This paper illustrates the value of applying the law of parsimony to canonical correlation analysis solutions. The primary purpose of parsimony is that the more parsimonious the solution, the more replicable the model will be. The ultimate goal is to estimate an equal or reasonable amount of variance with the smallest variable set possible. A…

This study examined the relationship between adolescent (10-17 years) criminal offending and adult (18-33 years) offending. The sample comprised 378 Canadian male offenders whose criminal trajectory was tracked for an average of 12.1 years, from adolescence into adulthood. Their man age at the time of the most recent follow-up was 27.5 years. The…

This longitudinal study of 248 male offenders examined the relationship between psychiatric disorders, diagnosed in adolescence, and subsequent adult criminal activity. Criminal offences were tracked for an average of 8.7 years from age 18-33. Cox Proportional Intensity regression analyses were conducted to predict the rates of adult offending of…

While other surveys seek to measure the extent of substance abuse in the population, the "CASA National Survey of American Attitudes on Substance Abuse VIII: Teens and Parents" probes substance-abuse risk and identifies factors that increase or diminish the likelihood that teens will abuse tobacco, alcohol or illegal drugs. This year,…

This study was undertaken to determine the relative risk of violence among students attending Hawaii schools with higher proportions of children of Department of Defense (DOD) personnel. Approximately 12% of the total public school enrollment of nearly 190,000 students is DOD connected. Forty-eight schools (8 high schools, 8 intermediates, and 32…

This paper focuses on handling measurement error in predictor variables using item response theory (IRT). Measurement error is of great important in assessment of theoretical constructs, such as intelligence or the school climate. Measurement error is modeled by treating the predictors as unobserved latent variables and using the normal ogive…

Investigation of second- and third-graders' achievement in mathematics and its correlation with nonmathematical variables yielded the following results: listening ability and vocabulary levels were significant variables, intelligence was significant for grade three but not grade two, and sex and the ability to conserve were not significant for…

Results of the present study indicate that investigation into sib-sib and parental-surrogate-sib interactions cannot be overemphasized. (Author)

While stepwise regression analyses revealed that the demographic variables of sex, school ability, age, and father's occupation increased the power of prediction the LTIT alone was judged to be a valid predictor of academic achievement. (Author)

This article illustrates the role multiple linear regression can play in developing prediction equations by providing examples of regression models that could be used in answering questions relative to the importance of a single predictor variable, interactions between predictor variables, and the cross-validation and generalizability of…

Setwise regression analysis is a new technique developed to allow a stepwise solution when the interest is in sets of variables rather than in single variables. (CK)