No description available.

No description available.

Im Rahmen der vorliegenden Arbeit wurde eine Metaanalyse der Daten aller verfügbaren Studien (n = 109) zur medikamentösen Behandlung der drei für den Kliniker wesentlichen Angststörungen (PDA, GAD, SAD) durchgeführt. In die Metaanalyse wurden 187 Studienarme sowie die Daten von insgesamt 28785 Patienten eingeschlossen. Eine vergleichbare Metaanalyse, die alle drei Angststörungen zusammengefasst untersucht hat, wurde in dieser Form bisher nicht durchgeführt, wobei neben der zusammengefassten Analyse im Weiteren auch eine separate Betrachtung jeder einzelnen Angststörung erfolgte. Während im Rahmen aller bisher durchgeführten Metaanalysen zumeist lediglich Treated-vs.-Control-Effektstärken berechnet wurden, wurden in der vorliegenden Arbeit darüberhinaus auch Prae-Post-Effektstärken bestimmt. Dies ermöglicht einen besseren Vergleich der Wirksamkeit verschiedener Medikamente. Es ergab sich folgendes: Die in die Metaanalyse eingeschlossenen Studien zeigten trotz ähnlicher Ein- und Ausschlusskriterien sowie oftmaliger Verwendung gleicher Skalen eine hohe bis sehr hohe Heterogenität. Alle Medikamente bis auf Citalopram, Moclobemid und Opipramol zeigten einen signifikanten Unterschied zu Placebo. Die höchsten unadjustierten Treated-vs.-Control-Effektstärken konnten für Phenelzin (d = 0,98), Lorazepam und Clomipramin (d = 0,87) sowie Hydroxyzin (d = 0,79) berechnet werden. Die höchsten Prae-Post-Effektstärken wurden für Benzodiazepine (z. B. Delorazepam: d = 3,54; Bromazepam: d = 2,86; Lorazepam: d = 2,53), Quetiapin (d = 3,39), Escitalopram (d = 2,67) und Hyd-roxyzin (d = 2,56) berechnet, wobei in die Berechnung dieser Effektstärken zum Teil nur sehr wenige Primärstudien eingingen, so dass diese Ergebnisse als weniger reliabel zu werten sind. Bei Betrachtung der einzelnen Stoffgruppen erreichten die SNRIs mit d = 2,25 die höchste Prae-Post-Effektstärke, gefolgt von den Benzodiazepinen (d = 2,14) und den SSRIs (d = 2,09). Bei der Wahl eines Arzneimittels sollte auf ein angemessenes Verhältnis seines Nutzens zu seinen Risiken (Nebenwirkungen) geachtet werden. Viele der Medikamente, für die in der vorliegenden Arbeit relativ hohe Effektstärken berechnet werden konnten, weisen ein ungünstigeres Nebenwirkungsprofil als beispielsweise SNRIs und SSRIs auf. Vor allem wird aufgrund des bestehenden Abhängigkeitspotentials nicht empfohlen, Benzodiazepine routinemäßig zu verordnen. Ebenso führen trizyklische Antidepressiva häufiger zu Nebenwirkungen als SSRIs (Bandelow et al. 2008a). Weiterhin konnte in der vorliegenden Arbeit gezeigt werden, dass die Effektstärken der Pillenplacebos zwischen 1983 und 2013 stark anstiegen. Die Studien wurden mit Hilfe verschiedener Methoden zur Detektion eines Publication Bias analysiert. Hierbei ergaben sich zwar für mehrere Medikamente Hinweise auf das Vorliegen eines Publication Bias, dies hatte jedoch nicht zur Folge, dass die Annahme einer vormals berechneten signifikanten Überlegenheit des Medikamentes gegenüber Placebo wieder verworfen werden musste. Für 50,8% von insgesamt 187 Studienarmen wurden Allegiance-Effekte angenommen. Die durchschnittliche Effektstärke der Studien mit angenommenem Allegiance-Effekt unterschied sich jedoch nicht signifikant von der ohne solche Effekte. Klinisch tätige Ärzte können sich an den Ergebnissen der Metaanalyse orientieren, um – unter Berücksichtigung von potentiellen Nebenwirkungen und Kontraindikationen – für ihre Patienten das Präparat mit dem günstigsten Nutzen-Risiko-Verhältnis auszuwählen.

