No description available.

The purpose of this study was to examine the effect of talker and listener dialect on the validity and reliability of word recognition scores from two sets of Mandarin speech audiometry materials. Four lists of bisyllabic words in Mainland Mandarin and Taiwan Mandarin dialects were administered to 16 participants of each dialect with normal hearing across two test sessions. The performance on materials presented in the native dialect was compared to performance on non-native dialect assessment to determine validity and reliability of test materials. Statistical analysis indicated significant differences between word recognition scores across test sessions, talker and listener dialect, and among lists. However it is unclear if such differences constitute clinically significant differences.

Each year, billions of dollars are spent on missiles and munitions by the United States government. It is therefore vital to have a dependable method to estimate the reliability of these missiles. It is important to take into account the age of the missile, the reliability of different components of the missile, and the impact of different launch phases on missile reliability. Additionally, it is of importance to estimate the missile performance under a variety of test conditions, or modalities. Bayesian logistic regression is utilized to accurately make these estimates. This project presents both previously proposed methods and ways to combine these methods to accurately estimate the reliability of the Cruise Missile.

The main purpose of this paper is to provide evidence of financial leverage, liquidity, profitability, and firm size ability to predict bankruptcy of Swedish small and medium-sized enterprises (SMEs), and to create a bankruptcy prediction model for Swedish SMEs. The sample consists of 1086 Swedish SMEs, among which 543 did go bankrupt between 2015 and 2019. The paper employs logistic regression and Mann-Whitney U-test to test the hypotheses. The independent variables are derived from previous research and further filtered in a selection process, resulting in a final set of six variables. Financial leverage, liquidity, profitability, and firm size is found to have significantly predictive abilities to determine SME bankruptcy. The model has an overall classification accuracy of 77.6% out-of-sample and is able to classify 82.2% of the bankruptcies correctly out-of-sample.

Arterial roads constitute the majority of the centerline miles of the Florida State Highway System. Severe injury involvements on these roads account for a quarter of the total severe injuries reported statewide. This research focuses on driver injury severity analysis of statewide multilane high speed arterials using crash data for the years 2002 to 2004. The first goal is to test different ways of analyzing crash data (by road entity and crash types) and find the best method of driver injury severity analysis. A second goal is to find driver, vehicle, road and environment related factors that contribute to severe involvements on multilane arterials. Exploratory analysis using one year of crash data (2004) using binary logit regression was used to measure the risk of driver severe injury given that a crash occurs. A preliminary list of significant factors was obtained. A massive data preparation effort was undertaken and a random sample of multivehicle crashes was selected for final analysis. The final injury severity analysis consisted of six road entity models and twenty crash type models. The data preparation and sampling was successful in allowing a robust dataset. The overall model was a good candidate for the analysis of driver injury severity on multilane high speed roads. Driver injury severity resulting from angle and left turn crashes were best modeled by separate non-signalized intersection crash analysis. Injury severity from rear end and fixed object crashes was best modeled by combined analysis of pure segment and non-signalized intersection crashes. The most important contributing factors found in the overall analysis included driver related variables such as age, gender, seat belt use, at-fault driver, physical defects and speeding. Crash and vehicle related contributing factors included driver ejection, collision type (harmful event), contributing cause, type of vehicle and off roadway crash. Multivehicle crashes and interactions with intersection and off road crashes were also significant. The most significant roadway related variables included speed limit, ADT per lane, access class, lane width, roadway curve, sidewalk width, non-high mast lighting density, type of friction course and skid resistance. The overall model had a very good fit but some misspecification symptoms appeared due to major differences in road entities and crash types by land use. Two additional models of crashes for urban and rural areas were successfully developed. The land use models' goodness of fit was substantially better than any other combination by road entity or the overall model. Their coefficients were substantially robust and their values agreed with scientific or empirical principles. Additional research is needed to prove these results for crash type models found most reliable by this investigation. A framework for injury severity analysis and safety improvement guidelines based on the results is presented. Additional integration of road characteristics (especially intersection) data is recommended for future research. Also, the use of statistical methods that account for correlation among crashes and locations are suggested for use in future research.

