Showing papers in "Educational and Psychological Measurement in 1978"
TL;DR: In this article, the authors analyzed background characteristics, school performance, and achievement test data for 788 third grade boys and 774 third grade girls who were known later to have become high school dropouts or graduates.
Abstract: Background characteristics, school performance, and achievement test data were analyzed for 788 third-grade boys and 774 third-grade girls who were known later to have become high school dropouts or graduates. As early as the third grade, dropouts differed significantly from graduates in age, IQ level, marks received in course work, parent's occupational and educational level, family size, marital status of parents, and tested reading, arithmetic, and language skill achievement. A combination of predictors produced a multiple correlation with dropout or graduation of .51 for boys and .49 for girls. A discriminant function yielded an overall correct classification of dropouts and graduates of 75%. Although prediction was less accurate for dropouts than for graduates, 6 or 7 of every 10 later school failures were correctly classified by characteristics exhibited in the third grade.
290 citations
TL;DR: For example, the occurrence of standardized regression coefficients greater than one in a model raises questions concerning the legitimacy of such coefficients, and poses a serious problem as discussed by the authors, and poses serious problem.
Abstract: For some investigators, the occurrence of standardized regression coefficients greater than one in a model raises questions concerning the legitimacy of such coefficients, and poses serious problem...
176 citations
TL;DR: In this paper, a generalisation of Rasch's simple logistic model for dichotomously scored achievement items is presented, which has consequences for both statement scaling and person measurement.
Abstract: The statement scaling emphasis associated with the Thurstone tradition for the study of attitude provides no direct consequence for attitude measurement, while conversely, the attitude measurement emphasis in the Likert tradition provides no direct consequence for statement scaling. Both aspects of the study of attitude are unified by a generalisation of Rasch's simple logistic model for dichotomously scored achievement items. The generalized model caters for the response category system of attitude questionnaires constructed and scored in the Likert tradition, but because both statements and persons are parameterised, the model has consequences for both statement scaling and person measurement. As with the simple logistic model, the two sets of parameter estimates, in this case those of the statements and of the persons, are independent of each other. An illustration involving attitudes to the professional roles of teachers is provided.
155 citations
TL;DR: In this paper, it was shown that if a behavior domain can be described by the common factor model with a finite number of factors, the squared correlation between the sum of a selection of items and the domain is bounded.
Abstract: It is shown that if a behavior domain can be described by the common factor model with a finite number of factors, the squared correlation between the sum of a selection of items and the domain tot...
143 citations
TL;DR: The Suinn Test Anxiety Behavior Scale (STABS) as mentioned in this paper was used to measure test anxiety in two groups of undergraduates: females had significantly higher MARS scores than males, but these sex differences did not replicate.
Abstract: The Mathematics Anxiety Rating Scale (MARS) was given to two groups of undergraduates. The first group also completed a questionnaire on their background in mathematics; the second, the Suinn Test Anxiety Behavior Scale (STABS). In both studies Humanities majors scored higher in mathematical anxiety than did Social Science majors who in turn had higher MARS scores than did Physical Science majors. In the first study females had significantly higher MARS scores than did males, but these sex differences did not replicate. In further validation of the MARS there was an inverse relationship between high MARS scores and (a) number of years of high school mathematics, (b) number of terms of Calculus, and (c) grades achieved in high school mathematics. Higher MARS scores were positively correlated with dislike of mathematics, a self-report of anxiety about the subject, the length of time such anxiety had persisted, and test anxiety as measured by the STABS.
130 citations
TL;DR: In this article, a similarity generating function composed of elevation, scatter and shape parameters is presented, and linear models for integrating these parameters either for euclidean distance or vector-product association indices.
Abstract: The aims of this paper are: (1) to present a similarity generating function composed of elevation, scatter and shape parameters; (2) to describe linear models for integrating these parameters either for euclidean distance or vector-product association indices; and (3) to suggest a computational strategy based upon the Eckart-Young (1936) theorem that has certain advantages for minimizing the effects of measurement error in estimating profile similarity. Given these developments, the investigator may differentiate the independent contribution of each parameter to more global indices of resemblance. A brief example from the classification of psychopathology is discussed.
125 citations
TL;DR: In this article, a procedure for estimating the reliability of a factorially complex composite is considered, where the reliability is defined as the probability of a composite composite composite having a given reliability.
Abstract: A procedure for estimating the reliability of a factorially complex composite is considered.
115 citations
TL;DR: Two investigations of students entering college (Ns = 84 and 262) and one of the students completing high school (N = 147) explored the presence of a psychological competence configuration as a co...
