Classificatory Analysis Research Articles

What kind of motivation drives medical students' learning quests?

To describe the patterns of medical students' motivation early in the undergraduate programme and to examine their relationships with learning features and motivational outcomes. The Academic Motivation Scale (AMS) was administered after the first medical year to 297 students of both sexes from consecutive classes within a 4-year timeframe. Measures of learner orientation and reflection in learning were also obtained. Academic achievement and peer tutoring experience were recorded during a 2-year follow-up. Quantitative approaches included analysis of variance, correlational and classificatory analyses of the data. The profile of the students' responses revealed higher levels of autonomous motivation than of controlled motivation although such measures were positively related. Correlation analysis showed significant association of autonomous motivation with higher levels of meaning orientation, reflection in learning, academic achievement, cross-year peer-tutoring experience, and intention to continue with studies. Classificatory analysis identified 4 student groups with distinct patterns of motivation. Analysis of variance revealed significant and consistent differences in learning features and outcomes among such groups. The findings indicate that medical students portray distinct patterns of autonomous and controlled motivation that seem to relate to the learners' frame of mind towards learning as well as the educational environment. Autonomous motivation had closer relationships than controlled motivation with measures of self-regulation of learning and academic success in the context of a demanding medical programme.

Rough Set on Trademark Images for Neural Network Classifier

Rough set theory (RS), introduced by Zdzislaw Pawlak in the early 1980s, is a methodology that concerned with the classificatory analysis of imprecise, uncertain or incomplete information or knowledge expressed in terms of data acquired from experiences or observations. It has the ability to distinguish between object and reason about the objects in the universe in which objects are perceived through the information that is available about them through the values for a predetermined set of attribute. The main advantage of RS is that it requires no additional information to the data represented in table. On the other hand, Supervised Neural Network learns by abstracting a mapping function from the training data for classification purposes. However the drawback of using a supervised neural network is that a large amount of training data must be provided for it to obtain an accurate mapping function. The problem is further aggravated if the data are in the continuous form (real values). Thus, in this paper we overcome the problem by transforming the training data in the continuous form into discrete values using Rough Sets theory and Boolean Reasoning technique. Here, global shape features are chosen to represent the logo images. The invariant features representing logo images are obtained by using the Geometric Invariant Moment Technique (Hu, 1962). The classification results prove that discretization using Rough Sets and Boolean Reasoning can reduce the training cycle and significantly increase the accuracy of the classification of logo images.

The Application of Rough Sets to Sightseeing Expenditures

The existing tourism demand forecasting models in tourism are unable to capture useful information from a database with numeric and nonnumeric data. This article presents a new approach that applies the rough set theory to form a forecasting model for sightseeing expenditures in Hong Kong. The rough set theory deals with the classificatory analysis of imprecise, uncertain, or incomplete knowledge (data) by incorporating the classical set theory. Based on officially published tourist sightseeing data, decision rules are generated to represent the relationships between the independent variables and the dependent variable. Experimental results revealed that the forecasting model can classify 94.1% of the testing cases, and that 87.5% of the classified cases were identical to their actual counterparts. There was no significant difference between the actual values and the forecast values. The advantages of using decision rules induced by rough set to forecast sightseeing expenditure were also offered.

A Numerical Taxonomic Study of Trema (Ulmaceae) from Togo, West Africa

Abstract The assignment of collections belonging to the genus Trema has been reassessed using numerical taxonomy methods. Observation in the field in Togo, West Africa, confirmed high phenotypic variation but not the previously asserted lack of character concordance within the T. guineensis complex. This study was carried out to assess the circumscription of taxa in the area. Distance between specimens was computed using the Gower's general coefficient of similarity and the non-parametric MODECLUS cluster analysis was used to discover how the specimens segregated. The existence of three clusters in the 158 specimen dataset using 44 morphological characters was suggested by MODECLUS. Canonical discriminant analysis supported the recognition of those three clusters using 40 morphological characters. Classificatory discriminant analysis showed that all specimens except one are identified correctly by the discriminant function. We suggest that three species be recognized in Togo and the neighboring countries:...

THE FOREIGN MARKET SERVICING BEHAVIOR OF MULTINATIONALCORPORATIONS: AN EMPIRICAL INVESTIGATION.

