Search

In a paper in this Journal, Watson (1961) has discussed the use of group screening methods for detecting effective factors. The methods are analogous to those suggested by Doff man (1943) for the medical problem of detecting a rare defect among members of a large population by using pooled blood samples. One step in Watson's derivation of the probability that an effective factor is declared significant is invalid. The purpose of this note is to derive the correct expression for this probability and show how the correction affects the subsequent theory. None of Watson's qualitative statements about the efficiency of group screening methods nor his numerical example are seriously affected. [ABSTRACT FROM AUTHOR]

The causal relationships in a two variable system are examined, using a simultaneous linear differential equation model. Emphasis is placed on the necessity for considering mutual interaction instead of looking for unidirectional causality. To demonstrate the usefulness of this approach an example is taken from the management science literature concerning the relationship between labour turnover and overtime. It is shown that, on certain assumptions, the data disconfirm a hypothesis previously thought to be confirmed. Certain informal observations made in the original paper were shown to receive confirmation from the data. [ABSTRACT FROM AUTHOR]

The "sign rule" for drawing implications from static theories is not always valid if the theory contains loops of reciprocal causation between variables. However, if the theory is the stable equilibrium of an underlying dynamic model and if it contains only positive or only negative loops, then the sign rule can freely be used to predict the relation between any pair of independent and dependent variables. The sign rule is not in general valid when a theory contains interesting positive (potentially explosive) and negative (dampening) causal loops. [ABSTRACT FROM AUTHOR]

This paper proposes a set of techniques for measuring the "importance" of each independent variable in a multivariate causal law (i.e., a law showing the combined effect of several independent variables on a single dependent vari- able). The techniques apply to both linear and nonlinear laws; when used with linear laws they reduce to the widely-used practice of using regression weights as measures of importance, as in path analysis. This paper can thus be viewed as an extension of some of the principles of path analysis to nonlinear laws. [ABSTRACT FROM AUTHOR]

Causal structures involving a dynamic relation with time lags present difficulties if approached with conventional multivariate cross-sectional techniques. It is shown that customary tests for causal structure, e. g. partial correlations, break down or yield ambiguous results. Some further implications of lagged structure, and an extension of the idea to variability in space, are discussed in the paper. [ABSTRACT FROM AUTHOR]

The viewpoints and findings of seven empirical studies of conflict within and between organizations are compared and contrasted. A distinction is made between frictional conflict within a stable organization structure and strategic conflict aimed at changing the organization structure. The importance of attitudes as causes and effects of conflict is also discussed. [ABSTRACT FROM AUTHOR]

The methodology of simulation involves a representational approach to observed phenomena. Systems theory has been, in the past, more terminology than theory. However, a logical classification of the possible representational forms available for any simulation would serve as a meta-theory in simulation methodology. This paper presents such a classification based upon an analysis of the possible kinds of relations explanation which can be used to connect elements of existential sets. Four types of relations are possible: (1) cause-effect, (2) producer-product, (3) genetic, (4) accidental. Of these, only two (the cause-effect and producer-product) have been used to construct formal representations in science: ateleological and teleological representations. This taxonomy, together with the notions of open and closed systems, provides a logical framework for any general form of a simulation. We show how familiar paradigms such as decision theory, allocation problems and stochastic processes are derived in this representational theory. [ABSTRACT FROM AUTHOR]

Economic development has occurred very unevenly, not only among countries and major parts of the world but also among communities within an economy or country. One of the central problems before economists is to account satisfactorily for the disparate rates of growth. In an attempt to explain this uneven growth, this research paper summarizes the results of a study of the long retardation, followed by a recent sharp acceleration, in the application of mechanized methods of production in the cotton region of the southern United States. It was observed that this region does not coincide precisely with the South as a cultural region, though most of the welfare problems associated with cotton production have been concentrated in portions of the Old South. Moreover, while cotton is losing its relative importance as the dominating enterprise in the South, the conditions under which it is produced still exercise a major influence in the rural parts of the region. The analysis utilizes three groupings of forces--institutional, technological, and exogenous factors in the process of development in giving an account for the non-uniform development.