Population living along the coast are at risk of losing land, households, and economic resources due to the hazards of coastal erosion. Scientific research has indicated that 70% of the planet's sandy coastal environment is being impacted by coastal erosion. Due to the different characteristics of the lands in the coastal zone versus other areas, it is important to understand how the hazard of shoreline erosion contributes to subsequent land use change in affected coastal regions. This study analyzes how the level of erosion, land loss, and developed land loss performs when added with the default land use change parameters such as existing developed land proximity, proximity to forested areas, population, transportation, etc. Sample points of 1020 from 10 years and 15 years of shoreline erosion data for the lower Meghna River estuary of Southeast Bangladesh have been obtained and from there, different erosion statistics have been developed. Developed land use data has been collected from ESA's World Settlement Footprint dataset and other datasets are also collected from secondary data sources. Logistic regression modeling shows that there are verified contributions of proximity to erosion and amount of land loss with the probability of developed land use conversion in the study area. Adding the variables of environmental hazards increases the prediction accuracy by 2-3% and overall, the models are at least 85% accurate. / Master of Science / Population living along the coast are at risk of losing land, households, and economic resources due to the hazards of coastal erosion. The coast of the Lower Meghna estuary in Bangladesh is a region experiencing chronic and severe shoreline erosion that causes the land to be lost to estuarine waters. This research quantifies the amount of land lost to erosion with a special focus on the amount of developed land that is lost. Developed land in this study is defined as a built-up area typically composed of buildings and roads. The research also evaluates the effects of lost land on the subsequent conversion of interior land from a non-developed to developed status. The main contribution is to quantitatively identify the association between the erosion-induced land loss to future land use conversion. Using statistical modeling and digital mapping methods, results show that loss of land is associated with the subsequent conversion of non-developed land to developed land use. In particular, conversion has a higher probability at sites that are located more distant from the eroding shoreline that also are proximal to shoreline sites with higher rates of erosion-induced land loss. These results are suggestive of a relocation process where previously lost developed land is reestablished at interior sites within five kilometers of the eroding shoreline.

This project deals with exploring 1) travel-related health risk perception, and 2) actions taken to mitigate that health risk. Ordered logistic regression models were used to identify factors associated with the perceived risk of contracting influenza at work, school, daycare, stores, restaurants, libraries, hospitals, doctor’s offices, public transportation, and family or friends’ homes. Based on the models, factors influencing risk perception of contracting influenza in public places for discretionary activities (stores, restaurants, and libraries) are consistent but differ from models of discretionary social visits to someone’s home. Mandatory activities (work, school, daycare) seem to have a few unique factors (e.g., age, gender, work exposure), as do different types of health-related visits (hospitals, doctors’ offices). Across all of the models, recent experience with the virus, of either an individual or a household member, was the most consistent set of factors increasing risk perception. Using such factors in examining transportation implications will require tracking virus outbreaks for use in conjunction with other factors. Subsequently, social-health risk mitigation strategies were studied with the objective of understanding how risk perception influences an individual’s protective behavior. For this objective, this study analyzes travel-actions associated with two scenarios during an outbreak of influenza: 1) A sick person avoiding spreading the disease and 2) A healthy person avoiding getting in contact with the disease. Ordered logistic regression models were used to identify factors associated with mitigation behavior in the first scenario: visiting a doctor’s office, avoiding public places, avoiding public transit, staying at home; and in the second scenario: avoiding public places, avoiding public transit, staying at home. Based on the models for Scenario 1, the factors affecting the decision of avoiding public places, avoiding public transit, and staying at home were fairly consistent but differ for visiting a doctor’s office. However, Scenario 2 models were consistent with their counterpart mitigation models in Scenario 1 except for two factors: gender and household characteristics. Across all the models from Scenario 1, gender was the most significant factor, and for Scenario 2, the most significant factor was the ratio of household income to the household size. / Master of Science

Recent inspections conducted on levee safety in the U.S. that participate in the United States Army Corps of Engineers (USACE) Levee Safety Program under Public-Law 84-99 have shown that the overwhelming majority (>95%) of these levees have at least some deficiency associated with them, and many being identified as having an unacceptable safety rating (≈30%). In the U.S., many levees were constructed using funding from the federal government, but the responsibility of operation and maintenance of the levees were turned over to local government bodies. Given the local funding of levee maintenance, the socioeconomic characteristics of these levee-protected communities may be useful in identifying which communities may not have the economic, social, and / or political capital to maintain their levees to an acceptable safety standard. Using the lens of socio-hydrology, this study examines the socioeconomic and demographic characteristics of levee-protected communities and their relationship to the safety rating of their levee system. Using GIS, socioeconomic data were compiled for each of the evaluated Illinois levee systems from the US Census and the USACE’s National Levee Database (NLD). In addition to socioeconomic data, the NLD contained information on a levee’s age, protection level, estimated value of structures within the levee-protected area, ownership, inspection status, safety rating, and other structural details. The value of levee-protected agricultural lands was also assessed using a soil productivity index used by the state of Illinois to assess taxes on farmland. This information was compiled to investigate the potential differences of socioeconomic characteristics of communities with unacceptable to those with an acceptable levee rating. To assess the differences between the compiled socioeconomic information, the Independent Samples U-Test was implemented to quantify differences between communities with an acceptable verse unacceptable safety rating. In this study, 71 levee systems were identified with a levee safety rating and socioeconomic data from which to perform the statistical comparison between levee systems with an acceptable versus unacceptable safety rating. Of these 71 levees systems, 28 had an unacceptable and 43 had an at least minimally acceptable safety rating. The results from the Independent Samples U-Test showed that five variables with substantial variance (α ≤ 0.2, 80% CI) between the levee safety ratings were, the age of the levee, property value per structure, the average soil productivity index, per capita income, and the percentage of population being black. Using these substantial variables, a binary logistic regression model was created to see if they could be used to realistically predict the levee system’s safety rating. The regression model was able to accurately predict 84% of the ‘acceptable’ group while only correctly predicting 25% of the ‘unacceptable’ group resulting in an overall accuracy of 61%. The inability of this model to predict a levee system’s safety rating underscores the complexities in trying to determine which socioeconomic factors are important for identifying a given levee system’s safety rating. This finding also suggests there are potentially other variables which may be more robust predictors of a community’s ability to adequately maintain their levee. Future research should investigate these complexities in identifying which communities can adequately maintain their levee system.