This paper classifies a set of the world's largest multinational corporations (MNCs) according to the strategic choice they make about servicing foreign markets: the three generic international strategies proposed here are global exporter, foreign producer and mixed mode. A fourth category includes firms with low levels of both exporting and foreign production. Applying this categorization schema focuses attention on a relatively neglected aspect of international strategy at the firm level. The classificatory analysis shows that almost a quarter (24.8%) of the 443 MNCs pursue a foreign producer strategy. However, dominant models of internationalization devote little descriptive or normative attention to managing such firms. Accordingly, the findings here suggest that a foreign production-centred firm strategy and its attendant organizational requirements are material for a significant subset of MNC managers and thus warrant specific research attention. In addition to industry and country effects, the findings also reveal significant within-industry variation in firm choice on this dimension of international strategy. As such, the research has implications for the relative importance of models based on industry versus firm choice of position within an industry, and for stages models of internationalization.

Morphological variation among members of theStellaria longipes complex:S. longipes, S. longifolia, andS. porsildii (Caryophyllaceae)

Morphological relationships were investigated among diploidStellaria porsildii, polyploidS. longipes, and diploidS. longifolia. Canonical discriminant analysis, based on a priori assumptions to maximize differences among groups, showed thatS. longipes clusters equally distant between the two diploid species along an axis connecting the diploids' centroids, but it differs along an axis perpendicular to this axis. The intermediacy along the former axis is evidence thatS. longipes is an amphiploid derived fromS. longifolia andS. porsildii. The divergence along the latter axis may be attributable to adaptively valuable heterotic traits which were retained following amphidiploidization. The only morphological discontinuity occurred between the two diploids, whereas the morphological range ofS. longipes overlapped the range of both diploids forming a continuum. The lack of discrete clusters is likely due to hybridization and introgression withS. longifolia on one hand, and convergence of traits betweenS. longipes var.monantha andS. porsildii on the other. High a posteriori assignments in classificatory discriminant analysis supports the separation ofS. longipes var.monantha from otherS. longipes specimens. AlthoughS. longipes var.monantha grouped close toS. porsildii, the two groups separate based on leaf shape traits. Overall results support, firstly, the hypothesis thatS. porsildii is a diploid parent species which by hybridizing withS. longifolia gave rise to polyploidS. longipes. Secondly, results suggest thatS. longipes var.monantha converged morphologically towardsS. porsildii relatively recently due to ecological specialization, and merits distinction at least as a variety ofS. longipes.

Computerized Interactive Morphometry and the Diagnosis of Lymphoid-Rich Effusions

An experimental computerized interactive morphometry system for the classification of lymphoid-rich effusions is presented. The relatively inexpensive microcomputer-based system was assembled in our laboratory with commercially available hardware and software. One hundred twenty-two effusions (86 benign lymphocytoses, 26 lymphomas, and 10 chronic lymphocytic leukemias) were studied. Lymphoid cells were selected randomly by the system from real-time images of Papanicolaou-fixed and stained cytospin smears of pleural, peritoneal, and pericardial effusions. Parameters of nuclear shape, area, and optical density were measured automatically. Multiparameter statistical procedures of discriminant classificatory analysis analyzed the distribution of lymphoid nuclear profile integrated optical density to yield three groups of effusions, each with a predictive value of diagnosis of 100%. Neither these statistical procedures nor a simple rule-based expert system accurately classified chronic lymphocytic leukemia effusions based on the distribution of lymphoid nuclear profile area. When chronic lymphocytic leukemia effusions were excluded, however, the predictive values of a diagnosis of lymphoma by statistical analysis and by rule-based expert system were 92.3% and 88.9%, respectively, whereas the predictive values of a diagnosis of benign lymphocytosis were 97.7% and 91.3%, respectively. Potential applications and limitations of this technology for the diagnosis of lymphoid-rich effusions are discussed.

