The Educational and Early Occupational Attainment Process: A Social Psychological Model
Critique of Blau and Duncan's (1967) Path Model
Blau and Duncan's Basic Model (1967:165–172):
Described the occupational attainment process of American adult males.
Initial Stratification Variables:
Father's educational attainment status.
Father's occupational attainment status.
Behavioral Variables:
Individual's completed educational level.
Prestige level of individual's first job.
Dependent Variable:
Person's occupational prestige position in 1962. (OccAtt in 1962)
Variance Explained by the Model:
The model accounted for a notable portion of the variance in educational attainment.
A significant portion of the variance in first job attainment was explained.
The model explained a substantial proportion of the variance in 1962 level of occupational attainment.
Additions: Various additions (nativity, migration, farm origin, subgroup position, marriage, assortative mating) were presented but none significantly improved the model.
Observations and Critiques by Sewell, Haller, and Portes:
Omission of Psychological Inputs: While father's stratification position is relevant for occupational prestige attainment, the model lacked psychological inputs.
Specifically, mental ability, a frequently referenced factor in differential occupational attainment literature (Lipset and Bendix, 1959:203-226; Sewell and Armer, 1966).
Note: This gap was partially filled later by Duncan (1968).
Omission of Social Psychological Mediating Factors: The model did not include social psychological factors that mediate the influence of input variables on attainment.
These factors are suggested by speculative theory and concrete social psychology research.
Examples of such intervening variables:
Reference groups (Merton, 1957:281-386).
Significant others (Gerth and Mills, 1953:84-91).
Self-concept (Super, 1957:80-100).
Behavior expectations (Gross et al., 1958).
Levels of educational and occupational aspiration (Haller and Miller, 1963; Kuvlesky and Ohlendorf, 1967; Ohlendorf et al., 1967).
Experiences of success or failure in school (Parsons, 1959; Brookover et al., 1965).
Advantages of a Social Psychological Model
Plausible Causal Argument: An explanation of a behavior system requires a plausible causal argument, not just path coefficients between temporally ordered variables.
Social psychological mediating variables can provide this possibility, as indicated by Duncan's (1969) work.
The Blau-Duncan model fails to explain why connections between father's education/occupation and respondent's education, first job, and 1962 occupation should be expected.
Social psychological positions agree that cognitions and motivations (knowledge, self-concept, aspirations) develop in structured situations (including others' expectations) and that actions (attainments) result from these orientations combined with situational factors.
Identifies Intervention Points: A valid social psychological model can suggest new points in the causal system where interventions can change attainment behaviors.
The Blau and Duncan volume did not address this issue.
Variables like the expectations of significant others offer possibilities for manipulating outcomes, including educational attainments.
Increased Variance Explanation: Such a model might further explain the variance in dependent variables (educational and occupational attainment).
The Proposed Social Psychological Model
Problem Statement: This report extends previous attempts (Sewell and Armer, 1966; Sewell and Orenstein, 1965; Sewell and Shah, 1967; Sewell, 1964; Haller and Sewell, 1967; Portes et al., 1968; Haller, 1966; Haller and Miller, 1963; Miller and Haller, 1964; Sewell et al., 1957) to apply social psychological concepts to explain variations in educational and occupational attainment.
Assumptions of the Model:
Certain social structural (initial stratification position) and psychological (mental ability) factors affect the influences of significant others on youth and the youth's self-observations of ability.
The influence of significant others, and possibly self-estimates of ability, affect the youth's levels of educational and occupational aspiration.
Levels of aspiration affect subsequent levels of educational attainment.
Educational attainment, in turn, affects occupational attainment.
All effects are assumed to be linear.
Social psychological variables perform only mediating functions.
Nature of the Model: Provides a logically consistent social psychological model with a plausible causal argument linking stratification and mental ability inputs through social psychological and behavioral mechanisms to educational and occupational attainments.
Compelling Feature: Some inputs can be manipulated through experimental or other purposive interventions, allowing for future experimental testing and practical policy applications.
Eight Causal Variables:
X_1: Occupational Prestige Level attained by the adult person (OccAtt - Occupational Attainment).
