While this dilemma has received considerable attention in studies of socioeconomic mobility across generations, it is also a central issue for intra-generational mobility, that is, the study of mobility through individuals' careers. Career selections take many forms, and they occur at various times. Selections occur when individuals enter the work force, and subsequent selections may periodically occur throughout their work careers. How are individuals' opportunities affected by the selections occurring at various times? This question stresses the importance of timing, and it asks not only which factors have effects, but also when they have their effects.
On a priori grounds, we have little basis for knowing how the conflict between opportunity and inequality is settled. Societal norms permit inequality of outcomes even as they prescribe equality of opportunity; and they fail to specify how the conflict can be reconciled. Moreover, social science theories do not specify how the conflict can be resolved. Economic theories, such as human capital theory, are built on assumptions of open opportunity which deny structural constraints, while structural theories contend that selections severely limit opportunity.
Although the conflict between opportunity and inequality exists throughout society, this conflict is likely to be particularly acute in a single organization. Careers are likely to be patterned by several aspects of organizational structures: authority and status classifications associated with jobs, hierarchical shapes and vacancy chains (White 1970a), and specific personnel policies about careers. However, the countervailing influence of opportunity norms is also likely to be stronger within organizations; indeed, company "cultures" strongly emphasize opportunity norms. Moreover, since attainments are easier to discern in organizations, individuals can more easily evaluate whether an organization is living up to these norms than would be possible in large labor markets. As a result, structural inequalities and opportunity norms may both be stronger within organizations than they are in larger society, and the conflict between opportunity and structure may be even more acute within organizations than it is in larger society.
By studying how this conflict is handled in a large corporation, this investigation combines two important fields of sociological interest- stratification and organizations-and it permits significant extensions of central issues in contemporary stratification research. Work carried out since 1970 has focused renewed attention on the structural proper-ties of occupational mobility (Sorensen 1977; Vanneman 1977; Breiger 1981, 1982) and on more fine-grained classifications of occupations (Stolzenberg 1975). However, even when occupations are divided into 435 detailed categories, the resulting categories are not very homogeneous: Nearly two-thirds of income variation is still within occupational groups (Jencks et al. 1972, pp. 226). Moreover, for individuals in many occupations, changes to other occupations are rare, and the main kind of advancement is within an occupational or organizational hierarchy (Lipset and Bendix 1952; Spilerman 1977; Spenner 1981). As a result of such findings, the study of structural effects on attainment has begun to focus on firms (Kanter 1977; Baron and Bielby 1980). By considering specific jobs in organizations, the researcher is able to investigate the compensation processes by which actual jobs and job classifications become related to earnings and the matching processes by which individuals become allocated to jobs (Granovetter 1981). This study extends this new thrust, and, by having information over several historical periods, it seeks to investigate the dynamic features of organizational structures and their effects on economic outcomes.
In addition, rather than studying attainments at a single point in time, this is a study of individuals' attainment over time: the study of careers as coherent entities of sociological interest. Careers have received increased attention since 1970. Some studies have suggested that psychological changes may be affected by organizational structure, but this research has rarely described such career structures in detail (Hall 1976; Van Maanan 1977; Schein 1978; Bailyn 1980). Internal labor market theory also describes distinct segments in organizations that offer different career opportunities, but these segments describe very gross distinctions (Doeringer and Piore 1971). For instance, the primary sector contains many different segments which offer different career lines, some of which even preclude further advancement (Spilerman 1977, p 583). Status attainment research provides some descriptions of intragenerational mobility (Blau and Duncan 1967; Jencks et al. 1972, 1979), but it ignores some important aspects of careers:
Jobs in that literature are treated implicitly as independent entities rather than as linked components of coherent trajectories,. . . the models are insensitive to the fact that changes in status and earnings ... are to some extent "scheduled," . . . [and that] education gets converted into status and earnings not in some vague or random manner but in relation to career lines. (Spilerman 1977, pp. 584-585)
To understand the ways that attainments change over time, it is necessary to study the properties of careers as coherent entities: how and when they are affected by background factors like education and how these relationships are affected by changes in historical circumstances.
This study describes attainments in terms of the detailed job classifications in an organization and analyzes the ways that individuals' careers evolve and are affected over time by various background factors and previous attainments. It investigates the dynamics of the social structure: how the process of matching individuals with jobs changes over different periods of a career and over different historical circumstances (Granovetter 1981). Moreover, as Spilerman notes, "career lines constitute an intermediary structure, relating the behaviors and life chances of individuals to 'macro' organizational units, thereby linking two important levels in sociological analysis" (1977, p. 586).
