A method for drawing inferences from complex samples is based on Rubin's approach to missing data in survey research. Standard procedures for drawing such inferences do not apply when the variables of interest are not observed directly, but must be inferred from secondary random variables which depend on the variables of interest stochastically.…

This paper analyses the measurement issues in education achievement, and integration of education goals and targets, in the context of human development in India. Measurement issues are distinguished by (a) choice of indicators and variables and (b) data used in estimation/projection/computation of indicators and variables in the global human…

Traditional predictor variables of SAT scores and high school grades seldom account for more than 16 percent of the variance in student performance in college and graduate school, and they account for even less in decisions of students to leave school. Three studies are presented that added nonintellective variables to the traditional combination…

Structural equation models incorporating unmeasured variables make possible the rigorous testing of theories previously difficult to test adequately because of fallible measures of the theoretic variables. This paper first discusses a simple causal model; incorporating a single unmeasured variable for the purpose of exposition. A substantive…

In a study of factors in the educational environment of the home, the school, and the peer group contributing to achievement and attitudes towards schooling at the upper primary and lower secondary school levels, an opportunity arose to examine the data available for configural and curvilinear relationships. Information was collected on 11…

Methodological challenges associated with structural equation modeling (SEM) and structured means modeling (SMM) in research on school violence and related topics in the social and behavioral sciences are examined. Problems associated with multiyear implementations of large-scale surveys are discussed. Complex sample designs, part of any…

Q-type factor analysis was used to re-analyze baseline data collected in 1957, on 47 men aged 65-91. Q-type analysis is the use of factor methods to study persons rather than tests. Although 550 variables were originally studied involving psychiatry, medicine, cerebral metabolism and chemistry, personality, audiometry, dichotic and diotic memory,…

The study assessed the self-concept of academic ability among 6th, 8th, 10th, and 12th grade Indian children, compared their level of self-concept with that of other children enrolled in the same classes to find out the role of socioeconomic and ethnic background in determining the level of self-concept, and examined the factors that correlated…

An efficiency study was conducted of the sampling design used for the National Longitudinal Study of the High School Class of 1972 (NLS), and the change variables used in the first followup. Nine alternative designs were compared, using a cost model based upon the number of schools and students sampled. Fourteen change variables and eight domains…

This survey attempted to obtain data on the probability of improving student achievement when a new instructional program is implemented in an elementary school. The purpose of estimating the variable probability of improving student achievement was for use in the specification of a decision model whose purpose is to rank educational goals in…

The first set of National Longitudinal Study (NLS) data was analyzed to derive composite scores to be used in subsequent analyses. The composites include a general ability measure, a socioeconomic status (SES) index, and factor scores from factor analyses of sets of question items measuring goals, attitudes, and other psychological attributes. The…

Five statistical techniques were used to analyze the data collected in a previous article (TM 511 223) on student achievement, attention, and motivation in secondary school mathematics and science: ordinary least squares regression, canonical correlation, factorial modeling, partial least squares path analysis, and linear structural relations…

To calculate the economic benefits of schooling and compensatory education, a statistical model estimated the effects of schooling and literacy on earnings and employment. The literacy measure was obtained for a probable sample of the U.S. population in 1972, from the National Reading Survey. The measure was administered with a background…

The Quality Measurement Project was established in 1956 by the New York State Education Department to determine the feasibility of measuring quality differences between schools and of identifying variables related to these differences. In the latest revision of the project, begun in 1964, 99 representative school systems, excluding the New York…

Since the mean score for a sample composed of several subgroups can be viewed as the sum of the mean of each subgroup weighted by the proportional size of the subgroup, then the mean change in a time period--in this case, from 1972 to 1980--is the sum of the differences between the means for each subgroup, with each mean weighted by its…