Guttman Scales Research Articles

A Method For Analyzing a Class of Related Problems in the Patterning of Central Place Activities

A METHOD FOR ANALYZING A CLASS OF RELATED PROBLEMS IN THE PATTERNING OF CENTRAL PLACE ACTIVITIES Thomas L. Bell and Phillip L. Frankland* THE CONTENT ANALYSIS DESCRIBED. Numerous empirical studies exist which attempt to describe the activity patterns of towns of varying sizes in varying spatial environments. (1) Many of these studies have displayed the collected data in the form of an incidence matrix with towns as rows and central place functions as columns. (2) The cells of this dichotomous matrix tell whether a particular central place possessed a particular central function or not. Researchers in different social science disciplines have adopted different research methodologies in order to analyze such matrices. It is the analysis of such data that serves as the point of departure for this paper. There seems to be a fundamen­ tal paradox existing among social science researchers that is perpetuated by strict disciplinary divisions of labor. Although the research endeavors of, say, rural sociologists and geo­ graphers are often the same, each field has developed its own methodol­ ogy for attacking the problem and is often ignorant, or chooses to ignore, the literature of the other field. This paradox certainly comes to light in the case of central place studies. Rural sociologists, from the time of Galpin (3) and Kolb (4) have been interested in the problem of analyz­ ing the functional content of central places. Many of these studies have assumed that the economic impact of the rural community on its sur­ rounding hinterland population has corresponding social impact on those dispersed rural residents. This social impact is manifested by commu­ nity identification, social ties, and contact fields. (5) This thesis of eco­ nomic and social interdependence has not, however, gone without chal­ lenge among rural sociologists. (6) Likewise, geographers from the time of Christaller (7) and Losch (8) have been concerned with the spatial distribution of central places and the effect of this distribution upon the interaction patterns of the surrounding hinterland population. (9) It is at the analysis stage that the paradox discussed above becomes apparent. Both geographers and rural sociologists are concerned with the analysis of incidence matrices of central places and the central func­ tions that they perform, albeit for alternative conceptual reasons. How­ ever, geographers have most often opted for direct factor analysis as * Mr. Bell Is instructor of Geography at the University of Tennessee, Knoxville, and Mr. Frankland is a Ph.D. candidate at the University of Iowa. The paper was accepted for publication in July 1972. V o l. XIII, No. 1 31 the way in which to analyze such matrices, whereas rural sociologists have utilized the technique of scale analysis as developed by Louis Guttman. (10) Although there has recently been explicit mention of the work of the rural sociologists by the geographer Marshall (11), the para­ dox still remains. Most researchers in one discipline seem unaware of the empirical work in the other. ADVANTAGES OF GUTTMAN SCALING AS A METHOD OF ANALYSIS. There are several reasons, made explicit in this section, why geographers should consider Guttman scaling as one of their re­ search tools. The following discussion concerns itself with the Guttman scaling model per se. (12) The Guttman technique was originally designed to rank (or scale) a group of respondents upon the basis of their attitude toward some sub­ ject area. (13) In order for a group of items to be considered “scalable,” they must possess two important properties. First, Guttman scales must be unidimensional. That is, the component items must all measure movement toward or away from the same single underlying object. The substance of the object is relatively unimportant— it may be racial preju­ dice, technological development of societies, or, in the context of this article, ubiquity of central place functions. Secondly, Guttman scales must be cumulative. This is the property which differentiates Guttman scales from almost all other types of scales. Operationally, a cumulative scale implies that the component items can be ordered by degree of difficulty and that respondents who reply positively to a difficult item will always respond positively to less difficult items and vice versa. There are several advantages of Guttman scaling over other tech­ niques that have been...

Monothetic and Polythetic Typologies and their Relation to Conceptualization, Measurement and Scaling

Past studies of the relationship between conceptualization and measurement in typology construction have depended upon the over-simplified heuristic-empirical (H-E) distinction. This paper presents the classification-identification and monothetic-polythetic distinctions. These are used not only to construct an extension of the H-E distinction (heuristic, empirical, classical and reduced classical types), but also to examine previously neglected relationships between types, scales, and forms of property-space reduction. Specifically, I show that all heuristic types and all categories of simple order scales (e.g., Guttman scales) are monothetic, while all empirical types and categories of partial order scales (e.g., Likert scales) are polythetic. I also show that a partial order scale can be constructed from a monothetic typology through Lazarsfeld's arbitrary numerical reduction, and a simple order scale can be constructed from a monothetic typology through functional reduction. Four characteristics of Weber's rational administration (as presented by Udy, 1958) are used to construct examples of the various typologies and scales.

