It is a challenging problem to make a general hand gesture recognition system work in a practical operation environment. In this study, it is mainly focused on recognizing English letters and digits performed near the steering wheel of a car and captured by a video camera. Like most human computer interaction (HCI) scenarios, the in-car gesture recognition suffers from various robustness issues, including multiple human factors and highly varying lighting conditions. It therefore brings up quite a few research issues to be addressed. First, multiple gesturing alternatives may share the same meaning, which is not typical in most previous systems. Next, gestures may not be the same as expected because users cannot see what exactly has been written, which increases the gesture diversity significantly.In addition, varying illumination conditions will make hand detection trivial and thus result in noisy hand gestures. And most severely, users will tend to perform letters at a fast pace, which may result in lack of frames for well-describing gestures. Since users are allowed to perform gestures in free-style, multiple alternatives and variations should be considered while modeling gestures. The main contribution of this work is to analyze and address these challenging issues step-by-step such that eventually the robustness of the whole system can be effectively improved. By choosing color-space representation and performing the compensation techniques for varying recording conditions, the hand detection performance for multiple illumination conditions is first enhanced. Furthermore, the issues of low frame rate and different gesturing tempo will be separately resolved via the cubic B-spline interpolation and i-vector method for feature extraction. Finally, remaining issues will be handled by other modeling techniques such as sub-letter stroke modeling. According to experimental results based on the above strategies, the proposed framework clearly improved the system robustness and thus encouraged the future research direction on exploring more discriminative features and modeling techniques.

This work describes three case studies conducted to address two major problems in the area of chemistry education research, the lack of reported psychometrics regarding instrument scores, and the need for well-characterized assessments to evaluate college chemistry curricula. The first case study describes a psychometric evaluation of the Colorado Learning Attitudes about Science Survey (CLASS), an instrument designed to assess student beliefs about the learning of chemistry. Results from this work suggest that the CLASS instrument provides fertile ground for short instruments with reasonable psychometric properties. Responses to a single scale instrument, created from CLASS, showed that students in an introductory general chemistry course tend to be slightly more expert-like than novice-like in their beliefs about chemistry towards the end of the semester. The second case study discusses the use of a two-tier diagnostic instrument in assessing student understanding of the particulate nature of matter and chemical bonding. In addition to examining psychometric properties of the instrument's scores, this study uses student responses to think about the role of a preparatory chemistry course in promoting understanding of the measured concepts. Results of this study showed that the performance of students with the preparatory chemistry course was slightly better than those without it. The third case study focuses on the development of the Targeted Misconception Inventory (TMI), a two-tier instrument designed to measure student understanding of Bond Energy, Ionic, Bonding, and Phase Changes. The TMI was used to create an instructional intervention. Results from the intervention suggested a learning gain for Bond Energy concept. The three instruments discussed above were multiple-choice given as paper and pencil tests in an introductory chemistry course. The work described in this dissertation showcase a method for examining psychometric evidence. The three case studies provide a significant addition to the psychometric information available on existing instruments. This work makes an emphasis on the importance of pilot testing instruments and gathering psychometric information to provide evidence that the instrument is functioning as intended when used with different samples. This work provides a model for researchers to follow when refining an instrument, and implications for the use of assessment tools in chemistry curricula evaluation.

Intelligence testing has had a relatively long and controversial history, beginning with what is generally considered the first formal measure of intelligence, the Binet-Simon Scales (1916). Questions regarding possible cultural bias in these measures arose virtually simultaneously (e.g. Burt, 1921; Stern, 1914). Over the course of the twentieth and early twenty-first centuries, an abundance of intelligence measures have been developed, with many of them having several revisions, but the issue of test bias remains an important one, both in the professional literature and in the popular press (Reynolds & Lowe, 2009). A current intelligence measure in use, the Differential Ability Scales, Second Edition (DAS-II, Elliott, 2007), is a test with growing popularity for assessment of children and youth, not only for its ease of use, but also for its appeal to young children and its nonverbal composite (among other things). Consequently, it is essential that there be empirical evidence supporting the use of the DAS-II as an appropriate measure of cognitive abilities for children of varying backgrounds. The test publishers conducted extensive research with a representative sample during test development in an effort to ensure that the measure met adequate reliability and validity criteria; however, the issue of test bias, particularly regarding cultural or racial/ethnic groups, was not explicitly addressed. This issue was raised and examined with the original DAS by Keith, Quirk, Schartzer, and Elliott (1999), but with the significant changes made from the first edition to the second, there is no guaranty that the evidence from the earlier would necessarily apply to the latter. The current study investigated whether the DAS-II demonstrates systematic construct bias toward children and youth of any of four ethnic groups: Black, Hispanic, Asian, and White. Multi-group confirmatory factor analysis using data from the DAS-II standardization sample was used to assess whether criteria for increasingly strict levels of invariance were met across groups. Outcomes of this research contribute to an existing body of literature on test bias, as well as provide evidence regarding cross-group construct validity in the DAS-II. Ultimately the results of this study can be used to evaluate the appropriateness of the DAS-II for clinical use with certain ethnic groups and will help to emphasize further the importance of exploring these issues with all standardized tests. / text

