La presente investigación tiene como finalidad encontrar la relación que existe entre el empleo materno y la desnutrición crónica infantil para niños entre los 13 a 59 meses en el Perú. Para ello, se utilizó la base datos de la Encuesta Demográfica y de Salud Familiar (ENDES) del 2019 donde se obtuvo la información de la madre y su hijo, así como sus medidas antropométricas. Adicionalmente, se emplearon las metodologías de Logit y Probit para cuantificar el impacto que tienen estas variables entre sí. En base a los criterios de selección del modelo, el modelo Probit fue el que mejor se ajustó para poder explicar la relación entre estas dos variables. El resultado que se obtuvo fue que para una madre el hecho de trabajar reduce la probabilidad de que su hijo sufra de desnutrición crónica. Asimismo, se encontraron otras variables que afectan a la desnutrición crónica infantil: nivel socioeconómico, urbanidad, número de hijos en el hogar y tipo de trabajo de la madre. / This research aims to find the relationship between maternal employment and chronic child malnutrition for children between 13 to 59 months in Peru. For this, the database that we used was the Demographic and Family Health Survey of 2019, where we can find the information of the mother and her child, as well as their anthropometric measurements. In addition, the Logit and Probit methodologies were used to quantify the impact that these variables have to each other. Based on the model selection criteria, the Probit model was the one that had the best adjusted to be able to explain the relationship between these two variables. The result was that for a mother, working reduces the probability that her child will suffer from chronic malnutrition. Also, this investigation found other variables that affect the chronic child malnutrition such as urbanity, socioeconomic status, urbanity, type of work of the mother and number of children in the home. / Trabajo de investigación

The overall goal of this study was to analyse the welfare effect of improved wastewater treatment with the view of making policy recommendations for sustainable urban and peri-urban irrigation agriculture in Kenya. This goal was achieved by investigating three specific objectives. The first objective was to assess the farmers’ awareness of health risks in urban and peri-urban wastewater irrigation. Second objective was to analyse the factors that affect the choice of low-risk adaptations in reuse of untreated wastewater for irrigation. The third objective was to estimate the value that urban and peri-urban farmers who practice wastewater irrigation impute to improvements in specific characteristics of the wastewater input in agriculture. In order to achieve the first objective, an ordered probit model was used to identify the factors that influence farmers’ awareness of health risks in untreated wastewater irrigation. The model was fitted to data collected from a cross-sectional survey of 317 urban farm households in the Kibera informal settlement of Kenya. Results of this study show that gender of household head, household size, education level of household head, farm size, ownership of the farm, membership to farmers’ group, and market access for the fresh produce significantly affect awareness of farmers about health risks in wastewater irrigation. Therefore, there is need for awareness programs to promote public education through regular training and local workshops on wastewater reuse in order to improve the human capital of the urban and peri-urban farmers. To achieve the second objective, the study used a multinomial logit model to analyse the farmers’ choice of low-risk adaptations in untreated wastewater irrigation. A survey of 317 urban and peri-urban farmers was conducted and measures for risk-reduction in wastewater reuse were analysed. The urban and peri-urban farmers were found to have adopted low-risk wastewater irrigation techniques such as cessation of irrigation before harvesting, crop restriction and safer application methods. Results of the study show that adoption of risk-reduction measures is significantly influenced by the following factors: household size, age of the household head, education of household head, access to extension, access to media, access to credit, farmers’ group membership, and risk awareness. Also, marginal analysis of the coefficients confirmed the socio-economic characteristics are key determinants in adoption of low-risk measures in wastewater reuse. The study recommends that policies in support of low-risk urban and peri-urban irrigation agriculture should disaggregate farmers according to their socio-economic and institutional characteristics in order to achieve their intended objectives. To achieve the third objective, the study employed the discrete choice experiment approach to estimate the benefits farmers impute to improvements in attributes of the wastewater irrigation input, whose aim is to reduce the health risks associated with untreated