Controlling for the wide variability in the physical health status of geriatric populations is important as severity of illness is known to both moderate and suppress relationships examined in psychosocial research. The purpose of the present investigation was to develop a uniform, easily administered quantitative index of illness severity, composed of disease-specific scales, that was independent of psychosocial factors and appropriate for use with a geriatric population. As well, the aim was to collect preliminary data on the reliability and validity of the scale. The development of the Severity of Illness-Geriatric (SOI­G) scale involved the adaptation of a previously developed severity of illness instrument Severity of Renal Disease Scale (SORDS). <p>The present investigation involved five programmatically linked studies. Study 1 involved the determination of the items to be included on SOI-G while Study 2 defined the severity criteria for each item. In Study 3, five geriatric specialists scaled each level of each item on the same underlying threat to life scale. There was a high level of initial agreement between the raters supporting the reliability of the severity values. The final scale consisted of 32 items. <p>In Study 4, archival data was collected on 61 patients admitted to the geriatric unit of a rehabilitation hospital. The SOI-G was compared to the Cumulative Illness Rating Scale-Geriatric (CIRS-G) and a global severity rating. <p>SOI-G inter-rater reliability estimates were low (likely due to rater error) but promising. SOI-G demonstrated support for content validity, face validity, and construct validity but evidence for convergent validity was not established. SOI-G scores were sensitive to differences among patients with respect to discharge outcome. The utility of SOI-G as a moderator variable in psychosocial research with the elderly could not be explored in Study 5 due to a limited sample size. <p>It was concluded that the present investigation demonstrated the potential usefulness of SOI-G in psychosocial research with the elderly but further research is needed before definitive conclusions can be made. The SOI-G offers researchers a tool for controlling disease variability that is not measured by psychological tests but must be accounted for in research designs.

<p> Probability forecasts play an important role in many decision and risk analysis applications. Research and practice over the years have shown that the shift towards distributional forecasts provides a more accurate and appropriate means of capturing risk in models for these applications. This means that mathematical tools for analyzing the quality of these forecasts, may it come from experts, models or data, become important to the decision maker. In this regard, strictly proper scoring rules have been widely studied because of their ability to encourage assessors to provide truthful reports. This dissertation contributes to the scoring rule literature in two main areas of assessment - probability forecasts and quantile assessments. </p><p>In the area of probability assessment, scoring rules typically studied in the literature, and commonly used in practice, evaluate probability assessments relative to a default uniform measure. In many applications, the uniform baseline used to represent some notion of ignorance is inappropriate. In this dissertation, we generalize the power and pseudospherical family of scoring rules, two large parametric families of commonly-used scoring rules, by incorporating the notion of a non-uniform baseline distribution for both the discrete and continuous cases. With an appropriate normalization and choice of parameters, we show that these new families of scoring rules relate to various well-known divergence measures from information theory and to well-founded decision models when framed in an expected utility maximization context. </p><p>In applications where the probability space considered has an ordinal ranking between states, an important property often considered is sensitivity to distance. Scoring rules with this property provide higher scores to assessments that allocate higher probability mass to events “closer” to that which occurs based on some notion of distance. In this setting, we provide an approach that allows us to generate new sensitive to distance strictly proper scoring rules from well-known strictly proper binary scoring rules. Through the use of the weighted scoring rules, we also show that these new scores can incorporate a specified baseline distribution, in addition to being strictly proper and sensitive to distance. </p><p>In the inverse problem of quantile assessment, scoring rules have not yet been well-studied and well-developed. We examine the differences between scoring rules for probability and quantile assessments, and demonstrate why the tools that have been developed for probability assessments no longer encourage truthful reporting when used for quantile assessments. In addition, we shed light on new properties and characterizations for some of these rules that could guide decision makers trying to choosing an appropriate scoring rule. </p> / Dissertation