X_2: Educational Level previously attained (EdAtt - Educational Attainment).
X_3: Occupational Prestige Level aspired to as a youth (LOA - Level of Occupational Aspiration).
X_4: Level of Educational Aspiration as a youth (LEA).
X_5: Influence for educational achievement exerted by significant others while in high school (SOI - Significant Others' Influence).
X_6: Quality of academic performance in high school (AP - Academic Performance).
X_7: Level of family in the stratification system (SES - Socioeconomic Status).
X_8: Mental Ability as measured while in high school (MA).
Causal Order of Variables:
X7 (SES) and X8 (MA) precede all other variables.
X5 (SOI) and X6 (AP) precede both aspirations and attainments.
X6 (AP) is generally assumed to precede X5 (SOI).
Youthful aspirations (LOA, LEA) obviously precede later educational and occupational attainments.
Pre-adult educational attainments (EdAtt) precede adult occupational attainments (OccAtt).
Hypothesized Causal Linkages (Figure 67.1):
Unanalyzed Correlation: A low positive relationship exists between mental ability (MA) and parents' socioeconomic status (SES).
Academic Performance (AP):
Anticipated substantial effect of MA on AP.
Debatable Path: A possible direct influence from SES to AP suggests that teachers might be influenced by parental prestige leading to a direct effect on grades, though this is not adequately demonstrated.
Significant Others' Influence (SOI):
SOI is theorized to be controlled by AP, SES, and exogenous factors.
Definition: Significant others are specific persons from whom an individual derives aspirations, either as models or through communicated expectations (Woelfel, 1967).
This term is preferred over "reference group" as it focuses on specific individuals rather than necessarily collectives.
Importance shown by research (Sherif, 1935; Herriott, 1963).
Hypothesized substantial direct influence from SES to SOI.
Hypothesized substantial indirect effect of MA on SOI via AP, because significant others base expectations partly on demonstrated abilities (MA through AP to SOI).
Levels of Aspiration (LOA, LEA):
Major effects of SOI on attainment are hypothesized to be mediated by its effects on aspirations.
Influence from SOI to LOA.
Influence from SOI to LEA.
LOA and LEA are highly related, as education is often seen as a prerequisite for high occupational attainment.
Educational Attainment (EdAtt):
Hypothesized pronounced effect of LEA on EdAtt.
Debatable Path: A possible direct influence from SOI to EdAtt suggests a direct influence, particularly for college education during late high school.
No reason to include a path from SOI directly to occupational attainment (as SOI refers to late high school influences).
The entire effect of LOA on occupational attainment is expected to be expressed through EdAtt.
Occupational Attainment (OccAtt):
A direct effect of LOA on OccAtt is hypothesized.
No hypothesized path from LOA to EdAtt (beyond its relationship with LEA).
Path Coefficient Calculation Strategy:
Out of 26 possible paths, 8 (or 10 including dotted lines) were hypothesized.
Due to the non-rigorous nature of the theoretical model at this stage, all 26 possible relationships were investigated.
Prediction: Hypothesized relationships should show considerably stronger influences than unhypothesized ones.
This approach also allows for the discovery of important unhypothesized causal lines.
Methodology
Sample: All high school seniors in Wisconsin in 1957.
Data Collection: An extensive questionnaire in 1957 (educational/occupational aspirations, related topics). Follow-up in 1964 by Sewell to collect data on later educational and occupational attainments from a one-third random sample.
Subjects for Analysis: 929 males for whom data were available at both 1957 and 1964, and whose fathers were farmers in 1957.
Statistical Analysis: Relationships between variables were computed on all 929 cases (with missing data acceptance). More detailed analysis was based on 739 cases with complete data on all variables.
Note: Initial relationships between variables for both 929 and 739 cases were practically identical.
Variable Operationalization:
Level of Occupational Attainment (X_1 - OccAtt): Measured by Duncan's (1961) socioeconomic index of occupational status.
Level of Educational Attainment (X_2 - EdAtt): Dichotomized based on 1964 data: at least some college education vs. no college education at all.
Level of Occupational Aspiration (X_3 - LOA): Duncan’s (1961) socioeconomic index scores assigned to the occupation the respondent desired to fill in the future (from 1957).