Abstract: Two investigations of students entering college (Ns = 84 and 262), and one of the students completing high school (N = 147) explored the presence of a psychological competence configuration as a co...
104 citations
TL;DR: In this article, specific category agreement levels for both nominally and ordinally scaled data are computed for both ordinally and nonsmooth data, and an option is available for collapsing the original scale to a small scale.
Abstract: This program computes specific category agreement levels for both nominally and ordinally scaled data. For ordinally scaled data, an option is available for collapsing the original scale to a small...
93 citations
TL;DR: In this paper, a semipartial correlation to the zero-order correlation is proposed, which can be easily extended from the two predictor case to the general p-predictor case.
Abstract: Since Horst (1941) initially introduced the concept, a number of investigators have suggested alternative definitions that both include a broader class of situations and are more precise. Following a conceptual framework suggested by Conger (1974), the present paper discusses a definition based on the relation of the semipartial correlation to the zero order correlation. This definition possesses a number of advantages, including the fact that it can be easily extended from the two predictor case to the general p-predictor case.
84 citations
TL;DR: In this article, reliability and validity data are reported for a questionnaire measure of individual differences in achieving tendency which is an expanded and improved version of Mehrabian's (1969) scales The fi
Abstract: Reliability and validity data are reported for a questionnaire measure of individual differences in achieving tendency which is an expanded and improved version of Mehrabian's (1969) scales The fi
TL;DR: In this paper, reliability and validity data are presented for three orthogonal measures of temperament: trait-pleasure, trait-arousal, and traitdominance, and the KR-20 reliability coefficients for the three scales are analyzed.
Abstract: Reliability and validity data are presented for three orthogonal measures of temperament: trait-pleasure, trait-arousal, and trait-dominance. The KR-20 reliability coefficients for the three scales...
TL;DR: The Nonsexist Personal Attribute Inventory for Children (NPAIC) as mentioned in this paper is a self-concept construct that is essentially free of sex bias and has a test-retest correlation of.62 (p <.0001) with the Piers-Harris scale.
Abstract: Thirty-two words that were found to be nonsexist in nature according to the responses of 686 children, were derived from the Personal Attribute Inventory for Children. Half of the words are defined to be positive; the other half, negative. Together, they make up the Nonsexist Personal Attribute Inventory for Children. This inventory was administered along with the Piers-Harris scale to 297 fifth through eighth grade children. A correlation of .49 (p <.0001) was found between these two scales. Over a one month interval a test-retest correlation of .62 (p < .0001) was obtained for the Nonsexist Personal Attribute Inventory for Children (NPAIC), which was based on the responses of 272 of the previously mentioned initial sample of 297 children. It would appear that a relatively reliable and valid measure of a broad self-concept construct has been achieved that is essentially free of sex bias.
TL;DR: In this paper, the power of F tests for numerator degrees of freedom was computed and the significance levels, alpha =.05,.10(10),.50, and phi = 2.3.
Abstract: In numerous applications of the analysis of variance it is necessary to compute the power of F tests having numerically high alpha (significance) levels. This article tabulates the power of F tests for numerator degrees of freedom, df = 1, 2, 3, 6, 9, 12; denominator df = 3, 6, 9, 12, 15, 20, 40, 60, 120; alternative hypotheses, phi = 2.(.2)3.0; and significance levels, alpha = .05, .10(.10),.50. The use of these tables is illustrated with a brief numerical example.
TL;DR: Two hundred twenty-two students at Michigan State University completed a questionnaire including the Rational Behavior Inventory (RBI), the Srole Anomia scale, Lane's Authoritarianism scale, a ten-...
Abstract: Two hundred twenty-two students at Michigan State University completed a questionnaire including the Rational Behavior Inventory (RBI), the Srole Anomia scale, Lane's Authoritarianism scale, a ten-...
TL;DR: In this paper, the Piaget Matrices and Order of Appearance tasks were administered to 189 children, 5-10 years of age, who represented three ethnic groups: Anglo, Black and Mexican-American.
Abstract: Piaget Matrices and Order of Appearance tasks were administered to 189 children, 5-10 years of age. The children represented three ethnic groups: Anglo, Black, and Mexican-American. Three conditions of testing were employed. Differences between conditions were detected for all age groups with performance in elaborated conditions exceeding performance under standard procedures. An interaction was identified for age × condition for both tasks. A tendency toward an interaction was identified for condition X ethnic group, with performance for Black and Mexican-American children tending to be below that of Anglos in the standard, but not in the fully elaborated condition. Results are discussed in terms of the importance of a differential testing approach to increase the validity of assessment of cognitive capabilities by reducing Type II error.