Lexical inferencing is considered a listening strategy that is commonly employed by advanced EFL (English as a Foreign Language) listeners and a factor that contributes to successful listening comprehension. However, investigations of the factors that influence inferencing success in listening as well as how much each factor contributes to success are scant, as more studies have been conducted exploring lexical inferencing in reading. In addition, even though idiomatic expressions such as smell a rat, jump the gun, and go cold turkey are ubiquitous in the English language, especially in oral communication, and they are considered crucial in both first language (L1) and second language (L2) acquisition, little is known about the effectiveness of inferencing strategies where idiomatic expressions are concerned.Three goals motivated the current study. The first goal was to investigate whether inferencing is an effective strategy in the case where the target item is an idiomatic expression. The second goal was to investigate how four person-level factors, familiarity, listening proficiency, listening vocabulary size and working memory, two sentence-level factors, lexical density and sentence length, and two lexical-level factors, L1–L2 congruency and semantic transparency, influence the inferencing success of English idiomatic expressions in listening. The third goal, related to the second goal, was to determine which of the two lexical component factors, L1–L2 congruency and semantic transparency, is more important to inferencing success. A mixed methods design, the explanatory sequential design (Creswell & Plano Clark, 2018), was employed in this study. Quantitative data were collected from 89 EFL Japanese university students using a Listening Vocabulary Levels Test, a Listening Span Test, and an Idiom Inferencing Elicitation Task. The collected data were examined using mixed-effects logistic regression. Twelve participants were invited to participate in follow-up interviews based on their response patterns on the Idiom Inferencing Elicitation Task. The quantitative results indicated that familiarity, listening comprehension skills, working memory, and L1–L2 congruency were significant factors influencing inferencing success and the qualitative results supported these findings. In addition, the qualitative analyses suggested that depth of vocabulary is another potentially important factor. Furthermore, listening comprehension moderated the L1–L2 congruency effect. The finding that semantic transparency is not an influential factor in successful inferencing of unfamiliar idiomatic expressions provides evidence that the semantic transparency of known idiomatic expressions formed after learners acquire the meaning of the expression is a different construct from the perceived semantic transparency of unfamiliar idiomatic expressions. In addition, even though the sentence-level factors were not statistically significant in successful idiom inferencing in this study, further studies are required in order to see if this result holds true when the characteristics of the listening tasks differ from those of the task used in this study. It is hoped that the findings provide insights into how to help Japanese university EFL learners improve their listening skills, especially in tasks that include unfamiliar idiomatic expressions. / Teaching & Learning

Alzheimer’s disease (AD) accounts for the majority of all cases of dementia and can be characterized as a disease that causes a progressive decline of cognitive functions. Detecting the disease at it’s earliest stage is important as medical treatments can be more effective if they can be applied before the disease has caused irreparable brain damage. However, making a correct diagnosis of AD can be difficult, especially in the early stage when the symptoms are still mild. Machine learning algorithms can help in this process, with the purpose of this study being to investigate just how accurately machine learning algorithms can detect early-stage AD. Three algorithms were selected for the study, Random Forest, AdaBoost and Logistic Regression, which were then evaluated on the accuracy of their predictions. The results showed that Random Forest had the best overall performance with an accuracy of 79.78%. AdaBoost attained an accuracy of 76.40% and Logistic Regression attained an accuracy of 74.16%. These results suggest that machine learning algorithms can be used to make relatively accurate predictions of AD even when the disease is in it’s early stage.