Some Issues in Sociological Measurement

For a discipline that ignored errors of measurement too long, suffering the consequences of low predicti-ve capability and spurious theoretical debate, sociology has recently shown a remarkable turnabout, investing an appreciable proportion of its scholarly pages to analyses of the measurement problem. Of course the concern is not entirely unprecedented: the team that produced the monumental The American Soldier showed a keen and sophisticated interest in measurement problems with eventual fruition in Paul Lazarsfeld's latent structure analysis and Louis Guttman's scalogram procedures (Guttman's later work seems less associated with sociometrics than with psychometrics). Yet in the years since World War II the potential impact of latent structure analysis was little realized because only a few passing through Columbia University were exposed to it (a situation remedied recently by Henry and Lazarsfeld, 1968). Guttman scales suffered another fate, becoming for a decade a fetish among American sociologists, applied so often where the underlying model was inappropriate that some metricians turned to discrediting the technique generally (Nunnally, 1967).

Correlation between family and infant food habits by scalogram analysis

370 randomly selected families from an urban and a rural communities located in an industrializing area of Northern Mexico were studied. The foods consumed by the family the day before the interview as well as indicators of education, occupation and economical level of the family were determined. Type and sequence of foods introduced to the infant's diet were also investigated in 199 cases. Guttman scales for the family diet and the infant diet at six months of age were found with coefficients of scalability of 0.80 and 0.78 respectively. A significant correlation between both scales was found (p ∼ less than 0.01) suggesting that the infant at six months participates in some extent in the family diet. Higher correlations were found between both scales and indicators of economic position of the family than with the educational level of both parents. It is suggested that in areas with some industrial progress, the economy becomes an important limiting factor in determining the dietary habits of the family and the infant.

Statistical Models of Senate Roll Call Voting

This paper uses statistical analysis to consider what factors influence the way senators vote and how important these influences are. The answers are specific statements about individual senators' decision processes, and quantitative estimates of the weights applied to the different variables in these models. The form of the voting models and the variables in them were developed from hypotheses about individual decision-making and descriptions of the legislative process. Examples of the variables are the preferences of the senator's constituency, the opinions of his party leader, and the views of the President. Quantitative measures of the variables were obtained from state demographic characteristics and by Guttman-scaling the votes on passage of the amendments to specific bills considered in 1961 and 1962. Linear regression analysis of these Guttman scales was then used to test the hypotheses and estimate the coefficients in each senator's voting model. After the results from these analyses are discussed, the explanatory power of these models on individual bills is evaluated by comparing the vote they estimate with those predicted using two alternative models.

Delinquency Potential and Attitudes toward the Criminal Justice System

A questionnaire dealing with attitudes toward the criminal justice system was administered to 626 seventh-grade boys. Guttman scales were developed for 9 attitude dimensions associated with the criminal justice system and the relationship of these attitudes to measures of delinquency potential was examined. All of the associations were negative as hypothesized, but of low magnitude. The relationship with delinquency potential was much stronger for some attitudes toward the criminal justice system than others. When race was controlled the relationships varied. The relationship between delinquency potential and attitudes toward the criminal justice system is very complex and many issues must be resolved before a complete acceptance can be made of the generalization that potential delinquents hold more unfavorable attitudes toward the criminal justice system than potential non-delinquents.