TL;DR: The authors examined the invariance of results from magnitude estimation and pair-comparison treatment of complete ranks scaling procedures and explored whether roughly-interval measurement is achieved by these methods.
Abstract: This article examined the invariance of results from magnitude estimation and pair-comparison treatment of complete ranks scaling procedures and explored whether roughly interval measurement is achieved by these methods. Magnitude estimation and ranking questionnaires were administered to the same set of 116 college students, and their responses were used to scale sets of 20 and 10 frequency expressions, respectively. A comparison of the results obtained with those from two earlier studies indicated that magnitude estimation scaling produced fairly invariant anchor points which appear to attain at least roughly equal interval measurement, whereas pair-comparison treatment of complete ranks did not. Implications for validity of measurement are briefly discussed.
TL;DR: Joreskog's procedure for the analysis of simplex models was used to test Humphreys' (1968) assertion that eight semesters of undergraduate grade-point averages have a simplex form as discussed by the authors.
Abstract: Joreskog's (1970a) procedure for the analysis of simplex models was used to test Humphreys' (1968) assertion that eight semesters of undergraduate grade-point averages have a simplex form Not only was this assertion confirmed but precise estimates of reliability and of unattenuated correlations were obtained
TL;DR: In this article, the universe of possible relationships among two predictors and a common criterion is analyzed following Conger's (1974) revised definition of suppressor variables, and a simple mapping of the possibl...
Abstract: Following Conger's (1974) revised definition of suppressor variables, the universe of possible relationships among two predictors and a common criterion is analyzed. A simple mapping of the possibl...
TL;DR: In this paper, a questionnaire measure of individual differences in dominance-submissiveness was proposed, based on an initial set of 457 items which was subsequently augmented by an additional 62 rewritten and new items.
Abstract: Reliability and validity data are reported for a questionnaire measure of individual differences in dominance-submissiveness. The measure was based on an initial set of 457 items which was subsequently augmented by an additional 62 rewritten and new items. The initial pool of items, which was carefully constructed to assure broad generality, contained 64 content groups representing various aspects of dominance-submissiveness. The final 48-item dominance-submissiveness questionnaire, which was balanced for response bias, had high internal consistency (Kuder-Richardson formula 20 coefficient was equal to .95). It was independent of social desirability (correlated .08 with the Crowne and Marlowe social desirability scale). As the scale also satisfied the requirements of low correlations with measures of trait pleasure ( r = .14) and arousability (r = -.10), it can thus be used with these two measures for a three-dimensional characterization of temperament. Correlations with other available measures of domina...
TL;DR: The reactions and opinions of 163 examinees who participated in a tailored testing pilot study conducted at the U. S. Civil Service Commission during the fall of 1975 are presented.
Abstract: In the last few years, computer assisted tailored testing has been shown to be technically feasible; the question of its acceptability to examinees therefore assumes increased importance. This report presents the reactions and opinions of 163 examinees who participated in a tailored testing pilot study conducted at the U. S. Civil Service Commission during the fall of 1975. The reactions of the examinees can accurately be summarized as overwhelmingly positive, thus disposing of yet another potential obstacle to the advance of this innovative technique.
TL;DR: The ability of the Visual-Motor Integration Test (VMI) to predict academic achievement was assessed by using three samples of pre-kindergarten pupils from three successive years in a large midwest region as discussed by the authors.
Abstract: The ability of the Visual-Motor Integration Test (VMI) to predict academic achievement was assessed by using three samples of pre-kindergarten pupils from three successive years in a large midwest ...
TL;DR: This paper found that SAT was the more valid predictor at a highly selective midwestern university compared to ACT in predicting first semester college grades at highly selective institutions, and made comparisons between the long and short form results of the SAT to determine equivalency of the tests.
Abstract: Previous research has suggested that although ACT and SAT are equally capable of predicting first semester college grades, ACT seems to predict first semester GPA more accurately at highly selective institutions. The present study found, instead, that SAT was the more valid predictor at a highly selective midwestern university. In addition, comparisons were made between the long and short form results of the SAT to determine equivalency of the tests as well as to determine equivalency of the results under different testing conditions.
TL;DR: For each of two revised forms of the Dimensions of Self-Concept (DOSC) measure (E form for elementary school pupils in grades 4, 5, and 6, and S form for secondary school students in grades 7 through 12), statistical information is presented concerning the intercorrelations of each of five factor scales, the reliability and standard error of measurement of each scale, and the results of item analyses as mentioned in this paper.