The paper first examines the relations between functional and causal analysis, and concludes that a statement about the function of an institution can be made relevant to a causal analysis of the development of that same institution only (legitimately) by additional assumptions concerning human motives or evolutionary selection or (illegitimately) by use of the postulate of necessary integration. Secondly, it argues that only events can properly be related by causal laws and that questions about the functions of institutions (considered as constructs from observed events) logically imply questions about the causal effects of the events from which they are constructed. It finally seeks to classify approaches to socio- logical analysis with respect to these two theses and to indicate those differences which stem from different methodological assumptions and those which reflect different value preferences. [ABSTRACT FROM AUTHOR]

The first part of this paper discussed the ideas of a theoretical domain and a causal locus in that domain as a philosophical basis for measurement theory. This part discusses the difficulties created for measurement theory by the fact that time plays a different role in different functional forms of equations for stating causal relations in a theoretical domain. We then turn to a redefinition and critique of traditional measurement concepts in terms of our philosophical position and of our analysis of the role of time in causal theories. [ABSTRACT FROM AUTHOR]

Econometric models are used for many purposes. The chosen purpose of a model dictates, to a large extent, both the method of constructing the model and the data required for estimating its parameters. The purpose of the model described in this paper is to develop statistical relationships, which explain a number of key economic variables. In so doing, authors are able to test a number of statistical hypotheses that arise from economic reasoning. However, it is worth emphasizing at the outset that they make no attempt to estimate the parameters of any a priori structural model. No claim is made that the direction of cause-and-effect relationships can be inferred from the numerical results. For this reason, the only forecasts that can be produced from this model are those which are conditional on the assumed ex post behavior of chosen variables. One purpose of a formal model of the kind reported here is to display in a systematic manner the statistical relationships among different economic variables. Although most of the attention is focused on those associations, which are economically meaningful and statistically significant, some of the more interesting observations pertain to relationships, which are widely alleged to be significant, but which, it turns out, are not.

The article presents the reply of the author to the criticism of his paper "Causation and Theory Construction," published in the March 1972 issue of the journal "Social Science Quarterly." According to the author, researchers Herbert L. Costner and Hubert M. Blalock, invoke an empirical criterion to judge logical validity; but logic has nothing to do with the empirical truth of propositions, and that point bears directly on their rules for relating symbols. Had Blalock and Costner not confused utility and logical validity, there would be no argument. At one point Blalock and Costner indicate that a theorist need only assume close relations to justify the sign rule, but elsewhere they say "if those conditions are met." If so, the criterion ignores the point that the empirical validity of the premises may be unknowable directly. At one point Blalock and Costner undertake to describe the "mathematical form of causal models and how this form corresponds to one common notion of a causal relation." But the "mathematical form" of causal models requires no further explication; rather, the issue is correspondence between that form and a conception of causation.

Usually a speaker at a conference treats the title of his paper as an examination question to be answered, or as an excuse for the discussion of any idea that seems relevant to the occasion. Everyone works in organizations, and every students would do likewise, as teachers, social workers, administrators, personnel managers, and so on. The tasks one performs in these organizations may be divided rather arbitrarily into two kinds. First there is the daily job of teaching people, helping people, providing a service, keeping the wheels of the organization turning. Each day's work might be unique or habitual, simple or highly skilled, but everyone knows what is required of them even if they cannot always deliver the goods. They are faced with a means-ends problem, capable of rational solution. The other kind of task demands a reappraisal of daily jobs, or of the aims and methods of the organizations in which one works, or of the relation between these organizations and others. On these occasions the ends themselves are in question, one has to think where they are heading for and ask themselves whether they ought to be going somewhere else.

The paper focuses on the participation of social planners in physical development programmers and highlights the reasons underlying the successful or unsuccessful fusion of the two types of planning in practice. [ABSTRACT FROM AUTHOR]

Generally speaking, scientific propaganda analysis necessarily employs many types of criteria. These include criteria of objectivity, adequacy, generalization, interpretation, causation, prediction, and interest. Of these, criteria that are frequently taken too much for granted and that usually demand most careful statement, assessment, and restatement are those associated with interest, interest in the sense of the individual, group, and general societal objectives or purposes served. The article deals with these interest criteria. In propaganda analysis, it has been common to describe interest criteria in a general way by saying that purposes of analysts are "to help the intelligent citizen to detect and to analyze propaganda." This is not to be taken as a narrow, purely individualistic or individualism-promoting approach. On the contrary, the conception is thrown into a broader perspective, as the more clear-eyed and intelligent citizens are, the more they can detect and understand actual issues at stake in a social agitation or conflict.