Level of Educational Aspiration (X_4 - LEA): Dichotomous variable reflecting the respondent’s 1957 statement of whether they planned to attend college.
Index of Significant Others’ Influence (X_5 - SOI): A summated score (range: zero to three) of three dichotomized variables from 1957:
Youth's report of perceived parental encouragement for college.
Youth's report of perceived teacher encouragement for college.
Respondent's statement on whether most close friends planned to go to college.
Rationale: These components reflect the same conceptual dimension (education-oriented influence) and were positively related, confirmed by factor analysis.
Quality of Academic Performance in High School (X_6 - AP): Self-reported rank in high school class, coded into 10 percentile ranges (e.g., top 10\%, next 10\%, etc.).
Socioeconomic Status (X_7 - SES): Based on Hollingshead’s (1957) Two-Factor Index, using father’s occupation and father’s education (1957).
Mental Ability (X_8 - MA): Raw score on the Henmon-Nelson Test of Mental Ability (1957).
Summary of Main Ideas
Critique of Blau and Duncan's (1967) Path Model: The original model for occupational attainment of American adult males, while explaining some variance in educational attainment, first job, and 1962 occupational attainment, was criticized for omitting crucial psychological inputs (like mental ability) and social psychological mediating factors (such as reference groups, significant others' influence, self-concept, and aspirations). These omissions limited the model's ability to provide a plausible causal argument for observed connections.
Advantages of a Social Psychological Model: Incorporating social psychological variables offers significant benefits. Such a model provides a more plausible causal explanation for how stratification variables influence attainment, moving beyond mere correlations. It also helps identify specific intervention points within the causal system where changes could alter attainment behaviors, and potentially explains a greater proportion of the variance in educational and occupational outcomes.
The Proposed Social Psychological Model: This alternative model integrates social structural (Socioeconomic Status - X7) and psychological (Mental Ability - X8) factors with social psychological mechanisms to explain educational (X2) and occupational (X1) attainment. Key mediating variables include Academic Performance (X6), Significant Others' Influence (X5), and Levels of Educational (X4) and Occupational (X3) Aspiration. The model assumes linear effects and posits that social psychological variables primarily serve mediating functions, leading to a testable framework for understanding and potentially influencing attainment processes.
Conclusion
The critique of Blau and Duncan's (1967) path model highlighted its significant omissions, particularly the lack of psychological and social psychological mediating factors essential for a comprehensive understanding of occupational attainment. While the original model successfully explained a portion of the variance in educational attainment, first job, and 1962 occupational attainment, its inability to provide a plausible causal argument limited its theoretical depth. The proposed social psychological model directly addresses these shortcomings by integrating social structural (Socioeconomic Status - X7) and psychological (Mental Ability - X8) inputs with crucial mediating variables such as Academic Performance (X6), Significant Others' Influence (X5), and Levels of Educational (X4) and Occupational (X3) Aspiration.
This revised framework offers several advantages: it establishes a more plausible causal chain from initial stratification positions to educational (X2) and occupational (X1) achievements, moving beyond mere correlations. Crucially, it identifies specific intervention points within the causal system, such as manipulating the influence of significant others, which could potentially alter attainment behaviors and inform practical policy applications. Furthermore, the model holds the promise of explaining a greater proportion of the variance in dependent variables. The empirical results from the Wisconsin high school senior sample (1957-1964) largely supported the hypothesized strong influences, with Academic Performance significantly impacting Significant Others' Influence and Mental Ability strongly affecting Academic Performance. Socioeconomic Status also showed a direct influence on SOI. This initial validation underscores the potential of a social psychological
The main arguments presented in the notes are: The critique of Blau and Duncan's (1967) path model for omitting crucial psychological (mental ability) and social psychological (significant others' influence, aspirations) mediating factors, which limited its causal explanatory power in occupational attainment. The advantages of a social psychological model, which offers a more plausible causal explanation, identifies intervention points for changing attainment behaviors, and potentially explains more variance in educational and occupational outcomes. The proposed social psychological model integrates social structural (SES) and psychological (Mental Ability) inputs with key mediating variables like Academic