Abstract: For each of two revised forms of the Dimensions of Self-Concept (DOSC) measure (E form for elementary school pupils in grades 4, 5, and 6, and S form for secondary school students in grades 7 through 12), statistical information is presented concerning the intercorrelations of each of five factor scales, the reliability and standard error of measurement of each scale, and the results of item analyses. Also included are a brief summary of data pertaining to factorial validity and preliminary evidence regarding the concurrent validities of each of the five factor scales of the DOSC with respect to scores earned on tests of reading comprehension and critical thinking. The conclusion was reached that the DOSC affords considerable promise of providing relatively reliable measures of individuals within the elmentary school, the junior high school, and the senior high school on each of five fairly homogeneous dimensions of self-concept, the internal and factorial validity of which has received substantial support.
TL;DR: In this paper, the E-(correlation) coefficient concept is considered and six characteristics of an E-coefficient are mentioned for 23 similarity measures of interval, ordinal, dichotomous, and nominal data.
Abstract: The term correlation coefficient has been defined in various ways. In this article the E-(correlation) coefficient concept is considered. Six characteristics of an E-coefficient are mentioned. For 23 similarity measures of interval, ordinal, dichotomous, and nominal data is considered whether they are E-coefficients or not. Finally, the importance of the concept is discussed.
TL;DR: The Arizona Course/Instructor Evaluation Questionnaire as discussed by the authors is the result of over 15 years of research and experience in the area of instructional evaluation, which is the most widely used evaluation questionnaire.
Abstract: Development of the Arizona Course/Instructor Evaluation Questionnaire is the result of over 15 years of research and experience in the area of instructional evaluation. The method of constructing the initial item pool, successive data analyses, subscale development, and normative data generation are described. Related research is presented demonstrating the reliability and validity of the Arizona Course/Instructor Evaluation Questionnaire.
TL;DR: In this article, the authors investigate the interrelationships between scores on selected scales of the Minnesota Multiphasic Personality Inventory (MMPI), those on the Cornell Critical Thinking Test (Level Z), and academic success as measured one year later by Grade Point Average (GPA) and total number of graduate units completed.
Abstract: Involving use of 100 graduate student subjects, this study was concerned with the interrelationships between scores on selected scales of the Minnesota Multiphasic Personality Inventory (MMPI), those on the Cornell Critical Thinking Test (Level Z), and academic success as measured one year later by Grade Point Average (GPA) and total number of graduate units completed. Measures of ego development and critical thinking were significantly related for the sample of females, but not for that of males. Measures of personality maladjustment were not significantly related in either sample to a measure in ability in critical thinking. Finally, in each sample, critical thinking ability was predictive of success in graduate school.
TL;DR: The strengths and weaknesses of these indices (mean difference, percent overlap, percent agreement, correlation, and variance of difference scores) are discussed in the light of a proposed new definition of fakeability based upon Naylor's (1967) model of measurement accuracy as mentioned in this paper.
Abstract: Past practice of operationalizing the concept of fakeability is reviewed. The strengths and weaknesses of these indices (mean difference, percent overlap, percent agreement, correlation, and variance of difference scores) are discussed in the light of a proposed new definition of fakeability based upon Naylor's (1967) model of measurement accuracy.
TL;DR: In this paper, Cohen's kappa, κ, the chance corrected percentage of agreement between actual and predicted group membership, was used to set confidence limits for the accuracy of the discriminant prediction and to test the difference in predictive accuracy for two independent samples.
Abstract: Most programs for performing discriminant analysis provide a summary table of hits and misses in predicting group membership by using the discriminant function. The interpretation of such tables can be enhanced greatly by computing Cohen's kappa, κ, the chance corrected percentage of agreement between actual and predicted group membership. The standard error of kappa can be used to set confidence limits for the accuracy of the discriminant prediction and to test the difference in predictive accuracy for two independent samples. This was demonstrated in this article, using data previously published in a more preliminary form.
TL;DR: In this paper, the authors investigated the cross-domain correlations between 20 ability variables and 14 personality traits from a sample of 138 late adolescent males and used Tucker's interbattery factor model to reveal the structure of cross-battery correlations.
Abstract: Linkages between the abilities and personality domains are explicated from the results of a recent study, in which cross-domain correlations between 20 ability variables and 14 personality traits were obtained from a sample of 138 late adolescent males. A model designed to reveal the structure of cross-battery correlations— Tucker's interbattery factor model—was employed, and three interbattery factors were isolated. These inter-domain factors are discussed and related to earlier-discovered second-order factors in the personality domain.