The aim of this thesis was to investigate the use of effect sizes to report the results of statistical credit rating models in a more practical way. Rating systems in the form of statistical probability models like logistic regression models are used to forecast the behaviour of clients and guide business in rating clients as “high” or “low” risk borrowers. Therefore, model results were reported in terms of statistical significance as well as business language (practical significance), which business experts can understand and interpret. In this thesis, statistical results were expressed as effect sizes like Cohen‟s d that puts the results into standardised and measurable units, which can be reported practically. These effect sizes indicated strength of correlations between variables, contribution of variables to the odds of defaulting, the overall goodness-of-fit of the models and the models‟ discriminating ability between high and low risk customers. / Statistics / M. Sc. (Statistics)

Både nationella och internationella undersökningar visar att svenska elevers studieresultat i fysik har försämrats. Syftet med denna studie är att undersöka om elever kan få bättre studieresultat om deras arbete enligt kursplaneringen kontrolleras kontinuerligt så att de är i fas med planeringen. Genom att kontrollera kontinuerligt får man dels möjlighet att upptäcka elever som halkar efter men också hur bra undervisningen har tagits emot av eleverna. Undersökningen gjordes i två parallella gymnasieklasser som läste grundkursen i fysik och pågick mellan två provtillfällen. Båda klasserna hade samma lärare, kursbok, kursplanering och prov. Ur planeringen valdes några grundläggande uppgifter och tilldelades ena klassen medan den andra utgjorde kontrollgrupp. Testgruppseleverna fick sina lösningar kontrollerade och fick återkoppling. Effektstorleken av interventionen bestämdes och jämfördes med referensgruppens. Resultatet visade att testgruppen hade fått bättre studieresultat efter ingreppet. Även om resultatet är signifikant så är det för osäkert med tanke på antalet elever (16 st) som deltog i studien för att kunna generaliseras. Eftersom deltagandet i undersökningen var frivilligt och icke-betygsgrundande så var antalet inlämningar inte så många som det hade förväntats. Arbetet är tidskrävande både ur elev- och lärarperspektiv men synliggör varför lärare bör kontrollera sina elever och hur detta kan genomföras. / Swedish and international studies show that Swedish students' results in physics have deteriorated. The aim of this study is to investigate whether students can achieve better study results if their work according to the course planning is checked continuously so that they are in phase with the planning. By continuously checking, one gets the opportunity to discover students who are lagging behind but also to see how the teaching has been absorbed by the students more generally. The study was done in two parallel upper secondary classes studying the basic course in physics and occurred between two test periods. Both classes had the same teacher, course book, course plan and exams. A few basic tasks where selected and assigned from the course plan and given to the test group but not to the control group. The test group students got their solutions checked and got feedback. The effect size of this intervention was measured and compared to the control group. The result showed that the test group had achieved a better result after the intervention. Even if the result was significant, it is too uncertain to be generalized, in view of the small sample size (16 students) who participated in the study. As participation in the survey was voluntary and non-grading, the number of submissions was not as many as it had been expected. The approach is time consuming both for teacher and student, but it makes visible why teachers should check their students and how this can be implemented.

The aim of this thesis was to investigate the use of effect sizes to report the results of statistical credit rating models in a more practical way. Rating systems in the form of statistical probability models like logistic regression models are used to forecast the behaviour of clients and guide business in rating clients as “high” or “low” risk borrowers. Therefore, model results were reported in terms of statistical significance as well as business language (practical significance), which business experts can understand and interpret. In this thesis, statistical results were expressed as effect sizes like Cohen‟s d that puts the results into standardised and measurable units, which can be reported practically. These effect sizes indicated strength of correlations between variables, contribution of variables to the odds of defaulting, the overall goodness-of-fit of the models and the models‟ discriminating ability between high and low risk customers. / Statistics / M. Sc. (Statistics)

In South Africa, Science and Technology Education faces many problems. Insufficient numbers of Science and Technology teachers, inadequate in–service training, large classes, instruction with the aim of narrowly orienting students towards examination passes an insufficient integration of technology in the curriculum, and insufficient physical infrastructure dominates the list. The Department of Education envisages the use of ICT as a tool for learning and teaching. ICT has the potential to improve the quality of education and training. If adequate resources are available, and teachers have confidence in the usefulness of ICTs, then the integration of Information and Communication Technology (ICT) may improve the teaching and learning of Mathematics and Science. A review of the literature indicated that the deployment of ICT resources alone will not bring about desirable pedagogical practices in the classroom. There exists a need for interventions that will enhance ICT pedagogical practices in South Africa. The following main research questions were formulated: What are the ICT pedagogic practices used by grade 8 Mathematics and Science teachers in South African classrooms? How do the barriers that grade 8 Mathematics and Science teachers encounter, as well as the support they receive, influence their pedagogical practices? What is the Principal’s role in promoting the emerging pedagogic practices using ICT in South African classrooms? This research comprises a secondary data analysis of the SITES 2006 South African data base. The population and sample for this study was based on the South African grade 8 Mathematics and Natural science teachers. In SITES 2006, the samples comprised more than 504 schools. Due to the fact that ICT is only significantly implemented in two out of nine provinces in South Africa, 25 strata were created to secure fair representation of the population with 666 Mathematics teachers and 622 Natural Science teachers. Bromfenbrenner’s Ecological Systems Theory and Engeström’s Activity Theory was used to investigate Natural Science and Mathematics teachers’ progress in their ICT pedagogical practices through the time–frame 2004 to 2013, as stipulated in the South Africa’s White paper on e–Education policy. Statistical analysis using Statistical Package for Social Sciences was used to address the research and sub–questions. The study found that South African Mathematics and Natural Science teachers’ level of ICT use is small; when they do use ICT, it is enhanced 21st century pedagogic practices. This is in accordance with findings from the international literature study. / Thesis (Ph.D. (Education))--North-West University, Potchefstroom Campus, 2012.

