David T. Morse

Minsize2: a Computer Program for Determining Effect Size and Minimum Sample Size for Statistical Significance for Univariate, Multivariate, and Nonparametric Tests

The American Psychological Association’s editorial style urges authors to provide effect size estimates. Several journals, including Educational and Psychological Measurement, have adopted author guidelines that call for determining the minimum sample size necessary for a given result to have been declared statistically significant. MINSIZE2, a computer program that permits the user to determine both effect size and the minimum sample size needed for the results of a given analysis to be statistically significant, is described. Program applications for statistical significance tests for univariate, multivariate, and nonparametric statistics are provided. Explanation of the program’s operation is given, and examples are furnished.

The Healthy Teen Girls Project: Comparison of Health Education and STD Risk Reduction Intervention for Incarcerated Adolescent Females

Adolescent girls incarcerated in a state reformatory (N = 246) were recruited and assigned to an 18-session health education program or a time-equivalent HIV prevention program. Cohorts were assigned to conditions using a randomized block design separated by a washout period to reduce contamination. Post intervention, girls in the HIV risk reduction program demonstrated the acquisition of risk-reduction behavioral skills and improved condom application skill. At a follow-up assessment approximately 9 months after release from the correctional facility, girls in both conditions reported fewer unprotected sexual intercourse occasions and less sex while under the influence of alcohol or other drugs.

MINSIZE: A Computer Program for Obtaining Minimum Sample Size as an Indicator of Effect Size:

Several journals, including Educational and Psychological Measurement, have adopted author guidelines that call for determining the minimum sample size necessary for a given result to have been declared statistically significant. MINSIZE, an MS-DOS computer program that permits the user to determine the minimum sample size needed for the results of a given analysis to be statistically significant, is described. Program applications for statistical significance tests based on z, t, r, F, or x2 statistics are provided. Explanation of the programs operation is given for each of the types of statistics, and examples are furnished.

Confirmatory Factor Analysis of the Gibb Experimental Test of Testwiseness.

The purpose of the study was to assess the stability of a two-factor model recently proposed for the Gibb Experimental Test of Testwiseness using confirmatory factor analysis. Participants were 173 undergraduate volunteers. Confirmatory factor analyses using LISREL 8 were performed using total scores on the seven skills as data. One- and two-factor models were compared. Results indicated that the data fit the two-factor model and the simpler one-factor model. For this sample, the Gibb test could be characterized as tapping a general proficiency in testwiseness.

The Relative Difficulty of Selected Test-Wiseness Skills among College Students

Although there is much research on test-wiseness, ease of exercising various test-taking skills has received scant attention. Participants in the present study were 243 undergraduate students who attempted the 70-item Gibb Experimental Test of Testwiseness, designed to measure seven test-wiseness skills. The results indicated the use of specific determiners as a cue to be significantly more challenging than eliminating irrelevant alternatives, selecting the alternative having the most information, or using grammar cues. It appears that not all test-wiseness skills are equally easy to employ. Researchers or trainers addressing test-wiseness should take into account these differences.

Preliminary Study of the Khatena-Morse Multitalent Perception Inventory

The Khatena-Morse Multitalent Perception Inventory has two forms of 50 items each designed to identify leadership, music, art, and creative talent of people age 10 yr. and over. It was developed on the basis of previous research and hypotheses about these areas of talent reported and used by the first author in the construction of earlier records for assessment of talent. Most subjects can complete the inventory in 10 to 30 min. Scoring is done by counting affirmative responses; the total serves as a versatility index. Internal consistency, alternate forms, and test-retest reliability, and indices of validity indicate the inventory is promising as a brief screening device for multitalent presently defined as versatility.

Hurricane Katrina's impact on the mental health of adolescent female offenders

Abstract Exposure to multiple traumatic events and high rates of mental health problems are common among juvenile offenders. This study draws on Conservation of Resources (COR) stress theory to examine the impact of a specific trauma, Hurricane Katrina, relative to other adverse life events, on the mental health of female adolescent offenders in Mississippi. Teenage girls (N=258, 69% African American) were recruited from four juvenile detention centers and the state training school. Participants were interviewed about the occurrence and timing of adverse life events and hurricane-related experiences and completed a self-administered mental health assessment. Hierarchical linear regression models were used to identify predictors of anxiety and depression. Pre-hurricane family stressors, pre-hurricane traumatic events, hurricane-related property damage, and receipt of hurricane-related financial assistance significantly predicted symptoms of anxiety and depression. Findings support COR theory. Family stressors had the greatest influence on symptoms of anxiety and depression, highlighting the need for family based services that address the multiple, inter-related problems and challenges in the lives of female juvenile offenders.

Factor structure of the WISC‐R for Mexican‐American students referred for intellectually gifted assessment

This study assessed the factor structure of the WISC‐R with Mexican‐American students referred for intellectually gifted assessment. The students included 68 Mexican‐American students who were referred for evaluation for an intellectually gifted program. The 12 WISC‐R subtest scaled scores were used as the variables to be factored. The method of factor extraction used was the maximum likelihood analysis followed by a varimax rotation. On each of the previously‐identified three factors, differences were observed.

ADDITIONAL EVIDENCE ON RELIABILITY AND VALIDITY FOR THE KHATENA-MORSE MULTITALENT PERCEPTION INVENTORY '

The Khatena-Morse Multitalent Perception Inventory has two forms of 50 items designed to identify leadership, music, art, and creative talent of people of age 10 yr. and over. Most subjects can complete the inventory in 10 to 30 min. Scoring is done by counting affirmative responses; the total serves as a versatility index. Using responses of 6,186 school-age children and 1,617 adults, this paper presents evidence of internal consistency, and alternate forms and test-retest reliability as well as validity indices based on the Khatena-Torrance Creative Perception Inventory and Thinking Creatively with Sounds and Words, the former giving an index of creative perception and the latter a measure of originality requiring processing of information for production. The data suggest the Khatena-Morse Multitalent Perception Inventory is promising as a brief screening device for versatility of talent.

Reliability estimates for total, factor, and group mean scores on the Khatena-Torrance Creative Perception Inventory

Responses of 31 samples of adults (N = 2503) and 37 samples of children (N = 7048) to the two parts of the Khatena-Torrance Creative Perception Inventory were used to estimate internal consistency reliabilities for total scales and factor scores and to estimate the reliability of group means. The information may be helpful to users of Something About Myself and What Kind of Person Are You? when deciding what typical reliabilities to expect or in planning studies involving the measures. The analyses suggest (a) only total scores for the inventories may approach acceptable reliability for making decisions about individuals, (b) factor scores tend to be less reliable so they may only be acceptable for making decisions about groups, and (c) reliabilities of group means for total scores are acceptable for groups of about 40 adults or 65 children.