The purpose of this study is to implement a rock fall hazard rating system to the cut slopes along Kizilcahamam-Gerede segment of D750 (Ankara-istanbul) Highway. The rating system developed by the Tennessee Department of Transportation was assessed for thirty six cut slopes which were selected based on a reconnaissance survey along D750 highway, between Kurtbogazi Dam (50 km northwest of Ankara) and AktaS village (15 km to Gerede town of Bolu province). The stages of the investigation consist of project conception, field investigations and application of this system, assessment and presentation of data. The cut slopes were classified by implementing this method which requires a scoring on an exponential scale assigned to various parameters related to the site and roadway geometry and geologic characteristics. The rating process was completed at two stages: Preliminary and Detailed Rating. Based on the Tennessee RHRS, nineteen cutslopes were assessed according to these two stages while the other seventeen cut slopes were able to be classified only with the preliminary rating stage. Different modes of slope failure (planar, wedge, toppling, rock fall with differential weathering, raveling) throughout the selected segments of the highway were investigated and the slope and highway related parameters such as slope height, ditch effectiveness, average vehicle risk, road width, percent desicion site distance and rockfall history were identified for these nineteen cut slopes. After the scoring process was completed all cut slopes were classified based on their hazard ratings from the point of the problems that they may cause in transportation. According to the rules of Tennessee RHRS, a total of thirty five cut slopes were rated. Among these slopes, nineteen of them are rated as A slopes which are considered to be potentially hazardous, while a total of seven are rated as C slopes which pose no danger. In placing a slope into a B category, it is considered that they are not as prone as A slopes to create a danger and a total of nine B slopes are detected. The detailed rating is accomplished for these nineteen A slopes and as a result of the scorings, it has been seen that the final RHRS scores range from 164 to 591. The slopes with scores over 500 can be counted as more hazardous slopes since they get very high scores both from site and roadway geometry and geologic hazard part.

Abstract According to a survey conducted by Rock (1984), the major factors of influenced credit risk are (1) the relationship with other creditor, (2) income, (3) loan-income ration, (4) profession, (5) immovable property, (6) check & deposit account. And, the sure way to score with lenders are (1) rules of thumb & subjective judgment, (2) credit rating system, (3) credit scoring system, and (4) expert system. The purpose of the present study is to examine the relationships between sex, age, income, profession, assets, the purpose of loan, employment information, credit references, credit limit, total installment loan account by the consumer, total number of inquiries and the consumer¡¦s payment records. The results of this search indicate that: 1. The previous stereo type thinking of banking industry always treat the military officials as wall as police officials are risky groups to consumer credit loan. However, this study found the contrary result. The payment over due rate is comparatively lower than that of other customer groups. The conclusion is that military officials and police officials are potentially good customers to banking system in terms of profit margin against risk. 2. From the credit scoring system of banking industry. That the customers are between 35 to 50s should be better than those age between 20 to 30s. However, this study demonstrates the other direction that customers with age below 35years old always better than those who over 35years old to the banking creditability actual performance. 3. The banking industry assume the married people will be a better group compared with non-married group on the money collect of the loan they made. However this study proves that creditability performance in sequence is (i) age below 35 and singer is the best. (ii) those married is the second while.(iii)age over 35 and non- married group is the worst one. 4. Most of people think those who have consumer credit loan from bank and would not want their family to be aware of their personal loan may have higher chance of payment over due. However, the statistics study from bank branch A indicates that this kind of customers (don¡¦t want family member know about loan) are the best group on payment over due (only 5.5%). While those who agree to let family member aware are the second (7.5%), and others with no comment are the worst.

This study was an attempt to add to existing research by estimating the ability of third-party examiners to assess whether or not an organization successfully implemented strategies based on the criteria of the Texas Award for Performance Excellence (TAPE). The TAPE is given each year by the Quality Texas Foundation and recognizes organizations that demonstrate superior performance as it is defined by customer satisfaction and continuous improvement. The TAPE is a state-level award for quality that uses the same criteria as the Malcolm Baldrige National Quality Award for Performance Excellence. This research was an analysis of the TAPE process at the level of examiners, also known as the Board of Examiners. The Board is made up of approximately 150 experienced professionals from several types of business sectors and is responsible for evaluating organizational self-assessments. In this quantitative study, data were converted from the Quality Texas Foundation into a database. Because the set of the TAPE applicants included in the study consisted of the entire population of TAPE applicants selected from 2001 to 2004, descriptive statistics were appropriate for producing informative data that could be analyzed for variation and stability in the scoring process. Exploration of patterns in descriptive statistics and multivariate analysis of variance were the primary tools used in this particular study along with Cronbach’s Alpha as an indicator of reliability. Since scoring for the TAPE is based on an individual examiner’s best subjective assessment, it was impossible to have one objective score against which all the other scores could be measured. The team consensus score was therefore used as the true score for measurement. Establishing reliability of examiners’ scores was a problem due to the fact that organizations and teams did not repeat. Results from the study led to the conclusion that there was insufficient evidence to make a determination on what influences examiners’ scoring consistency. More data will need to be collected in such a way so as to make it possible to identify that impact consistency of examiner scores.

