps take a job that does not match their desires and aspirations or to seek alternative means (e.g., migrate) to pursue career goals. Furthermore, as noted above, the eventual occupational status a youth attains is dependent in large measure on his plans for addition- al training after high school. College training, then, can be seen as either a means for enhancing one's chances of finding a "good" job in the home region or as a preparatory "internship" before seeking employment outside the home area. In either case, the uniquely patterned opportunity structures and the balance between educational and work opportunities in a region undoubtedly influence decisions on the local level and subsequently structure the nature and range of career options. These regional modalities specify the social context or set of parameters through which and by which educational and occupational plans are shaped, molded and eventually crystallized. The above thinking is consistent with research reported by Lazarsfeld in his classic study of German and Austrian youth. Trans- lating and summarizing Lazarfeld's findings, Lipset and Bendix make the following observations: 5 In reanalyzing the occupational choices of school youth in a number of German and Austrian cities, Paul Lazarfeld reported that "local variations in occupational choices are parallel to differences in the economic structure." Thus, the larger the proportion of persons working in a particular kind of job in a city, the greater the number of l4-year-old school youth who desire to go into that occupation. Lazarfeld interpretted this finding as follows: "The nature of occupational choice is not determined primarily as an individual decision, but rather is a result of external influences. For the occupational im- ‘ pressions offered by daily life are proportional to the actual occupational distribution. The greater the number of metal workers, the more frequently will young people hear about the occupation, and the greater will they be stimulated to choose it." (Lipset and Bendix 1960: 22l) Although the social structural characteristics of a region can be seen as having direct bearing on the career plans of rural youth, the true importance of regional context is recognizd only in terms of the influence it exerts on the manner by which the determinants of plans, namely, social class origin, grade rank, and sex "come together" to explain the nature of career tracking. In other words, the im- portance of any one or combination of these variables emerges only within a given regional setting. The relative predictive power of social class and/or grades in a causal model is contingent on the opportunity structures and the balance between educational and work opportunities in a region. Thus, even though a region may be relatively depressed and characterized by a limited occupational opportunity structure, its effect on the life chances of an individual is mediated and molded by the immediate pressures (i.e., social conditioning) exerted by the family and school. In the above case, region is viewed as a unique "causal universe" in which sets of variables assume distinctive patterns of interrelationship. Differing social contexts then, especially on the regional level, may lead to different styles of ambition, career goal setting and the manner by which talent is selected to fill job 6 roles in the larger society. In short, we believe that equivalent contextual situations are synonomous with homogeneous career planning models, but, as regional circumstances vary, the career setting model is modified and variant mechanisms emerge to sort young people out for career roles. LITERATURE OVERVIEW The patterning of occupational choices and the structuring of educational and migrational plans in rural areas is and, of course, can be viewed as being determined by a wide variety of social, psycho- logical and even environmental factors. In this research the effects of two variables (grades in school and social class origin) and two interrelated social contexts (community and regional context) on the educational and migrational plans of boys and girls will be explored. Because normative expectations in the patterning of career ambitions differ for boys and girls, sex is employed as a contextual control. Grade point average in high school and social class origin are introduced as possible explanatory variables and will be treated as attributes whose determinisitic importance emerges only within a given social context. Local community context and regional setting, on the other hand, are indicative of particular institutional con- figurations, normative climates and opportunity structures and as such define the social mileau in which career plans are formulated. Social Class The effects of social class have been widely documented (Sewell 1957; Rosen 1956; Empy l965; Schwarzweller l960; Burchinal l962).1 Their results have consistently shown that the higher the social class origin of a student the better the base he has from which to build a satisfying career. Even when other variables known to 7 to be associated with both social class and career plans, such as performance in school and Value orientations are controlled, social class has been found to have an independent effect on educational and occupational plans. A person's social class position, ascribed at birth provides a frame of reference from which career goals emerge. Working class families of America often lack many of the material amenities attributed to upper class standing (e.g., family income, etc.). These monetary aspects provide a youngster with a material resource base necessary for college. More important, perhaps, working class families do not provide strong normative o- motivational support for college. This support, a nonmaterial aspect of class position, includes such things as parental encouragement, favorable attitudes toward education, and, in general, an appreciation of the benefitis education may provide. Grade Rank Grade performance has also been shown to be associated with plan for college. A youngster with "high" grades is more likely to enroll in college than a youngster with "low" grades (Kahl I953; Burchinal, et. al. l963; Sewell and Shah l968). Past grade average can be conceived as an achieved status attained within the school and hence as a "stepping stone" on the path toward higher education and a niche in the work world. However, it should be noted that grade rank is not an ab- solute standard of achievement, as might be argued for 1.0. and various standardized achievement tests. The distribution of "ability" and the criteria used in evaluating a youth's academic performance may vary not only between regions, but also within a particular region. According to Schwarzweller, though: 8 Regardless of the kind or quality of information utilized in establishing a student's scholastic rank, and regardless of the criteria emphasized (creative ability, motivation, aggres- siveness, behavioral conformity, scores on tests), the end product is the formation of a stratification system within the mini-society of a school class. As with any other stratif- ication system in other community contexts, the rankings in- variably become a basis for the differential allocation of rewards and facilities. (Schwarzweller 1974: 3) Thus academic rank is indicative of a developmental process that affects a young person's life chances, perception of attainable goals, and consequently his level of abbition (Schwarzweller 1974). £92: Sex is almost universally used as a "contextual" control variable in studies of career plans. From time of birth each sex tends to be treated somewhat differently regarding what is expected, what is taught, and what opportunities are presented to them by society. Generally, boys have "higher" plans than girls (Kuder 1960; Schwarzweller 1960; Sewell 1966). Not only do the sexes differ in terms of status goals, but also in terms of the "types" of jobs aspired toward. In a study of the Kuder preference record, based on responses of 3,418 boys and 4,466 girls, it was found that more interest in outdoor, scientific, and mechanical pursuits is generally expressed by boys than girls. Greater interest in musical, social service, and clerical pursuits is generally expressed by girls than boys (Kuder 1960). It has also been noted that many occupations which tradition ascribes to women are those practices performed by the "ideal" mother, such as nursing and teaching (Mathers and Drabick 1965; Fleege and Malone 1946). Clearly, then, there are important sex differences in the manner by which career plans are formulated. Normatively sanctioned 9 role expectations serve as important structuring mechanisms in deter- mining the life chances and career patterns of boys and girls. Regional/Community Context As stated in the introduction, the importance of what are generally rearded as deterministic variables (i.e., sex, grades, and social class) is contingent upon the nature of the social context in which they operate. The social context thesis has been well articu- lated in numerous forms (Sewell 1957; Sewell and Armor 1966; Rogoff 1961; Michael 1961; Bogie 1970; Schwarzweller 1973). The method commonly used to isolate contextual or structural effects requires that a relationship between the distribution of a given characteristic in various collectivities or settings and the effect criterion is isolated while this characteristic is held constant for individuals (Blau 1960). Linz (1971) provides the social researcher with a succinct, empirically grounded definition of social context: Persons in apparently the same "objective" situation -- in terms of occupation, skill, income, education, social origin, religion, etc. -- will think differently about their position and react to it differently depending on social context. Context is defined by attributes such as whether those so placed are in a majority or minority in their community, whether they work in large or small plants, and whether they interact or are isolated. (Linz 1971: 107-108) Contextual analysis in career planning studies has often centered around rural-urban differences, community size, and peer group associ- ations. At present, however, the one area of contextual analysis that seems to have aroused the most concern and attention deals with the socioeconomic character of the immediate school or community en- vironment. It is generally assumed that a normative climate associated with the class composition of the local community or school district 10 is influential, over and above, a yputh's actual social class position in determining his level of educational and occupational ambition. Turner's classic study, Social Context of Ambition, found that when high schools in the Los Angeles area were classified according to the average socioeconomic status of the student body, the educational and occupational aspirations of individual students were influenced in the direction of the majority. That is, working class students attending primarily middle class schools were more likely to aspire to college than their counterparts attending primarily working class schools. The reverse process for middle class students was also noted. Even when a person's social class and intelligence were controlled, he found that a residual school effect of .16 for boys and .12 for girls (Turner 1959). The social composition of the school has been shown not only to be related to the level of aspiration, but also indirectly through its influence on grade rank (Nelson 1972). Nelson finds that on a theoretical level school context has a real and important influence on the plans of young people: In context analysis we may be developing a central principle of social stratification -- that individuals may, involuntarily, be ranked at the bottom of an unequal system which, talent and ability notwithstanding, can adversely effect their aspirations and behavior. (Nelson 1972: 147) At least one study indicates that the relationship between local residence and occupational and educational choices may be spurious. Sewell and Armor (1966) found that neighborhood context has little effect on college plans when social class and intelligence are controlled. Using multiple regression techniques, they found that neighborhood status results in an absolute increase of 1.8 percent in the explained variance of college plans beyond the effects ll of sex, socioeconomic status, and intelligence. However, they also noted that the effect of neighborhood context is considerably more important in some populations than in others, a fact not apparent in multiple correlation analysis. Rural-urban contextual differences have also been widely examined (Rogoff 1961; Burchinal 1961; Sewell 1964, 1965; Schwarzweller 1973). With few exceptions, these researchers have found that, re- gardless of the criteria for categorizing communities, youth from small, more rural neighborhoods are less likely to aspire toward high occupational status than youth from urban areas or large towns. The opportunity structure in rural environments tends to be both narrower and more restricted than that found in cities, thus reducing the chances for career advancement. Additionally, most past researches have failed to grapple with the wider socio-cultural contexts within which structural sub- units such as peer groups, school districts, neighborhoods, and communities exist (e.g., the broader socio-regional contexts that are exemplified by such “bounded" areas as Appalachia, the cut-over area of the Great Lakes, etc.).2 Socio-regional context may have important influences on the manner by which goals are formulated and plans implemented. Regional context, a central concern of this exploratory research merits further conceptual clarification. REGION AND REGIONALISM The concepts of “region" and "regionalism" have a rich heritage in many of the social sciences. Geography, anthropology, political science, economics, and sociology. of course, have made frequent use of regional divisions for the purposes of collecting, arranging, 12 analyzing and interpreting various kinds of data about nations and states. Among the social sciences, however, although common threads of meaning exist, there is only limited consensus on the nature and specification of regional entities. This is understandable, for the conceptualization of such entities is heavily dependent upon the problem focus of the discipline and disciplinary needs. Never- theless, for the purposes at hand, it is appropriate and perhaps useful to search out the themes that characterize the various disciplinary approaches.3 Anthropology: The Search for Cultural Areas Central to anthropology, for example, is a concern with "cultural areas". According to Amos Hawley: "... a cultural area ... is essentially a diagrammatic construct rather than an actual portion of the physical landscape occupied by population." (Hawley 1950: Cultural areas, in other words, are conceptualized primarily in terms of human use factors, not physical characteristics. Nevertheless, the effects of physical environment on culture has not been ignored (Wissler 1916; Kroeber 1939). No culture is wholly intelligible with- out reference to non-culture (i.e., environemental factors. Wissler, for instance, states that: Environment does not produce culture, but stabilizes it. Be- cause at many points that culture must be adapted to the environ- ment, the latter tends to hold it fast. Cultures. therefore, incline to change slowly once they have fitted themselves to a setting, an to enter a new environment with difficulty than to spread over the whole of the natural area in which their form was worked out. (Wissler 1916: 6) Many sociologists (Taylor 1949; Mangus 1940; Zimmerman 1947) have employed various forms of the culture area concept to delimit rural regions in the United States. However, these attempts have been severely criticized by Herskovits: 13 Experience has shown that ... (a cultural area) is not adapted to use where the distribution of geographical differences is overridden by stratification resulting from ... larger population aggregates ... Industrial centers do differ from agricultural regions, but all of them in the United States, for instance, have such preponderant cultural similarities that it obscures rather than brings out the significance of differences between them to regard them as of the order of cultural areas. (Herskovits 1960: 409) The use of cultural areas in anthropology has waned in recent years. Increasingly, anthropologists have turned their attention from folk to more complex societies and from a culture area approach to a structural-functional alternative (Steward 1950). Physical and Cultural Areas in Geography The geographer, on the other hand, views region as a "natural area", the primary elements of which are structure and relief, climate and vegetation. Despite the emphasis on natural areas, however, many early geographers betrayed a naive faith that natural regions would also be found to coincide with regions defined according to cultural patterns (Hall 1935). As a result of translations and interpretations of European work prior to World War I, a number of ppmap_or cultural regional studies were produced (Kaplan 1964). Gradually a reapproachment occurred between the pure natural area and the cultural area approaches. Culture came to be seen as the agent while natural area was seen to be the medium of regional geographic studies (Hall 1935). The integration of physical and cultural approaches to region has been an uneasy one in geography. There has not yet been any agreement on the relative weights of factors in the fusion of features that depict human occupance of an area (Whittlesey 1954). The present state of geography with regards to region is best summarized by James: 14 Geography seeks precision; that is, it is concerned more than bordering fields of study with the exact measurement of areas and patterns, and from these more exact studies of area differences, it brings to light additional understanding of the role of rel- ative location. Geography cannot find all of the answers to economic, social political or military problems of a practical nature; but it can contribute to an understanding by applying its own individual kind of analysis. (James 1952: 215) The Politcal Science Perspective The political scientists, likewise, deal with "region" in a manner consistent with their specific subject matter needs. Thus in the political science literature region is very often treated as a politically defined administrative area. The administrative area is any area delineated for the purpose of applying a policy or maintaining a public service. Among the distinctive characteristics of adminstrative areas are their arbitrary determination, their limited purpose use, and the inflexibility of their boundaries (Hawley 1950). Regional Economics In economics the concept of region takes several forms, but the overall emphasis has some unity: It insists that factors independent of the physical environment critically condition spatial configurations and that these factors are largely economic in nature. (Isard 1956: 14) Regional economists tend to center their studies on primary trade areas, metropolitan regions (i.e., a large metropolitan center and surrounding territory), regions of convenience (i.e., economic areas delimited by the census), and economic development regions. In economic development region: . . the focus is upon economic problems and economic opportunites which can be realized, described, and met primarily on an areal basis. The emphasis is on the development of policies, programs and actions to move the region from where it is economically toward pre-determined economic objectives ... sometimes it is an economic problem area ... sometimes it is an economic opportunity area ... (Fisher 1955: w6-7) 15 Of all of the fields concerned with region as a unit of analysis, economics has made the most use of quantitative statistical delineation. Such analytical tools as factor input-output analysis, regional bal- ance of payments, gross regional product, benefit cost analysis, and multiplier effects have been employed or proposed by the regional economist. These measurement devices make economic region delineation exceedingly precise. Regional Science: A Synthesis of Perspectives The culmination of efforts in the social sciences can be seen in the emergence of a "regional science." Walter Isard, a leading proponent of the regional science concept states: Regional science is, primarily, a social science field. It focusses upon spaces and systems of spaces, regions and systems of regions, locations and systems of locations. Distance -- physical, economic, social and political -- are all key concepts. Basic phenomena embrace, among others, transportation and communi- cations, population and commodity flows, and land, water, and other resource use and conservation. Some significant character- istics of concern are population numbers, administrative area structures. Change -- growth and decline -- comprehensive interdependence and linage analysis strongly colors its approaches, both behavioral and mechanistic. (Isard 1956: 13) Although the stated goal of regional science is to synthesize and utilize the theories and methods of many of the social sciences, it unfortunately has a strong economics bias attached. Attempts to quantify regional characteristics often result in a loss of the richer, sometimes more useful considerations of socio-cultural context. The meaning of "region" to the regional scientist is too often couched in "hard science" terms of input-output models, econometric problems, stochastic chains, and the like. REGIONAL APPROACH IN SOCIOLOGY In sociology "region" is often conceived as a geographical area, the inhabitants of which share some common social, economic, 16 political or demographic characteristics, or combination of these, which deistinguish them from residents of other surrounding geographic areas. A somewhat more analytical definition is offered by Howard Odom, the sociologist whose name is intimately associated with regional studies: Region means the composite societal region combining a relatively large degree of homogeneity measure by a relatively large num- ber of indices available for a relatively large number of purposes or classifications. This means it must comprehend both the natural factors and the societal factors which must, of course, include the American states and prevailing historic