wastewater irrigation. Urban and peri-urban farmers who practice wastewater irrigation drawn from Motoine-Ngong River in Nairobi were randomly selected for the study. A total of 241 farmers completed the presented choice cards for the choice model estimation. A random parameter logit model was used to estimate the individual level willingness to pay for wastewater treatment. The results show that urban and peri-urban farmers are willing to pay significant monthly municipality taxes for treatment of wastewater. Conclusion of this study was that, quality of treated wastewater, quantity of treated wastewater and the riverine ecosystem restoration are significant factors of preference over policy alternative designs in wastewater treatment and reuse. / Thesis (PhD)--University of Pretoria, 2013. / gm2014 / Agricultural Economics, Extension and Rural Development / unrestricted

Due to enormous growth rates during the last couple of years CarSharing has become an interesting field of research. This Master Thesis focuses on the analysis of CarSharing users. For this purpose more than 37.000 participants of the "Große SrV-Vergleichsstädte"-Survey were taken into account. After a short literature review and some general information about the statistical methods, the thesis tries to work out some major characteristics of the typical CarSharing user. It will be shown, that a high income and a young age are of central importance. Also when analysing other variables different user quotas are observed, but can mostly be explained with different age or income levels. At the end a binary logit model is sugested to differ between users and non users, by using the previous analysed observed characteristics as input.:Inhalt Tabellenverzeichnis iii Abbildungsverzeichnis v 1 Einleitung 1 2 Literaturüberblick 3 3 Methodische Grundlagen 7 3.1 Korrigierter Kontingenzkoeffizient nach Karl Pearson 7 3.2 Kolmogorov-Smirnov-Test 8 3.3 Logistische Regression 9 4 Vorbetrachtungen 12 4.1 Datenaufbereitung 12 4.2 Vergleich von tatsächlichen Nutzern und angemeldeten Personen 13 5 Der Typische CarSharing-Nutzer 17 5.1 Alter 17 5.2 Einkommen 22 5.3 Haushaltsgröße 27 5.4 PKW-Verfügbarkeit 36 5.6 Höchste Berufsausbildung 42 5.7 Zusammenfassung 45 6 Logistische Regression 46 7 Fazit 53 Anhang I Literaturverzeichnis III Ehrenwörtliche Erklärung V

Artificial Intelligence in form of Machine Learning classifiers is increasingly applied for travel choice modeling issues and therefore constitutes a promising, competitive alternative towards conventional discrete choice models like the Logit approach. In comparison to traditional theory-based models, data-driven Machine Learning generally shows powerful predictive performance, but often lacks in model interpretability, i.e., the provision of comprehensible explanations of individual decision behavior. Consequently, the question about which approach is superior remains unanswered. Thus, this paper performs an in-depth comparison between benchmark Logit models and Artificial Neural Networks and Decision Trees representing two popular algorithms of Artificial Intelligence. The primary focus of the analysis is on the models’ prediction performance and its ability to provide reasonable economic behavioral information such as the value of travel time and demand elasticities. For this purpose, I use crossvalidation and extract behavioral indicators numerically from Machine Learning models by means of post-hoc sensitivity analysis. All models are specified and estimated on synthetic and empirical data. As the results show, Neural Networks provide plausible aggregate value of time and elasticity measures, even though their values are in different regions as those of the Logit models. The simple Classification Tree algorithm, however, appears unsuitable for the applied computation procedure of these indicators, although it provides reasonable interpretable decision rules for travel choice behavior. Consistent with the literature, both Machine Learning methods achieve strong overall predictive performance and therefore outperform the Logit models in this regard. Finally, there is no clear indication of which approach is superior. Rather, there seems to be a methodological tradeoff between Artificial Intelligence and discrete choice models depending on the underlying modeling objective.

