Psychometricians and test developers use DIF analysis to determine if there is possible bias in a given test item. This study examines the conditions under which two predominant methods for determining differential item function compare with each other in item bias detection using an effect size statistic as the basis for comparison. The main focus of the present research was to test whether or not incorporating an effect size for LR DIF will more accurately detect DIF and to compare the utility of an effect size index across MH DIF and LR DIF methods. A simulation study was used to compare the accuracy of MH DIF and LR DIF methods using a p value or supplemented with an effect size. Effect sizes were found to increase the accuracy of DIF and the possibility of the detection of DIF across varying ability distributions, population distributions, and sample size combinations. Varying ability distributions and sample size combinations affected the detection of DIF, while population distributions did not seem to affect the detection of DIF.

Thesis (M.A.)--University of Ottawa, 1999. / Includes bibliographical references (p. 131-137). Also available on the Internet. MODE OF ACCESS via web browser by entering the following URL: http://www.nlc-bnc.ca/obj/s4/f2/dsk2/ftp01/MQ38757.pdf.

The current investigation examined the effects of a differential attentional training task on subsequent emotional reactivity in response to a task that was aimed to either induce positive or negative affectivity. Specifically, the study employed a dot-probe attentional training paradigm to train attention toward positive images (Attend-Positive condition), toward neutral images (Attend-Neutral condition), or to not train attention at all (Control condition). The hypothesis was that individuals whose attention was trained toward positive images would exhibit faster response times toward positive images (compared to the Attend-Neutral and Control conditions) at post assessment, individuals whose attention was trained toward neutral images would exhibit faster response times toward neutral images (compared to the Attend-Positive and Control conditions) at post assessment, and individuals in the Control condition would maintain similar response times from baseline to post assessment, aside from general practice effects. It was also hypothesized that those in the Attend-Positive condition would better regulate emotion, as measured by less negative affect in response to a stress task and more positive affect in response to a positive mood induction task, compared to individuals who have engaged in a control task involving no attentional bias training (Control condition). It was also hypothesized that those in the Attend-Neutral condition would better regulate emotion, as measured by less negative affect in response to a stress task and more positive affect in response to a positive mood induction task, compared to individuals who have engaged in a control task involving no attentional bias training (Control condition). Last, it was hypothesized that individuals in the Attend-Positive condition would report differentially less negative affect in response to the stress task (failure anagrams) than those in the Attend-Neutral condition, and that individuals in the Attend-Positive condition would report differentially more positive affect in response to the positive mood induction task (success anagrams) than those in the Attend-Neutral condition. In all conditions, the dot-probe attentional training did not effectively modify biases in the hypothesized directions. While there was differential affectivity change for individuals who underwent a Failure Anagram task versus a Success Anagram task, there is no way to definitively interpret the meaning of these changes given the failure of the attentional manipulation. The findings from the current investigation provide no evidence for single-session dot-probe attentional bias modification procedures to manipulate attentional bias toward positive stimuli or toward neutral stimuli. Possible explanations for these results, including lack of reliability of the dot-probe task, are discussed.

This dissertation is a cross-sectional study of the acquisition of tense and aspect by 45 Persian speakers as the experimental group and 15 native English speaking children as the control group. This study specifically investigates the hypothesis of Primacy of Aspect (POA) that claims there is; (1) a strong association of past/perfective morpheme with achievement and accomplishment verbs, (2) a strong association of progressive morpheme with activity verbs, (3) no overextension of progressive inflection to stative verbs, and (4) strong association of the present morpheme '-s' with stative verbs. The study also argues for the semantic implications of the present data for the Distributional Bias Hypothesis (Andersen 1990), that the distinction of verb type in the input is skewed so as to create the acquisitional pattern found in studies of the POA in language acquisition, and for the Language Bioprogram (Bickerton 1981), that aspectual values are the knowledge a child is born with as the sources of the POA in SLA. The study also examines the syntactic implications of the data for the 'initial state' hypotheses in SLA: The Minimal Trees Hypothesis (Vainikka & Young-Scholten 1994, 1996a, 1996b), The Full Transfer/Full Access hypothesis (Schwartz & Sprouse 1996) and The Weak Parametric Transfer (Valueless Features) Hypothesis (Eubank 1993/94, 1996). The results supported the findings of the POA and the Minimal Trees Hypothesis.

The Canadian Precipitation Analysis (CaPA) generates gridded precipitation data outputs based on the assimilation of both observation and climate model data. CaPA outputs are highly valuable to modelling efforts dependent on precipitation inputs, and as such the quality of CaPA outputs is crucial. Two improvements to CaPA were investigated: reducing transformation bias though correction against moving-window averaged CaPA output that avoids transformation, and enhancing semivariograms through anisotropy and convection considerations. Accounting for convection in the semivariogram proved ineffectual, while the bias correction technique and anisotropic semivariograms both reduced bias and improved related metrics. No methods improved the Equitable Threat Score. If implemented separately, the bias correction or anisotropic semivariogram approaches will yield targeted benefits for CaPA users, particularly for applications focused on extreme precipitation values. Improvements were not so comprehensive as to warrant adoption in the operational CaPA configuration, although availability in experimental versions is recommended.

Thesis (Ph.D.)--University of Delaware, 2008. / Principal faculty advisor: Joseph J. Glutting, School of Education. Includes bibliographical references.

Thesis (Ph. D.)--Virginia Polytechnic Institute and State University, 1993. / Vita. Abstract. Includes bibliographical references (leaves 257-266). Also available via the Internet.

Thesis (Ed. D.)--West Virginia University, 2009. / Title from document title page. Document formatted into pages; contains vii, 114 p. : col. ill. Includes abstract. Includes bibliographical references (p. 83-86).

Master-Arbeit Univ. St. Gallen, 2009.

Master-Arbeit Univ. St. Gallen, 2009.