Delinquency Potential and Attitudes Toward the Criminal Justice System

A questionnaire dealing with attitudes toward the criminal justice system was administered to 626 seventh-grade boys. Guttman scales were developed for 9 attitude dimensions associated with the criminal justice system and the relationship of these attitudes to measures of delinquency potential was examined. All of the associations were negative as hypothesized, but of low magnitude. The relationship with delinquency potential was much stronger for some attitudes toward the criminal justice system than others. When race was controlled the relationships varied. The relationship between delinquency potential and attitudes toward the criminal justice system is very complex and many issues must be resolved before a complete acceptance can be made of the generalization that potential delinquents hold more unfavorable attitudes toward the criminal justice system than potential non-delinquents. KEYWORDS: Juvenile justice; juvenile delinquency

Institutional Differentiation in Guatemalan Communities

Previous articleNext article No AccessInstitutional Differentiation in Guatemalan CommunitiesJohn W. DurstonJohn W. Durston Search for more articles by this author PDFPDF PLUS Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinkedInRedditEmail SectionsMoreDetailsFiguresReferencesCited by Economic Development and Cultural Change Volume 18, Number 4, Part 1Jul., 1970Part 1 Article DOIhttps://doi.org/10.1086/450454 Views: 1Total views on this site Citations: 3Citations are reported from Crossref Copyright 1970 The University of ChicagoPDF download Crossref reports the following articles citing this article: John D. Early Ethnography as an Interpreter of National Censuses: The Guatemalan Case, Journal of Anthropological Research 36, no.11 (Sep 2015): 71–86.https://doi.org/10.1086/jar.36.1.3629553Eileen Stommes Village complexity and the delivery of agricultural extension services: Lessons from El Salvador, Agricultural Systems 3, no.44 (Oct 1978): 265–280.https://doi.org/10.1016/0308-521X(78)90013-6Berkley Spencer Community Differentiation and the Fallacy of Intenectorial Causation, Human Organization 32, no.11 (Mar 1973): 59–71.https://doi.org/10.17730/humo.32.1.d23343x373323808

Infant feeding and weaning practices in a rural preindustrial setting. A sociocultural approach.

SummaryThe major purpose of this study was to assess the infant feeding pattern and weaning habits in a group of 125 infants in relation to their mothers' social characteristics and other presumably related background factors. Specific objectives included the utilization of research methods and techniques of social science in the field of nutrition and the exploration of traditional feeding practices and food ideology within a cultural context.The investigation was carried out in a village located 65 miles southwest of Mexico City, with a population of about 6000. A multiple factor theory of biosocial factors affecting malnutrition provided the theoretical framework for this research.The present report is based on the data collected in 125 families, part of the total sample of a longitudinal ecologic study of the relation between nutrition, mental development and learning in a preindustrial society. Guttman scaling, rank correlations, cross‐tabulations and “gamma” tests of association were the statistical methods used in the analysis of the data.The findings of the present study support the hypothesis that the characteristics of typical familial complex of preindustrial societies, seems to be the underlying phenomenon of malnutrition problems in those areas: large families, high parity, low levels of education, minimal contact with contemporary knowledge through mass communication media and health personnel, and a strong attachment to the traditional ways of life.The findings of the study also support the conclusion that breast‐feeding practices and weaning habits during the first six months of life are strongly associated with these general familial, social and cultural factors.Early introduction of feeding supplements, a trend toward lessening the breast‐feeding period, as well as attitudes concerning acceptance of bottle‐feeding practices were found.Pre‐scientific conceptions of food, health and disease held by the majority of the women studied strongly influence and still determine the feeding patterns. Severe food restrictions during healthy and sick stages imposed on chil dren and lactating women seem to confirm the above consideration.A significant negative association was found between mother's age and length of lactating period of the youngest child, with older mothers weaning at later periods. Similarly, negative correlations were found with many of the breast‐feeding practices and mother's age and educational level.Two six‐step Guttman scales for the infant and family food consumption data were found with high coefficients of reproducibility.The family food scale theorized as indicator of dietary complexity correlated well with other indicators of complexity in the family's “way of life”. The infant food scale departed completely in this sense, indicating a dualism which supports Reeder's concept of specificity of elements of decision making and social action. Associations were found between the family food scale and social position, education and communication variables, as well as with indicators of “modern” health practices.

Historical Inferences from Guttman Scales: The Return of Age-Area Magic?

