Hubert M. Blalock

Conceptualization and Measurement in the Social Sciences.

Introduction Fundamental Measurement, Scaling, and Dimensionality Issues The Comparability of Measures Categorical Variables, Conceptualization, and Comparability Some Implications of Omitting Variables from Causal Explanations The Confounding of Variables and Oversimplified Interpretations Aggregation and Measurement Error in Macro-Analyses Appendix Assessing Comparability with Multiple Indicators

The Identification Problem and Theory Building: The Case of Status Inconsistency

The identification problem arises when there are too many unknowns for solution. Identification can be achieved in one-way causal situations by making assumptions about the error terms, but these are often unrealistic. Even this procedure breaks down when exact mathematical relationships are assumed among some of the variables, as is the case where status inconsistency (or social mobility) is taken as a difference between two statuses. In such situations one must make additional restrictive assumptions; otherwise the separate effects of the component factors cannot be isolated. In the case of status inconsistency, identification may be achieved by: (1) making a priori assumptions regarding the signs or magnitudes of some of the coefficients; (2) finding and measuring additional exogenous variables that appear in some but not all of the equations; or (3) postulating certain kinds of nonlinear relationships.

Four-Variable Causal Models and Partial Correlations

Simons method for making causal inferences from correlational data is applied to the various possible four-variable causal models. Prediction equations are given for forty-one models, so that the goodness of fit of any particular model can easily be evaluated without the use of tedious computations. Certain suggestion are also made for handling problems involving a larger number of variables. The major purpose of the paper, however, is to investigate what happens to the correlations between two variables when controls are made for variables which are causally related to these variables in different ways.

Multiple Indicators and the Causal Approach to Measurement Error

Since the practical utility of the kind of multiple indicators approach discussed in Costners paper depends on the ways it can be broadened to include a diversity of causal situations, the purpose of the present paper is to extend the argument in three respects: (1) to show that it holds generally in any recursive system; (2) to note the circumstances under which a single indicator of one or more of the variables can be used; and (3) to point out that the use of multiple indicators can be combined with an instrumental-variables approach that has been discussed in the econometrics literature.

Basic Dilemmas In The Social Sciences.

The Challenge Complications produced by Multiple Causation Further Ambiguities in the Real World Can we move from many Facts to Fewer Lawlike Propositions? The Simplification Process Dilemmas and Strategies Some Defects in Our Intellectual Culture Some Positive Steps

The Real and Unrealized Contributions of Quantitative Sociology

There are real trends in the quality of data analysis, owing to the computer revolution, interdisciplinary borrowing, and the availability of large data sets. Less obvious are improvements in the quality of data collection, conceptualization and measurement, or the ignoring of data gaps. Nor has the quality of our training or our students improved. The result is an increasing communication gap and failure of practices to keep pace with the technical literature. A case for quantification is made. Ideally, hidden assumptions are made explicit, common sense can be clarified and refined, systematic search procedures developed, intractable problems located, and new theoretical insights obtained. But there are disciplinary obstacles to improving training, increasing the quality of graduate students, improving editorial policies, and promoting on the basis of quality rather than quantity of publications. We need to face this quality problem head on.

Evaluating the Relative Importance of Variables

atically deviate from this sequence because of the source of capital (government) which influences the commitments of appointive trustees (commissioners), and the character of the administrative tasks. The scheme presented here, when used in conjunction with the concept of operative goals, may provide a tool for analyzing the dynamics of goal setting and goal changing in all complex organizations.

Status Inconsistency and Interaction: Some Alternative Models

Since an identification problem arises in attempts to test a status inconsistency theory, one cannot associate an inconsistency effect in any simple way with the empirically determined interaction term. The existence of interaction can, however, be taken as weak evidence in favor of an inconsistency effect, provided that alternative explanations for interaction can be eliminated. Several mathematical models of inconsistency are explored,with a view to finding implications of rather weak assumptions regarding the directions and order relationship among main and inconsistency effects.

Understanding Social Inequality: Modeling Allocation Processes.

Why Study Allocation Processes? Properties of Allocated Goods General Factors Influencing Allocation Decisions Allocator Decision Processes Eligibility Pools Decision Processes in Sequential Pools Reactive Processes Equity and Distributive Justice Toward an Agenda for Research

Probabilistic Interpretations for the Mean Square Contingency

Abstract The purpose of this paper is to interpret θ2, the mean square contingency, as a function of probabilities associated with proportional prediction and in so doing to provide some additional evidence on the difficulty of giving useful operational interpretations to measures of association based on chi-square.