economic and culture traits. (Odom 1964: 170) Carl Zimmerman also views region as an important aid to understanding the social activities and solving the social problems of the nation. He thus proposes a "regional sociology" as one method of dealing with area context. Regional sociology is an approach to the location of man's values and activities on this globe to the extent that indefinite areas become understandably definite from a combination of geographic factors, working in fairly recognizable similarities of social time and social space. (Zimmerman 1947: 15) The beginnings of significant though on the concept of region in sociology in the U. S. can be traced to several sources. Charles Galpin laid the foundation for region studies in rural areas, while Burgess and Park and others from the Chicago School dealt with area context using a metropolitan frame of reference. As early as 1911, Galpin reporting on his research in upstate New York found: Village and open farming country form a community of homes which seems to be a sort of social drainage basin beyond whose border every home drains off into some other basin. The big discovery of this survey is the fact of a real rural community ... It appears that this community takes in parts of four townships and ignores in its social dealings the voting precincts set by law. (Galpin 1911 in Warren 1963: 23) In these early researches a careful distinction between 17 region and community was not made. Theoretically, both region and community can be viewed as conceptually similar contextual variables. The region, however, is necessarily a larger geographical area en- compassing at least one, and usually more, "communities." Galpin used the term "region" in its narrowest sense, defining one community and its surrounding "region." Today, "region" (e.g., Appalachia) generally refers to a broad geographic area composed of numerous "community/regions." This early interest by Galpin and others in delineating rural community/regions on the basis of process soon gave way to an emphasis on ennumeration of larger homogeneous and quantitatively distinct geographical areas. Even today studies appear in the literature delimiting "types of farming" regions, historical regions, and the like. The scientific validity and usefulness of such studies can be seriously questioned since the extent of homogeneity and the number of regions to be delimited depend solely on the purposes of the investigator. Regional composition and boundaries, if arbitrarily determined, exist only as constructs in the researchers mind and thus cannot be construed as "real" regions or social facts. Alvin Bertrand provides a systemmatic introduction to the concept of region for sociological research purposes. In his definition of regional study he distinguishes between emphasis on region as a means of categorization and as a system of social structures and relations. According to Bertrand: Regional study is a special branch of sociology primarily con- cerned with (l) the delineation of homogeneous regions and sub- regions, (2) the study of all forms of human association within a specific regional environment, and (3) the comparative study of regional social systems. (Bertrand 1958: 105) The importance of region, then, involves several levels of 18 attack developed along an ascending scale of complexity and integra- tion. Regional analysis at the most elementary level calls for the de- lineation of homogeneous single and multiple factor geographic areas. At more elaborate levels of complexity it is concerned with the con- struction of social theories and the exploration of change and process patterns through multiple factor structural-functional regions. Let us now turn our attention to a more thorough examination of each of these three concerns of regional study that Bertrand lists. Our aim will be to examine the nature, meaning, limitations, and socio- logical usefulness of "region'l as it is conceived in different forms. The Problem of Delineating Homogeneous Regions There are many ways, of course, to divide the United States, or any large nation, into meaningful subareas of relatively homogeneity with respect to certain specified dimensions. Depending on the problem under investigation and the resources available, the researcher must at some point decide on the kinds and number of indices he wishes to employ. The least complex strategy in regional analysis is the description of single factor regions. Bertrand notes four types of single factor regions: 1) Natural regions are distinguished from one another on the basis 6fla given geographical characteristic, such as topography, rainfall or type of soil. 2) Cultural regions are set apart on the basis of differences in any material or nonmaterial cultural trait, such as character- istic literature, esthetic expression, architecture, or dress. 3) Service regions are distinguished for administrative purposes. There are innumerable examples of such regions in the fields of government and economics, such as Federal Reserve Bank regions and sales regions. 4) Types-of—farming regions are based on marked differences in agricultural endeavors. (Bertrand 1958: 106) A somewhat more complex delineation of rural regions involves 19 construction of regional configurations or universes based on several variables or characteristics since, except for purposes of specialized study, the multiple factor, composite region has greater theoretical significance and impact than single factor regions. Tow methods are commonly employed to delimit complex regional structures. First, one may attempt to reconstruct the process from which a region has been built up out of its component parts (e.g., farm market regions, service areas, etc.). Second, one may use statistical measurement to delimit areas homogeneous in certain economic, social, and cultural characteristics (Vance 1951: 127). An example of the second method is A. R. Mangus' (1940) delineation of a complete set of rural regions for the United States. He employed such quantitative measures of culture as a plane of living index, percent of farm tenancy, type of non-farm industry, etc. Although his methodology has been criticized on the grounds that he fails to measure distribution of culture within the areas, Mangus contends that quantitative traits or elements, if carefully selected, can be considered valid indicators of more fundamental, unmeasureable aspects of culture. On a more micro-level, delineations and descriptions of areas of socio-cultural homogeneity have been done for numerous states (e.g., Ohio, Louisianna, Missouri). The delineations generally differ from nation or state "type-of—farming" maps used by agricultural economists and state economic areas of the Bureau of Census, since additional factors of a more “social" nature (e.g., school expenditures, racial composition, age indices, etc.) are also considered. 20 Intra-Regional Analysis Delineation of regions by measures of common modes of life (e.g., types-of—farming, etc.) can be advanced beyond the categor- ization stage if factors employed to describe a region are taken to determine structural-functional relationships operating within a geographic area. The logic here is fairly clear. According to Gregory, "... the fabric of modern society is composed of variable traits that are correlated among themselves." This approach to regional research focusses on an interactive socio-cultural complex and homogeneity of region, as contrasted with others, is one of association rather than kind. Obviously, if no socio-cultural trait were related to any other, a region would have no meaning except in terms of that trait (Gregory 1949). In the present mode of analysis, then, regions move from being merely a classificatory scheme to be- coming an analytical scientific tool. The main concern at this level of analysis is simply stated by Vance: Whatever factors are found to fuse together to give us the regional economy, the regional culture, and the region's consensus of opinion -- these, obviously, are the factors which in the be- ginning should be selected for use in drawing boundaries between regions. (Vance 1951: 133) Thus, region is seen as a more or less homogeneous entity, a basic configuration of human life. Region is viewed as a unique causal universe in which sets of variables may assume different patterns of interrelationship and, consequently, form unique structural configurations. Bertrand provides us with a list of forces he con- siders important in weaving the various threads of an area into a unified fabric. The three forces he considers significant are: 21 Geographic and Physiographic Characteristics. The natural environment ... is readily recognized as one of the basic region- making forces. This is true because man's physical setting usually conditions his social adjustment to an observable extent. Thus differences such as in occupation, dress, architecture, and recreation may often be linked to topography, climate, soils, etc.... History. Accidents of history can become the most powerful region-making force operative among a people because a strong consciousness of identity is bred through shared experiences and problems.... Cultural Experiences. It is inevitable that cultural materials that are invented, discovered, or borrowed in a specific area are shared only unevenly with the rest of the national group. This fact can be accounted for in terms of the rate of acceptance of any given cultural trait and its usefulness in various areas. The extent to which an item of culture will be useful varies directly with such factors as the economic structure and geographic conditions. Rates of diffusion of culture are, of course, de- pendent on the values, attitudes, opinions and experiences of the people. Obviously, the amount and kind of cultural material accepted help to determine the socio-cultural characteristics of a people and thus constitute a significant region making force. (Bertrand 1958: 107-108) In additon to the three "forces" listed by Bertrand, the impact of larger structural entitites (i.e., institutions on the national and international level) as region-making forces have been noted by Steward: . . a nation is a great deal more than the sum total of its subcultures. In addition to various subcultural groups based on regional, occupational status, ethnic, and other differences, there are national economic, political, military, and other institutions which influence each local group somewhat differently. Thus, regional subcultures must be understood in relation also to world trade markets, growing industrialization and expanding governmental controls and services. (Steward 1955: 298) Unfortunately, empirical assessment of the exact weight of a given set of forces in developing a particular region is very difficult. Research has indicated that in one instance one type of force will be the most important in regional development; whereas in a second instance another kind of force will direct the growth of a region. The important point, however, is that regions can 22 be viewed as distinct entities manifesting internal organizational differences. To view region as an analytical tool as opposed to a clas- sificatory scheme requires a dynamic rather than static approach. Preoccupation with regional "structure" and the task of defining regional boundaries on a map tends toward the impression that region is a fixed and static thing. An understanding of the structure and functioning of social groups and process becomes the major objective of dynamic regional studies (Vance 1951: 134). Heberle states it another way: The local groups of neighborhood and community; the groupings on the basis of interest; the class structure characteristic of a certain region; migration, its forms, causes and consequences; and dissociation; and finally ... public opinion ... all of these and many other phenomena will constitute essential subjects of inquiry aiming at the understanding of a region as a societal entity. (Heberle 1943: 283) Additionally, Heberle continues, one must examine the "character" of the people of a region in terms of "types." These types are determined by occupation, ethnic origin, religious affiliation, etc., and by the social values adhered to, and the normative conduct resulting therefrom (Heberle 1943: 283). Thus, people in a region are "bound" together by common interests, common traditions and history, and the common socio-cultural structures which have merged from their interests, traditions and history. W. A. Anderson carried regional analysis one step further by integrating areal context into a broader analytical framework. Region is viewed as one of four structural entities that compose society. According to Anderson: "... when individuals interact with each other, they form ... distinctive structural forms that can be recognized and described as such." (Anderson 1947: 350) 23 Anderson called these structural forms -- "hurelure” -- and distinguished the following four classes: 1) social groups, 2) ecological entities, 3) instituitions, and 4) collectivities. Each of these is a distinctive type of structural form produced by relationships between men, and each is a reality in any society and therefore essential to describe it. Briefly: A social group consists of two or more people meeting in the same enviFonment or overcoming distance by some means of communication, who are influencing each other psychologically. An ecological entity (region, community, neighborhood) is a pop- ulation aggregate occupying a contiguous territory integrated through common social and economic activities and so able to and often carrying on many of the functions of a corporate entity. An institution is a definitely patterned, relatively fixed, and socially sanctioned arrangement of relationships between people which at a given time and place seems to be the most apt way of carrying out some basic need of society that has been crystal- lized in the mores. A collectivity consists of a number of people whose behavior is specifically polarized around some temporary center of attraction and stimulus that gives them interaction and unity, which interaction and unity exists only as long as the center of attraction exists. (Anderson 1947: 352) To Anderson, ecological entities function as social contexts in which "social groups" interact. Regions, the largest of the eco- logical entities, are composed of similar, but by no means identical, smaller social contexts (i.e., neighborhoods and communities). Although communities and neighborhoods within a given region may posses somewhat distinctive social configurations and situational circumstances, they are tied together in a regional setting by "... an interwoven economic life that established, in addition to the spatial order, other relationships between people." (Anderson 1947: 353) In conclusion, Anderson states that a complete system of sociology must include: 24 . a description of the structure of the above forms, a con- sideration of how surrounding environmental forces condition their structure and operation, a statement of the principles by which the forms act, a description of the reciprocal relation- ships between the forms, how social controls result from the actions of these forms, how these relationship structures change in time, and how, as a result of their operation, cultures and)persona1ities are produced and changed. (Anderson 1947: 354— 355 Comparative Study of Regional Social Systems The concept of region in sociological research achieves its greatest utility in the comparative study of regional social systems. Research at thi stage embodies the goal to which all social science aspires: formulation of universal propositions and the creation of valid and significant theories of society and human behavior. The purpose and method of cross-regional analysis can be seen as closely analagous to cross-cultural, cross-national or comparative studies. According to Schwarzweller: . . greater attention to cross-regional comparative research and the development of an efficient cross-regional method in the social sciences would not only stimulate the formulation of more universally applicable sociological theory, but would function also to enhance the development of an efficient cross- national comparative method. (Schwarzweller 1968: 15) Cross-regional analysis necessarily involves specialized knowledge of the social organization and linkages between regional social systems. According to Marsh, data relevant for such com- parisons "... are derived from human interaction.“ (Marsh 1967: Employing a framework developed by Parsons, Marsh states that data for comparative analysis: . can be classified under one of the following Systems of Action: social systems, cultural systems, and personality systems. Also relevant are the data referable to other systems with which these three systems mentioned have "boundary" ex- changes: the human organism, demographic and ecological systems, and the physical environment. (Marsh 1967: 16) The above scheme and the nature and types of data it suggests 25 for analysis is very similar to those presented by Heberle, Vance, and others in the previous section. The use of region as a "unit" for comparison in comparative research is a relatively recent phenomena. Anthropologists and sociologists have long been concerned with the problem of defining the socio-ecological boundaries of a social group. It is genrally assumed that the units for comparison (e.g., societies, regions, communities, etc.) should have the same characteristics. This problem of sampling "units" was raised by Murdock (1949) and, in a somewhat different form by Radcliffe-Brown. Radcliffe-Brown writes that sociologists: ...commonly talk of societies as if they were distinguishable, discrete entities as, for example, when we are told that society is an organism. Is the British Empire a society, or a collection of societies? Is a Chinese village a society, or is it merely a fragment of the Republic of China? If we say that our subject is the study and comparison of human societies, we ought to be able to say what are the unit entities with which we are concerned. (Radcliffe-Brown 1940: 4) To show the importance of "region" as a unit entity in comparative research we must make a distinction between intra-societal regions and supra-societal regions. Intra-societal regions refer to distinguishable regional situations -- cultural, economic, social, etc. -- of one kind or another yithip_a particular national society. Intra-societal regional areas may differ markedly in terms of historical development, economy, cultural orientation, politics, and even language. However, these "regions" in no way can be considered separate and distinct societies (Schwarzweller l968). Supra-societal regions, on the other hand, refer to regional situations that transcend societal or national boundaries and are composed of locality groupings and perhaps even whole societies that are similar with respect to selected classificatory criteria 26 (Schwarzweller 1968). Schapera cites as an example of a supra- societal region: . the many hundreds of politically separate Bantu speaking "tribes" in south Africa can be divided on the basis of marriage regulations into three major groups, characterized respectively by comprehensive exogamy, preferential matrilaterial cross-cousin marriage and marria e permissable with any kind of first cousin. (Schapera 1953: 3597 In discussing the importance of supra-societal regional classification, as a unit for comparative research, Schapera argues that by making an intensive study of a given "region" (i.e., supra- societal): . . embracing all the peoples living there about whom information is available ... (and) ... By carefully comparing the forms taken among those peoples by the particular social phenomena with which we are concerned -- as in this instance, kinship and marriage -- we try to establish, by a process of generalization, one or more basic types into which the various forms can be classified ... such an approach ... will enable us in the first place to ensure that all the known "societies" of people inhabiting a given region are discussed, and will therefore do away with the need for sampling or any other arbitrary form of selection. (Schapera 1957: 359 In other words, we gain greater universality and prediction through a process of data reduction on a grand scale (i.e., much the same effect and principle as in the collapsing of categories in survey research) (Schwarzweller 1968). The problem of unit sampling and reduction into regional units becomes more complex when one is dealing with modern industrial societies (e.g., the United States) rather than with village or tribal societies as Schapera does. Within industrialized nations the socio- cultural system of any region or local area cannot be considered autonomous or sovereign. All areas within the geo-politcal boundary of a nation state share a common set of social, legal, and cultural norms. However, one can distinguish certain important regional 27 differences within even the smallest nation-states and certain sim- ilarities, in much the same way as suggested by Schapera. In other words, one can delineate a number of distinct intra-societal regions. According to Schwarzweller: In making comparisons, then, between national societies one must attned to the problems of internal sociocultural differences within such societies. If cross-sectional samples are not drawn, then the scope of generalization must be carefully delineated and specified; and even when random sampling procedures are employed, one often encounters difficulty in specifying the applicability of the results of research to regional circumstances within one or the other nations included in the project (Schwarzweller 1968: 18-19 Additionally, it is important to reemphasize that no region is a completely homogeneous entity. Differences among sub-regional units (e.g., communities and neighborhoods) exist within each area. Thus, to assure the greatest validity and reliability in research findings, not only must cross-regional samples be drawn, but variations within those regions must also be accounted for. Julian Steward cites possible cources for intra-regional heterogeneity: ... a region is (not) a territory of cultural uniformity, a cultural subarea in which all persons share the same behavior patterns. Partially differing ways of life will characterize town and rural populations -- the former related to marketing, distributing and servicing functions and the latter to various roles in the agrarian productive system ... In the highly industrialized nations which have advanced education, transpor- tation, and mass communications, the sub-cultural differences between city and rural dwellers are diminishing. This is not to say that sub-cultural differences are being leveled. If appro- priate studies were made they would probably show that subcultures are increasingly related to occupations. (Steward 1955: 298) In a nutshell, if one theory could be made to fit all conceivable areas and situations, there would not exist the need for comparative theory that regionalism promises at this level. However, until one is sure that all specifying conditions that account 28 for, the creation of useful and valid sociological theories and the generalizability of sociological knowledge is at stake. Thus according to Vance: .. fortunately or not, regionalism in one form or another seems destined to review -- if not to undermine -- all those theories that speculate about some sort of abstract man, abstract culture, or abstract economy. Theories developed in one culture area or in one stage must be tested against those developed about other areas or stages. (Vance 1951: 136) 1. 