none

Previous research investigations in the area of spelling development have adopted two approaches, the broad approach and the narrow approach. The broad approach suggests that spelling develops in sequential stages whereas the narrow approach focuses on individual linguistic patterns. However, research findings have revealed that children’s spellings do not exhibit errors pertaining to specifically one stage or reflecting one linguistic element, yet a research void exists in resolving how these two approaches might intermix. This study examined the spelling errors of typically developing children in first through fourth grades (N = 400) to determine the quantitative and qualitative differences in misspellings among grade levels. Each grade level had an equal representation of children (N = 100) and male and female participants. The spelling errors were extracted from two writing samples completed by the children, a narrative and expository sample. In an attempt to combine the broad and narrow approaches, a coding system was designed to evaluate the linguistic category (phonological, orthographic, morphological) and specific features (letter name spelling, vowel error, digraph, etc.) of the spelling errors. The findings revealed a significant interaction between grade level and error type for phonologically-based spelling errors (1 st graders made more errors than 2nd and 4th graders) and a greater number of morphological errors was noted in 4 th vs. 2nd grade. No significant effects were noted for writing genre or gender. Analysis of performance patterns for specific linguistic category errors within and across grade levels revealed that all four grade levels committed the most phonological errors in the PSE (phonological – silent /e/) and PSON (phonological – sonorant clusters) categories. The OLN (orthographic – letter name) and ODI (orthographic – digraph) errors also occurred frequently in all four grades with first graders demonstrating significantly more occurrences of the OLN than ODI error. Morphological findings revealed that first graders made significantly more MINF (morphological – inflection) than MHOM (morphological – homonym) errors and all four grades had significantly more MINF than MCON (morphological – contraction) errors. A qualitative analysis regarding the most frequently misspelled words and most frequently encountered codes was also performed. The clinical and educational implications of these findings are discussed.

Manual sleep scoring, executed by visual inspection of the EEG, is a very time consuming activity, with an inherent subjective decisional component. Automatic sleep scoring could ease the job of the technicians, because faster and more accurate. Frequency information characterizing the main brain rhythms, and consequently the sleep stages, needs to be extracted from the EEG data. The approach used in this study involves a wavelet filter bank for the EEG frequency features extraction. The wavelet packet analysis tool in MATLAB has been employed and the frequency information subsequently used for the automatic sleep scoring by means of an artificial neural network. Finally, the automatic sleep scoring has been employed for epoching the fMRI data, thus allowing for studying brain resting state networks during sleep. Three resting state networks have been inspected; the Default Mode Network, The Attentional Network and the Salience Network. The networks functional connectivity variations have been inspected in both healthy and narcoleptic subjects. Narcolepsy is a neurobiological disorder characterized by an excessive daytime sleepiness, whose aetiology may be linked to a loss of neurons in the hypothalamic region.

When an exam is administered across dual platforms, such as paper-and-pencil and computer-based testing simultaneously, individual items may become more or less difficult in the computer version (CBT) as compared to the paper-and-pencil (P&P) version, possibly resulting in a shift in the overall difficulty of the test (Mazzeo & Harvey, 1988). Using 38,955 examinees' response data across five forms of the National Nurse Aide Assessment Program (NNAAP) administered in both the CBT and P&P mode, three methods of differential item functioning (DIF) detection were used to detect item DIF across platforms. The three methods were Mantel-Haenszel (MH), Logistic Regression (LR), and the 1-Parameter Logistic Model (1-PL). These methods were compared to determine if they detect DIF equally in all items on the NNAAP forms. Data were reported by agreement of methods, that is, an item flagged by multiple DIF methods. A kappa statistic was calculated to provide an index of agreement bet ween paired methods of the LR, MH, and the 1-PL based on the inferential tests. Finally, in order to determine what, if any, impact these DIF items may have on the test as a whole, the test characteristic curves for each test form and examinee group were displayed. Results indicated that items behaved differently and the examinee's odds of answering an item correctly were influenced by the test mode administration for several items ranging from 23% of the items on Forms W and Z (MH) to 38% of the items on Form X (1-PL) with an average of 29%. The test characteristic curves for each test form were examined by examinee group and it was concluded that the impact of the DIF items on the test was not consequential. Each of the three methods detected items exhibiting DIF in each test form (ranging from 14 items to 23 items). The Kappa statistic demonstrated a strong degree of agreement between paired methods of analysis for each test form and each DIF method pairing (reporting good to excell ent agreement in all pairings). Findings indicated that while items did exhibit DIF, there was no substantial impact at the test level.

A predictive validity approach has been employed to find some implications to support evidences for Automated Essay Scoring (AES) systems. First, using R² values from multiple linear regression models, validity indices are compared first between multiple choice scores and essay scores across four AES systems. Secondly, R² values from models using only essay scores, the validity indices of four AES systems are hypothetically compared to see if how well AES systems could predict student outcome such as GPA. / text