The challenge to the principle of causality in other disciplines has been taken over by some sociologists, who have criticized its utility for sociological inquiry and who suggest in some instances that the term should be dropped from the sociological lexicon. The critiques are grounded in four somewhat overlapping perspectives: the positivist, the statistical, the interactionist, and the conceptualist. A formulation of what constitutes causality is presented, which takes account of these critiques and which also attempts to preserve continuity with past definitions. The basic elements are ontological status, productivity, conditionality, uniformity, and asymmetry. The importance of this subject, why the notion should be retained by sociologists, and the difference that this formulation makes are discussed. [ABSTRACT FROM AUTHOR]

This article presents some exploratory ideas developed while considering the historic controversy associated with the meaning and use of the term "cause." Certain scholars have insisted that cause is a fundamental and indispensable concept for understanding the relationships among phenomena. Others have contended that the concept is based on assumptions that are beyond proof and therefore of questionable scientific value. The confusion is increased when the scholar attempts to bridge the communication gap between his own specialized observations and theoretical ideas, and the ability of others to comprehend. Scientists have traditionally spent their time and energy looking for the causes and effects of the events they perceived or believed were occurring. Most scientists recognized the difficulty associated with the use of the ambiguous term cause within the scope of their own studies, but attempts to explain the problem to others often resulted in confusion rather than clarification. Some scientists inspired the classification of sequential relationships between recurring events which can be served as a "causeless" system of scientific logic.

The article compares two teaching procedures with respect to their efficacy in increasing the ability of the student to identify the latitude of cause and effect relationship. The two teaching procedures are the text-demonstration instruction procedure, and an experimental procedure. The term "cause and effect relationship" is used in the context of discussion on scientific attitudes. It is defined on the basis of statements by various famous scientists including J.M. Baldwin, Max Planck and Karl Pearson. Questions like whether the experimental teaching procedure is more efficacious in increasing the ability to identify the different degrees of cause and effect relationship, whether it is more helpful for brighter students, and whether it is lesser effective for girls, are addressed. Different studies on characterization of scientific attitudes, their measurement and their different elements were conducted by various scientists. A well chalked procedure was followed to reach to conclusions on the two procedures.

This article presents information on models of community conflict. In terms of the peace issue, a conflict model implies that peace can only be gained through conflict resolution; a consensus model implies that peace can be gained by conflict control. Conflict implies resolution because it assumes victory of one side over the other, hence the elimination of the cause of the conflict. Consensus implies control because it assumes agreement among the parties rather than the elimination of one of them; it frequently assumes that agreement equals conflict control or management, which equals conflict resolution, since conflict is not inherent in the situation. Conflict is perceived in the consensus model as an incidental dysfunction, in the conflict model as inherent in the situation. If it is incidental, it can be resolved through management; if it is inherent, it can be resolved only through abolition. The consensus model is usually an elitist model because it implies the continued existence of systems as they are and attempts to reconcile theft differences by negotiation. It is not inherently elitist; but it is elitist where elites rule and is democratic only where decisions are democratically arrived at.

The article presents the personal narrative of Otto R. Frisch and his involvement in nuclear fission. He considers his part in its study as a result of chance. He mentions his interest in physics started with his aunt Lise Meitner's influence. He attributes to chance his presence in the development of Niels Bohr's idea that the atomic nucleus was like a minute drop of liquid. He reveals his summary article on nuclear physics for the 1939 annual report on the progress of chemistry is again the result of chance.

In an experiment with 292 housewives, need for social approval and cognitive task importance was used to divide respondents into groups with predominantly problem-solving or social goals. The first group reacted more to competence of the source of a persuasive message, the second more to its social attractiveness. Rejection of the message led to a less favorable view of the source by subjects with social goals. [ABSTRACT FROM AUTHOR]

The article presents a discussion on a sociological study made on delinquency area situation involving the conflict of values. Analysis of the problem of the causation of delinquency in urban areas of high rates of delinquents is frequently confused by allusions to statistical data which suggest that while in proportion of delinquents in these areas is high, this class nonetheless represents a distinct minority of the age eligible population. Examination of delinquency statistics indicates that no conclusive judgements can be made on the basis of these statistics. The statistics do support the proposition that urban areas of high rates of delinquents are characterized by a duality of conduct norms rather than the dominance of either a conventional or a criminal culture.