29 FOOTNOTES CHAPTER 1 For a summary of various studies dealing with social class and other structural determinants of educational mobility see James T. Hornor, James G. Butterbaugh and J. Judith Carefoot, "Factors Relating to Occupational and Educational Decision Making of Rural Youth," Department of Agricultural Education, Report 1, University of Nebraska, Lincoln, Nebraska, April 1967. . Two exceptions are Schwarzweller (1973) and Bogie (1970). Schwarzweller's study dealt with the influence of regional socio-economic circumstances on the educational plans of rural youth from Germany, Norway and the United States. Bogie concerned himself with the educational and occupational plans of rural youth from three areas of Kentucky. . An annotated bibliography that was especially useful in organizing this section was: Paul F. Kaplan, "The Region as a Unit of Analysis: Emerging Consensus," Cornell Journal of Social Relations, 1 (October 1966), pp. 135-54. CHAPTER II RESEARCH METHODS AND PROCEDURES Exploring the career choosing patterns of high school seniors from five rural areas of the United States, this research deals with the interrelated influences of sex, social class origin, scholastic rank, and local and community setting on educational and migrational plans. Chapter II focusses on a detailed description of the regional areas selected for study, the operationalization of the major research variables, data collection techniques and the analytic strategy and statistical design. DESCRIPTION OF REGIONAL SETTINGS Earlier we examined some of the ways by which region is conceptualized and used as a research tool in the social sciences. In sociology, generally speaking, region may be conceived as a geo- graphic area, larger than a community, the inhabitants of which share some common social, economic, political or demographic characteristics, or combination of these, which distinguish them from residents of surrounding geographical areas. As employed here, however, region is not viewed solely as a socially, economically, politcally or demographically homogeneous geographical area, but rather as a discrete, well defined "social system." That is, following the thinking of Heberle (1943), Vance (1951) and others, region is conceptualized as a geographically bounded social system distinguish- able from other social systems in terms of the manner in which sets of career planning variables assume different patterns of inter- relationship and consequently form unique structural/ecological 30 31 configurations. For our purposes, then, a rural county or group of counties sharing similar structural/ecological configurations can be con- sidered indicative or representative of a regional modality or setting. We believe that the uniquely patterned "opportunity structures" and the balance between educational and work opportunites in a region influence decisions on the local level and subsequently structure the nature and range of a young person's career options. Empirical indicants of a region's opportunity structure are, of course, both numerous and varied. Perhaps the most com- prehensive compilation of factors relating to the nature and range educational and occupational opportunities in an area can be found in the census. Selected census material presented in Table l, for example, give some indication of the social and economic real- ities of the "regions" drawn into this study. These data, and material from other sources will be used to describe the situational circum- stances confronting young people at the point where their educational and migrational plans are crystallized. However, a more rigorous and complete definition of a region”s l'opportunity structure" should include, among other things, quantitative data on geographic proximity of higher education facil- ities, school drop-out rates, per-pupil school expenditures, avail— ability of financial aid for college, and types of school curriculums (e.g., vocational. college prep, etc.). In addition, qualitative, ethnographic material relating to local value orientations and normative climates would further add to an understanding of the contextual environment in which career plans are formed. It can be argued, of course, that the greater the number 32 of indices and characteristics used to define "opportunity structures," the less important "region" becomes as an explanatory variable. Thus, according to Prezworski and Teune: The basic assumption is that names of nations, or of social systems in general, are treated as residua of variables that influence the phenomenon being explained but have not yet been considered. Thus such concepts as "culture," "nation," "society," and "political system,‘ are treated as residua of variables, which can be incorporated into a general theory.... If we accept this residual nature of names of social systems, we can then attempt to replace these names by variables. When we find that societies differ with regard to a particular char- acteristic, we can ask what it is about these societies that causes this difference. (Prezworski and Teune 1970: 28) However, we believe one can still employ regional names if these names are used not in an explanatory sense, but rather as descriptors of opportunity structure "types." Regional analysis, in other words, is a useful tool in isolating and categorizing different contextual situations. The five settings to be examined here include two socio- cultural areas of Appalachia (non-coal/subsistence agriculture and coal mining) and three socio-cultural areas outside of Appalachia (rural industrialized/Western Kentucky, commercial agriculture/ Central Kentucky, and the copper-pulp area of Ontonagon County, Michigan). Appalachian Non-coal Area Eight Eastern Kentucky counties (Breathitt, Lee, Menifee, Owsley, Magoffin, Elliot and Wolfe) compose the non-coal region of Appalachia. All of these counties, with the exception of Breathitt, are located in State Economic Area 8 (Eastern Kentucky Hills Area). The economy of this Appalachian region has its roots in agriculture. Over half (55.5 percent) of the high school seniors in this area indicated they lived on farms, contrasted with 37.3 percent in 33 o.mp N.m mfim.N FN¢.w P.¢ w.N m.n m.o m.o¢ v.05 m.Fm «.mn o.om N.¢P m. n.om o.N_ w.¢ m.m¢ m.mm m.¢m m.m¢ m.o_ o.m _.N_ m._P mrcwu mxom mFLwo mxom Axcwmzccw vmwwwmcm>wo .mCA Axcpmzucw 15cm; FmrocmEEouv xxozpcmx pmcpcmu q_:a\cmaqouV cmmwgowz coma: 92 Ema mg 3 0.; mi 58 RN :3 :3 mg 5: m2: m_2ww axom €an -_mwcpm:ucw apw>mmg .chzcuwsmmv xxozpcmx ccmpmwz augm>oa zoPmm acousma Amy msoocchmPuwz noxopgemca “gouge; mocon Loam; :mwpw>mu cw mmmpcwocma mcwcspommacmz 8.2.2,. menopauwcm< ”cw cmzopasm Ampmsmuuwpmzv wagon xcoz mo mmmucmucmm cm>o use mm wm< .mmpmsumsw Poocom ;m_: wmmpcmocma Lm>o ucm mm wm< .nmpmpaeou Foocom :mwumz .mcovmmc xuzum mcu mo muwpmwcmpomcmso oweocoum new meoom umuomme ._ m—aMH 34 _.mm w.n¢ xgcw>oa zopwm acmocma mmo.m mam.m Amy aeoocH cawumz o.“ M.“ “.8 o.o_ nmxo_asmc= Sewage; w.PN o.Po m.om o.om mucou scam; cmmpw>wo cw mmmpcmocma P.o «.mp mewcauomwzcwz _.mm ~.m mcwcwz o. e.m mcsp_=owcm< ”cw umxopasm Ampmsmmumpmzv mono; xcoz mo mmmucmocwa m.¢m m._m o.m_ o.w_ Lw>o new mm mm< .mwpmzcmco Foocom cmw: manpcwucma m.@ m.@ m.m o.w Les ace mm am< .umpmpqeoo Foocum serum: m_gww mxom mpcww mxom Amcvscmm wucmumwm Amcw:_a Fmoov loam .Fmouucocv .m> .3\»xu:pcmx Axozpcmx :cwgmmm .A.n.p:ouv _ m_nmp 35 Central Kentucky, 27.0 percent in Western Kentucky, 2.9 percent in the coal area, and 11.2 percent in Ontonagon County, Michigan. Furthermore, many of these farms have little non-farm income due to the lack of industrial development in the area. In addition to being characterized as a marginal agri- cultural area, the Appalachian non-coal region is often described as a "familistic" type of society. Schwarzweller and Brown, for example, describe the social organization of Beech Creek in the following way: A crucial aspect of the eastern Kentucky situation, confronting any student of the Appalachian mountain areas, is the importance attached to kin relationships in the everyday life of its people. Familism, as a value orientation, tends to permeate the local society and stamps all institutions with its mark. (Schwarzweller, Brown, and Mangallum 1970: 85) The median income for 1970 ($3,693) and the percent of the population living below the poverty level (47.8 percent) place this region at the bottom, economically, of all the regions in the study. In fact, it is one of the poorest regions in America. Given the general economic conditions of the area and the limited opportunities for securing a "decent" job, it is not surprising that at least 66. 5 percent of the senior boys and 76.9 percent of the senior girls (Table 2) plan to leave the region after graduation from high school. The socio-economic character of the region is also manifested in the small percentage of seniors who intend to continue their formal education beyond the high school level. Table 2 shows that only about one-third of the boys and 37.8 percent of the girls plan on college. The economic resources of the area and the general normative/motivational climate associated with a poverty area effectively mediate against upward educational mobility. 36 This region, then, can be described as having a relatively low standard of living, high unemployment and, in general, very limited opportunities for occupational and educational advancement. The opportunity structure can be considered narrow and restricted and the social mileau from which plans formed, depressed. Appalachian Coal Area The Appalachian coal region is represented by Floyd, Pike, and Leslie Counties in eastern Kentucky and by Mingo County across the Tug River in West Virginia. It is also a rural low income area, though of a different character from the subsistence farming region. The economic resources are based primarily on an extractive industry: coal mining. Due in part to the nature of coal mining as an occupation, the relatively standardized wage scale of miners, the character of the mining community, and the way of life in the coal camps, the coal region emerges as a unique entity in comparison with families in the other study regions. The nature of the mining industry, mechanization in the mines, and the lack of non-coal occuaptional alternatives have lead to high rates of out-migration, especially among the young people. Union wage sclae permits the coal miner to live at an adequate, but not affluent level. Non-coal manual workers in this region typify a varied level of living status. Some manual non-coal families survive on virtually a nonexistent income that is likely supplimented by social security, food stamps or welfare payments. Those manual workers living in or near the larger towns of the region, however, are more likely to find and hold higher paying jobs. In generaly, the overall socio-economic climate has a de- pressing effect on the educational chances of young people. Table 2 37 shows that only about a third of the high school seniors in the present study population plan on additional formal education beyond the high school level. Thus, this area of Appalachia is in many respects very similar to the Eastern Kentucky non-coal region. Both Appalachian areas are primarily rural and are characterized by relatively low standards of living, limited chances for upward educational and occupational mobility, and high out-migration rates. What distinguishes the coal area from the subsistence farming area is the somewhat different structural exigencies embodied in the regional economy. Life style characteristics and residential patterns are dependent on the dominant mode of production and the viability of the regional economy. An intra-Appalachian comparison of the determinants of career plans, then, will provide an interesting and enlightening component to the overall research effort. The Commercial Farming Region of Central Kentucky The commercial farming region of Central Kentucky is re- presented by the counties of Jessamine, Anderson, Clark, and Scott. It is the most prosperous non-metropolitan section of Kentucky. About 90 percent of the land area is in farms and almost all of the land is used for agricultural purposes. Unlike the subsistence farms (non-commercial, rural residences) in Eastern Kentucky, Central Kentucky farms are larger, the soils more productive and the topography is more suitable for commercial purposes. Bluegrass farms produce a great diversity of !roducts (e.g., tobacco, grain, dairy, beef, etc.) and are able to provide a stable and adequate income for their operators. Tje metropolitan area of Lexington, plus several smaller cities in and around the region provide a diversity of occupational opportunities for 38 the young people of Cnetral Kentucky. The percent of the seniors planning to migrate from Table 2 (boys -- 52.9, girls -- 57.1) and the percent planning college (boys -- 43.3, girls -- 37.7) place this area in the middle of our study regions. Although the area is relatively prosperous and characterized by a favorable opportunity structure, the absence of a large metropolitan area within the immediate region no doubt has a supressing effect on the career plans of young people living in Central Kentucky. (This would be consistent with findings by Burchinal 1962, Sewell 1957 and others.) The Rural Industrialized Western Kentucky Region The industrial, semi-rural counties of Henderson and Daviess represent the Western Kentucky region. Once primarily an agricultural region, this area is the most economically developed of those drawn into the study. Owensboro, the county seat of Daviess County, is home to a number of important industries including the manufacturing of radio parts, whiskey distillers, and a relatively large steel plant on the city's outskirts. In addition, there are tobacco manufacturers, steel fabricators, food processors, chemical plants, plastic manufacturers, and furniture making concerns in the immediate area. Because of its geographical location and relatively economic prosperity, there is a broad range of diverse occupational opportunities for young people, and, subsequently, the rate of planned out-migration (Table 2) is low compared with other regions. A further reflection of the prosperous circumstances of the area is found in the number of seniors planning college: over half of the boys and girls plan to contin- ue their formal education after high school. 39 The Copper/Pulp Area of Upper Michigan: Ontonagon County Ontonagon County, Michigan is a rural area located on the shore of Lake Superior in the western portion of Upper Michigan. Agriculture, once an important activity in the area, now occupies only a small fraction of the work force. This is due to the poor soil, short growing season and the remoteness of the area from po- tential markets. The most important economic enterprises in the area today center on the pulp industry and copper mining. The White Pine copper mine, largest single employer in the area, has over 3,000 workers drawn from as far away as 90 miles. In addition, about 1,000 men and women are employed by Hoener-Waldorf Paper Company and related concerns. 2 Although the economic and employment situation is bright, the remoteness from urban areas and the lack of many essential services and facilities impose a considerable hardship on the local inhabitants. The nearest cities are between 60 and 70 miles away. Furthermore, indoor recreation facilities, shopping centers, and public transportation are either very limited or entirely absent (e.g., there are nor movies in the county). Ontonagon County is somewhat unique since it is characterized by high out-migration (66.7 percent boys, 74.6 percent girls) gpg_ high educational aspirations (48.9 percent boys, 53.4 percent girls plan on college). Given the balance between educational and work opportunities in the area, however, this is not surprising. The limited range of occupational opportunities (mainly White Pine mine or Hoener Waldorf paper), the remoteness from more diversified occupational opportunities and the lack of entertainment and recreational facilities force many of the young people to migrate. 40 However, the area's prosperity and generally favorable attitude toward education provide the economic and motivational resources necessary for the stimulation of upward educational mobility. Thus, although a sizeable majority of the young people aim to leave the county, about half of the boys and girls plan to go to college. To summarize, the "regions" selected for examination in this study represent a wide range of rural socio-economic circumstances. They can be considered somewhat representative of area variations across the United States -- economically depressed, remote rural areas; more prosperous commercialized agricultural areas; rural areas that are becoming industrialized or are already in advanced stages of rural industrialization. STUDY POPULATIONS The study population consists of 2987 rural high school seniors from five areas in Kentucky, West Virginia, and Michigan. Table 3 presents the areas to be considered directly in the study, the number of cases in each region, the number of schools represented, the year the data were collected, and the principal investigator. With the exception of Ontonagon County, Michigan, all studies were organized in line with the aims of a larger project (supervised by Harry K. Schwarzweller) and data were collected in a comparable fashion by means of a self-administered questionaire in the schools.3 The Ontonagon portion is one part of another larger research project directed by J. Allan Beegle, Jon Rieger, and Harry K. Schwarzweller. Although the foci of the two parent projects are somewhat different, comparable data on career ambitions (educational and migrational plans) and determinants of ambitions (social class, grade rank, and rural-urban residence) are available. For purposes of this study, 41 cmmm_m cow\m_mmmm :m__< .a mom_ m Amva :mmwco?z ..ou commcopco mwmom n—mcoa\:mmum2mmumw wmmw M Mwmww .mbumwcwzowcmwmww Fmou cmegum_maa< a_mom u_m=oo m©m_ m Amaov _eou-=oz cawgua_aaa< memom upmcoo m©m_ m Au_mv Axuabcmx _mcpcmo memom u_mcoo wom_ N Amemv Axuspcmx campmmz Loammwpmm>cH Pmawucwca me> mpoozom mo Langsz .dmw .mwum .uwcmaeoo meowpm_:aoa zoapm chowmmm .m mpnmp o.mo o.~m m.mm m.oo mumcmwz o» mcwccmpm ucmocma m.o¢ 0.0m m.mm m.mm mmmppou mcwccmpa ucmocma aflmm MUHmflmm .w> .3\xxu:pcwx xxozpcwx ccmummm _.Nm a.Nm N.