Fatal child neglect: Characteristics,causation, and strategies for prevention

Research in child fatalities because of abuse and neglect has continued to increase, yet the mechanisms of the death incident and risk factors for these deaths remain unclear. The purpose of this study was to systematically examine the types of neglect that resulted in children's deaths as determined by child welfare and a child death review board. This case review study reviewed 22 years of data (n = 372) of child fatalities attributed solely to neglect taken from a larger sample (N = 754) of abuse and neglect death cases spanning the years 1987–2008. The file information reviewed was provided by the Oklahoma Child Death Review Board (CDRB) and the Oklahoma Department of Human Services (DHS) Division of Children and Family Services. Variables of interest were child age, ethnicity, and birth order; parental age and ethnicity; cause of death as determined by child protective services (CPS); and involvement with DHS at the time of the fatal event. Three categories of fatal neglect – supervisory neglect, deprivation of needs, and medical neglect – were identified and analyzed. Results found an overwhelming presence of supervisory neglect in child neglect fatalities and indicated no significant differences between children living in rural and urban settings. Young children and male children comprised the majority of fatalities, and African American and Native American children were over-represented in the sample when compared to the state population. This study underscores the critical need for prevention and educational programming related to appropriate adult supervision and adequate safety measures to prevent a child's death because of neglect.     

Screening for school readiness: The influence of birthdate and sex

School readiness has been measured in many ways; chronological age, sex, and developmental age all have been used, with varying success, to predict young children's readiness for school. One hundred and fifty-two children were classified by birthdate and sex, and compared on their performances on the Gesell Screening Test, the Gesell School Readiness Test (GSRT), and the Stanford Achievement Test. The Gesell measures were sensitive to the different birthdate groups; however, these differences diminished as the children aged. There were no significant differences between birthdate groups on the Stanford Achievement Test. Females scored higher than males on the GSRT at the end of kindergarten, but no interactions between birthdate and sex were found. Implications of the varying ways to determine school readiness are discussed.     

Predicting elementary science learning using national assessment data

This study made use of data collected during 1981—1982 from a random sample of 1960 nine-year-old students from 124 elementary schools involved in a national assessment of educational progress in science sponsored by the National Science Foundation. This data base was used in secondary analyses which probed the validity of a model of educational productivity involving a set of nine aptitudinal, instructional, and environmental variables which require optimization to increase student learning. When controlled for other factors, ability, motivation, class environment, home environment, amount of television viewing (negative direction), gender, and race were all found to be significantly related to achievement. For an attitude outcome, the factors linked with attitudinal attainment were ability, motivation, class environment, and race. Overall the findings supported the model of educational productivity and suggested that elementary science students' achievement and attitude are influenced jointly by a number of factors rather than one or two dominant ones. Also the study attests to the potential value of science education researchers performing secondary analyses on the high-quality random data bases generated as part of this national assessment.     

Author Correction: Digital Citizenship: you Can’t Go Home Again

The original version of this article unfortunately contained a mistake. Figure 3 on the published version of the article was wrong. The correct Fig. 3 is now presented below.     

Estimation of Classification Consistency When the Probability of a Correct Response Varies

The purpose of this study was to examine the effect that large, within-examinee item difficulty variability had on estimates of the proportion of consistent classification of examinees into mastery categories over two test administrations. The classification consistency estimate was based on a single test administration from an estimation procedure suggested by Subkoviak (1976). Analyses of simulated data revealed that the use of a single estimate for an examinee's probability of a correct response, even when that probability varied greatly within a test for an examinee, did not affect the estimation of the proportion of consistent classifications.