The National Research Council (NRC) outlines an assessment design framework in Knowing What Students Know. This framework proposes the integration of three components in assessment design that can be represented by a triangle, with each corner representing: cognition, or model of student learning in the domain; observation, or evidence of competencies; and interpretation, or making sense of this evidence. This triangle representation signifies the idea of a need for interconnectedness, consistency, and integrated development of the three elements, as opposed to having them as isolated from each other. Based on the recommendations for research outlined in the NRC's assessment report, this dissertation aims to conduct a dimensionality analysis of Programme for International Student Assessment (PISA) mathematics items. PISA assesses 15-year olds' skills and competencies in reading, math, and science literacy, implementing an assessment every three years since 2000. PISA's mathematics assessment framework, as proposed by the Organisation for Economic Co-operation and Development (OECD), has a multidimensional structure: content, processes, and context, each having three to four sub-dimensions. The goal of this dissertation is to show how and to what extent this complex multidimensional nature of assessment framework is reflected on the actual tests by investigating the dimensional structure of the PISA 2003, 2006, and 2009 mathematics items through the student responses from all participating OECD countries, and analyzing the correspondence between the mathematics framework and the actual items change over time through these three implementation cycles. Focusing on the cognition and interpretation components of the assessment triangle and the relationship between the two, the results provide evidence addressing construct validity of PISA mathematics assessment. Confirmatory factor analysis (CFA) and structural equation modeling (SEM) were used for a dimensionality analysis of the PISA mathematics items in three different cycles: 2003, 2006, and 2009. Seven CFA models including a unidimensional model, three correlated factor (1-level) models, and three higher order factor (2-level) models were applied to the PISA mathematics items for each cycle. Although the results did not contradict the multidimensionality, stronger evidence was found to support the unidimensionality of the PISA mathematics items. The findings also showed that the dimensional structure of the PISA mathematics items were very stable across different cycles. / text

Fundamental to the debate of culture and its impact is the identification of the dimensions that comprise it. The impact of culture as an explanatory variable can be found in various social, scientific, and economic arenas, such as social perception, economic development, and the organization of industries and companies. By identifying and measuring these dimensions, researchers can then organize cultures empirically and develop complex descriptions of various cultures. The study aimed to test the structure of the dimensions proposed by Ho and Chiu (1994) by means of scale development. Test-item writers involved psychology graduate students of various nationalities with the purpose of addressing reliability issues of previous measures by virtue of increased content breadth. The study also aimed to investigate the notion that cultural tendencies vary by dimension across geographical regions. Phase-I factor analysis results indicated that a 5-factor solution (responsibility, affiliation, social welfare, religion, and achievement) should be retained. Phase-II involved administering the scale to an international and American student sample that formed the basis for group comparisons. The results for the group comparisons were illuminating, providing evidence for the conceptualization of individualism and collectivism as worldviews and that the groups varied in their worldview depending on the pertinent dimension being measured. Implications for organizational research are discussed within the framework of linking individualism and collectivism to workplace variables. This study hopes to spur further empirical research in the area to catch up with the progressing theoretical development through expanded cultural dimensions, theory refinement, determining the process(es) by which cultural factors are linked to work behaviors, and uncover the various areas of applicability and research.