¢N “.80 “.mm 0.0m abacm_z o“ mcwccmpa pcmocwa N.nm m.m¢ «.mm m.wq m.Pm ¢.om mmmFPoo mcwccmpm ucmogwm flag flag 3% xxozpcmx _mcpcmu :mmwcuwz comma xxozpcmx :memmz .cowmmc xuzum sumo cw mumcmws on mcwccmpa pcmucma use mmmppoo mcwccmpa pcmocwa .m mPQMH the n can t STUDY Caree in ti a lon ultim choic more ”educe tYD01c occupa 0f Plar School, Elsewhe Employme the give iaMes h. 42 the method of data collection, measurement procedures, and techniques can be considered equivalent. STUDY VARIABLES Career Plans Choosing a career is not an event that occurs at one point in time. Rather, it is best viewed as a series of decisions made over a long period which lead upt to, and in a sense "predict" an individual's ultimate occupational choice. Two of the more important pre—occupational choice decisions concern plans to leave the home area in search of more favorable occupational opportunities. These two career decisions -- "educational plan" and "migrational plan" -- form an interactive typology that helps to delimit the nature and scope of a young person's occupational options and ultimately his life chances. In this research, educational plans are classified in terms of plan to go to college or to terminate formal education after high school. Migrational plans refer to a youth's expectation to relocate elsewhere in an area with easier access to favorable occupational employment opportunities than the place of origin (i.e., outside the given regional context). For the purposes of this study, the var- iables have been dichotomized. For all regions except Ontonagon County, Michigan, information on a young person's educational plan was solicited by the question: "Do you actually expect to go to college? (no; maybe or don't know; yes)." In dichotomizing the variable, a secific "yes" response is treated as "high" and all others as "low." In Ontonagon County a young person's educational plan is determined by the patterned response to two simultaneous questions: "Do you intend to get further training after high school? (yes; 43 no; don't know)". "What do you plan? (college; trade school; ap- prenticeship; other)". Those seniors who indicated a specific intention to go to college were classified as having "high" educational plans; all others were treated as "low” for the pruposes at hand. To determine migrational plan, seniors in the Kentucky and West Virginia were asked: "Do you expect that someday you will move away from this area (the county) and live somewhere else? (no, definitely not; maybe; yes, perhaps; yes, definitely)". The "yes, perhaps" and "yes, definitely" responses were treated as indicants of plan to migrate (or "high” migrational plan); the other two categories were reduced to a "low" migrational plan category. In the Michigan study, migrational plan was ascertained by the question: "How eager are you to stay or move from your home community after graduation? (Eager to stay; probably stay, but not eager to stay; probably leave, but not eager to leave; eager to leave)". A response of "probably leave, but not eager to leave" pp_"eager to leave" is treated as a "high" migrational plan. Social Class Origin Social class origin is determined by the father's social class standing, since for the most part a young person assumes the social status of his family and little intergenerational mobility takes place until a youth has completed his formal schooling. The indicant of social class used in this research is based on father's occupational status. (In all of the areas a modified Edward's (1943) type scale was utilized for this purpose. Census divisions and systems of categorization developed by other researchers were employed as criteria of ranking occupations. For our purposes, the ordinal Edward's 44 scale is reduced to a dichotomy comparing "manual” and "nonmanual" occupations. Grade Rank In all areas, intellectual ability is assessed by a youth's grade rank or relative status within the classroom. Information regarding grade point average or grade rank was gathered for each respondent from school records by the field worker. Each senior is ranked in relation to his classmates wiphig his high school. A dichotomy separating the "lower" half from the "upper" half is employed in this study. Community Context/Rural-Town Setting In addition to regional setting, rural/town context is employed as an indicant of the immediate situational circumstances confronting as youth and as a possible source of career plan variation within each region. For example, it has been shown that the social environment of the mountain coal camps and subsistence farming neigh- borhoods of Appalachia differ in many ways from the larger towns and city centers of the region and that these differences are manifested in the career paths young people from the Appalachian regions take (Schwarzweller 1973; Bogie 1970). Although research distinctions have been made between rural/farm and rural/nonfarm residence, rural residence, in general, whether farm or Open country, has a distinct effect of the career plans of young people. Consequently, an indicant of local setting based on a rural/town dichotomy is employed in this study. In all regions the following question was asked: "Where do you live? (on a farm; open country, but not on a farm; village or town)". For our purposes, responses "on a farm" ggg_"open country, but not on a farm" are collapsed into a "rural" category, while 45 “village or town" represents the “town" category. STATISTICAL PROCEDURES AND ANALYTICAL STRATEGY To empirically test the effects of variations in regional and community contexts on the career planning model, percentage differ- ences observed in contingency tables will be employed. This method is cpmmonly referred to as "elaboration by partials." By this tech- nique, relationships between independent and dependent variables are first determined. After these relationships have been established, controls are introduced and the interrelationships among the control measures, independent variables and dependent variables is determined. It should be noted that in our statistical design, regional setting and sex are treated as "contextual controls" (e.g., supra- control variables) and will not be entered into the elaboration process in the usual manner. The overall research design, then, can be conceived as a multiple replication and elaboration of an identical career planning model in five distinct regional/ecological contexts. James A. Davis' (1971) method of non-linear causal analysis will be used to ascertain the independent and interrelated effects, for boys and girls, of social class origin, residence place, and grade rank on career plans. Davis' technique is closely analagous to multiple regression/path analysis. However, unlike regression an- alysis, which assumes linear relationships among the variables under study, the Davis method is based on Yule's Q coefficient and is specifically designed to handle dichtomous variables. The analytic strategy suggests the following approach: (1) Educational plans and propensity to migrate will be viewed as dependent variables. These two variables, representing individual components of the broader term "career plans", will be treated both 46 as independent indicants and as forming an interactive dependent variable typology. (2) Social class origin and grade rank will be introduced individually and jointly as "predictors" of career plans. They will be viewed as independent and control variables in the statistical desing. (3) Rural/town setting will be employed as a contextual control and, where appropriate, be introduced into the elaboration and prediction procedures. (4) Regional context will be treated as a supra-control variable and, as noted above, will not be introduced directly into the statistical procedures. (5) Sex will also be viewed as a control variable. Career planning models will be positted for both boys and girls. The purpose here, and the main objective of the study, is to establish empirically the extent to which regional/environ- mental contexts effect the manner in which social class, grade rank, and local residence place individually and jointly structure career plans. 47 FOOTNOTES Chapter 2 l. A large majority of these farms are small, marginal operations. Most farms in the region are unable to support even a subsistence standard of living. 2. The economic situation in Ontonagon County has been radically altered since data were collected in 1968. In January 1976 White Pine Copper was forced to lay off two-thirds of its labor force due, in part, to a drop in the world price of copper and also due to a U. S. Government ruling against a proposed merger of White Pine with another large copper concern. At present, plans are being made to study the impact of this economic calamity on the residents of the copper/pulp area of Upper Michigan. 3. Although the Mingo County and Eastern Kentucky data were collected by different researchers, for somewhat different purposes, examin- ation of the general socio-economic characteristics of each area shows them to be very similar (i.e., low income, coal mining, depressed areas). Hence, because identical research techniques were employed, for the purposes of this research these two study populations will be combined to form the "Appalachian Coal Region.“ CHAPTER III ANALYSIS OF FINDINGS This chapter reports the findings. The analysis is organized to explore inter-regional variations in the patterning of career plans among rural high school seniors. We begin with an examination of educational and migrational choices by regions and sex. This overview provides a basis for elaboration by social class background, grade rank, and intra-regional residence place. BASIC INTERRELATIONSHIPS College Plan and Migrational Plan High school seniors from the two Appalachian areas (coal and non-coal) are less likely to pursue further academic training, beyond high school, than are high school seniors in the non-Appalachian areas. This fact is evidenced quite clearly in Table 2. In terms of migration plans, however, the Appalachian -- non-Appalachian distinction does not hold. The copper/pulp area of Upper Michigan is also characterized by very high out-migration and in this respect is strikingly similar to Appalachia.1 To help interpret and elaborate on the above pattern, the two dependent variables, educational and migrational plans, are combined into an interactive typology (Table 4). This analytic strategy reveals that a large proportion, nearly half, of "migration oriented" Appalachian youth plan to leave the immediate area without the benefit of college preparation. At least 45 percent of the Appalachian youth (and probably more, considering the hogh drop-out rate from high school) will enter the labor market with no more than a high school 48 49 $va 0.2: 98 ¢.Nm o.m— n.¢m 38V 0.2: Nam m.mm m.m~ m.mm A8: 0.2: 8.3 o.mm m.o~ N.P¢ 83 0.2: Q: v.¢m o.mp N.Nm $5 0.2: :N m.¢m _o.mm «.mm 303 92: gm m.wp N.mm m.—m sz Pogo» ucmugwm _muok mpmcmve on www.mmmli mmmppou op :o om uoz 38.35 3 flflll vamp—cu 0p :9 cm #02 mumgmwe op.mmm.mmm mmmFFoo on :o om mpmcmw5.mmm mmmppou op co ow P.mm N.Nm mpgwu m.mm m.m¢ mxom axozucmx Facucmu N.¢~ ¢.mm mpcwo m.©© w.~¢ mxom cmmwzowz Loam: N.Nm m.Fm mpg?w N.om 5.0m mxom zxospcmx ccmummz Acommmc esp m>mw~ ..m.wv mumcmwz mmm__ou ow co ow no“ mcwccmpa ucmucmm .mucmuwmmc mo cowmmc can xmm xn uwcwneou use xp_m:uw>wucw mmumcmwe op mewccmpa “smegma cam wmmp_oo o» co mcwom pcmucma .a mpnmh 50 Ammav o.cop “.m, _.m¢ m.mp m.mm m.m_ o.m~ Ap_mv o.oo_ m.oF m.m¢ N.w m._m Aommv o.oo_ m.m~ m.mv e.m N.¢N sz PapOH “coucma Pouch mumcmws ou.mmm.mmmlu wmmPFou on :0 cm poz mpmcmws o».mmWII .mmmppou op :0 on poz muwcmws op.mmm.mmm mmmppou op :0 cm mpmcmwe.mmm mmwFFou op co ow m.~m _.mm mpgww m.mo .a> .3\x¥o=pcm¥ m.©m m.nm mpcww m.o© xxozpcwg :memmu Acommwc mg» w>mmp ..w.wv mumsmwz mmmPFoo op :0 cm ”op mcmccopm pcmocmm .A.u.p=oov a mpaeh 51 diploma. In contrast, seniors from the copper/pulp area in Michigan who plan to migrate are far more likely to be college bound. Michigan young people from a remote rural area in some ways similar to Appal- achia, obviously will enter the labor market in a significantly more competitive position than their Appalachian counterparts. A comparison between Appalachia and Upper Michigan clearly illustrates the importance of regional normative and economic mileaus in career planning. Families in the depressed Appalachian areas are often unable to provide the material or motivational support necessary for their offspring to pursue a college education. Because of inadequate hogh school preparation, limited occupational opportunities, and limited social and economic resources in the home area, Appalachian youth are, in effect, "pushed” out of the region with little tange- able support or marketable training. In Michigan's copper/pulp region, on the other hand, although a sizeable majority of high school seniors plan to leave the area after graduation, a socio-economic climate that facilitates achievement and advancement and a school system that, in general, maintains high academic standards assures a large proportion of these young migrants the chance to obtain at least some college training or other vocational training before they confront the work world. In both the heavily industrialized Western Kentucky area and the commercial farming area of Cnetral Kentucky, for boys es- pecially, college education is associated with plans to move out of the home region and, conversely, a decision not to continue on toward college seems to be linked with plans to find employment in the home area. This pattern is especially prevalent where blue collar/ farm related occupational opportunities are relatively abundent. 52 The availability and accessability of stimulating and rewarding manual occupations in these regions offers a realistic alternative to out-migration and/or a college degree. To briefly summarize: career planning choices vary by region. The more prosperous regions (e.g., Western Kentucky, Central Kentucky, and Upper Michigan) are characterized by larger proportions of seniors planning college than the less affluent Appalachian areas. Additionally, the balance between educational and work opportunities in a region appears to have direct bearing on migrational plans. Seniors from areas with unfavorable or limited occupational structures (e.g., Appalachia and Upper Michigan) are more likely to anticipate moving away from the home area than young people residing in regions where employment chances are both more numerous and varied. Father's Occupational Status and Career Plans Given that regional context has an immediate influence on the career choices of high school seniors (Table 4), we are now in a position to begin a more direct exploration of the regional- effect hypothesis. Our thesis holds that regional setting is best viwed as a distinct social system in which sets of career determinants assume different structural configurations. Table 5 shows, not surprisingly, that father's occupational status is highly associated with further academic plans in all regions. In the case of boys, correlations range from .45 in the copper/pulp area of Michigan to .63 in the non-coal sector of Appalachia. Al- though the relative differences among the regions are not great, a comparison of absolute percentage differences lends support to our hypothesis. The educational ambitions of senior boys from the more affluent regions (Western Kentucky, Upper Michigan, and to a lesser 53 mpcww Nc. 25$ :4 88 go —o. :93 we.” 83 m8 mxom xxoapcmx Pmcucmu Pmcowumazuuo m_cm;ume x3 mumcmwe ow mavccm_a pcmucma ocm mmmFPoo ca :0 mcwom pcmocmm 3. 8. 3: at 3: 58 $3 $3 0.3 58 8. 3. at a: 9% fig 33 a: 0.3 58 22:0 mxom cmmwcuwz Lang: mm. 8m: N3 33 98 88V .92 as we: mpgww om. A83 :3 A mm mxom xxuzucmx :memmz o 98:; :3 _m:=mz c: Pmacmscoz mzpmpm Pacowpmazuuo m_cm;“md mpmcmwz op mcwccm—m “smegma o 98:; 9: Peace: 2,: Fascmscoz mspmpm chowpmazuoo m.cw;umm ammFPou mcwccmFQ pcmocma .mucmvwmmc to common can .xmm .mzpmum .m «Pack 54 om. m_.- Ammmv Aemmv N.No m.mo Amo_v Afiwv N.ma m.Nm om. mm. Ammmv Aummv P.0m m._m Aoo_v Ammv N.m© N.oo mptwo axon .m> .3\xxu=bcm¥ NF. mm. Ammmv Amqmv m.NN m.¢o A543 Amev o.mm m.¢m mm. mm. Ammmv Amewv m.mm «.mm Aeuv Afiuv m.mm w.mo m_cee axom xxuspcmx cgmpmmm o m.a_:> sz Peace: sz szcmscoz magnum chowpmazouo m.cw;pmm mpmcmwz op mavens—a pcmocma o m.a_=> sz Peace: sz szcmEcoz mzpmpm Fm:0wpma:ouo m.cm;pmm mmuPFOUImcmccm—a ucmugma .Au.8=ouv m apnap exte seni aDPY effe plan Mich are ' mOdEI blueg A lac Michi senio COOtr Centr Optio coal SUb~p Sugge for u CornDa and r1 that the f‘ Mlchig betWEE 55 extent Central Kentucky) are proportionately higher compared to seniors from Appalachia. Boys from the depressed Appalachian areas approach high school graduation from a socio-economic mileau that effectively dampens chances for further academic training. The patterning of migrational plans, unlike that of educational plans, varies regionally. In Appalachia and the copper/pulp area of Michigan the correlations between social class and propensity to migrate are low or negative, while in Central and Western Kentucky they are moderately high. Relative differences range from .42 in the Central bluegrass area of Kentucky to -.13 for senior coal boys from Appalachia. A lack of blue collar work options and/or opportunities in Upper Michigan and Appalachia apparently force many lower class high school seniors out of the home region in search of career fulfillment. In contrast, blue collar boys from the more prosperous Western and Central Kentucky regions are confronted with far more occupational options, thus reducing the necessity to leave the home region. It is interesting to note that nonmanual boys from the coal region are less inclined to migrate than any other Appalachian sub-population (including nonmanual/non-coal boys). This pattern suggests that occupational aspirations and expectations, at least for upper class boys, are more easily attained in the coal area compared with the non-coal area. We suspect that the mining industry and related concerns offer career possibilities and alternatives that are absent in the non-coal area and consequently act to dampen the flow of out-migration. In the case of girls, Table 5 shows that, although Upper Michigan is somewhat different, the relative and absolute relationships between educational plans and father's occupational status resembles 56 the boy's pattern. Girls from lower class homes in Western Kentucky and Upper Michigan manifest proportionately higher educational plans than their Appalachian counterparts. Interestingly, upper class senior girls from the copper/pulp area have very low educational ambitions, especially compared with girls from Western and Central Kentucky. It may be that the small study population (N=25) of nonmanual girls gn Upper Michigan may, in part, account for the aberant pattern. But then, again, the pattern may not be an artifact and indeed a phenomenon meritting further study. The migration plans of girls in all regions closely paralell the boy's pattern. Girls from blue collar homes in Appalachia lean more toward migration than their counterparts in Western and Central Kentucky. Furthermore, the propensity to migrate is greater for nonmanual girls in all study areas. To summarize: inter-regional variations in the association between father's occupational status and career plans (especially migration plans) along with absolute percentage differences supports the contention that regional circumstances have an important impact on the manner by which family background variables (i.e., social class) influence educational and migrational plans. Let us now focus our attention on another important predictor of career plans -- grade rank. Grade Rank and Educational Plan High academic performance in school is generally regarded as an important "stepping stone” for admission to college. Grades are accepted as the legitimate sorting criterion in our educational system. A young person with "high" grades has, in effect, attained the endorsement of the school system to pursue higher educational goa stal or 1 pott educ plan show leve' for I senic Kentu does b0ys I gradUé from A non-AD among . to 90 1 dlmjnjS llgibie bOYS an Sanctlo. 0f r9910 Plays 1n 57 goals. However, our research thesis suggests that regional circum- stances exert pressure on the youth that may inhibit, supplement, or facilitate the importance of grade rank. In other words, we hy- pothesize that the importance of scholastic achievement vis-a-vis educational plans varies as situational circumstances vary. While correlations between grade standing and educational plan are substantial for both boys and girls in all regions, Table 6 shows that absolute inequalities exist among regions at all grade levels. "Low" scholastic performance is especially detremental for Appalachian youth; less than 20 percent of the poor achieving seniors plan on college. In contrast, in Upper Michigan and Western Kentucky, and in particular among boys, a "low" academic placement does not manifest the same consequences; around 40 percent of the senior boys plan on college even though they rank in the bottom half of their graduation class. Even among the better qualified students, seniors from Appalachia plan to go to college less frequently than their non-Appalachian counterparts. Furthermore, sexual inequlaities are evident in all regions among the "high" achievers. Boys with "high" grades are more likely to go to college than similarly qualified girls. These inequalities diminish, however, among the "low” performers in Appalachia. Neg- ligible percentage differences between the less academically gifted boys and girls in Appalachia suggests the absence of normatively sanctioned expectations at the lower end of the achievement continuum. Findings from Table 6, then, further confirm the importance of regional context on career plans. The role that academic performance plays in selecting and sorting talent for high academic placement appears to be contingent, to some extent, on the nature of the x 3.. 33 : 3. P a L u c 3 c c. 3.: u E. #095 Easesmlm. E 3133 .cup_m.U.meL $0 :21me sec 325. 5:2: EELS. Xe 203:0... «5 to 9:03. fruuLUl .0 «£05 58 ,0. wk. Amopv ANKNV 8.5F o.u_ Amwmv Aowpv o.N¢ m.mo m_Lco axom .m> .3\»¥u=pcm¥ am. mm. mm. mm. Aqmpv AmN_v Ammv ANNV m.¢_ m.mm m.~m m.mm Akm_v Ao__v “mow Afimv o.¢m N.wo N.oo m._w mpceo mxom m—wa mxom zxuzucmg Pmcucmu :mmwcowz Loan: mm. mm. o m.mF:> Am_Fv Amway sz _.m_ ¢.m_ 304 Ama_v Am_Pv sz o.om o.mo cow: xcmc mumcw mpcww mxom ummr_ou.m:wccmpm pcmocmm axuzpcmx :cmpmmm ON. we. a m.m_=> Amo_v A©N_v sz N.©N m.~e 20; Aom_v Ae__v sz m.mm m.mN sow: 1cm; mecca m_cwo mxom «moppou mcwccmrm pcwucma xxuspcmx :cmpmwz .wocmuwmmc to cowmmc use .xmm .xcmc mnmcm an mum—poo on :o mcwom pcmocmm .o m—nmk 59 regional environments or mileaus. That is, the meaning and importance of a specified grade rank vis-a-vis a youth's academic ambitions is interpretted and mediated in relation to his place in the regional social/opportunity structure. Before positting multiple control and intra-regional causal models, we will now examine the influence of local residence place on career plans. Local Residence Setting and Career Plans Regional settings were selected to represent a wide range of rural socio-economic circumstances. Within all regions, however, community and neighborhood differences also exist, of course. The car- eer patterns and life chances of young people from the more predom- inantly rural, isolated tracts (e.g., open country or farm) within a region may vary significantly from young people living in the larger towns or urban communities. In this section we wish to examine the effect of intra-regional residence context on educational ambitions and migrational plans. As shown in Table 7, youth residing in one of the larger towns or cities of an area are more likely to plan on college than their more rural counterparts. Positive relationships hold for all areas except the copper/pulp region. Moderately high correlations exist for both boys in Appalachia and for boys and girls from the Central Kentucky bluegrass region, while somewhat lower associations hold for seniors in Western Kentucky and among Appalachian girls. Regional effects are most apparent among the "rural" segments of our study population. Seniors living in the rural areas of Western Kentucky are more oriented toward college than seniors residing in the mountain neighborhoods of Appalachia or in the farm/open country : istic 253.23.: .225: xxuadmv...-50.»-ch .mocouwmcg to :o_awL Ute .xmm .mom~1 30:2:va 2.3: x: muccatz 34 apficcfa fiSOLQQ etc 330:.30 Cw :0 32:3: acousol .k 07:: 60 mm. me. «0.: wF.u cm. Fm. o m_m~:> A_m_v AN©_V Amev Ammv AFN_V Amk_v sz m.om _.m¢ m.mn m.FN N.o¢ m.m¢ Paczm Amm_v Amm_v Aqmv Ammv Auo_v An__v sz m.m© ¢.©© 2.4“ m.m© ©.mo m.mm _eL=L-:oz mum—a mocmuwmmm mpmcmvz on mcwccmpa pcwucma _e. _m. 04.- “0.- m_. ON. a m.m_=> A_m_v AN©_V Amav Ammv ANN_V AmN_v sz m.mN m.©m m.m© 0.0m “.54 m.Nm _a23m Amm v A©N_v Aumv Ammv Amo_v A~__V sz ©.ma 4.Nm ©.Ne 8.04 0.xm «.mo Fm232-=oz mumpa mocwuwmmm mpcww mxom m_L?w mxom mpcww mxom wmw__ u wwwccmhm pcmucma xxusucmx Facucmu :mmwcowz Loan: xxozpcmx ccwummz .mocmuwmmc yo cowmmc use .xmm .mumpa mucmv_mmc PmUOF xn mpmgmve op mcwccm_a ucmocma new mmmppou o“ co mcwom pcmugmm .N m—nm» 61 mm. m_. m_.