In South Africa, Science and Technology Education faces many problems. Insufficient numbers of Science and Technology teachers, inadequate in–service training, large classes, instruction with the aim of narrowly orienting students towards examination passes an insufficient integration of technology in the curriculum, and insufficient physical infrastructure dominates the list. The Department of Education envisages the use of ICT as a tool for learning and teaching. ICT has the potential to improve the quality of education and training. If adequate resources are available, and teachers have confidence in the usefulness of ICTs, then the integration of Information and Communication Technology (ICT) may improve the teaching and learning of Mathematics and Science. A review of the literature indicated that the deployment of ICT resources alone will not bring about desirable pedagogical practices in the classroom. There exists a need for interventions that will enhance ICT pedagogical practices in South Africa. The following main research questions were formulated: What are the ICT pedagogic practices used by grade 8 Mathematics and Science teachers in South African classrooms? How do the barriers that grade 8 Mathematics and Science teachers encounter, as well as the support they receive, influence their pedagogical practices? What is the Principal’s role in promoting the emerging pedagogic practices using ICT in South African classrooms? This research comprises a secondary data analysis of the SITES 2006 South African data base. The population and sample for this study was based on the South African grade 8 Mathematics and Natural science teachers. In SITES 2006, the samples comprised more than 504 schools. Due to the fact that ICT is only significantly implemented in two out of nine provinces in South Africa, 25 strata were created to secure fair representation of the population with 666 Mathematics teachers and 622 Natural Science teachers. Bromfenbrenner’s Ecological Systems Theory and Engeström’s Activity Theory was used to investigate Natural Science and Mathematics teachers’ progress in their ICT pedagogical practices through the time–frame 2004 to 2013, as stipulated in the South Africa’s White paper on e–Education policy. Statistical analysis using Statistical Package for Social Sciences was used to address the research and sub–questions. The study found that South African Mathematics and Natural Science teachers’ level of ICT use is small; when they do use ICT, it is enhanced 21st century pedagogic practices. This is in accordance with findings from the international literature study. / Thesis (Ph.D. (Education))--North-West University, Potchefstroom Campus, 2012.

Classic parametric statistical tests, like the analysis of variance (ANOVA), are powerful tools used for comparing population means. These tests produce accurate results provided the data satisfies underlying assumptions such as homoscedasticity and balancedness, otherwise biased results are obtained. However, these assumptions are rarely satisfied in real-life. Alternative procedures must be explored. This thesis aims at investigating the impact of heteroscedasticity and unbalancedness on effect sizes in two-way fixed-effects ANOVA models. A real-life dataset, from which three different samples were simulated was used to investigate the changes in effect sizes under the influence of unequal variances and unbalancedness. The parametric bootstrap approach was proposed in case of unequal variances and non-normality. The results obtained indicated that heteroscedasticity significantly inflates effect sizes while unbalancedness has non-significant impact on effect sizes in two-way ANOVA models. However, the impact worsens when the data is both unbalanced and heteroscedastic. / Statistics / M. Sc. (Statistics)

MURP / Department of Urban and Regional Planning / Rural municipalities in South Africa have been grappling with delivering quality services to communities due to shrinking revenue bases over the last ten years. The study explored the relationship between consumer debt recovery and municipal service delivery in three selected municipalities in Vhembe District, Limpopo Province. The implications of revenue collection in rural areas was examined through the computation of the Coefficient Effect Size Model which aimed at proving whether rurality had an effect on payment of rates and taxes. A combination of Goal Achievement Matrix and Correlation Quotient analysis were used to explain the linkage between service delivery and the resource base of rural municipalities. The Rurality Quotient and the Coefficient Effect Size were both developed in the study to link the rurality nature of municipalities to the viability of rural municipalities to provide services. The review of literature incorporated an appraisal of municipal policy documents such as the Integrated Development Plans, Spatial Development Plans, Annual Budget Reports and Service Delivery Budget Implementation Plan. The study findings revealed that consumer debts resulted in rural municipalities failing to generate revenue thus affecting internal revenue pockets. Furthermore it was discovered that rural municipalities were recording a negative effect size as well as a negative Rurality Quotient thus proving that rurality directly affected the level of payment of services. The projection showed a significant probability of a steady increase in consumer debts in rural municipalities in the next ten years if the issues and challenges are not addressed effectively. Therefore a Rural Consumer Debt Recovery Framework tool was proposed to provide clear guidelines on how rural municipalities can deal with non-payment of rates and taxes. In addition to this a review and update of all revenue enhancement policies such as the Credit Control and Debt Collection Policy to match the changes occurring over time in rural municipalities is necessary to change the current status of revenue collection in municipalities. / NRF