Microarray technology has revolutionized the field of molecular biology by offering an efficient and cost effective platform for the simultaneous quantification of thousands of genes or even entire genomes in a single experiment. Unlike southern blotting, which is restricted to the measurement of one gene at-a-time, microarrays offer biologists with the opportunity to carry out genome-wide experiments in order to help them gain a systems level understanding of cell regulation and control. The application of bioinformatics in the milieu of gene expression analysis has attracted a great deal of attention in the recent past due to specific algorithms and software solutions that attempt to illustrate complex multidimensional microarray data in a biologically coherent fashion so that it can be understood by the biologist. This has given rise to some exciting prospects for deciphering microarray data, by helping us refine our comprehension pertinent to the underlying physiological dynamics of disease. Although much progress is being made in the development of specialized bioinformatics software pipelines with the purpose of decoding large volumes of gene expression data in the context of systems biology, several loopholes exist. Perhaps most notable of these loopholes is the fact that there is an increasing demand for software solutions that specialize in automating the comparison of multiple gene expression profiles, derived from microarray experiments sharing a common biological theme. This is no doubt an important challenge, since common genes across different biological conditions having similar expression patterns are likely to be involved in the same biological process and hence, may share the same regulatory signatures. The potential benefits of this in refining our understanding of the physiology of disease are undeniable. The research presented in this thesis provides a systematic walkthrough of a series of software pipelines developed for the purpose of streamlining gene expression analysis in a systems biology context. Firstly, we present BiSAn, a software tool that deciphers expression data from the perspective of transcriptional regulation. Following this, we present Genome Interaction Analyzer (GIA), which analyzes microarray data in the integrative framework of transcription factor binding sites, protein-protein interactions and molecular pathways. The final contribution is a software pipeline called MicroPath, which analyzes multiple sets of gene expression profiles and attempts to extract common regulatory signatures that may be implicating the biological question.

For many years researchers have been troubled with the evaluation of Information Systems (IS) concluding to a lack of understanding as to the time, methods and tools for measuring the impact of IS on organisational performance. Motivated by this lacuna in the literature, this thesis explored the relationship between Information Systems’ Effectiveness and Organisational Performance. The theoretical framework is based on the Delone and McLean’s (D&M) IS success model which is widely adopted in the IS research. Based on a comprehensive literature review on the older and more recent studies the research framework incorporated several new items used by researchers for the measurement of the D&M dimensions. Following a web survey on 168 Greek firms this study sheds some light into the IS field by focusing on how IS effectiveness measures affect Organisational Performance. Data for this study was collected by means of a web-link questionnaire and a sample of 700 companies of different sizes operating in various industries. Many descriptive statistics of academic and managerial importance were produced. Following a correlation analysis and Exploratory Factor Analysis, 15 factors were used for Multiple Regression analysis conducted to test a number of hypotheses around the relationship between the dependent construct (organisational performance) and the independent construct (IS effectiveness). This thesis contributes to existing research in the following ways. First, this study extends our knowledge on IS effectiveness as we adapted and modified DeLone and McLean's model of IS success to incorporate new variables from recent research. The results indicate a significant statistical link between IS effectiveness and performance measures. Second, it provides a holistic framework for measuring Organisational Performance with financial and non-financial variables. Finally, the study presents findings from Greek companies that have adopted IS providing practioners with advice for the practices that can lead to possible and realistic benefits.

The last two decades have seen a surge in published research on primate personality. This surge contrasts with the paucity of research over the preceding century. People interested in primate personality research come from a broad range of fields, but they are all interested in measuring primate personality in a way that is reliable, valid, and practical. This dissertation aims to describe the development and evaluation of the construct validity of a new rating scale in chimpanzees. The scale is based on a bottom-up approach to scale development and was developed using steps from both Uher (2008a,b) and Gosling (1998). As described in Chapter 3, the scale was evaluated by using it to rate 143 chimpanzees at the University of Texas MD Anderson Cancer Center Facility in Bastrop, TX. Twenty-one people who have worked with the chimpanzees between 6 months to 20 years rated the chimpanzees. Chapter 4 describes how inter-class correlation coefficients (ICCs) were used to calculate the reliability of the items on the scale. There was only one item (predictable) that turned out to not be reliable. The other 40 items were included in subsequent analyses. An exploratory factor analysis, as described in Chapter 5, was performed in order to determine the structure underlying the scale. Five methods were used to determine that a six-factor solution fit the data best. The six factors were labeled Reactivity, Dominance, Openness, Extroversion Agreeableness, and Conscientiousness based on the degree to that they correlated with other previous chimpanzees scales that used those labels. The convergent and discriminant validity of the factors was evaluated, as described in Chapter 6, by looking at the predicted relationships between each of the six factors and the variables of sex, age, rearing history, behavior in reaction to a novel stimulus, general behavior, injuries, illnesses, blood chemistry, and cortisol. The results indicate that there is a lack of evidence for convergent validity, but some evidence for discriminant validity of the new chimpanzee rating scale. The discussion in Chapter 7 focuses on the findings from the study as well as strengths and limitations of the new chimpanzee rating scale. / text