- No. 0 m.m_:> Amomv Aommv Ammmv Aewmv sz o.po p.00 o.mp m.oo pmazm Ammpv Apmpv Ammo pogo sz p.wp m.mo _.mp ¢.po pma32-:oz oompo mocoopmom mpogmpz op mopccopo pcmocoo _m. mm. o_. mm. o w_m_=> Amomv Aommv Aoomv Amwmv sz _.om m.oN m.om p.mm pmezm Ammpv Amm_v Ammo noqv sz N.ma m.mm N.¢e m.oo pm232-:oz mompo wocmopmom mpppu mxom mpgpm mxom mmo—Poulmopccopo pcmupoo .o> .3\>xoop:m¥ xxoopcmx :memom .A.o.pcoov A mpnmp 62 areas of Central Kentucky. Appalachian boys are especially hurt by a rural home origin. Interestingly, however, young people living in the larger towns or villages, regardless of regional setting, manifest strikingly similar educational ambitions. Excluding Upper Michigan, the absolute percentage differences among the study regions varies only 10 percent for boys and 12 percent for girls. Viewing college plans from an intra-regional perspective seems to show that opportunities for up- ward educational mobility are more equitably distributed in Western Ken- tucky than in Central Kentucky or the Appalachian areas. In the copper/pulp region of Upper Michigan an inverse relationship between rurality and college plans is evident. The relative isolation of the area and the absence of any sizeable towns in Ontonagong County3 operate to equalize residence place differences and make the schools, in a sense, less localistic and more integrated on the regional level. Unlike Appalachia, concentra- tions of wealth and/or opportunities are not found in the largest town of the area (i.e., Ontonagon -— the county seat), but rather, what economic and social resources exist in the county tend to be rather widely dispersed. Propensity to migrate is positively correlated with non- rural residence in Western and Central Kentucky (Table 7). In Appalachia and Upper Michigan, on the other hand, clear patterns do not emerge. As previously noted, the Appalachian areas and Upper Michigan are typified by extremely high out-migration rates. The findings from Tables 5 and 7 seem to indicate that because of the heavy exodus of young people, two of the more generally recognized predictors of migration plans (i.e., social class origin and local 63 residence place) exhibit little explanatory power in these areas. In the more prosperous areas of Western and Central Kentucky, however, both social class origin (Table 5) and residence place (Table 7) are linked with out-migration plans. Young people residing in the more rural environs of Western and Central Kentucky are less likely to leave the home region than their town/city counterparts. Clearly, the greater the availability and accessability of employment opportunities in a region, the more likely a young person will be to find a satisfying career in the home area. However, when occupational opportunities are limited or lacking in an area, mass out-migration is a normal phenomenon. As the present study shows, this out-migration cuts across both class and residence lines. It appears that only in regions where the probability of migration is tempered by the chance of finding suitable employment nearby do other determinants such as social class and local residence place become important predictors of migration plans. A summary of findings to this point provides a base upon which to build three and four variable causal models of career plans. Generally speaking, as hypothesized, regional circumstances tend to condition the manner in which career plans are formulated. The regional socio—economic environment and the balance between ed- ucational and work opportunities in a region shape and mold the rel- ative importance of family and school based pressures in a career planning model. Furthermore, the extent to which social and economic resources are evenly distributed throughout a region affects the life chances and career possibilities of a youngster. Regions such as Appalachia, where available opportunities and resources, though limited, tend 64 to concentrate in the larger cities and towns, reveal significant intra-regional differences in the patterning of career plans. Seniors living in the rural/isolated neighborhoods of Appalachia manifest considerable lower academic aspirations than their more "urban" cousins. In the heavily industrialized Western Kentucky area and the copper/pulp area of Upper Michigan, on the other hand, intra-regional resources are more equitably distributed. Consequently, local residence place distinctions, as determinants of educational plans, are less important. THREE AND FOUR VARIABLE ANALYSIS BY REGION AND SEX Having examined regional variations in a number of two variable relationships, we are now in a position to combine predictor variables (i.e., social class origin, grade achievement, and local residence place) into more interesting and complex three and four variable causal models. These models will be probed for regional effects using percentage differences observed in contingency tables and James A. Davis' method of non-linear causal modelling. It should be noted that in some instances, especially for the Upper Michigan study population, cell size becomes extremely small undercertain multiple control conditions; hence some of the effects noted may be of relatively low reliability. In these cases, only descriptive statements concerning the direction of the relationship can be made. Model 1: Father's Occupational Status -- Grade Rank -- College Plan The first set of interrelationships involve social class origin (father's occupational status), grade rank, and college plan. The causal model for all regions can be most simply diagrammed as follows: 65 Social Class College Plan Grade Rank Partial correlation coefficients (Yule's Q) from Table 8 reveal that in all regions social class origin and scholastic achieve- ment have independent effects on educational plans. Although father's status is positively associated with grades, and hence the model suggests that grades may operate as an intervening variable between class and plan, only among two sub-populations (Western Kentucky girls and Central Kentucky girls) does grade achievement function to reduce the importance of class background on educational ambitions. Even among these sub-populations the reduction is very slight. Although the above model is in large measure free of inter- actional effects and regionall consistent, careful examination of Table 9 yields some important relative and absolute regional differences and some interesting intra-regional (e.g., sex differences) var- iations. First, correlations between grades and educational plans are stronger for Appalachian boys and weaker for Appalachian girls than in other regions. An academically unqualified Appalachian boy, regardless of class origin, is significantly less likely to pursue a college education than similarly qualified boys from the more affluent regions (i.e., Western Kentucky, Central Kentucky, and Upper Michigan). ”High" scholastic achievement, then, functions as a critical prerequisite for upward educational mobility among boys from Appalachia. For Appalachian girls, however, grades are somewhat less important in the quest for college. This reversal phenomenon appears to be a function of the usually high level of college ..- _. ..\~r\._+:c\‘ :LCLVHV\A .oocoopmw; Co :opaoc oco xom x a :opa oae~_oo ozo m.Lozpcw azoso Ac m .xzog mnoca .mzpopm .o_:>v m:cppcpcccco ”mpmxpecc Noteppoqoooo ppmzcu Lcczxsnzc: escapk=> OSLQK .2 z~cak 66 coppopmcpoo Poppcoo mopopm Foocme cospod mopopm poocmsco: cospou «Pompcm> Pocpcou capo oomppoo oco xcmc moocw coppmpmccou cmoco oLoN mm. mm. F“. on. mo. om. NR. we. Rm. Pm. ox. mo. om. mm. mm. mm. mpcpw mxom xxoopcmx Pmcpcmu mm. o“. mm. mm. mm. mm. oo. oo. om. oo._ mm. mm. mo. pp. mm. mm. mo. um. mu. «m. mm.- «a. mm. mm. mm. oo.p No. ,4. No.- oo. _o. pm. mpapo axom mpcpo axom cowpgopz Logo: xxoopcmg cpmpmmz coppopoccou Poppcoo mooocm zoo mooocm cap: opnoppo> Pocpcou capo moo—poo oco coppoooooo m.cm;pou coppopmcpoo cooco oLmN .oocmopmmc mo copmoc oco xmm he capo mmmppoo oco .xcoc moopm .mopopm Focoppoooouo m.co;pow mcoEo Ao m_mpo>v mooppopmpcoo "mpmzpoco pomooo Lomcppuco: apnopco> moose .w mpoo» . o > -...z \..A.v..o.:-p.cov. xv? 3.5.x E 3 mom. 4 .U. p230» .3. arises 67 mm. mm. Fm. mm. mm. Fm. mm. mm. coppmpoccou poppcoo mopopm Foocos cospom mopmpm poocoscoc gospmu mpaopco> —oppcou capo oomppoo ocm xcog ounce coppopmpcou pooco oLmN Fm. mm. No. No. mpepo mm. om. mm. om. mxom .a> .z\»xu=pcm¥ mm. _m. om. pm» mm. mm. om. pm. om. pm. ma.- om. mu. mm. mm. mm. mpapo axom xxoopcox ccopmom coppopoccou Foppcoo mmomcm zoo mmoocm cop: mpnopco> Focpcoo capo mumppoo oco coppoooooo m_po;pou coppopopgoo cooco opoN .A.o.p=oov m apamp 68 00000 0.00 0000 0.04 00000 0.00 0000 0.00 00000 mxom .0>,.300000000¥ 00000 00000 0000 4.00 0.00 0.00 0000 0000 000 0.04 0.00 0.00 00000 000V 0000 0.04 0.00 0.40 0000 0040 0000 0.00 0.00 0.00 00000 0000 00000 0000 0.40 000 0.00 0000 0.00 000 0.004 mxom xxoopcmx Pogpcmo l 00mpcowz Long: 0000 0.00 0000 0.0 00400 4.04 0000 4.00 mpgww sz 003:0: 020 Foocoacoz sz 000:0: sz Poocoecoz xxoopcmx m.om 00000 4.00 0000 0.00 00000 00p0pm .ooo 0.0wcp00 020 000:0: sz Fmocoecoz sz 003:0: 020 pmocmecoz xxoopcox :memoz mopopm .ooo 0.0mcpou .mocmopmmg mo cowmms oco .xwm .xcme oomgm .mop0p0 pocoppoqzooo m.gmcp00 >0 mmmFFoo op co mcwom pcoocmm 3o4 204 000: 000: monocw zoo 204 cap: 000: mmompw .m @4500 69 aspiration manifested by lower class girls. Similar findings have been noted (Schwarzweller 1973, 1974; Bogie 1970), though reasons for its persistence, at present remain unclear. In contrast to the Appalachian case, percentage differences from Table 9 indicate that, in the industrialized Western Kentucky area, in the commercial farming area of Central Kentucky, and in the copper/pulp area of Upper Michigan, "high" grade achievement enhances the chances for college among seniors from both manual and nonmanual homes. However, unlike Appalachia, poor achievers in these more prosperous areas have surprisingly high academic expectations. We suspect that other normative and structural mechanisms exist in these non-Appalachian areas to keep more young people on the academic track.4 Secondly, the "normative climate" and economic realities associated with Appalachian life mediate against all but the most academically gifted and economically affluent young people (i.e., those boys and girls from nonmanual homes that are ranked in the top half of their graduation class) to supress academic aspirations. Interestingly, the educational ambitions of youth from the upper strata of Appalachia transcend regional boundaries. The magnitude of their plans is comparable to their non-Appalachian counterparts. However, in general, Appalachian youth who perform poorly in school, come from lower class thresholds, or both, manifest lower ambitions than comparable ranked young people in Western Kentucky, Central Kentucky, and Upper Michigan. Clearly, then, regional variations exist in this three variable career planning model. Although the associations noted in Table 8 are similar in magnitude and direction for seniors in 70 all regions, absolute percentage differences suggest that regional mileaus manifest an influence which implicitly or explicitly affect a young person's career plans and ultimately his life chances. Furthermore, as evident in Appalachia, the regional opportunity structure may be perceived, interpretted, and reacted to differently by boys and girls. Model 2: Father's Occupational Status -- Residence Place -- College Plan The second model we will pursue concerns the independent and interrelated effects of social class origin and intra-regional residence place on college plan. Zero order correlations for the regions, in general, yield the following causal paths: > C0] 1699 P1 an The search for regional variations in Model 2 requires Social Class Residence Place introduction of local or intra-regional residence place (rural/town) as a possible explanatory factor of the relationship between father's occupational status and college plan (see Table 5). Conversely, since the relationship between class and residence is viewed as asymmetrical, the "effect" of class must also be partialled-out from the residence place/college plan relationship. Turning first to the intervening influence of residence place, Tables 10 and 11 reveal that intra-regional residence var- iation has a negligible effect, in all regions, on the association between class and plan. In general, young people from nonmanual homes are more prone toward college, regardless of local community context, than young people from lower class backgrounds. 71 N4. FF. 0m.- NN.- m4. NF. cowpopmggou pmwpgma mo. mo. 0a.: om.u m—. m_. mopmpm Fwocme gmspmu m0. mm. mm. mm. 00.: mo. mzpmpm poacmsco: stpmm 0000000> Pogpcoo 04. 00. 04.- 00.- 00. 00. 0000 0000000 0:0 mo00o mocmowmmm mcowpmpmgpou Loogo oLwN IIDIIIIDIIII 00. 00. 40. 00. 00. 00. 00000000000 0000000 04. 40. 00.- 00. 00. 00. 00000 000000000 00000 00. 00. 40. 00. 40. 40. 00000 000000000 00000-002 0000000> Pocpcoo e. o. 5.- e. e. 0. 2.5:.efiamwfiw .0000m .0000 .00000 .0000 .00000 .0000 000000000000 00000 0000 0Joopcmx ngpcou com0cowz Loam: zxoopcmg :memmz .mocmowmmp mo cowmog 0:0 xmm 00 0000 mmmPFou 0:0 .oo0_o mocoowmmp Fooop .mopopm chowpmooooo 0.000p00 mcoeo Ac m.mpo>v moo0p00oggoo ”000000:o 000:0o 000:00-co: o000000> mogch .o— 0000» 72 :owp00000ou 000p000 mzp0pm 400:05 Logp0u mop0pm 00::0E:o: :05p00 0400000> 000p:ou :000 mmmFPoo 0:0 00000 00:000000 :o0p0PoLLou :00:o oLmN 00. 04. 00. 04. 00. 00. 00. 00. 00. 04. 44. 00. 00. 00. 00. 40. 00:00 0000 .0> .3000000000 mo. cm. 00. Pm. mm. an. op. om. mm. 0m. mm. mm. om. m4. 0m. mm. 00000 mxom axoopcmx ::mp00m :ovp00000ou 000p000 000—0 00:000000 000:0 00000 00:00000: 00000-:oz 0000000> Fogpcou c000 mmop—oo 0:0 :o0p0azooo 0.:mgp00 :o0p0pmggoo 0000o opoN .0.0.0000v 00 00000 00 00000 00000 4.00 4.40 0000 0000 0.00 0.00 0000 A000 0.00 0.00 0000 0040 0.40 0.00 00000 0000 xxoapcmx 00:p:mu 00:00000: 00oo0 .mop0p0 00:o0p00:ooo 0.:o:p0m >0 mmoppoo op :o mcwom pcoULma 00000 0.00 0440 0.00 0400 0.00 0000 0.00 mpgww 00400 0.00 0000 0.04 0400 0.04 0000 0.00 mxom .0> .3000000000 0000 0.00 0000 4.00 0040 0.00 0400 0.40 m—va 0000 0.00 000 0.00 0040 0.00 :0000002 Loam: 00000 04000 020 0.0m m.om 000:0: 00:00 0000 0000 020 0.0m 0.0m 000:0sooz 00:00 0000 0000 020 0.0m 0.00 000:0: 00:=:-:oz 0000 0000 020 0.m0 0.m0 00::0E:oz 00:00-:oz 00:00 0000 00p0pm .ooo 0.:mgp00 00:000000 xxoopcmx :mem0m 04000 04000 020 c.04 0.00 000:0: 00:0: 000V 000V 020 0.00 0.00 00::0scoz 00:00 0000 0000 020 0.04 0.40 000:0: 00000-:oz 000v 0040 020 0.00 0.00 000:0e:oz 00:=:-:oz 00:00 mxom mop0pm .ooo 0.:m:p00 00:000000 .0xozpcmx :memmz .moco0wmo: wo :opmm: 0:0 .000 .00000 .00 00000 74 Reordering the independent variables so that social class is treated as a first order control reveals interesting and often complex interactional effects. In the case of Appalachian girls and Western and Central Kentucky boys and girls, Tables 10 and 11 show, for instance, that class related factors reduce the correlation between residence place and college plan. In other words, local residence conditions are relatively unimportant determinants of the educational ambitions among these study populations. However, for Appalachian boys and for boys and girls from the copper/pulp area of Upper Michigan, intra-regional variations persist or increase when class origins are controlled. Non-rural Appalachian boys, regardless of class origin, are significantly more likely to aspire toward college than their counterparts living in the more predominatly rural areas of the region (Table 11). A lack of educational oppor- tunities and career alternatives in the more isolated rural neighbor- hoods of Appalachia appears to dampen the aspirations and expectations of boys from all social class levels. Interestingly, boys from the larger Appalachian towns and villages not only manifest higher educational ambitions than their rural cousins, but, they also posess ambitions comparable (or greater) than non-rural Western and Central Kentucky boys. Looking somewhat more closely into the Appalachian case, we note that nonmanual/rural boys from the coal area are less likely to plan college than their non-coal counterparts. This is an en- lightening and intriguing finding. However, although we suspect that occupational opportunities associated with the recent revital- ization of industry in the coal fields may account for this difference, we will wait until the next section, where we examine migration 75 plans by class and local residence, to more fully probe and elaborate the career patterns of nonmanual/rural coal and non-coal boys. In the copper/pulp area of Michigan, Table 11 reveals negative associations between residence place and college plan, and an increase in magnitude of these relationships when class origins are controlled. These high negative correlations are a function of the large number of rural seniors from blue collar families planning college (71 percent girls and 50 percent boys). The manual labor force in Ontonagon County, though concentrated primarily in two industries -— mining and paper -- is well trained and highly paid. Many manual class families are able to (and apparently do) provide the monetary resources for college. Thus, given the re- stricted job structure in the copper/pulp area, especially for girls, and the accessability of college, it is not surprising to find many blue collar seniors pursuing further academic training. Model 2 alerts us to the fact that variations in a career planning model may exist at more than one contextual level. In other words, assurances of validity in regional research may be ques- tioned if the assumption is made that geographical boundaries delineate homogeneous socio-cultural entities. Not only must regional differences be taken into account, but intra-regional residence variations (e.g., rural farm/rural non-farm/urban) must also be noted. In Appalachia for example, we found interesting and important independent intra-regional effects on career plans. Rural residence in Appalachia significantly dampens college expectations. As we previously noted, the distribution of wealth, resources, and opportunities tends to be concentrated in the more "urban" areas of the region and subsequently facilitates "high" 76 academic aspirations among non-rural seniors. In Western Kentucky, Central Kentucky, and to a lesser extent Upper Michigan, on ther other hand, intra-regional economic and social homogeneity seems to foster a more "equalitarian" system of educational chances. Model 3: Father's Occupational Status -- Residence Place -- Nfigrational Plan The final three variable model we wish to examine deals with migration plan and the predictor variables -- social class origin and residence place. Although causal paths for this model are regionally unstable, in general: Social Class > migration Plan Correlation coefficients reported in Table 12 generally Residence Place indicate that both social class origin and intra-regional residence place are independent determinants of migration plan. The pattern of results found in Tables 12 and 13 are very similar to those pre- viously noted in Tables 5 and 7. However, careful analysis of Table 13 allows us to describe somewhat more clearly the migration patterns for specific social class/residence place groupings. In Western and Central Kentucky it is the nonmanual/non- rural seniors who are most prone to leave the home area. Not sur- prisingly, Table 11 shows this subpopulation to be the most upwardly educational mobile. It appears that the aspirations and life ex- pectations of these highly motivated young people extend beyond the opportunities available in the home region. Seniors least likely to leave Western or Central Kentucky come from rural, working class homes. Table 11 further shows these 77 mo.- 04.- o—. 00.- cm. Fm. mm. mm. DDIIIIIIII :o0p00000ou 000p000 m=p0pm 000:05 :00p00 00p0pm 00::0E:o: Locp00 0400000> Fogpcou c000 :o0p0000e 0:0 00000 00:00000: 00oo0 0:o0p000:0ou :00:o oLmN mm. mm. mm. mm. mm. mm. mm. 40. 40. mm. 00. MP. 00. mm. 40. 04. 00:00 mxom .xxoop:o¥ 00Lp:mu mo. mo.- my. 00. 04006 00.- 00. 00.- 00. 00.0- 00. 00.- 40. 00. 00. 00.0 00. 40.- 00. 00. 00. 0000 00:00 :0m0zowz 00mm: mm. mm. mm. om. mxom 000:p:0¥ :Lopmoz :o0p0FmLLoo 000p000 00000 00000000; 000:0 00000 00000000: 40:30-:oz 0000000> 000p:ou :000 :o0p0gmpe 0:0 :o0p0qoooo 0.0o:p00 0:o0p0000:ou 000:0 oLoN .mo:m0000: mo :o0000 0:0 000 00 :000 00:o0p0gmps 0:0 000000 00:00000: 00oo4 .mop0pm 00co0p0aouoo m.:0:p0w m:oE0 Ac m.mpz>v m:o0p0pm::oo ”000000:0 400000 000:00-co: mpn0wg0> 00050 .N_ 00000 78 0m. 0m. mm. mm. mo. Po. mm. 00. mm. o—.- mm. mo.- mo. . mm.- om. m0.- 00000 0000 .0> .3000000000 00.- 00. 00.- 00. 04. 00.- 00.- 00.- 00. 04. 00. 00. 00. 40.- 00. 00. 00:00 0000 xxoopcmg :Lopm0m :owp00000ou 000p000 mop0pm 000:05 Lw:p0u 00p0p0 00::0s:o: Lo:p0u 0000000> Fogpcou c000 :o0p0gmpe 0:0 00000 00:000mo: 00000 0:00p00000ou 000:0 oLmN :o0p0FmLLou 000p000 00000 00:000000 00:00 00000 00:00000: 00:00-:oz 0000000> Fogpcou :00: :o0p0gmps 0:0 :o0p0qzooo m.go:p0m m:o0p0_m::ou :m0go oLoN .0.0_00000 00 00000 79 ANm_v A_mfiv N.m¢ o.N¢ Aomv Afimv “.mm ¢.m¢ “New Amxv N.¢© N.©m Ammv Am¢v m.m© m.Fw m_wa mxom xxuzpcmx Fmgpcmu Ampmv Aoqmv m.m© o.oo m N Aqu Am V m.~N _. m A¢mv Awmv ¢.NN o.NN “gov Ammv m.om N.Nm mygwo mxom .m> .3\»xu=pcm¥ Ammv Acmv m.