Separating the Odds: Thresholds for Entropy in Logistic Regression

Abstract

Researchers are often reluctant to rely on classification rates because a model with favorable classification rates but poor separation may not replicate well. In comparison, entropy captures information about borderline cases unlikely to generalize to the population. In logistic regression, the correctness of predicted group membership is known, however, this information has not yet been utilized in entropy calculations. The purpose of this study was to, 1) introduce three new variants of entropy as approximate-model-fit measures, 2) establish rule-of-thumb thresholds to determine whether a theoretical model fits the data, and 3) investigate empirical Type I error and statistical power associated with those thresholds. Results are presented from two Monte Carlo simulations. Simulation results indicated that EFR-rescaled was the most representative of overall model effect size, whereas EFR provided the most intuitive interpretation for all group size ratios. Empirically-derived thresholds are provided.     

The Relationship between Parenting Types and Older Adolescents' Personality, Academic Achievement, Adjustment, and Substance Use

The purpose of the present study was to examine Baumrind's T3 conceptual framework using a multiple informant design and an older adolescent population. With 178 college students and their families as participants, the present study found many of the predicted relations between parents' child-rearing style (Authoritative, Democratic, Nondirective, Nonauthoritarian-Directive, Authoritarian-Directive, and Unengaged) and their adolescent children's behavior in the 4 domains assessed: personality, adjustment, academic achievement, and substance use. The differences between parenting types on the criterion measures were not as large as reported in Baumrind's study, and significant effects were predominantly due to the poor scores from children with Unengaged and Authoritarian-Directive parents. The results are discussed in terms of their implications for the Authoritative parenting type, the utility of using a typology, and areas for future research.     

Auditing for Score Inflation Using Self-Monitoring Assessments: Findings From Three Pilot Studies

Test-based accountability often produces score inflation. Most studies have evaluated inflation by comparing trends on a high-stakes test and a lower stakes audit test. However, Koretz and Beguin (2010) noted weaknesses of audit tests and suggested self-monitoring assessments (SMAs), which incorporate audit items into high-stakes tests. This article reports the first three trials of SMAs, evaluating whether SMAs can detect inflation that had already been documented. The studies were conducted with mathematics tests in three grades. Despite severe conservative biases, the audit component functioned as expected in many of the trials. The difference in performance between nonaudit and audit items was associated with factors that earlier research showed to be related to test preparation and score inflation, such as scoring just below the Proficient cut in the previous year and school poverty. However, a number of null findings underscore the need for additional research into the design of audit items.     

Integrating control generated by positive and negative reinforcement on an operant baseline: Appetitive-aversive interactions

The interaction of opposing motivational states was measured within a design in which rats barpressed for food in one stimulus and to avoid shock in another. Tone and light discriminative stimuli (Sds) were counterbalanced over incentive conditions. Extinction eliminated responding when neither Sd was present. To minimize the influence of competing peripheral operants or reinforcer-elicited behaviors during appetitive-aversive interactions, contingency parameters were manipulated to generate similar rates and patterns of barpressing in both Sds and stimulus-compounding tests were administered in extinction. On these tests, rates in tone, light, and tone plus light (T+L) were equivalent. In contrast, when the same reinforcer (i.e., food or shock avoidance) maintained comparable training rates in tone and in light, in testing, T+L controlled double the rates of the single stimuli—strong additive summation. These results were strikingly similar to those of single-incentive experiments concerned with the contribution of excitatory and inhibitory incentive states to the results of stimulus compounding. Simultaneously presenting two Sds whose implicit stimulus-reinforcer (S-S^r) contingencies were arranged to make them, respectively, conditioned appetitive and aversive exciters (present experiment) produced test results comparable to those of two Sds whose implicit S-S^r contingencies were arranged to make them both conditioned inhibitors. Reciprocal antagonism between these two motive states more than neutralizes them. It appears to produce a negative (i.e., an inhibitory) motive state.     

Children at play: the role of peers in understanding the effects of child care

Although the studies of the National Institute of Child Health and Human Development (NICHD) Early Child Care Research Network (this issue) and of Watamura, Donzella, Alwin, and Gunnar (this issue) identify the potential of child care to affect children's adjustment, how these effects are produced remains unclear. In this commentary, it is argued that there is a need to expand child care research by considering one of the most important, but unrecognized, contributors to child care effects-peers. Research is reviewed that suggests that children's interactions in same-sex peer groups at child care affect their short- and long-term adjustment. We consider how research on early peer influences can inform the findings of the NICHD Early Child Care Research Network and Watamura et al. studies, aswell as contribute to the next generation of child care research.     

Fostering the Development of Reading Skill through Supplemental Instruction: Results for Hispanic and Non-Hispanic Students

This paper reports the effects of a two-year supplemental reading program for kindergarten through third grade students that focused on the development of decoding skills and reading fluency. Two hundred ninety-nine students were identified for participation and were randomly assigned to the supplemental instruction or to a no-treatment control group. Participants' reading ability was assessed in the fall, before the first year of the intervention, and again in the spring of years 1, 2, 3, and 4. At the end of the two-year intervention, students who received the supplemental instruction performed significantly better than their matched controls on measures of entry level reading skills (i.e., letter-word identification and word attack), oral reading fluency, vocabulary, and comprehension. The benefits of the instruction were still clear two years after instruction had ended with students in the supplemental-instruction condition still showing significantly greater growth on the measure of oral reading fluency. Hispanic students benefited from the supplemental reading instruction in English as much as or more than non-Hispanic students. Results support the value of supplemental instruction focused on the development of word recognition skills for helping students at risk for reading failure.     

Pre‐schoolers’ feature placement on own and others’ person drawings

Young children find meaning in the drawings they create that may not be apparent to an adult observer. The purpose of this study is to access the children’s views using a drawing change task. Seventy‐three pre‐schoolers were asked to draw a person and then draw a person with a belly button. It was anticipated that tadpole (no separate body) drawers would include a body to accommodate this new feature. Instead belly buttons were included without modifications to the figure suggesting that in the children’s view tadpole figures are not as deficient as they might appear. When placing a belly button on a figure drawn by someone else 40% of tadpole drawers responded differently compared to their own. Possible explanations considered are the effect of active involvement in creating a drawing, the respect for another’s drawing intention, and lack of awareness of differences due to working memory limitations.     

Relations Among the Home Language and Literacy Environment and Children’s Language Abilities: A Study of Head Start Dual Language Learners and Their Mothers

Research Findings: This study explored the relations between Spanish–English dual language learner (DLL) children’s home language and literacy experiences and their expressive vocabulary and oral comprehension abilities in Spanish and in English. Data from Spanish–English mothers of 93 preschool-age Head Start children who resided in central Pennsylvania were analyzed. Children completed the Picture Vocabulary and Oral Comprehension subtests of the Batería III Woodcock–Muñoz and the Woodcock–Johnson III Tests of Achievement. Results revealed that the language spoken by mothers and children and the frequency of mother–child reading at home influenced children’s Spanish language abilities. In addition, the frequency with which children told a story was positively related to children’s performance on English oral language measures. Practice or Policy: The findings suggest that language and literacy experiences at home have a differential impact on DLLs’ language abilities in their 2 languages. Specific components of the home environment that benefit and support DLL children’s language abilities are discussed.