Η παρούσα διπλωματική ερευνητική εργασία μελετά τους παράγοντες που διαμορφώνουν την ποιότητα των υπόγειων νερών του νομού Ηλείας. Στην εργασία παρουσιάζονται τα αποτελέσματα που προέκυψαν από τις εργαστηριακές αναλύσεις 99 δειγμάτων νερού, που περιλαμβάνουν σημεία υδροληψίας από το μεγαλύτερο μέρος του νομού. Τα δείγματα αυτά συγκεντρώθηκαν στο πλαίσιο της τεχνικής έκθεσης με τίτλο «Έλεγχος της ποιότητας των υπόγειων νερών ύδρευσης των πλέον επιδεκτικών σε ρυπάνσεις πυρόπληκτων περιοχών του νομού Ηλείας» που ανέθεσε η Νομαρχία Ηλείας το 2008 στο εργαστήριο Υδρογεωλογίας, με επικεφαλής τον καθηγητή κ. Νικόλαο Λαμπράκη. Το γεωλογικό υπόβαθρο του νομού Ηλείας διαμορφώνεται από την Ιόνια ζώνη, τη ζώνη Ωλονού - Πίνδου, τη ζώνη Γαβρόβου Τρίπολης, Τριτογενείς γεωλογικούς σχηματισμούς και Τεταρτογενείς αποθέσεις. Στους Νεογενείς σχηματισμούς που εντοπίζονται και καλύπτουν το μεγαλύτερο μέρος του νομού, αναπτύσσονται υδροφόρα στρώματα που ταξινομούνται βάσει του προτύπου IAH/UNESKO του 1995. Οι σημαντικότεροι υδροφόροι ορίζοντες αναπτύσσονται στους σχηματισμούς Βούναργου και Χελιδονούς - Ολυμπίας, Καλαθά και στις ποτάμιες αναβαθμίδες Αλφειού και Ζαχάρως. Η περιοχή εμφανίζει έντονη τεκτονική με σημαντικότερες ρηξιγενείς ζώνες, τη ζώνη του Βούναργου, του Πράσινου, του Αλφειού και του Πελόπιου - Ολυμπίας. Με την επεξεργασία των φυσικοχημικών παραμέτρων των δειγμάτων στο πρόγραμμα Phreeqc 3.0 καταλήξαμε σε τρεις υδροχημικούς τύπους για τα νερά της Ηλείας. Επικρατεί ο τύπος Ca-HCO3 που φανερώνει φρέσκα νερά και κατανέμεται σχεδόν ομοιόμορφα στην περιοχή μελέτης. Ο τύπος Ca-HCO3-SO4 οφείλεται κυρίως στη διάλυση της γύψου των Νεογενών, στο κεντρικό τμήμα του νομού. Στον τύπο Ca-Na-HCO3 η ύπαρξη Νατρίου αποδίδεται σε διαδικασίες ιοανταλλαγής, παρουσία εβαποριτών της Ιόνιας ζώνης ή γειτνίαση με τη θάλασσα. Κανένα από τα παραπάνω στοιχεία δεν ξεπερνά ωστόσο τα όρια ποσιμότητας (Oδηγία 98/83/EC). Για τη στατιστική επεξεργασία των δεδομένων χρησιμοποιήθηκε R-τύπου παραγοντική ανάλυση, με τη βοήθεια του προγράμματος SPSS 17.0. Για τη χωρική κατανομή των παραγοντικών τιμών που προέκυψαν χρησιμοποιήθηκε η μέθοδος Ordinary kriging στο ArcMap 9.3. Επιλέχτηκε ένα τετραπλό παραγοντικό μοντέλο που εκφράζει το 75% της αρχικής πληροφορίας. Ο πρώτος παράγοντας συνδέει τη θερμοκρασία του υπόγειου νερού με το Λίθιο και το Βόριο και κατανέμεται στην ερύτερη περιοχή του Κατάκολου. Πρόκειται για μια γνωστή σχέση που ευνοείται από θερμομεταλλική κυκλοφορία υδάτων, έντονη τεκτονική, μεγάλους χρόνους παραμονής και βαθείς υδροφόρους. Ο δεύτερος παράγοντας συσχετίζει τον Σίδηρο και το Μαγγάνιο με αναγωγικά περιβάλλοντα στους υδροφόρους της περιοχής Αμαλιάδας - Πηνειού. Ο τρίτος παράγοντας απηχεί την αυξημένη συγκέντρωση Χρωμίου, Ουρανίου και Αρσενικού σε αναγωγικές συνθήκες, στο κεντρικό τμήμα του νομού. Τον τέταρτο παράγοντα αποτελεί το δυναμικό οξειδοαναγωγής που εμφανίζει αρνητικές τιμές στα αναγωγικά περιβάλλοντα και συνδυάζεται με τα αποτελέσματα του δεύτερου και του τρίτου παράγοντα. / The present master’s thesis examines and interprets the factors that influence the groundwater quality in Elia Prefecture, Peloponnese. Moreover it elaborates the results of the laboratory analysis of 99 water samples that have been collected in the frame of a technical assessment carried out in 2008 by the Hydrogeology Laboratory and its head professor Nikolaos Lamprakis. The assessment entitled “Drinking groundwater quality assessment of the most susceptible of contamination regions in Elia Perfecture that have been affected by fire” was assigned to the Hydrogeology Laboratory by the Regional Council of Elia. The geological background of Elia Prefecture is formed by the Ionian zone, the Olonos - Pindos zone and the Tripolis zone, as well as Tertiary formations and Quaternary alluvial. Inside the Neogene formations several water-bodies have been developed. These are classified according to the IAH/UNESCO standard of 1995. The most productive aquifers are those of Vounargon, Helidon - Olympia, Kalathas, Alfeios and Zacharo fluvial conglomerates. Intense tectonic activity occurs in the study area. The most important fault zones are those of Vounargon, Prasinon, Alfeios and Pelopion - Olympia. The physical and chemical parameters of the samples have been processed by Phreeqc 3.0 resulting in three hydrochemical types of groundwater in Elia. The prevailing type is Ca-HCO3 that indicates fresh water, which is evenly distributed over the study area. The dilution of the gypsum of the Neogene formations causes the Ca-HCO3-SO4 type to occur. The Potassium concentration in the Ca-Na-HCO3 type is attributed to ion-exchange processes, to the presence of evaporates or to the proximity of the aquifers to seawater. Nevertheless, the aforementioned concentrations were below the permissible limits (Directive 98/83/EC). The method used for the statistical analysis was R-type factor analysis that has been performed in SPSS 17.0. To acquire the spatial distribution maps of each factor Ordinary kriging interpolation method was used, in ArcMap 9.3. The statistical analysis resulted in a four-factor model that represents the 75% of the initial-values information. The first factor associates the groundwater-temperature to the presence of Lithium and Boron. It is the dominant factor over the area around Katakolo. It indicates a well-known relation which is a function of thermometallic water circulation, tectonic structures, long residence time and deep aquifers. The second factor expresses the relation of Iron and Manganese to reducing conditions in the aquifers of Amaliada and Pineia. Redox potential is the third factor. Its negative values point out reducing environments that are confirmed by the examination of the distribution maps of the second and third factor.