©N 0.0“ fl__v Amv N.NN o.oo_ Ao¢v Amqv m.NN m.qo fi¢_v Aofiv m.m~ m.Nm m_;ww mxom snagguwz mem: 03A qfi \OA [\N mpgww xxuzpcmx :Lmummm Ammfiv Amm_v m.m¢ m.mm Ammv Ammv _.mm m.mm Amov ANNV $.00 o.mm Ammv Aogv ©.N© m.N© m_gww mxom xxuzpcmx :Lmummz .mucmuwmmg mo cowmmg use .xmm .mompa mucmu_mmL FmUOF .mspmpm Pocowumazuuo m.gm;pmm >3 mumgm?s op mcwccmFQ “smegma sz Pusan: Fmgzm sz szcwscoz Fmgzm sz Panza: Fagsgucoz sz —m::mscoz ngzgucoz magnum .uoo m.Lm;pmm mucmvwmmm sz Fmscmz —mgzm sz Fmacm5202 Pagam sz Pmacmz ngzgucoz sz pmzcmscoz Fmgzgucoz magnum .uuo m.Lm:pmm wucmvwmmm .m— mpnmh 80 young people to have low academic expectations. Most likely, the availability and accessability of a wide range of blue collar/ semi-skilled work opportunities in the area dampen both migration plan and the necessity of obtaining a college degree before entering the labor force. In the copper/pulp region, although girls are somewhat more likely to leave the home area than boys, class/residence distinctions among migrants are relatively unimportant. As Tables 5 and 7 showed, Upper Michigan is typified by extremely high out-migration among its young people. The seniors that do leave, however, often plan on obtaining some college training before they enter the labor force (Table 4). In Appalachia, with the exception of a few subpopulations of boys, seniors manifest migration patterns very similar to Michigan youth. Because few educational and/or occupational alternatives exist for upwardly mobile Appalachian youth, only by leaving the region can seniors, especially girls, measureably enhance their career ex- pectations. Furthermore, the regional out-migration rate is generally so high that neither social class nor local residence place distinctions are useful indicants of a young person's propensity ot migrate. The interesting case of nonmanual/rural boys from the coal area should be noted here. In the previous section, Table ll revealed that this subpopulation manifested lower educational ambitions than their non-coal counterparts. From Table 13 we see that these boys, along with nonmanual/non—rural boys from both regions, have relatively low migration expectations. Combining the findings from Tables ll and 13 leads us to speculate about the interrelationship of regional economic and opportunity structures and their effect on the career 81 plans of various subpopulations of Appalachian senior boys. Earlier we suggested that the "opening-up" of occupational opportunities and employment chances in the coal fields has served to dampen somewhat plans for further academic training. From Table 13 it appears that this revitalization has also slowed the rate of regional out-migration. Furthermore, Table 13 suggests that it is the nonmanual Appalachian boys who are most likely to take advantage of opportunities in the home area. Within Appalachia, then, a pattern emerges regarding the distribution and claim on occupational opportunities. Employment chances "filter" down through the stratification system. Nonmanual/ non-rural boys have "first choice" among available jobs. Hence, the migration rate among these subpopulations is low. As the system "opens-up", additional opportunities are made available to other subpopulations (e.g., nonmanual/rural boys). It is the manual back- ground youth who are most hurt in a poverty environment. These boys are ranked at the bottom of an unequal system of rewards and opportunities and are thus forced to leave the home region, though often ill equipped, to compete for jobs in the larger society. Model 4: Father's Occupational Status -- Grade Rank -- Residence Place -- College Plan The final model we will explore deals with the interrelated and independent effects of three independent variables (social class, grade achievement, and intra-regional residence place) on educational plans. This model is represented below. All relationships are asymmetrical withsthe exception of the association between residence and social class. 82 Residence Place Social Class __A College Plan \ Grade Rank/ Table 14 presents the correlation coefficients necessary to cross-regionall analyze the four variable model. Since only social class is directly related to all variables in the model, multiple interactional effects emerge only when social class is treated as the zero order independent variable and local residence place and grade standing are introduced as controls. An ordering of variables such that either intra-regional residence place or grade rank is treated as the independent sero order variablg would produce results identical to those presented in Tables 7 and 8. Finally, to more fully under- stand and interpret Model 4, Table 15 presents the relative per- centage differences in educational plans among young people ordered by social class, scholastic performance, and intra-regional residence setting. Correlation coefficients for Model 4 (Table 14) show that social class origin exerts a rather strong and independent effect on educational plans in all regions. Even when scholastic performance and residence place are considered, second order partial correlations for boys range from .48 in the coal sector of Appalachia to .56 in the copper/pulp area of Upper Michigan. For girls the associations are somewhat weaker, though significant, ranging from .17 in Upper Michigan to .51 for girls for industrialized Western Kentucky.7 Though the correlations among the regions are similar in magnitude and direction, relative percentage differences (Table 15) reveal some interesting "universal" patterns as well as regional "specifications" in the educational ambitions of our study populations. 83 Pm. mo.u ¢o.n mo. mm. mv. ¢_.- om. mm. mm. mo.u Fo.u mm. mm. Fm. mm. mm. mm.. 00. mo. mo.n mo. No.1 00.- co. RN. up. mo. «0. mm. mm. mm. mm. NF. om. Fm. mm. mm. «0.: no. co. wF.- mo.u co.u Fo. Pm.- mm. om. no. No.1 mo. Fm. um. 3% ”mafia mpgwo mxom .xxuzpcwx Pmcucmu :mmwcuwz Loan: Axuzpcmx ccmpmmz .mucwuwmmc w mepcwa emcee ucoumm oumccm paced pommmw mocmuwmmm pummmm mmmpu CMPQ mmmPPou can xcmc mumcw cowpmpmccou cmuco oLwN mepgma cmuco ucoumm oomccm “econ pommwm mucmuwmmm uommmm mecca :m—a mom—Poo new cowpmasuoo m.cm;umu cowpmpmcgou cmuco oLmN o cowmwg vcm xmm an cmpa chowpooaum new .mum_q wocmuwmmc FmooF .xcmc mcmcm .m:p¢pm chowpmazuuo m_cmcpmw macaw Ac m.m_:>v mcowpmpwccoo nmwszwcm mesmu me:w_ucoc mpamwcm> Lao; .¢_ mpnm» 84 cm. w~.n om. FF. Fm. m—wa ow. ¢m. ¢o.u mo. mm. co. co. co.: mm. mm. mxom mpcww .m> .z\»xu:pcm¥ we. mo.u mp. NF. mm. mxom xxuspcwx :memmm vapcma emcee ecoomm pumccm p:_on pomwmm mucmu_mmm pumwwm mumc¢ cm_a mmmprou new cowpmazouo m.cm;pmu scrumpwccou cmuco oLmN mm. mp. m_.u Fq. Pv. mFwa mP. mo. w—.- Pu. mm. mxom xxosucmx chpcmu m¢.- m—. N—. mm. mm.u mpgww n_.n FF. mm. m_. mo.u Po. mm. mo. Fo.- my. mxom mrcwu cmmwzowz Luna: or. —o.- we. mp. wF. mxom zxuzucmx :memmz _mwpcmm cmuco ucoomm bummed pcwoa pumwmm mumgw uommmm mmm_u cm_a mmmppou new womFa mucmvwmmc Paco; :owpwpmccou emcee ocmN .A.u.p=ouv ¢_ m_aac 85 MF. em. ee. mm. meggma Lmuge ucoumm me. me. me. me. pummmm pcwoe PF.- me.- me. me.- pummmm mumee Ne. NP. mF. my. pummmm mmmfie a. a. e. e. 3.3%”..Wmuofifi cormmchcoe emcee ogmN km. me. om. a“. aMwmcmmlmmuuenumqmmw me. me. me.. m_.u pummwm ucwoe —~.u we.u me. me. pummmm mucmuwmmm me. eF. RN. me. ummmmm mmm_e em. we. ee. Fe. cape memppoo ecm xcmc mange .mfluwm .wxmm .mflmww .mxmm cowmemLLoe emcee oLmN .m> .3\xxo:pcmx xxuzucmx :cmpmmm .A.e.p=oue e_ m_nah 86 at 0.2 Ame $3 0.; :3 9% $3 3: a: 33 83 0.3 $4; «.00 m—Lwe a: 0.2 4m: 0.0m 3xozucm¥ Fmeucme a: 0.8 E 3: as 0.3 E 0.3 S: 0.8 CV 33 m? C: :3 2:0 :3 2% E 0.02 E 0.: 2: SS 0.: E 0.8 S: 0.2 a: 0.02 mxom cmmwgowz cmmem :3 :3 3: 3.0m E a: 0.3 500 at :3 5% NE :3 3: N00 0.30 83 $3 mam gm :3 3: 500 N33 :2 $3 0.; 0.5 83 3% I: :0 a a xxmapcmx ccmpmmz :3 .mocmewmme m0 commmc 0:0 .xmm .400; mumce .mompa mocmevmme FmooF .mspmum chowummzouo m.gm;pmm >0 mmmFF00 00 :0 mewoe ucmogmm F0000: 304 Fmgmm :e P00005002 304 chsm 2: pmzcmz new: chzm :e szcmecoz gee: Fmezm e: —0::02 304 _0g:L-coz :e _0:cmEcoz 304 chsgucoz c: P0000: new: Fmeaeucoz 2e P00005002 sew: ngsgucoz mzpmpm .uue mmemce mmcmevmmm m.emcp0m .mp m—nm» 87 3330 30030 3.03 0.33 330 3030 0.03 0.03 30030 3000 0.00 0.00 3000 3030 0.00 0.03 3000 3000 0.03 3.30 3030 3030 0.00 0.30 3300 300e 0.00 0.33 3300 3300 0.00 0.00 .mflmflm .mxmm .0> .333xu0mcmx 030 0.03 303e 0.03 33330 3.00 30Ne 0.00 3330 0.03 330 3030 0.30 3030 0.03 mpewe mxom 0xozpcm¥ :Lmummm Aze 300:0: Aze 000005002 32v F0000: Aze 303005002 32v 00:00: 320 000005002 32v 000:0: 32V 300005002 mspmum .uoe 0.000003 304 chzm 304 emcee 003: 30000 new: _0L:m 304 Pmezcucoz 304 30303-002 :00: Fmgseucoz new: Focacncoz memce mocmewmmm .3.0.0000e 03 03003 88 For example, neither regional setting nor intra-regional residence place influence the plans of academically superior nonmanual boys. Cross-regionally, over 75 percent of these boys expect to go to college. Also consistent cross-regionally for boys is the effect of a "low" grade standing and/or l'1ow" social class rank on educational plans. In other words, lower class boys and/or boys with poor academic records are less likely to pursue a college education than seniors from higher class thresholds and/or high academic standing. Table 14, however, also presents evidence of regional variability in the four variable model.' The effect of local residence place on educational plans, for example, varies by region. There are fewer chances for upward educational mobility in rural Appalachia (and to a less extent Central Kentucky) compared with the rural sectors of Western Kentucky and Upper Michigan. The relationship is especially apparent among boys who perform poorly in school. For girls, regional effects are manifested for all sub- populations except upper class, non-rural girls who perform well in school. This cohort displays remarkably similar educational ambitions. Among most other classifications, senior girls from the Appalachian areas exhibit lower ambitions than their counterparts in the more affluent Kentucky and Michigan regions. To conclude this chapter we will briefly focus our attention solely on an intra-Appalachian (i.e., coal versus non-coal) comparison of career plans. An Intra-Appalachian Note Throughout this research we have generally found the two Appalachian study "regions" manifesting very similar career plan configurations. However, as stated in Chapter 2, the non-coal area 89 of Eastern Kentucky and the coal area of West Virginia and Kentucky represent two rather distinctive socio-economic modalities within what is generally recognized as "Appalachia." Our purpose here is to probe for patterned regularities and systemmatic differences in the structuring of career ambitions between high school seniors living in the predominantly coal area and seniors residing in the non-coal area. In general, the "opportunity structure" and the balance be- tween educational and work opportunities in Appalachia are not con- ducive for upward social mobility. The educational aspirations and expectations of all but the most talented and ambitious upper class youth are affected by the marginal economic structures and the norm— ative climate associated with a rural/poverty environment. Young people form manual/blue collar home are in an especially disadvantageous position when it comes to planning and implementing occupational and educational career goals. They have limited access to the normative and material resources that facilitate high educational attainment. Furthermore, the chances of finding viable unskilled or semi-skilled employment in the home area are indeed slim. Given the structuring of educational and work opportunities for lower class Appalachian youth, then, it is not surprising that a large proportion of these young people expect to leave the home area in search of employment. Compared with boys, lower class Appalachian girls are even less likely to formulate high career aspirations in the home region. Our data indicate that boys seem to have "first call" on both educational resources and available jobs. Not surprisingly, the regional out- migration rates is higher for girls than boys. 90 Having briefly examined commonalities in career planning con- figurations for coal and non-coal youth, let us know examine patterned differences aming various intra-Appalachian populations. For ex- ample, nonmanual/rural coal boys are significantly less likely to pursue a college education and are also less likely to leave the home region than their non-coal counterparts. As noted earlier, we suspect that the recent revitalization of the mining industry in the coal area has created a more favorable occupational opportunity structure and consequently has provided a realistic alternative to continuuing on in school and/or eventually migrating in search of work. Intra-Appalachian regional differences are also evident for senior girls. Specifically, upward educational mobility for Eastern Kentucky/non-coal girls (especially those from upper class homes) is contingent upon "good" grades. In the coal area, on the other hand, with few exceptions, college ambitions are not as tightly tied to academic status in the high school. It appears that the reemergence of coal mining as a viable economic enterprise provides a material resource base upon which ”coal" girls, regardless of academic credentials, can plan college. In the non-coal/subsistence agricultural area, on the other hand, economic resources are less available and when they are allocated to girls, seem to be given to the most talented, nonmanual segment of the population. In conclusion, our findings seem to indicate that living in a rural/poverty environment has a dual effect on the process of career development: 1) it dampens the college ambitions of all but the most academically talented and economically advantaged young people; and 2) it amplifies propensity to migrate among all 91 subpopulations that have limited access to job opportunities in the home region. Furthermore, our data suggest that career planning models, at least in the two Appalachian regions surveyed here, are relatively unaffected by modal economic structures, per se. The patterning of career ambitions seems to be more a function of the balance between educational and work opportunities within a region. Thus, although the coal and non-coal areas have different socio-economic bases, their "opportunity structures" are very similar and subsequently, career planning models closely resemble each other. 92 FOOTNOTES Chapter 3 . Again it should be restated that "plan to migrate" implies a move out of the home region and does not include intra-regional relocation. . A change in plans by two girls from ”low" to "high" would increase the percentage planning college from 52.0 percent to 60.0 percent. . The village of Ontonagon, county seat and largest town in the area, has a population of only 2,432 (1972). . Researchers have suggested the importance of other determinants of educational aspirations. For example, parental attitudes toward education were probed by Sewell and Shah (1968), Schwarzweller and Lyson (1974). Value orientations were explored by B. C. Rosen (1956) and Alan B. Wilson (1959). Finally, the influence of peer group culture was examined by Alexander and Campbell (1964) and James A. Davis (1966). Research dealing with the interrelationship of regional circumstnaces and factors such as those mentioned above is clearly needed. . It should be reemphasized that under certain multiple control conditions, cell size becomes extremely small, and in some cases vanishes entirely. Thus, in certain instances, certainty of validity and reliability concerning the magnitude of associations may be low. . Appendix A contains complete computations for the four variable model. Results obtained by employing either grades or residence as the zero order independent variable are consistent with the three variable analysis. . Again the interesting case of Upper Michigan girls should be noted. Lower class girls from this area manifest unusually high levels of college ambition. Evidence from other studies in Ontonagon County (1957 and 1974) suggest that these results are both valid and reliable. CHAPTER IV SUMMARY AND CONCLUSIONS This research was designed to explore the impact of regional contexts on the career planning configurations of rural high school seniors. We attempted to link social-psychological determinants of career ambitions to the more macro socio-cultural variables that characterize regional situations. In other words, regional circum- stances and the balance between educational and work opportunities within a region were viewed as specifying conditions in the career development sequence. Specifically, our concern was with the manner by which (1) social class origin, (2) academic performance in school, and (3) local residence setting combine to form regionally distinct career plan "models." SUMMARY OF FINDINGS The main empirical findings are summarized in outline form as follows: 1) Seniors from the two Appalachian areas (coal and non-coal), on the whole, manifest lower college expectations than seniors from Western Kentucky, Central Kentucky, or Upper Michigan. II) Proportionately more seniors from the Appalachian regions and from Upper Michigan plan to migrate away from their home areas than seniors from Western and Central Kentucky. III) In Western and Central Kentucky and Upper Michigan, migrational plan is more tightly linked to college plan than in Appalachia. Conversely, the proportion of Appalachian seniors migrating without planning on a college education is much higher than in 93 94 the other three study regions. IV) The zero order correlations between educational plan and father's occupational status, grade rank, and local residence place are, as follows: A) Father's occupational status and educational plan: 1) Father's occupational status is positively associated with plans for college for all segments of the study pop- ulation, except Upper Michigan girls. 2) Taking father's occupational status into account, absolute percentage differences reveal that Appalachian seniors manifest consistently lower levels of educational ambition than their regional counterparts. B) Grade rank and educational plan: . 1) Grade achievement is substantially correlated with educational plan among all regional study populations. 2) Taking grade rank into account, absolute percentage dif- ferences reveal that Appalachian seniors at all grade ranks are less likely to plan on attending college than their regional counterparts. C) Non-rural residence (e.g., city or town) is positively as— sociated with college plans for all areas except Upper Michigan. V) The zero order correlations between migrational plan and father's occupational status and intra-regional residence place are, as follows: A) Father's occupational status and migrational plan: 1) In Appalachia and Upper Michigan the association between father's status and migrational plan are low or negative. 2) In the two Kentucky regions the associations between father's 95 status and migrational plan are moderately high. B) Local residence place and migrational plan: 1) Non—rural residence is positively assocaited with plans to migrate in Western and Central Kentucky. 2) In Appalachia and Upper Michigan, because of very heavy out- migration, clear patterns do not emerge. VI) The relative and independent effects noted in Model 1 (social class, grade rank, and educational plan) are, as follows: A) Both social class origin and grade rank exert independent in- fluence on college plans in all regions. B) Correlations between grades and educational plans are rela- tively stronger for Appalachian boys and relatively weaker for Appalachian girls compared to other regions. C) In Appalachia the supressing effect of either a low social class standing gr_poor grades gr_both on college plans are more dramatic than in the other study regions. VII) The relative and independent effects noted in Model 2 (social class, local residence place, and educational plan) are, as follows: A) In all regions, local community context has a negligible effect on the association between class and college plan. B) The original association between local residence place and college plan disappears among seniors in Western and Central Kentucky and, likewise, for Appalachian girls when class is introduced as an intervening variable. C) The original association between local residence place and college plan persists or increases in Upper Michigan and for Appalachian boys when class is introduced as a control variable. 96 VIII) The relative and independent effects noted in Model 3 (social class, local residence place, and migrational plan) are, as follows: A) Both social class and intra-regional residence place have independent effects in Central and Western Kentucky. 1) Nonmanual/non-rural seniors are the most likely to migrate from Western and Central Kentucky. 2) Manual/rural seniors are those least likely to migrate from Western and Central Kentucky. B) In Upper Michigan and the two Appalachian areas class/ residence place distinctions among seniors expecting to migrate are relatively unimportant. IX) The relative and independent effects noted in Model 4 (social class, grade rank, local residence place, and educational plan) are, as follows: A) Social class exerts a rather strong and independent effect on educational plans in all regions, even when the effects of. grade achievement and intra-regional residence place are controlled. B) "Low" grade rank and/or ”low” social class origin dampen educational ambitions in all regions. C) Career planning patterns for boys reveal: 1) Local residence place and regional setting are relatively unimportant in influencing the plans of academically gifted seniors from upper class homes. 2) Demonstrated academic competence is relatively more im- portant vis-a-vis educational chances in Appalachia and Central Kentucky than in Western Kentucky and Upper Michigan. 97 3) Seniors in rural_Appalachia, regardless of their grade achievement or class background, exhibit lower educational expectations than their counterparts in the other regions. 0) Career planning patterns for girls reveal: 1) Only upper class/non-rural seniors who perform well in school are unaffected by regional pressures. 