Inferences about historical sequences are frequently drawn from Guttman scales, on the supposition that each item in the scale serves as a "functional prerequisite" for successive items. The validity of such inferences is here called into question on both theoretical and empirical grounds. Logically, such inferences are subject to the same limitations as inferences about the age of a trait based on its geographic distribution (the discredited "age-area" hypothesis). This is true even if specific scale items are seen as indices of broader evolutionary "stages," or if the unidimensional nature of the Guttman scale serves only as the basis for inferring some underlying unilinear evolutionary process. To test these notions empirically, 40 Mexican communities, varying widely in location, population, economic base, and social attributes, were investigated. Within each, data were gathered on the presence or absence of 24 traits typically selected for Guttman scaling, together with their actual historical sequence. A Guttman scale of high reproducibility was constructed from these trait lists, though some issues concerning the reliability of scaling were also raised. The developmental sequence inferred from this scale was then correlated with the historical sequences. Few correlations were negative, but the majority fell below .7, thus accounting for less than 50% of the variance. An alternative explanation for even this level of correlation was provided in the association found between population size and scale step (.83). An examination of the historical data also suggested that the actual processes occurring in Mexican communities resemble more closely a "multilinear" or a "systems" model of developmental growth than the unilinear model which Guttman scales are sometimes thought to imply.

Composite measurement

Several systems of fundamental measurement are available in the literature, which involve two (or more) sets of entities to be measured simultaneously. Well-known examples are: Guttman scales; additive conjoint measurement; Coombs systems. This paper is a first step towards a general analysis of such systems for which the name “composite measurement” is proposed. An extension principle plays a central role; it is used in three proofs and is discussed in general in the concluding section. Three examples of composite measurement are analyzed in detail: (i) bi-classificatory systems, (ii) bi-quasi-series or Guttman scales, (iii) bi-semiorders. As far as we know, the analysis of bi-semiorder is entirely new.

An Index of Riot Severity and Some Correlates

While certain variables do not correlate with the presence or absence of riots in American cities, they do correlate with riot severity. Such variables are more influential in determining the severity of a riot, once it has begun, that they are in determining the outbreak of that riot. The Index-a Guttman-type scale developed from materials describing disorders in 1967-suggest that the events that constitute riots and civil-criminal disorders are not bizarre, non-patterned, or randomly generated. On the contrary, employing the properties of Guttman scales, we may predict the sequence of events for levels of riot severity.

Guttman Scales and Evolutionary Theory: An Empirical Examination Regarding Differentiation in Educational Systems

Guttman Scales and Evolutionary Theory: An Empirical Examination Regarding Differentiation in Educational Systems

An Empirical Study in the Sociology of Knowledge*

Investigations in the sociology of knowledge usually take as their point of departure some substantive property of a mode of thought. They then seek to delineate and understand the social origins of that mode of thought. This paper suggests the possibility of supplementing the substantive approach with structural materials. Portions of two different intellectual systems are treated empirically to ascertain the manner in which their demonstrations of proof are structured. These structures are uncovered through the use of Guttman scale analysis. While any set of branching deductions may be considered a partly ordered set, the findings reported here show that the portions of the intellectual system chosen for analysis converge to an almost perfectly ordered set (a Guttman scale). Finally, known properties of Guttman scales are assigned to the structural dimension, and implications for the sociology of knowledge are drawn.

A Note of Extreme Caution on the Use of Guttman Scales

Since the change occurence of the various scale types is a function of the joint probability of the occurrence of their constituent responses, it is possible that Scalograms meeting currently acceptable levels of Scalabity and Reproducibility can occur by chance alone. Even Chilton's method, which determines whether the observed Reproducibility is significantly better than chance, does not guarantee that a scale is homogeneous; significant Reproductibility can occur even when three of six items are unrelated. Although very strict adherence to all of Ford's criteria might eliminate most chance-derived scales, the rule-of-thumb origins of these criteria limit their usefulness as rational proof that Guttman Scales are homogeneous.