Spatial and seasonal patterns in fish larvae distribution in the Banda Sea (Indonesia)

A study of horizontal and vertical distribution, abundance, biomass, and taxonomic composition (down to family level) of the fish larvae in the eastern Banda Sea was based on samples obtained with a Rectangular Midwater Trawl 1+8 in the upper 300 to 500 m during August 1984 (southeast monsoon) and February/March 1985 (northwest monsoon). No significant seasonal differences in abundance and average weight of the total population were found. At night on average two-thirds of the fish larvae occurred in the upper 100 m of the water column. During the day the larvae exhibited diel vertical migration, down to depths of 400 to 500 m, and there was evidence of net avoidance. In total 78 families were identified. Among the 61 families found in the southeast-monsoon samples, the Myctophidae and Gonostomatidae were the most abundant. In the northwest-monsoon samples larvae of 66 families were found of which Acanthuridae, Myctophidae and Gonostomatidae were the most abundant. In the southeast monsoon about 60% of the larvae were from oceanic taxa, against only 30% in the northwest monsoon. Classificatory analysis based on presence or absence and abundance of the families demonstrated major discontinuities in faunal composition for both seasons, possibly related to intrusion of water and fauna from outside the area studied.

An investigation of the taxonomy of Hordeum capense and H. secalinum (Poaceae: Triticeae)

That Hordeum capense, a South African species, and H. secalinum, a mainly European species, are conspecific, has been the prevailing view for the last 80 years because of a lack of distinguishing markers. In the present paper, morphological separability is demonstrated by means of cluster analysis, classificatory discriminant analysis, logistic discrimination, and canonical discriminant analysis. The performance of the linear classification functions are evaluated by the bootstrap and discussed. Lodicules and epiblasts were found to be good distinguishing markers. The nomenclatural type of H. secalinum has been designated as lectotype instead of the previously designated neotype.

Differentiation of fourHordeum sect.Stenostachys spp. (Poaceae: Triticeae) using multivariate morphometrics

Field collections and 296 herbarium sheets were examined for 27 morphometric variables. A priori species identifcation was based on geographical distribution except forH. californicum, a diploid species primarily occurring in California and differing from the much more widespread tetraploidH. brachyantherum that thrives in N. America and N.E. Asia;H. capense grows in S. Africa andH. secalinum mainly in Europe. Various cluster analyses were used followed by cluster recovery verification. Classificatory discriminant analysis and validation by the bootstrap yielded 85–90% overall total correct classification of the four species. Canonical analysis revealed thatH. californicum occupies an intermediate phenetic position among the other three distinct species. Factors of shape differences were unravelled and portrayed by shearing. A revised key to species was drawn up.

Classification of cultivated barley (Hordeum vulgare). 2. Elaboration of cultivar groups

Two barley (Hordeum vulgare L.) cultivar groups were established from 329 world-representative accessions of named cultivars scored for 37 morphometric characters. The two groups are similar to the popular division of two rowed and six rowed, one group containing mostly two-rowed barleys but also a few of the other kind and the second group mostly six rowed barleys together with some two-rowed. Both groups include naked and covered types, "iregular," "deficiens," and other types. The implications of the two groups for barley cultivar nomenclature are discussed. The two cultivar groups were discerned by the application of various hierarchical cluster analyses in combination with three different stopping rules as indicators to cluster numbers, supported by classificatory discriminant analysis with cross validation, and their relationships illustrated by canonical discriminant analyses.

Computerized classification of nuclear profiles in non-Hodgkin's lymphomas.

The authors present experimental technics for the diagnosis of non-Hodgkin's lymphomas, based on instrumental classification of nuclear profiles using a video-based system for computerized interactive morphometry (CIM). In their system, the real time video image of a specimen is superimposed to a graphics overlay generated by a computer, consisting of a test area with four visual markers for random sampling of cells and a menu with several options to send direct commands to the system. Using a touch-sensitive screen mounted on a video monitor as an interactive peripheral, a trained observer traces 100 randomly selected lymphoid cells, counts mitoses in 25 microscopic fields, and categorizes the lesion as diffuse or nodular. Each cell is instrumentally classified into either small cell noncleaved, small cell cleaved, or large, based on the length of their nuclear profiles, their enclosed nuclear area, and a circularity factor. Thereafter the computer provides a "diagnosis," based on hierarchic analysis of the data. The morphometric data are also interpreted by alternate statistical methods of discriminatory classificatory analysis that provide a diagnosis and a probability statement derived from matching unknown cases with a data base. Forty-two lymphoid lesions have been categorized with the CIM system. Studies of interobserver and intraobserver variations in data collection are discussed. The potential advantages of CIM for the objective classification of non-Hodgkin's lymphomas are discussed.