2) Appalachian seniors, regardless of their qualifications or background, generally manifest lower ambitions than their regional counterparts. X) An intra-Appalachian comparison of career planning configurations revealed few patterned differences between the coal and non-coal areas. To briefly summarize: empirical findings from Tables 7 -- 16 clearly supported our exploratory hypotheses. The relative importance and predictive power of social class and/or grades and/or local residence place in a career planning model was contingent upon the distribution of opportunities within a given region. Although correlation coefficients often masked important relative inter- regional percentage differences, in general, regional effects were noted in zero order as well as multiple control relationships. Our research reached its greatest complexity in Models 3 and 4 (Tables 12 -— 15) where the influence of multiple career planning determinants were untangled and interpretted. GENERAL CONCLUSIONS One finding central to the raison d'etre of this thesis merits careful attention. It concerns a basic component of the career plan configurations: the allocation and distribution of educational chances cross-regionally. To meaningfully interpret our findings, however, it is necessary to return to the literature 98 for a moment and lay some additional groundwork. Specifically, we must examine recent developments pertaining to issues of social equality/inequality, achievement and the structuring of educational and occupational opportunities. Seymour Martin Lipset hits at the heart of an ideological dilemna facing America today when he details the tension between the values we place on achievement and the somewhat contradictory values we place on equality. According to Lipset: We believe all persons must be given respect simply because they are human beings: we believe that the differences between high -- and low -— status people reflect accidental and perhaps temporary variations in social relationships .... Achievement is a corollary to our belief in equality. For people to be equal, they need to have a chance to become equal .... Achievement is a function of equality of opportunity. (Lipset 1967: 2) Milner states the relationship in a slightly different manner: If anything equality is a derivative of achievement rather than vice versa. Our committment to achievement is primary, and our committment to equality is in large measure a result of the former. At the very least, achievement is on a par with equality. Moreover ... equality of opportunity_is probably best understood as a mechanism for compromising and reconciling the contradictory aspects of equality and achievement. (Milner 1972: 12) Most Americans resolve the tension between two partially contradictory basic ideals (achievement and equality) by affirming a belief in "equality of opportunity." The American social and econ- omic system is oriented toward selecting and rewarding the most competent and talented individuals. (e.g., Scarce educational resources are allocated to the best academically qualified.) The system rests on the notion that, all things being equal, high status attainment in the educational and occupational sector is dependent Upon how hard an individual "works." That is, achievement and 'Euetting ahead" are seen primarily in terms of individual effort. 99 Again, Milner makes the point well: If people were given an equal chance at the beginning, their final destination would be determined by how hard they chose to work. How hard one worked was seen as a thing which each in- dividual could control. Therefore, it was perfectly fair to reward those who chose to put out the most effort. (Milner 1972: 14) This image of social justice, based on equality of opportunity, assumes an individual posses a high degree of freedom to choose among the alternatives confronting him. In reality, however, an individual's chocies are not free, but rather are dependent upon his location in the social environment. For example, in the quest for college, a young person's aspirations and expectations are influenced by his social class origin. High SES youth posess material and normative advantages not generally found in lower class homes. Additionally, the larger social context (e.g., region) in which determinants operate must also be reckoned with. Not all regional mileaus are "equal." Social and economic conditions vary from one area or region to another and may act to dampen (or exaggerate) the effects of structural determinants such as class, grades, etc. on educational plans. Having taken a cursory look at the structuring of social opportunities in American society, we have set the stage to add our findings to this body of theory and speculation. A tabular form of the Lorenz curve (Table 16) best illustrates the degree of regional inequalities in the distribution of educational opportunities. This form of measurement allows us to define a society that is completely free of regional inequalities, a society where gross regional dis- parities exist, or any degree of variation in between. In general, Appalachian high school seniors manifest lower educational ambitions than their counterparts in the heavily lOO 0.003 0.003 0.003 0.000 0.000 0.003 0.003 0000000 30003 .wqu .wqu 0.03 0.00 .mqwa .qum .mqmfl .00 .0330000000 0.33 0.03 0.00 0.30 0.00 0.30 0.33 30000000 0000000 0.00 0.00 0.03 0.03 3.03 0.00 0.30 30000000 3000000 0.00 0.30 0.00 0.00 0.30 0.03 0.00 00000002 00000 30.00 30.00 30.00 33.00 33.00 33.00 30.00 30000000 0000003 wauflm 0.000 0.003 0.003 0.003 0.000 0.003 0.000 0000000 00003 0404 .mqmfl .quH .3430 .0400 .mqm3 .0400 .00 .3330000000 0.00 0.00 0.30 3.30 0.00 0.00 0.00 30000000 0000000 0.00 0.00 0.03 0.03 0.00 0.00 0.00 30000000 0000000 3.00 0.00 0.03 0.00 0.00 0.00 0.00 0000000: 00000 30.33 30.00 30.00 30.30 30.03 30.00 30.00 30000000 0000003 F0000 00000-002 Mama .mmwm ~0=c0z P00005002 .flmm .mxmm m00_0 mocmewmmm 0:01 m000e 0:000m chowummmuoe 0.000000 .0000000000 0000000 0m0000m00 00000m0 000 0000m0 30 00000 000 0300 000 00m000000 P000000000m -- 0m>00e Ncm0040 .ep m—nmh lOl industrialized Western Kentucky area, the commercial farming are of Central Kentucky, or the copper/pulp area of Upper Michigan. With few exceptions, regional inequalities persist even when father's occupational status, grade achievement, or local residence place are considered. College expectations are especially depressed for manual status, low achieving, or "rural" coal and non-coal boys and for manual status or "rural" coal girls. These findings suggest that Appalachian high school seniors are ranked at or near the bottom of an unequal system of opportunities, which, ability and social origin notwithstanding, adversely effects their aspirations, expectations, and ultimately their life chances. The interesting case of Central Kentucky girls must also be noted. Compared with other manual status, low scholastic achiev- ing girls, or with girls living in the predominantly rural areas, Central Kentucky girls manifest extremely low educational ambitions. Reasons for this pattern point toward a differential assessment of regional career opportunities. It is generally assumed that girls from blue collar homes, girls who perform poorly in school, or girls from rural areas have relatively limited normative and/or material resource bases upon which to plan college. Consequently, they generally opt for manual or low status white collar jobs in or around the home area. In Appalachia and Upper Michigan, however, employment op- laortunities for girls are very limited. Economic necessity thus 'Forces many of these less favored girls to reorder their career priorities and consider college as a "real" career alternative. In Western Kentucku, on the other hand, extensive rural industrialization seems to enhance the quest for higher education. 102 The expectation patterns for all sub—groups of Western Kentucky seniors are high. 0f the regions surveyed here, only in the commercial agri- cultural area of Central Kentucky does it appear that educational plans are formed somewhat independently of the regional economic situation. That is, educational and occupational choices are less ”determined" by the economic realities of the region. Rather, we believe, family and school based pressures are "freer" to influence plans. The relatively low educational expectations of manual status or poor achieving or rural Central Kentucky girls, compared with those in the other study regions can be explained by looking at characterisitics of the larger (regional) economic system. Blue- grass girls form their plans in a mileau free of those socio-economic pressures and conditions that tend to increase or inflate expectation patterns. Overall, these findings suggest that, ability and social origin notwithstanding, when regional circumstances are considered, equality of educational opportunity is more myth than reality. Educational ambition is dependent not only upon an individual's position in the social order, but also upon the needs and socio- economic character of the immediate regional environment. In other words, the effects of social class origin or school achievement on educational ambitions are shaped by conditions unique to a given region. An additional set of findings dealing with career development and notions of "equality of opportunity" must be examined. These findings concern the relationship of regional opportunity structure and propensity to migrate. To aid in this discussion, we will again employ a tabular form of the Lorenz curve. l03 Table l7 clearly shows regional imbalance in migration plans. Appalachian seniors and seniors from the copper/pulp area of Upper Michigan are significantly more likely to migrate than their counterparts in Western or Central Kentucky. This pattern holds regardless of social class origin or intra-regional residence place. One must conclude that Appalachian and Upper Michigan youth are quite aware of the restricted occupational environment afforded them in their home region. In order for them to enhance their life chances in the career world and subsequently utilize their skills and abilities to an optimum extent, they must consider relocating in an area with a more favorable opportunity structure. Although our data show that both Appalachia and Upper Michigan manifest very high out-migration rates, we believe that the reasons for leaving the home area may be quite different. In the economically depressed Appalachian areas families are often unable to provide financial support toward a college education and consequently there is a great pressure on young people to find work. This pressure is often felt even before high school graduation (and may be one of the major causes of the high drop out rate). In the copper/pulp area of Upper Michigan, on the other hand, college is seen as more a "preparatory internship", an added boost, before a young person is forced to find work. Socio—economic conditions in the region are more conducive for upward educational mobility than in Appalachia. Thus, although propensity to migrate is "high". college ambitions are also "high." In conclusion, although many Americans believe that career selection and advancement are based on principles of "equality of opportunity" and individual achievement, the fact remains that l04 0.000 .0400 0.00 0.00 0.00 30.00 o.oo_ 0m.m_ .000mm 0.000 .0400 0.00 0.00 0.00 30.30 0.000 0400 0.00 0.00 0.00 30.00 F0000ucoz 000—0 000000000 o.oo_ o.oo_ .0400 .0400 —.¢N 3.NN m.o_ _.w_ _.mm w.om 3m.m— 03.o_ o.oo— o.oo_ M400 .0400 ¢.NN 3.NN ¢.o— o.—N N.mm m.om 3m.m— 0m.w_ ~20ch FMDCMECOZ 00000m _0cowumazuuo 0.000000 0000000 P000» .00 .3330000000 30000000 0000000 300:0:00 300pcwu 00000002 0000: 30000000 0000003 mPLTw 0000000 00000 .0> .3330000000 30000000 0000000 30000000 3000000 00000002 0000: 30000000 0000003 0000 .mcowpwucou 0000000 000000000 0000000 000 000000 30 03000 000 0300 0ow 000000000 000000005 .. 0m>0zu 0000000 .3_ 03003 l05 ascribed attributes such as social class origin rather rigidly structure a person's choices and options. Furthermore, patterns of career development are shaped and molded by the needs of the regional social system. The regional context provides a framework of alternatives, a framework that encourages or discourages individual orientations. Patterns of behvior, then, can be seen as affected by in- fluences at two social levels: the individual (e.g., social class origin, academic achievement, etc.) and the contextual (e.g., community, region, society, etc.). The interrelationship of factors and conditions at both levels helps to delimit the range and scope of an individual's opportunites and ultimately his life achievement. Only when we understand the intricate and complex interaction between social structure and social context can we hope to formulate theoret- ically sophisticated and universally valid career planning models. LIMITATIONS OF STUDY AND SUGGESTIONS FOR FUTURE RESEARCH Certain limitiations in this study should be noted here so that future research on the problems of career development may be strengthened. One of the more important decisions we had to make in organ- izing this research was in defining and delimiting regional areas. For analytical purposes we considered a rural county or group of counties sharing similar structural/ecological configurations in- dicative or representative of a regional modality or setting. Region was viewed in terms of uniquely patterned opportunity structures and the balance between educational and work opportunites in the immediate area. There are other ways of conceptualizing and operationally l06 defining region, of course, that may prove more useful in a study of this sort. Two alternatives come to mind. First, rather than defining region in terms of socio-economic/opportunity structures, emphasis could be placed on identifying and measuring normative climates. One could, for example, relate a youth's career plans and perceived life chances to the prevailing climate of opinion held by parents, teachers, peer groups, and other significant others in a region. Second, one might examine "regional” patterns of school organization. Here the concern would be with the sensitivity of the school system to the developmental needs of the young people in the area. Factors such as the nature and type of school admin- istration, curriculum content, school financing, guidance counselling, etc. could be correlated with educational and migrational plans. A second limitiation of the study concerns the number of "regions" surveyed. Ne tapped only five sets of regilnal circumstances; numerous other mileaus in the United States, each characterized by a somewhat distinctive "opportunity structure", could be identified. Although the aim of this research is not to formulate a model for every identifiable region, the inclusion of additional regional settings would aid the search for universal generalizations by in- creasing the validity and reliability of our findings. Also, potentially helpful statistical techniques, such as analysis of covariance which allows the researcher to untangle the effects of a specific social structure (e.g., region) after controlling for individual inputs, can only be usefully applied in research employing more regions than we studied here. Along this line, all of the regions we examined were drawn 107 from the United States. Variations in the organization of educational systems, methods of selection for higher education (e.g., contest versus sponsored systems), etc., were not taken into account. We believe that it would be very interesting and highly useful to re- plicate this study in different societal contexts where structural differences in the type of educational institutions and the nature of academic selection exist. Another limitation of the study involves the social class variable. Because data for this study were collected in two un- coordinated projects, father's occupational status is the only indicant of social class origin that can be utilized with some degree of cross-regional validity. Unfortunately, a measure of this sort is closely tied to regional socio-economic circumstances. For example, a manual worker (e.g., miner) in the coal fields of Appal- achia posesses certain values, a atandard of living, and a life style that is undoubtedly different, in some ways, from a manual worker (e.g., miner) in the copper/pulp area of Upper Michigan. In future studies it would be helpful to run a parallel analysis using indicants of social class that are somewhat independent of regional context (e.g., a combined scale of father's educational status and level of living). THEORETICAL IMPLICATIONS This study attempted to elaborate and clarify the importance of regional context as a conditional variable in the formation of career plans among rural high school seniors. Our research problem was somewhat atypical, especially for American sociology, where the cross-regional approach is seldom utilized. Implicitly or explicitly, sociologists in the United States seem to operate on 108 the premise of a "mass society." Although replication studies may be specifically designed to "tailor" some piece of sociological knowledge to local situational circumstances and thus elaborate hypotheses formulated from research in other areas of the United States, rarely are studies designed to anticipate regional variations in institutional configurations, socio-economic circumstnaces or culture. Because sociological research in the U. S. is not generally organized in the comparative vein, but rather as research seeking universal (e.g., national) generalizations about human social behavior, data necessary to explain regional deviations away from commonly accepted hypotheses is rarely made available. Thus, according to Schwarzweller: We assume, even though our judgements often run counter to common sense, a certain functional uniformity or culture or of social circumstances within a geographic area. Currently, of course, we are paying more attention to regional variations, as for example, between the South and the North, between Appal- achia and the North Central States. Nevertheless, American Sociology tends to cling to the operational assumption of a "mass society” ... (Schwarzweller 1968: 18) It is somewhat of a bewilderment to note that, although the regional concept was first utilized successfully as a research tool in the early l900's by Galpin and others, 60 years later it still has not gained wide spread recognition or acceptance as an important aid in reporting and interpretting sociological knowledge. For example, findings from a study of rural young people, say in Wisconsin, are accepted with few qualms as being generally valid and applicable elsewhere in the United States. In the present career plan research, regional setting was treated as a supra-control variable. Regional opportunity structures (i.e., the balance between educational and work opportunities in a lO9 region) were viewed as specifying conditions which mediated the im- mediate pressures exerted by the family and school on educational and migrational plans. Clearly, our findings show that regional variations exist in the manner in which plans are formed and goals implemented. In short, we assume that differing institutional configurations, normative climates, and socio-economic circumstances on the regional level lead to different styles of ambition and the manner by which talent is selected to fill high status roles in the larger society. Based on our research results, we suggest that the social context thesis, elevated to the level of regional analysis, merits further attention by American sociologists. It is incumbant upon the social researcher to be aware of and account for the effects of variations in regional social structure on the hypotheses, models, and theories he is testing. The usefulness and applicability of sociological generalizations is dependent upon specifying all con- ditions that account for a given behavior pattern. Clearly, region is one such specifying condition. PRACTICAL IMPLICATIONS Implicit in this research is a conception of the United States as a system of functionally interdependent regions. Although related to the needs of the larger national society, each study area is characterized by somewhat unique and particular socio—economic circumstances and institutional configurations. The circumstances and configurations are representative or indicative of "opportunity structures" which can facilitate or inhibit a young person's career ambitions and life chances. From a practical standpoint, educational and developmental llO programs aimed at helping young people make intelligent career decisions must reckon with, and be tailored to, the social conditions of the home region. Educational administrators, guidance counsellors, and other school officials should institute programs and curriculi that are sensitive to the intra-regional balance between educational and work opportunities and that prepare young people for satisfying and financially rewarding work careers in the larger society. For example, vocational education programs organized to meet regional and/or national labor market needs should be emphasized. Students must be made aware of and trained in the skills that are readily transferrable to available jobs. Too often vocational offerings exist in fields with limited employment possibilities. Along this line, it is important that Federal funds for vocational education be equitably distributed on an inter-regional basis. Branscome, for instance, notes: “Although Appalachia has l3 percent of the national enrollment in secondary vocational education, it receives only 7.3 percent of the federal funds available." (Branscome l972: 226). More importantly, perhaps, the intricate and complex interrelatiOnships among regional context, locus of control of educational resources, and the structuring of career ambitons must be explored. In research that examines the above interrelationships from a cross-cultural perspective (Norway, Germany, and the United States), Schwarzweller notes: In American society, where emphasis is placed on local authority and local resources in developing educational facilities, rural industrialization enhances the quest for higher education and appears to coincide with the opening-up and modernizing of educational facilities. On th other hand, in areas of limited economic opportunity such as Appalachia, educational opportunities are also limited and, as a result, the upward educational mobility lll ambitons of lower class youth are markedly depressed. Since industrialization means that the entire system of opportunities is pushed upwards, the net effect of rural industrialization in American society may foster greater regional inequalities. (Schwarzweller l973: l56) Planning at the regional level that coordinates and supplements the efforts of state and local authorities is clearly needed. Young people from depressed or isolated rural areas such as Appalachia must obtain the knowledge and skills that allow them to compete on a par with young people from other more "affluent" areas. 1. 