When a causal model includes a feedback loop, ordinary least-squares (OLS} is an inappropriate, biased estimation technique. Furthermore, since the system is nonrecursive, a simple application of path analysis or regression is precluded. If the causal feedback is indirect, however, or involves one or more weak causal links, the bias will tend to be small and in some cases may be negligible. If the cousal system is correctly specified to include only indirect feedback-i.e., with no direct reciprocal causation between any pair of variables, it is possible to calculate unbiased estimators. Even in other instances, it is often possible to set a reasonable upper limit to the bias encountered by estimators affected only by indirect feedback despite their inclusion in a system in which there is both direct and indirect feedback. [ABSTRACT FROM AUTHOR]

Complexity is a concept of which the first definition can only be negative, though it is not simple. A simple object is an object which can be conceived as an indivisible elementary unit. A simple concept is one whereby such an object can be conceived clearly and distinctly, as an entity which can be isolated from its surroundings. A simple explanation is one whereby a composite phenomenon can be reduced to its elementary units and the whole can be conceived as the sum of the characteristics of the units. In simple causality cause and effect can be isolated and the effect of the cause can be predicted in a strictly determinist manner. Simplicity precludes what is complicated, uncertain, ambiguous, contradictory and the theory that corresponds to simple phenomena is a simple theory. However, a simple theory may be applied to complicated, ambiguous and uncertain phenomena. In this case a process of simplification takes place. The problem of complexity is that of phenomena which cannot be reduced to the simple thought patterns of the observer. In other words, complexity will first manifest itself, for this observer, in the form of obscurity, doubts, ambiguity, or even paradox or contradiction.

20 21 and 3-year-old children were presented with simple questions and command in the context of an at-home play situation. The questions and commands were spoken in varying degrees of word-order distortion in order to evaluate the communicative effectiveness of word order in the children's comprehension of meaning. The majority of the non fluent children responded appropriately to both normal and scrambled sentences. Only the fluent children's response scores were significantly lower to scrambled sentences than to normal. Data suggest that young, non fluent children's receptive language processing is focused on familiar semantic elements rather than on syntactic framework. [ABSTRACT FROM AUTHOR]

Semantic and cognitive factors governing passive-voice comprehension were explored. 96 children, divided among 3 grade levels (kindergarten through second), identified the object referred to by nonsense-syllable actors and recipients in passive sentences and reversed active sentences and a classification matrix. Comprehension of passives did not require reversal to an active form since children who understood passive sentences did not necessarily pass the language and matrix reversal tasks. Level of comprehension of passives, but not backward repetition, was affected by actor content. But, there were significant inter correlations among the tasks, and, therefore, some overlaps in linguistic and cognitive performance. [ABSTRACT FROM AUTHOR]

Researcher R. De Charms stated, "man's primary motivational propensity is to be effective in producing changes in his environment. Man strives to be a causal agent, to be the primary locus of causation for, or the origin of, his behavior; he strives for personal causation." De Charms differentiated between origins, people who perceive their behavior to be determined by their own choosing, and Pawns, people who perceive their behavior to be determined by external forces. The distinction is, of course, continuous, depending upon the situation. A person gains personal knowledge concerning his ability to be a causal agent through experience. According to De Charms, if a person has previously experienced much success through his own efforts, he is likely to perceive himself as an Origin. According to Social Learning Theory, reinforcement acts to strengthen expectancies that a particular behavior or event will be followed by a particular reinforcement in the future; a casual relationship develops in the mind. The expectation can be further strengthened through future reinforcements or weakened by lack of reinforcement. The issue whether internals and externals respond differently to reinforcements with regard to expectation has not therefore, been decided.