Fowler's Type G Penannular Brooches Reconsidered

BASIC, but rigorous, classificatory analysis is still a principal requirement in the archaeology of western and northern Britain in the early post-Roman period. The largest group of mainland penannular brooches—type G in Mrs Elizabeth Fowler's classification—is therefore reviewed in the light of recent discussions, and for the first time most are illustrated. Clear distinctions of type in date and distribution are confirmed, notably between those from Scotland and northern Ireland (7th to 9th centuries) and those from further south (4th(?) to 6th centuries). The latter are examined in greater detail, and consideration of their dating involves an excursus on late Roman ring brooches.

Macrobenthic Communities of the Eastern Mediterranean Continental Shelf

Abstract. The results of a year long study in which soft-bottom epibenthic invertebrates were collected by grab, dredge and beam-trawl, along the southern Mediterranean coast of Israel, are described in this paper. The classificatory analysis used for both normal and inverse analyses used two measures of dissimilarity - Canberra metric and Bray-Curtis, and both group-average and nearest-neighbour clustering. The results were displayed as dendrograms. Four site groups and five species groups characterised a total of 58 site samples and 245 species. The prominent species in each of the five species groups are mentioned. The applicability of “community concepts” and the effects of sediment properties on community structure are discussed.

Craniometric Variation In The Northern And Central Plains

Cranial measurements of 13 male and 12 female samples from the Central and Northern Plains region were subjected to canonical analysis. The samples include historic or protohistoric crania that can be ascribed to the Arikara, Mandan, Pawnee, Ponca and Omaha tribes. In addition, two samples belong to the archaeologically defined St. Helena Focus. Both sexes yielded five significant canonical variates, although only four were readily interpretable. The first canonical variate is clearly a Siouan-Caddoan discrim inator and reflects variation in cranial vault height. St. Helena sites associate with the Arikara on this axis, supporting previous craniometric analyses which suggest a relationship between these two groups. Subsequent canonical variates deal with more particular aspects of craniometric variation among groups, but are still interpretable in historic or evolutionary terms. The classificatory analysis shows that the Arikara sites are closely related. A major exception to this is the Sully site, which frequently misclassifies with non-Arikara groups. This suggests that the Sully crania have little collective reality; and that there may be non-Arikara components represent ed at the Sully Site.

Appraisal by Numerical and Statistical Techniques of the Interaction of Moisture Stress and Phosphorus in the Development of the Tropical Legume Macroptilium atropurpureum (Dc.) Urb

The interaction of two levels of soil moisture and four levels of phosphorus on the development of Macroptilium atropurpureurn was studied for 18 harvests at 2-day intervals beginning on day 29 after sowing. Variates studied were fresh weight. dry weight, nitrogen and phosphorus contents for the laminae. stem plus petioles, root and nodules. The responses to treatment of these variates were shown by preliminary analyses of variance to contain complex patterns of interaction. The important facets of these interactions were then appraised by an extension of the usual univariate methods and the overall pattern of the interaction was delineated by a classificatory analysis. On the basis of the 14 attributes studied, there were six groups of responses to treatment. The lowest of these groups corresponded to failure to grow at low levels of phosphorus. The next group represented the response of nodules to treatment. Subsequent groups represented higher growth potential following efficient functioning of the nodules. Effects on nodulation were thus primary to subsequent patterns of plant development. Water stress greatly impaired both plant and nodule development, particularly at the higher two levels of phosphorus, whereas active growth occurred in normally watered plants. At these two phosphorus levels the effect of water stress was on development rather than on chemical composition, there being large absolute but small relative effects on chemical data. The lower levels of phosphorus supply contrasted with the two higher levels in having much lower rates of growth and of water loss.