112 FOOTNOTES Chapter 4 See for example research on career plans, status attainment, etc., by Sewell, Haller and Strauss (1957), Sewell (1964), Sewell and Armor (1966), Sewell and Orenstein (1965), Burchinal (1961, 1962), and Michael (1961). Although the above researches have made im- portant and significant theoretical and methodological contributions to American Sociology, the extent of cross-regional validity and reliability of their findings generally remains unanswered. BIBLIOGRAPHY BIBLIOGRAPHY Alexander, Norman C. and Ernest Q. Campbell, "Peer Influences on Adolscent Aspirations and Attainments,” American Sociological Review, 29 (August 1964), pp. 568-75. Anderson, w. A., "Rural Sociology as Science," Rural Sociology, XII (Winter 1947), pp. 347-56. Bertrand, Alvin L., Rural Sociology, New York: McGraw-Hill Book Company, Inc., 1958. Bishop, Carol, et. al., “Vocational and Educational Goals of Rural Youth in Virginia," Bulletin No. 568, Blacksburg, Virginia: Virginia Polytechnic Institute, Agricultural Experiment Station, 1964. Blau, Peter, “Structural Effects," American Sociological Review, XXV (April 1960), pp. 178-98. Bogie, Donald w., "Sociocultural Differences Among Three Areas in Kentucky as Determinants of Educational and Occupational Aspirations and Expectations of Rural Youth," unpublished Ph.D. dissertation, University of Kentucky, 1970. Bogue, Donald J., "Economic Areas as a Tool for Research and Planning," American Sociological Review, XV (Spring 1950), pp. 409-16. Boocock, Sarane S., "Toward a Sociology of Learning: A Selective Review of Existing Research," Sociology of Education, XXXIV (Winter 1966), pp. 1-40. Boyle, Richard P., "Community Influence on College Aspirations: An Empirical Evaluation of Explanatory Factors," Rural Sociology, XXXI (September 1966), pp. 277-92. Branscome, James, "The Crisis of Appalachian Youth," in Appalachia in the Sixties, David Walls and John B. Stephenson (eds.), Lexington: University of Kentucky Press, 1972, pp. 224-31. Burchinal, Lee G., "Differences in Educational and Occupational Aspirations of Farm, Small Town and City Boys," Rural Sociology, XXVI (June 1961), pp. 107-21. Burchinal, Lee G., "Career Choices of Rural Youth in a Changing Society," Bulletin No. 548, Minneapolis: University of Minnesota Agricultural Experiment Station, 1962. Burchinal, Lee G., et. al., "The Money to Get Started: What do Farm Boys Expect?" Iowa Farm Science, 17 (August 1963), pp. 3-5. 113 114 Davis, James, "The Campus as Frog Pond," American Journal of Sociology, 72 (Winter 1966), pp. 568-75 Davis, James, Elementary Survey Analysis, Englewood Cliffs: Prentice Hall, 1971. Edwards, Alba M., "Comparative Occupational Statistics for the United States," Washington, D. C.: U. S. Government Printing Office, 1943. Elder, Glen H., "Family Structure and Educational Attainment: A Cross- National Analysis," American Sociological Review, XXX (February 1965), pp. 82-96. Empy, L. T., "Social Class and Occupational Aspirations: A Comparison of Absolute and Relative Measurement," American Sociological Review, XXI (December 1956), pp. 703-09. Fisher, Joseph L., "Concepts in Regional Economic Development," Papers and Proceedings of the Regional Science Association, 1 (1955), W2-W20. Fleege, Urban H. and Helen J. Malone, "Motivations in Occupational Choice Among Junior and Senior High School Students," Journal of Educational Psychology, XXXVII (1946), pp. 77-86. Galpin, Charles, in Country Life Conference, University of Wisconsin Bulletin Series No. 472, General Series 308, Madison (February 1911), quoted in John H. Kolb and Edmund De S. Brunner, A Study of Rural Society, Fourth Edition, New York: Houghton Mifflin, 1952. Gregory, C. L., "Advanced Techniques in Delineation of Rural Regions,” Rural Sociolggy, XIV (March 1949), pp. 59-63. Hall, R. B., "The Geographic Region: A Resume," Annals of the Association of American Geographers, 25 (1935), pp. 122-36. Hawley, Amos H., Human Ecology, New York: The Ronald Press Co., 1950. Heberle, Rudolf, "Regionalism: Some Critical Observations," Social Forces, XXI (March 1943), pp. 281-86. Herskovits, Melville J., "The Cultre areas of Africa," Africa, 3 (1930), pp. 59-77. Isard, Walter, "Regional Science, the Concept of Region, and Regional Structure," Papers, The Regional Science Association, II (1956), pp. 13-26. Isard, Walter, "The Scope and Nature of Regional Science," Re ional Science Association: Papers and Proceeding§, Vol. 6 (1960), pp. 9-34. 115 James, Preston E., "Towards a Further Understanding of the Regional oncept," Annals of the Association of American Geographers, 42 (1952), pp. 195-222. Kahl, Joseph A., “Educational and Occupational Aspirations of 'Common Man' Boys," Harvard Education Review, XXIII (Summer 1953), pp. 186-203. Kaplan, Paul F., "The Region as a Unit of Analysis: Emerging Consensus," Cornell Journal of Social Relations, 1 (October 1966), pp. 135-54. Kroeber, Alfred L., "The Culture-Area and Age-Area Concepts of Clark Wissler," in Methods in Social Science, Stuart A. Rice (ed.), Chicago: University of Chicago Press, 1931, pp. 248-65. Kroeber, Alfred L., "Cultural and Natural Areas of Native North America," University of California, Publications in American Archeology and Ethnology, 68 (1939), pp. 1-242. Kroeber, Alfred L., “Cultural Area," in Encyclgpedia of the Social Sciences, E. R. Seligman (ed.), New York: The MacMillan Company, 1941, pp. 646-47. Kuder, Frederic G., Kuder Preference Record Vocational, Chicago: Science Research Associates, Inc., 1960. Linz, Juan J., "Ecological Analysis and Survey Research," in Quantitative Ecological Analysis in the Social Sciences, Mattei Dogan and Stein Rokkan (eds.), Cambridge: The M.I.T. Press, 1969, pp. 91-132. Lipset, Seymour Martin, "Social Mobility and Urbanization," Rural Sociology, XX (September-December 1955), pp. 220-28. Lipset, Seymour Martin and Reinhard Bendix, Social Mobility in Industrial Society, Los Angeles: UCLA Press, 1960. Lipset, Seymour Martin, The First New Nation, Garden City, N. Y.: Doubleday, 1967. Loomis, Charles P. and J. Allan Beegle, Rural Social Systems, New York: Prentice Hall, 1950. Mangus, A. R., Rural Regions of the United States, Washington: U. S. Government Printing Office, 1940. Marsh, Robert M., Comparative Sociology, New York: Harcourt, Brace and World, 1967. Mathews, Robert G. and Lawrence W. Drabick, "Reasons for Selection of Expected Occupations by Race and Sex," Educational Research Series, Raliegh, N. C.: North Carolina State University Department of Agricultural Educationa and Rural Sociology, No. 7 (November 1965), pp. 10-14. 116 Michael, John A., "High School Climates and Plans for College," Public Opinion Quarterly, XXIV (Winter 1961), pp. 585-95. Milner, Murray Jr., The Illusion of Equality, San Francisco: Josey- Bass Inc., 1972. Murdock, George Peter, Social Structure, New York: MacMillan, 1949. Nelson, Joel I., "High School Context and College Plans: The Impact of Social Structure on Aspirations," American Sociological Review, XXXVII (April 1972), pp. 143-48. Odom, Howard W., American Regionalism, New York: Henry Hold and Co., 1938. Odom, Howard W., Folk, Region and Society, Chapel Hill, N. C.: University of North Carolina Press, 1964. Prezworski, Adam and Henry Teune, The Logic of Comparative Social Inquiry, New York: John Wiley and Sons, 1970. Radcliffe-Brown, A, “On Social Structure," Journal of the Royal Anthropological Institute, 70 (1940), pp. 1-12. Reiner, Thomas A., "The Region as a Unit of Social Analysis and Its Place in Development Planning,“ in Social Development, Manfred Stanley (ed.), New York: Basic Books, 1972, pp. 170-202. Rieger, Jon H., J. Allan Beegel and Phillip N. Fulton, "Profiles of Rural Youth: A Decade of Migration and Social Mobility," East Lansing, Michigan: Michigan Agricultural Experiment Station, Research Report 178, 1973. Rogoff, Natalie, "Local Social Structure and Educational Selection," in Education, Economy and Society, A. H. Halsey, J. Floud and C. A. Anderson (eds.), Glencoe, Il1inois: Free Press, 1961. Rosen, B. C., "The Achievement Syndrome: A Psychocultural Dimension of Social Stratification," American Sociological Review, 21 (April 1956), pp. 203-11. Schapera, I., "Some Comments on the Comparative Method in Social Anthropology," American Anthropologist, 55 (1953), pp. 353-62. Schuessler, Karl, "Covariance Analysis in Sociological Research," in Sociological Methodology 1969, Edgar F. Borgatta (ed.), San Francisco: Josey-Bass Inc., 1969, pp. 219-44. \ Schwarzweller, Harry K., "Sociocultural Factors and the Career Aspirations and Plans of Rural Kentucky High School Seniors," Progress Report No. 94, Lexington, Kentucky: Agricultural Experiment Station, September 1960. 117 Schwarzweller, Harry K. and James S. Brown, "Education as a Cultural Bridge Between Eastern Kentucky and the Great Society," Rural Sociology, XXVII (Fall 1962), pp. 357-73. Schwarzweller, Harry K., "Community of Residence and Career Choices of German Rural Youth," Rural Sociology, XXXIII (March 1968), pp. 46—63. Schwarzweller, Harry K., "Comments on the Comparative Method," unpublished paper presented at the Second World Congress of Rural Sociology, The Netherlands, August 1968. Schwarzweller, Harry K., James S. Brown and J. J. Mangalam, Mountain Families in Transition, University Park, Pennsylvania: The Pennsylvania State University Press, 1971. Schwarzweller, Harry K., "Regional Variations in the Educational Plans of Rural Youth: Norway, Germany and the United States," Rural Sociology, XXXVIII (Summer 1973), pp. 139-58. Schwarzweller, Harry K. and Thomas A. Lyson, "Social Class, Parental Interest and the Educational Plans of American and Norwegian Rural Youth," Sociology of Education, 47 (Fall 1974), pp. 443-65. Sewell, William H., A. O. Haller and Murray A. Strauss, "Social Status and Educational and Occupational Aspirations," American Sociological Review, XXII (February 1957), pp. 67-73. Sewell, William H., "Community of Residence and College Plans," American Sociological Review, XXIX (February 1964), pp. 24-38. Sewell, William H. and Alan Orenstein, "Community of Residence and Occupational Choice,’l American Journal of Sociolqu, LXX (March 1965), pp. 551-63. 1 Sewell, William H. and Michael Armor, "Neighborhood Context and College Plans," American Sociological Review, XXXI (April 1966), pp. 159-68. Steward, Julian H., Area Research: Theory and Practice, New York: Social Science Research Council, Bulletin 63, 1950. Steward, Julian H., "Region -- An Heuristic Concept," Rural Sociology, XIX (Fall 1955), pp. 297-98. Taylor, Carl C., Rural Life in the United States, New York: A. A. Knopf, 1949. Turner, Ralph, The Social Context of Ambition: A Study of High School Seniors in Los Angeles, San Francisco, San Francisco: Chandler Publishing Company, 1964. 118 Vance, Rupert P., "The Regional Concept as a Tool for Social Research," in Regionalism in America, Merril Jensen (ed.), Madison, Wisconsin: University of Wisconsin Press, 1951, pp. 119-42. Warren, Roland L., The Community in America, Chicago: Rand McNally and Company, 1963. Whittlessey, Derwent, et. al., "The Regional Concept and the Regional Method," in Inventory and Prospect, Preston E. James and Clarence F. Jones (eds.), Syracuse: Syracuse University Press, 1954, pp. 21-68. Williamson, Jeffrey G., "Regional Inequality and the Process of National Development: A Description of Patterns," Economic Development and Cultural Change, XIII (July 1965), pp. 3—84. Wilson, Alan B., "Residential Segregation of Social Classes and Aspirations of High School Boys," American Sociological Review, 24 (December 1959), pp. 836-45. Wissler, Clark, The American Indian, New York: Oxford University Press, 1922. U. S. Bureau of the Census, Census of Population: 1970 General Social and Economic Characteristics, Final Report PC(1) C-l9, Kentucky; C-24, Michigan; C-50, West Virginia; Washington D. C.: U. S. Government Printing Office, 1972. Zimmerman, Carle C., Outline of American Regional Sociology, Cambridge, Mass.: Phillips Book Store, 1947. APPENDIX Complete Computations for Four Variable, Non-linear Causal Analysis S=Residence 58.3.1.1- tied X differs 3 variable 0(SX '\/ tied T differs 3 variable - Q(SY) tied T differs 3 variable Q(TX tied S differs 3 variable 0 TY) tied S differs 3 variable tied S differs 3 variable WESTERN KENTUCKY BOYS APPENBIX T=Grades X=Social Class Y differs tied 3 variable .037 -.O46 -.002 .195 .086 .144 .107 .013 .063 Y differs tied 3 variable .372 .282 .323 .357 .323 .341 363 .301 .332 X differs tied 3 variable .254 .102 .165 .267 .126 .188 .261 .114 .177 Y differs tied 3 variable .495 .156 .336 .511 .178 .355 .503 .167 .346 X differs tied 3 variable .676 .633 .650 .663 .672 .668 .669 .651 .659 ' T differs tied 3 variable .596 .529 .565 .606 .542 .577 .601 .535 .571 Y=Col1ege Plan Second order partial-- X effect -133 Y effect .084 Joint effect -024 Second order partial T effect .041 Y effect -090 Joint effect-~056 Second order partial T effect ‘023 X effect ‘15] Joint effect—,010 Second order partial S effect .022 Y effect .339 Joint effect-.005 Second order partial S effect .039 X effect .042 Joint effect-.051 Second order partial S effect .013 ' T effect .067 Joint effect .002 46 O .282 .102 .156 .633 .529 S=Residence Q(ST tied X differs 3 variable Q(SX) ~— tied T differs 3 variable Q(SY) —-——— tied T differs 3 variable filw'v\ ‘\.O‘! tied S differs 3 variable Q(TY) tied S differs 3 variable Q(XY) tied S differs 3 variable T=Grades X: Y differs tied -.128 -.142 .214 -.178 .042 -.157 Y differs tied .596 .428 .504 .457 .546 .441 X differs tied .559 .149 .426 -.033 .486 .049 Y differs tied .462 .178 .476 .013 .469 .089 X differs tied .695 .702 .653 .653 .673 .679 T differs tied .594 .551 .675 .663 .637 .613 CENTRAL KENTUCKY BOYS 120 APPENDIX Social Class 3 variable -.135 .064 -.O43 3 variable .515 .486 .500 3 variable .352 .195 .266 3 variable .354 .299 .325 3 variable .699 .653 .676 3 variable .575 .670 .627 Y=College Plan Second order partia1-. X effect-.035 Y effect .014 Joint effect .377 Second order partial T effect .029 Y effect .169 Joint effect-.121 Second order partial T effect-.182 X effect .410 Joint effect .049 Second order partial S effect'-164 Y effect ~410 Joint effect ~178 Second order partial S effect-.049 X effect-~007 Joint effect .008 Second order partial S effect .112 T effect .043 Joint effect-,030 142 .428 .149” .178 .702 .551 121 APPENDIX S=Residence T=Grades X=Socia1 Class Q(ST) Y differs tied 3 variable tied .365 -.424 .131 X differs .562 .162 .439 3 variatfle .473 -.094 .301 9.5x) v differs tied 3 variable tied .876 .684 .778 T differs .793 .809 .798 3 variable -825 -735 .789 Q(SY) x differs tied 3 variable T tied .695 .523 .624 differs .692 .472 .592 3 variable .693 .489 .604 Q(TX) v differs tied 3 variable tied .559 .155 .426 S differs .619 .240 .502 3 variable .585 .190 .458 91111 x differs tied 3 variable tied .857 .789 .808 S differs .790 .876 .828 3 variable .828 .806 .814 Q(XY) T differs tied 3 variable tied .602 . 81 .561 S differs .733 .666 .708 3 variable .659 .569 .627 EASTERN KENTUCKY BOYS Y=College Plan Second order partia1-. X effect .587 Y effect .789 Joint effect-.389 Second order partial T effect -125 Y effect -193 Joint effect'-209 Second order partial T effect—.051 X effect .172 Joint effect .047 Second order partial S effect .086 Y effect .405 Joint effect-.026 Second order partial S effect .087 X effect .068 Joint effectr-154 Second order partial S effect-185 ' T effect -121 Joint effect? 053 424 .684 .523 .155 .789 .481 122 APPENDIX S=Residence T=Grades X=Socia1 Class Q(ST) Y differs tied 3 variable tied .257 -.O93 .116 X differs .533 -.135 .352 3 variable .391 -.108 .219 Q(SX Y differs tied 3 variable tied .651 .585 .613 T differs .704 .495 .648 3 variable .686 .554 .631 Q(SY X differs tied 3 variable tied .664 .497 .567 T differs .636 .423 .526 3 variable .646 .453 .541 91:29 v differs tied 3 variable tied .398 .079 .278 S differs .560 -.046 .396 3 variable .496 .019 .345 OLD/.1 X differs tied 3 variable tied .845 .750 .781 S differs .861 .712 .784 3 variable .855 .733 .783 Q(X )_ T differs tied 3 variable tied .461 .397 .438 S differs .644 .616 .634 3 variable .570 .526 .555 Y=College Plan Second order partia1-, X effect—.041 Y effect .351 Joint effect .317 Second order partial T effect-.090 Y effect .066 Joint effect .144 Second order partial T effect-.074 X effect .168 Joint effect .046 Second order partial S effect-~125 Y effect ~320 Joint effect ~287 Second order partial S effect—,o37 X effect ,095 Joint effect .054 Second order partial S effect..219 T effect .064 Joint effect-.036 KENTUCKY/W. VA. BOYS 093 .585 .497 .079 .750 .397 123 APPENDIX S=Residence T=Grades X=Socia1 Class Q(ST v differs tied 3 variable tied -.158 ---- - 090 X differs .597 .449 .534 £85.19 Y . differs tied 3 variable T tied .595 .859 .744 differs .955 .857 .914 3 variable .803 .858 .830 96.1.1). X differs tied 3 variable T tied .111 -.168 -.101 differs .642 -.201 .076 3 variable .416 -.l84 -.013 Q(TX_ Y differs tied 3 variable tied .667 .321 .542 S differs .552 .306 .448 3 variable .620 .314 .501 Q(TY) X differs tied 3 variable tied .885 .711 .780 S differs .866 .692 .749 3 variable .877 .702 .766 (XY T differs tied 3 variable tied .781 .565 .696 S differs .687 .515 .614 3 variable .742 .543 .661 UPPER MICHIGAN BOYS Y=College Plan Second order partial ---- X effect -449 Y effect‘-158 Joint effect -306 Second order partial .859 T effect_.002 Y effect_.253 Joint effect .351 Second order partial-.158 T effect-,o33 X effect .279 Joint effect .564 Second order partial .321 S effect-.015 Y effect .346 Joint effect-.099 Second order partial .711 S effect-.019 X effect .174 Joint effect-.001 Second order partial .565 S effect-.049 ‘ T effect .217 Joint effect--045 S=Residence T=Grades 9&1; Y differs tied -.008 X differs .271 3 variable .116 Q(SX)_ Y differs tied .443 T differs ~388 3 variable .413 Q(SY X differs tied -186 T differs .326 3 variable .262 917:: Y differs tied .706 S differs .698 3 variable .702 91111 X differs tied .766 S differs .731 3 variable .748 9099 T differs tied -702 S differs ~724 3 variable ~713 124 APPENDIX X=Socia1 Class tied 107 .097 031 tied 367 .346 .359 tied 108 .116 .112 tied 481 .480 .480 tied 676 .615 .649 tied O7 .463 .484 3 variable -.053 .204 .054 3.462iab16 .372 .388 3 variable .142 .209 .178 3 variable .624 .614 .619 3 variable .711 .667 .690 3 variable .616 .604 .610 WESTERN KENTUCKY GIRLS Y=College Plan Second order partia1-. X effect .204 Y effect .098 Joint effect .076 Second order partial T effect--021 Y effect .076 Joint effect-~034 Second order partial T effect .007 X effect -078 Joint effect .132 Second order partial S effect—.00] Y effect .225 Joint effect-.008 Second order partial S effect-.062 X effect .089 Joint effect .027 Second order partial S effect--044 T effect .195 Joint effect .066 107 .367 .108 .481 .676 .507 S=Residence Q(ST _—-—..—- tied -.108 X differs 3 variable 9_§ll_ tied T differs 3 variable ;§LL LO tied T differs 3 variable 0(TX) tied S differs 3 variable Q(TY) tied S differs 3 variable . ENAX). tied S differs 3 variable . T=Grades 125 APPENDIX X=Social Class tied 3 variable -.203 -.159 .009 .296 -.125 .057 tied 3 variable .660 .711 .527 .662 .606 .688 tied 3 variable .220 .448 .044 .381 .128 .412 tied 3 variable .549 .661 .364 .559 .450 .603 tied 3 variable .710 .744 .667 .700 .691 .721 tied 3 variable .454 .606 .653 .731 .574 .681 Y=College Plan Second order partia1-, X effect .211 Y effect .094 Joint effect .341 Second order partial T effect-.132 Y effect .096 Joint effect .108 Second order partial T effect-.176 X effect .407 Joint effect .192 Second order partial S effect_.185 Y effect .186 Joint effect .110 Second order partial S effect-,043 X effect .085 Joint effect..025 Second order partial S effect .199 T effect .285 Joint effect-.138 CENTRAL KENTUCKY GIRLS 203 .660 .220 .549 .710 .454 S=Residence Q(ST) tied X differs 3 variable Q(SX) tied T differs 3 variable 9(5 tied T differs 3 variable MIX.) tied S differs 3 variable Q(TY tied S differs 3 variable Q(XY) tied S differs 3 variable 126 APPENDIX T=Grades X=Socia1 Class Y differs tied 3 variable .198 .165 .181 .432 .426 .430 .278 .234 .257 Y iffers tied 3 variable 305 .249 .276 .293 .238 .270 .299 .245 .273 X differs tied 3 variable .149 .002 .054 .310 .080 .159 .223 .039 .103 Y differs tied 3 variable .396 .143 .284 .596 .389 .509 .457 .214 .351 X differs tied 3 variable .770 .501 .577 .799 .561 .643 .779 .516 .595 T iffers tied 3 variable 302 .340 .321 .454 .399 .425 .348 .360 .354 Y=College Plan Second order partial X effect .262 Y effect .034 Joint effect-.028 Second order partial T effect--011 Y effect -055 Joint effect-~001 Second order partial 1 effect .078 X effect .147 Joint effect .083 Second order partial S effect .246 Y effect .252 Joint effect-.045 Second order partial S effect .060 X effect .269 Joint effect-.031 Second order partial 5 effect -059 T effect'-039 Joint effect -094 EASTERN KENTUCKY GIRLS .165 .249 .002 .143 .501 .340 127 APPENDIX S=Residence T=Grades X=Social Class Q(S ) Y differs tied 3 variable tied -.223 -.199 -.209 X differs .104 -.345 -.O72 3 variable-.056 -.250 -.150 Q(SX) Y differs tied 3 variable tied .593 .517 .558 T differs .597 .438 .535 3 variable .595 .485 .548 9.81.. x differs tied 3 variable tied .547 .129 .345 T differs .418 .015 .221 3 variable .486 .075 .286 Q(TX) Y differs tied 3 variable tied .489 .244 .383 S differs .368 -.048 .205 3 variable .421 .092 .286 QiTY) X differs tied 3 variable tied .639 .572 .598 S differs .575 .464 .521 3 variable -503 .526 .560 sum T differs tied 3 variable tied .611 501 .549 S differs 623 .696 .662 3 variable .618 .605 .611 KENTUCKY/W. VA. GIRLS Y=College Plan Second order partial -.199 X effectr.l46 Y effect-.024 Joint effect .473 Second order partial .517 T effectr.078 Y effect .076 Joint effect -083 Second order partial T effect-.114 X effect .418 .129 Joint effect .015 Second order partial .244 S effect-.292 Y effect .245 Joint effect.l71 Second order partial .572 S effectr.108 X effect .067 Joint effect.-044 Second order partial .501 S effect .195 T effect -110 Joint effect~183 128 APPENDIX S=Residence T=Grades X=Socia1 Class 91.31). Y ' differs tied 3 variable tied -.278 -.011 -.176 X differs .351 .671 .500 3 de~iable".091 .244 .046 91.10 Y differs tied 3 variable ” tied .221 .259 .245 1 differs -.181 -.256 -.216 3 variable .023 .083 .057 Q(SY)- x differs tied 3 variable tied -.221 -.062 —.480 T differs .096 -.480 -.309 3 variable-.065 -.539 -.393 ‘81.). Y differs tied 3 variable tied -.O84 .067 -.O31 S differs ---— -.O73 -.O34 3 variabie-.050 -.006 -.032 Q(TY) X differs tied 3 variable tied .853 .585 .708 S differs .681 .637 .650 3 variable .783 .615 .678 9110 T differs tied 3 variable tied -.143 .167 -.010 S differs -.064 -.015 -.O39 3 variable-.111 .078 -.023 UPPER MICHIGAN GIRLS Y=College Plan Second order partial-.011 X effect -682 Y effect-~257 Joint effect ~053 Second order partial .245 T effect.,515 Y effect-.039 Joint effect .114 Second order partial-.480 T effect .122 X effect .381 Joint effect .194 Second order partial .067 S effect-.140 Y effectr.151 Joint effect -224 Second order partial .585 S effect .052 X effect .268 Joint effect-.224 Second order partial .167 S effect-.182 T effect-.310 Joint effect .261 "‘11111111111111“