On preservation of mean square continuity under zero memory non-linear transformation

Mean square continuity of a random process is of considerable theoretical and practical importance. Mean square continuous random processes passed through zero memory non-linearities are considered. Several conditions are given relating to mean square continuity of the output. Bounded zero memory non-linearities are considered with arbitrary random processes as inputs, and conditions on both the zero memory non-linearity and on the random process are examined. Also, general zero memory non-linearities are considered with Gaussian random processes as inputs.     

Asymptotic memoryless detection of random signals in dependent noise

Design of detectors for strong mixing signals in strong mixing noise is considered, where a large degree of dependency may occur between the signal and noise. Under the criterion of asymptotic relative efficiency, it is shown that this design reduces to determining the solution of an integral equation, where only knowledge of the second order statistics of the randon processes involved is required. In particular, if the signal is independent of the noise and has nonzero mean, the optimal detector is the same as in the known constant signal case. It is also shown that it is possible to delete several regularity conditions which may be difficult to check in practice in the slightly more restrictive case where the maximal correlation coefficients of the signal and noise tend to zero.     

The moderating influence of instructional intensity and word type on the acquisition of academic vocabulary in young English language learners

This study used a within-subjects design to explore two instructional conditions for developing vocabulary in second-grade Spanish-speaking English learners (ELs)—extended instruction and embedded instruction implemented during shared interactive reading. Words assigned to the extended condition were directly taught using a multifaceted approach that included visuals, bilingual definitions, examples, spelling, and partner-talk about the words’ meaning. Words assigned to the embedded condition were taught solely through brief definitions to embedded text, writing activities, and songs. In the control condition students heard the target words read aloud during shared interactive reading but without definitions or direct instruction. The study also explored the interaction between instructional condition and word type. Four types of words—abstract cognates, abstract noncognates, concrete cognates, and concrete noncognates—were randomly assigned to each condition. Nine teachers in four schools and 187 second-grade ELs participated in this within-subjects intervention, which took place in transitional bilingual classrooms. Findings indicated that across all word types, both extended and embedded instruction were more effective than the control condition in helping ELs acquire vocabulary. Findings also indicated that extended instruction was more effective then embedded instruction for all word types except concrete cognates suggesting that these young Spanish-speaking ELs were advantaged by word knowledge in their home language. Finally, while embedded instruction was less effective than extended instruction, it was much more effective than a control condition. Embedded instruction takes significantly less preparation and instructional time than extended instruction, offering teachers a practical way to teach more vocabulary to ELs.     

A meta-analysis of the effects of various science teaching strategies on achievement

This study is a meta-analysis of the question “What are the effects on achievement of different teaching techniques?” Twelve categories of teaching techniques were specified. Among these are questioning, wait-time, testing, focusing, manipulative, presentation approach, inquiry or discovery, audio-visual, and teacher direction. A coding form was then developed to allow for the uniform examination and recording of 76 variables from each study. One or more effect sizes were calculated and coded for each study. Over 300 microfilmed dissertations covering the past 30 years were examined. Some 2000 ERIC science abstracts were reviewed and suitable studies obtained. Finally, journal articles were reviewed. A total of 400 effect sizes representing 160 studies were produced. The main effect size overall was 0.34. Thus, for all samples considered the experimental science teaching techniques on the average resulted in one-third of a standard deviation improvement over traditional techniques. Mean effect sizes also were calculated for each technique. More than 20 other variables such as class size, community type, and science subject area were cross-tabulated with effect size.