Delimitation of subtidal macrobenthos associations at the mouth of the Brisbane River

One hundred and fifty dredge samples of macrobenthic animals from the mouth of the Brisbane River were analysed by zonation analysis, Petersen-type community analysis, and hierarchical classificatory analysis. The methods were compared in terms of the faunal distribution patterns that they yielded. Hierarchical analysis gave the most acceptable results. Although the normal dendro- gram could be interpreted meaningfully to at least 16 sample groups, it has been interpreted to six groups here in preparation for subsequent, more detailed analyses. Spatial and temporal distribution patterns have been recognized and the consideration of the sample groupings as faunal states has been emphasized.

Second-order isomorphism of internal representations of familiar faces

Following a suggestion by Shepard and Chipman (1970), photographs and names of 15 well-known faces were used as stimuli in judgments of similarity. Ss ranked all pairs of stimuli in terms of facial similarity either from memory when presented with pairs of names, or from photographs. There was a highly significant positive correlation between these two types of judgment. A simple classificatory analysis of the 15 faces in terms of the physical features provided evidence that judgments based on the internal representation of the faces utilized pictorial information. These results extend the generality of Shepard and Chipman’s findings and confirm the usefulness of an approach to the problem of mental images based on the concept of second-order isomorphism.

Classification by Multivariate Analysis with Objectives of Minimizing Risk, Minimizing Maximum Risk, and Minimizing Probability of Misclassification

THE PURPOSE of this study is to I describe and illustrate certain developments in the field of classificatory analysis which are po tentially useful in education. The problem un der consideration arises when a number of measurements are made on an individual and it is desired to classify the individual into one of sev eral categories on the basis of these measure ments. It is assumed that the i n d i v i d ual has been randomly drawn from one of the populations and that in each population there is a probability distribution of the measurements. In problems of this type it frequently may be observed that the consequences of wrong decisions are not equally undesirable. Somewhat unique is the assumption that it is possible to specify the loss of utility resulting from a wrong decision or misclassification. Loss may be thought of as the penalty paid by the statistician when his guess is incorrect. In this investiga tion it serves to differentiate the seriousness of the possible errors of classification. By the risk of committing a certain error in classification is meant the probability of that error multiplied by the corresponding loss. Risk thus corresponds to the expected value of the loss or the expected disutility in the long-r un use of some classification rule. In considering the problem of classifying an individual into one of several populations there is a risk or expect ed loss associated with each possible choice or classification. Important in the solution of problems of class ification are the a priori probabilities whose values postulate the respective chances of drawing an individual from each of the popu lations under consideration. The c r i t e rion by which classification rules are chosen va r i e s with the presence or absence of the a priori probabilities. If they are assumed to be known, rules are selected which minimize overall risk I of misclassification. If in a particular problem it is not possible or appropriate to es tima te a priori probabilities, the solution requires a dif ferent criterion of goodness. Under such condi tions it is possible to choose classification rules which minimize the maximum risk of misclassi fication. This is a conservative approach and represents an application of the minimax princi ple. Anderson (1) discusses these criteria in detail. Rao (6), Brown (2), and Welch (9) are useful ref erences. More formal reasoning is gener ally expressed as follows: Assume, for exam pie, that an individual I with the set of measure ments xx-,xq has been randomly drawn from one of two populations nx or n2 with prob ability distributions of the measurements ix(xlf ...., xq) and f2 (xx,_, Xq), respectively. The choice of a classification rule R corresponds to a division of the q-dimensional sample space into two regions Rx and R2. If the random point corresponding to the individual falls in Rx the de cision is made that I is from TLl9 and if the point falls in R2 the decision is made that I is from n2. Let L(2/l) be the loss incurred in classifying I into H2 when in fact I belongs to Ul9 and let L(l/2) be the loss incurred in classifying I into nx when in fact I belongs to n2. For any rule R let rx(R) be the conditional risk or ex pected loss when I belongs to Ux. It is defined as the probability of I falling in R2 multiplied by the loss associated with this mistake. Simi larly, let r2(R) be the conditional risk when I belongs to H2. It is defined as the probability of falling in Rx multiplied by the corresponding loss. Suppose there exist a priori probabilities that I comes from the two populations, say probabil ity Pi that I does in fact belong to Ul9 and prob ability p2 that I belongs to II2, where px + p2 =