All the general explanations of crime are, due to the heterogeneity of criminal behaviour, failures. Typological approach makes causal analysis feasible, provided that proper criteria of classification are adopted; the two suggested here are: external characteristics of criminal behaviour, and its meaning. From the social policy viewpoint, components strategic for crime prevention are to be learned first. The multi-type components--each of them essential for various types of criminal behaviour--are the most strategic. The adopted usage of `cause', and the notion of the multi-type components, make it possible to restate and to specify the differential association, anomie, and Marxist theories of crime, and to reconsider the hypotheses of deterrent and of educative force of legal sanctions. Crime cannot be explained in terms of one or a few sufficient conditions. Due to the heterogeneity of criminal behaviour the number of its sufficient conditions and their essential components is almost unlimited, even at a close level of the causal distance.

The aim of this article is to investigate the relationships between causality and path and regression analysis. It could be maintained that path analysis is a technique used in the causal analysis of phenomena by means of linear relations. The aim of the article is to show that the connection between causality and path analysis is far from obvious. Although there is an enormous amount of different definitions of causality or causal relation, most of them take as their point of departure the concepts of necessary and sufficient condition. According to regularity theory, in order to be entitled to say that a causal connection obtains among phenomena one has to show that a universal law connects the phenomena. There is, however, a logical possibility to conceptualize causality independently of law concept. The theories of causality can be classified according to the form of description of relations. The relations can be described as deterministic or as stochastic. The article also discusses causality as a mode of determination in the social sciences.

The purpose of this study was to find evidence regarding the internal inhibition factor postulated by criminosynthesis theory as one of the factors involved in crime causation, in addition to need frustration, external inhibition, contact with reality, situational crime potential, and potential satisfaction. Five samples of offenders and two samples of nonoffenders were administered a measure of internal inhibition. The hypothesis predicting lower internal inhibition in offender than in nonoffender samples was supported, with the exception of the comparison between the male nonoffender and the adult female offender samples. Within both adult offender and nonoffender samples males were characterized by lower internal inhibition than females. The severity of alienation from society variable appeared also to be related to internal inhibition. [ABSTRACT FROM AUTHOR]

This article examines the social aspect of accident causation, or the social threshold of accident possibility. The factors contributory to accident causation of all kinds are divided into three groups: intrinsic, extrinsic and natural, and since certain of these factors, especially social, are variable, the accident situation is dynamic not static. Accident proneness is constantly altering as the social environment changes. A study concluded that social stresses give rise to emotional changes which in turn disorganize skilled activity. This effect of social stresses can be resisted by social conditioning, but beyond a certain level or field of social conditioning, here called the social threshold, it is believed that individual becomes accident prone. There are two psychological methods of approaching the problem of accidents, both employed in the reduction of accident rates. The segregation of an accident-prone segment of a plurel, either by psychological testing, or on the basis of accident histories. And the education of operators, where they are liable to accident or not. The social threshold of accident possibility at any given moment, may bear a relationship to the curve of behavior.

The article examines the relationship of society and its evolution with psychology. Important problems of method arise, involving the whole conception of original history. The problem of method that is central here is that of causation. A new notion of causation, that is, of proof of what causes what, has to be developed and accepted. The answer is that the new notion of causal relation between social phenomena, like all notions of causation, depends for its acceptance on the attitude of the perceiver. It seems, therefore, that all forms of proof in sociological studies boil down to two related types of proof, by convergence and by analogy. It is assumed here that the general cause of social psychological changes is to be found in changes in the material conditions of society. A general pattern can be traced through all the various types of behaviour as they spread down into society and gradually become second nature. Certain instincts disappear altogether. Large sections of society have lost any desire whatever to spit.

An examination of empirical evidence relating to current theory on delinquency raises three fundamental questions. First, studies of undetected delinquency suggest that the inverse relationship between social class and violation of the law may be less potent than current theory implies. Comprehensive study of delinquent behavior within as well as between classes is needed. Second, whether delinquent groups are, internally cohesive, bound by ties of loyalty, permanence, and stability or whether they are held together more by external pressures is an open question. Further studies on group cohesiveness are needed because of their obvious relevance to inquiries having to do both with causation and with control. Finally, studies have not revealed the kinds of singularly-focused delinquent subcultures which have been hypothesized, raising the question as to whether delinquent subculture is oppositional in character, unique and autonomous, or part of a more amorphous and widespread tradition of deviance. Some of these questions cannot be an- Second, whether delinquent groups are internally cohesive, tion on a broad cross-section of adolescents and of determining the effects of official identification and processing on the official picture of delinquency presented. [ABSTRACT FROM AUTHOR]