Assimilation-Contrast Effects and Item Selection in Thurstone Scaling

Previous research shows that Thurstone, Guttman, and Likert scales constructed from a common pool of items differentially select items having a reference to self, and furthermore that prediction to behavior is related to the proportion of self-reference items in the scale. In the present study, the systematic underselection of self-reference items in Thurstone's successiveinternal measurement, is replicated on a new subj ect population. This underselection is explained by showing that assimilation-contrast effects occur even when the usual criteria for eliminating judges for incompetence are employed, and that such effects are most pronounced for self-reference items. It is also suggested that whether a judge's own attitude is a popular or an unpopular one in the social environment is related to assimilation-contrast effects and thus to the type of item appearing in a Thurstone instrument. For years much attitude research has relied upon Thurstone scaling procedures to construct measuring instruments. Questions have been raised, however, about the adequacy of some of these procedures. One recent investigation compared the predictive power of a Thurstone successive-interval scale, a Likert scale, and a Guttman scale, all of which were developed from a common pool of items. The three instruments were designed to measure the same attitude and were evaluated in terms of their utility for predicting behavior patterns. The Thurstone scale proved to be the poorest; the Guttman scale performed somewhat better; and the Likert scale proved to be the best predictive tool. It was suggested that the poor performance of the Thurstone procedure could be accounted for by certain intrinsic features of the Thurstone technique. It was reported that relative to the other techniques the Thurstone method resulted in underselection of items which contain a specific reference to self. It was found that the proportion of selfreferent items included in each type of scale was closely related to the relative predictive power of the three scales.' The present research was undertaken to determine whether the underselection of self-referent items by the Thurstone procedure is replicable using a different subject population. Since the replication was successful, the data were further analyzed in an attempt to specify why the Thurstone procedure results in this systematic underselection of self-specific items. A satisfactory answer to this question would provide a basis for improving attitude measurement, as well as suggest some ideas for the development of a theory of measurement. Our working hypothesis, in explaining the underselection of self-specific items by the Thurstone technique stems from the research of Sherif and Hovland.2 They found evidence of an assimilation-contrast effect when subjects are asked to perform a judging task. A judge tends to assimilate, or perceive a stimulus closer to an internal anchor, whenever a discrepancy between the two is small. A contrast effect involves a movement away from an internal position when the discrepancy between that position and the stimulus is great. In the Sherif and Hovland data this process was manifest in a marked tendency for judges with extreme attitudes to place items in extreme categories. Additional studies have substantiated the influence of assimilation-contrast effects in judging 1 Charles R. Tittle and Richard J. Hill, Attitude Measurement and Prediction of Behavior: An Evaluation of Conditions and Measurement Techniques, Sociometry, 30 (June 1967), pp. 199-213. 2 Muzafer Sherif and Carl I. Hovland, Social Jtudgment (New Haven: Yale University Press,

Assimilation-Contrast Effects and Item Selection in Thurstone Scaling

Previous research shows that Thurstone, Guttman, and Likert scales constructed from a common pool of items differentially select items having a reference to self, and furthermore that prediction to behavior is related to the proportion of self-reference items in the scale. In the present study, the systematic underselection of self-reference items in Thurstone's successive-internal measurement, is replicated on a new subject population. This underselection is explained by showing that assimilation-contrast effects occur even when the usual criteria for eliminating judges for incompetence are employed, and that such effects are most pronounced for self-reference items. It is also suggested that whether a judge's own attitude is a popular or an unpopular one in the social environment is related to assimilation-contrast effects and thus to the type of item appearing in a Thurstone instrument.

School Identification and Delinquency Proneness

Is a lack of identification with social institutions important in the etiology of delinquency? Guttman scales were developed to measure six dimensions of the adolescent's identification with educational institutions: Capacity to Learn, Value of Education, Legitimacy, Teachers—General, Teachers—Personal, and Teach ers—Academic. The findings indicate delinquents are less likely to identify with the school than are nondelinquents and specify the dimensions along which identification is particularly absent. No differences between delinquent and nondelinquent boys were found in two dimensions, Value of Education and Attitude to ward Teachers—Academic.

Hayes, "Item Order and Guttman Scales"

Hayes, "Item Order and Guttman Scales"

Item Order and Guttman Scales

One deduction from balance theory is that respondents will tend to integrate their answers to a set of attitude items. A second deduction, from classical figure-ground relations, is that the first item of a set of attitude items will, under certain conditions, influence responses to subsequent items. Two experimental studies dealing with these deductions for Guttman scales are reported. The results show that individual responses to certain Guttman scales may be systematically manipulated by item rearrangement. The implications of these findings for interpreting scale patterns and the role of contextual factors in attitude measurement are discussed.