Αντικείμενο της παρούσας διπλωματικής εργασίας είναι η μελέτη των περιβαλλοντικών παραμέτρων της λιμνοθάλασσας της Κλείσοβας, η οποία βρίσκεται στο ανατολικό άκρο του λιμνοθαλάσσιου συστήματος του Μεσολογγίου. Στο πρώτο στάδιο γίνεται επεξεργασία των περιβαλλοντικών δεδομένων με χρήση μεθόδων πολυδιάστατης στατιστικής ανάλυσης με σκοπό τον καθορισμό των κύριων διεργασιών που αναπτύσσονται στις υδάτινες μάζες της λιμνοθάλασσας με ιδιαίτερη έμφαση στον εντοπισμό διάφορων ρυπογόνων διεργασιών. Στο δεύτερο στάδιο γίνεται η επεξεργασία των περιβαλλοντικών δεδομένων με χρήση του αλγορίθμου CLEAN καιεφαρμόζεται πολυδιάστατη στατιστική ανάλυση στα επεξεργασμένα δεδομένα. / Subject of the current work is the study of the physico-chemical parameters of Kleisovas'lagoon which is located in SE part of Mesologgi lagoon complex. In the first part are defining the most important lagoon process with emphasis to pollution processes through the execution of multidimensional statistical methods. In the second part is implemented a spectrum analysis of the environmental data through the Clean algorithm. Through execution of multidimentional statistical methods of the cleaned data are defining, in correlation with the analysis of the raw data, the dominant lagoon processes which effect the water quality of the ecosystem.