Free recall of an array of pictures followed by sequential location recall was observed in children ages 4 to 12 under 4 main presentation conditions arrays of pictures differing in salience of categories (contiguous, systematic, random grouping of category instances) versus noncategorical (control) array--and 2 task conditions, overt labeling versus pointing. Additionally, successive versus simultaneous presentation was studied. Grouping effects on memory were found to be systematically related to salience of grouping and to age. 2 labeling effects on memory performance were observed: (1) a direct primary effect at all ages that facilitated attention to the static configuration of all memory items, and (2) an indirect developmental effect of organizing that aided the discovery of categories under specific grouping and age conditions. Effect (1) was observed mainly on location recall, (2) on free recall. [ABSTRACT FROM AUTHOR]

Focuses on a study which determined the causal reasoning of children about natural phenomena. Subjects and method; Results; Discussion and summary.

Status integration as an explanation of suicide has been systematically developed and investigated by Professors Gibbs and Martin in a series of studies culminating in a recent monograph. The logical structure of this theoretical formulation and the operational definitions employed are examined. The empirical validity of the theory is assessed by reanalyzing previously reported data and by the presentation of new data. This appraisal raises questions about the logical structure of the theory as well as about its ability to account for suicide. [ABSTRACT FROM AUTHOR]

The article compares sociologist Kenneth Polk's and Hubert M. Blalock's contention in connection with Herbert A. Simon's asymmetric causal model about the relation between causation and correlation. Polk contends that the Simon-Blalock approach involves estimates that are inconsistent with the asymmetric models from which they are derived. The Simon model is a device for introducing ancillary information into a situation that would ordinarily be treated by regression analysis, its purpose being to use that knowledge to attain more meaningful and powerful results than those provided by regression analysis. For three or more variables, however, the use of coefficients defined in direct contradiction to what the derivation implies is essential in deriving the conditions by which Blalock distinguishes between models by empirical test, and the use of the coefficients in the multivariate case requires both explanation and justification. What is needed instead is a demonstration that what has been set equal to zero is not a regression coefficient and yet functions meaningfully in the derivation of the regression and correlation coefficients that appear as estimates or conditions for the model.

The article discusses asymmetric causal models concerning the relationship between correlation and causality. The technique developed by sociologist Herbert A. Simon in his model involves the use of simultaneous equations, rather than traditional partialling techniques, for tracing the relationships among a set of variables. In summary, difficulties may occur in the use of the Simon model because: First, linear regression models will be symmetrical even when the relationship between the variables is asymmetrical in the causal sense; and Second, a causal relationship cannot be established by the mathematical equation suggested by Simon, since causation can only be established by the researcher's theoretical knowledge of the nature of the variables included in the analysis. It has been argued that causal asymmetry should not be confused with predictive symmetry. The test of an asymmetrical "causal" hypothesis may involve the use of a prediction technique which is symmetrical. Although this technique cannot be used as a "causal criterion" in empirical investigations, it does not follow that the method proposed by Simon should be ignored, since it is difficult to conceive of a theory which does not involve assumptions of causal priority.

The article discusses asymmetric causal models concerning the general relationship between correlation and causality. Sociologist Herman Wold points out that there are two fundamentally different uses for regression equations: First, as prediction or estimating equations, and second as causal models. Some variables may be only indirectly related to each other through the operation of other variables. In such instances, the correlation between any two indirectly related variables should be weaker than the correlations between either of these variables and some intermediate variable. Prediction may also be used in the sense of estimation, where temporal or causal sequences are irrelevant. In this sense of the term, from the dependent to the independent variable just as readily as the other way around can be predicted. Thus we may estimate a person's IQ from his grades or, perhaps absurdly, we may even estimate rainfall from the wheat. No causal models can predict exact values of coefficients except in the case where the correlations disappear.

Two distinct criteria are discussed for evaluating the relative importance of several independent variables in determining the variation in a dependent variable. The quantitative criterion is used primarily with numerical data, whereas the causal criterion often appears in theoretical arguments. Simon's method for making causal inferences from correlational data may offer potentialities for combining these criteria. The indiscriminate use of partial correlations and a single multiple regression equation can yield misleading conclusions in evaluating importance. instead, an entire set of simultaneous equations is needed. [ABSTRACT FROM AUTHOR]

The supply of mental ability is important in many ways to human welfare and happiness, in the near future and the long run. Contrary to traditional beliefs, the resent limits of ability in our society are not set by genetic factors, but to an important extent by sociological conditions, which support a sort of "collective ability." The prospects for an early important increase in the collective ability in the United States are favorable. Some of this will follow automatically if resent economic and educational trends continue. Some depends on the acceptance of responsibility by sociology for systematic research in the field of social causation of high ability. [ABSTRACT FROM AUTHOR]

The art of discovering causal relationships depends first on familiarity with particular data, then on trying all likely causal conditions that suggest themselves, and screening them down to the smallest effective number. Systematic studies employing correlation techniques, occasionally aided by special methods such as multiple factor analysis, but with little attention to causation as such, have already been made to a limited extent in some sociological areas, for example, prediction of success in marriage and on parole, and prediction of juvenile delinquency, but the best known election poling, at least before 1948, have apparently regarded such procedures as unnecessary and academic. Though the methods used in predicting, say, success on parole, have taken a variety of forms, with scoring devices especially popular, the essential logic in the case of attributes can best be shown by a contingency or experience table. Election poling commonly predict the vote from statements of voting intention. This is justified only if such statements are causally related to voting behavior. The accuracy of prediction will turn in part on the closeness and reliability of the relationship, and some estimate of it must be made before any faith can be placed in the prediction.

The article presents a description of the method of analytic induction in sociological research. Analytic induction is discussed as a research procedure, as a method of causal analysis and as a method of proof. As a research procedure, analytic induction begins with an explanatory hypothesis as well as a definition of something to be explained. The hypothesis is compared with facts and accordingly, modifications are made. As a method of causal analysis, analytic induction consists in taking a number of instances in which the phenomenon occurs and finding certain conditions which always accompany that phenomenon. As a method of proof, analytic induction is regarded as a method of proving that the generalizations to which it leads will be applicable to all instances of the phenomenon.

The article offers information on a sociologiocal study conducted on the conflict of values in delinquency areas by the Illinois Institute for Juvenile Research. It is based on a hypothesis that in the disorganized city areas delinquency is primarily a product of cultural rather than of personality or psychological processes. Some of the official statistics of delinquency seem to suggest that most children conform to the legal norms of the wider society even in those urban areas where the culture of the local community is relatively favorable to the tranmission of delinquent conduct patterns. The statistics infers that even in this situation variables other than culture are of possibly greater importance in delinquency causation than the customary sociological explanations would concede.

Most of the variables utilized in past participation research have been of a "static" nature. Variables such as age, occupation, family status, and education are "static" in the sense that they cannot be readily changed by group members, leaders, or the professional worker. The study of "dynamic" factors is suggested. "Dynamic" variables such as member under- standing of the basic principles of organization and operation, satisfaction, identity, having a "say" in decision making, and feeling of responsibility were found to be highly related to member participation. These factors are "dynamic" in the sense that it is possible for group members and leaders to change them through group action. The use of such factors will broaden the scope of participation research and may implement the testing of causality. [ABSTRACT FROM AUTHOR]

Comments on the article of M.P. Miracle and B. Fetter on causal models of economic behavior in Africa. Motivation through economic rationality; Role of objective situation; Impact of degree of authority and egalitarianism in rural society on supply price.

Let me begin with three assertions. First, causal analysis is an important and viable method for theory construction in sociology. Second, the Merton variety of functional analysis in sociology is a unique method of theory construction and is the major alternative to causal analysis. Third, inherent methodological difficulties in Mertonian functional analysis in sociology argument for its abandonment. The first of the above assertions constitutes my reason for this attempt to spell out a notion of causal analysis as a method for theory construction in sociology. The appreciation of the second and third assertions requires some discussion of the work of Robert K. Merton and Kingsley Davis. [ABSTRACT FROM AUTHOR]