Climate change, increasing fine dust, changes in values and the accessibility of Carsharing are well discussed topics nowadays in combination with the vehicle ownerships in German households. This paper aims to characterize the vehicle ownership and to evaluate their effects. National and international literature discusses the vehicle ownership in different ways like car ownership as status symbol or the variable ‘vehicle ownership’ as a mediating variable. Basis of this analysis is a survey called ‘SrV - Mobilität in Städten’. The used data contains information about households in the 25 ‘großen SrV-Vergleichsstädte’. This information is available on different levels, which means that the information is available in separate datafiles for levels of ways, persons and households. The basis level for this analysis should be the household level. To get this level it is necessary to aggregate the information. As a result, we get several socioeconomic and alternative specific variables which must be investigated with descriptive and correlation methods in order to prove their suitability for the binary logit model. This model allows it to evaluate metric, nominal and categoric variables with the aim to find characteristics about vehicle ownership. Some results are for example that the vehicle ownership is more probable in households with more persons than in single-person households. Furthermore, the income and missing accessibility of alternatives have a positive effect on vehicle ownership. In addition, this model offers the possibility to predict the vehicle ownership in households. An interesting result is, that some variables have another effect than assumed. These results were compared with the findings of other papers. As a result, one can find some parallel and additional structures.:Abbildungsverzeichnis VII Tabellenverzeichnis IX Abkürzungsverzeichnis XI Symbolverzeichnis XIII 1 Einleitung 1 2 Literaturübersicht 3 3 Methodik 5 3.1 Deskriptive Analyse 5 3.1.1 Lage- und Streumaße 5 3.1.2 Zusammenhangsmaße 5 3.2 Binäre logistische Regression 7 3.2.1 Allgemeines 7 3.2.2 Modellformulierung 8 3.2.3 Schätzung der logistischen Regressionsfunktion 9 3.2.4 Prüfung des Gesamtmodells 10 3.2.5 Prüfung der Merkmalsvariablen 13 3.2.6 Residuen-Analyse 14 3.2.7 Interpretation der Regressionskoeffizienten 15 4 Daten 17 4.1 Datensatz 17 4.2 Aufbereitung der Daten 17 4.2.1 Zusammenhänge in der Multilevelstruktur 18 4.2.2 Wegedaten 18 4.2.3 Personendaten 19 4.2.4 Haushaltsdatei 20 4.3 Datengrundlage 21 5 Deskriptive Analyse 23 5.1 Vorgehen 23 5.2 Streu- und Lagemaße für kardinal skalierte und klassierte Variablen 23 5.2.1 Alternativenspezifische Variablen 23 5.2.2 Sozioökonomische Variablen 27 5.3 Korrelation zwischen den metrischen Variablen 29 5.4 Relative Häufigkeiten kategorialer Variablen 29 5.4.1 Höchste Schulausbildung im Haushalt 30 5.4.2 Höchste Berufsausbildung im Haushalt 30 5.4.3 Geschlecht 30 5.4.4 Altersklassen 31 5.4.5 Erwerbstätigkeit 32 5.5 Nominale Variablen 32 5.6 Beurteilung der Variablen anhand des korrigierten Kontingenzkoeffizienten nach Pearson 34 6 Binäres Logit-Modell 35 6.1 Schätzung der Regressionskoeffizienten 35 6.2 Prüfung des Gesamtmodells 37 6.2.1 Informationskriterien und Log-Likelihood-Wert 37 6.2.2 Likelihood-Ratio-Test 37 6.2.3 Pseudo-R2-Statistiken 37 6.2.4 Klassifizierung neuer Elemente 38 6.2.5 ROC-Kurve 38 6.3 Prüfung der Merkmalsvariablen 39 6.4 Residuen-Analyse 39 6.5 Interpretation und Diskussion der Regressionskoeffizienten 40 6.5.1 Metrische Variablen 40 6.5.2 Nominale Variablen 41 6.5.3 Kategoriale Variablen 42 6.5.4 Konfidenzintervalle 44 7 Fazit 45 8 Diskussion und Literatur 47 9 Kritische Würdigung und Ausblick 49 Anhang XVII Danksagung XXXI / In Zeiten des Klimawandels, erhöhten Feinstaubwerten, geänderten sozialen Wertevorstellungen und der Verfügbarkeit von Carsharing rückt der Pkw-Besitz in Haushalten immer wieder in den Fokus der Berichterstattung. Das Ziel dieser Arbeit ist es, Charakteristika zu finden, die den Pkw-Besitz beschreiben, und deren Wirkungen zu beurteilen. Der Besitz eines Pkws wird in der Literatur auf verschiedene Weise im Hinblick auf die Bedeutung als intervenierende Variable oder als Statussymbol untersucht. Als Grundlage dienen die Daten aus der Umfrage ‚SrV - Mobilität in Städten‘, wobei die Ergebnisse der 25 großen SrV-Vergleichsstädte verwendet werden. Diese Daten besitzen eine sogenannte Multilevelstruktur, das heißt, dass die Daten auf Wegeebene, Personenebene und Haushaltsebene separat vorliegen, wodurch eine Aggregation auf das Haushaltsniveau erforderlich wird. Der sich daraus ergebende Datensatz mit sozioökonomischen und alternativenspezifischen Variablen wird mithilfe deskriptiver Methoden sowie mit Zusammenhangsmaßen auf die Eignung als Variablen für die Anwendung des binären Logit-Modells untersucht, um aussagekräftige Ergebnisse generieren zu können. Mithilfe dieses Modells werden kardinale, kategoriale sowie nominale Variablen betrachtet und bewertet. Daraus lässt sich beispielsweise ableiten, dass der Pkw-Besitz in Haushalten mit zunehmender Personenzahl wahrscheinlicher ist, als bei Singlehaushalten. Auch das Einkommen und der fehlende Zugang zu Alternativen hat einen positiven Einfluss auf den Pkw-Besitz. Das Modell kann neben der Bestimmung der Eigenschaften dazu beitragen, den Pkw-Besitz in Haushalten zu prognostizieren. Interessant dabei ist, dass nicht alle Variablen die erwartete Wirkung entfalten. Die gefundenen Ergebnisse des Modells werden mit Erkenntnissen aus der Literatur verglichen, woraus sich einige Parallelen und Ergänzungen ergeben.:Abbildungsverzeichnis VII Tabellenverzeichnis IX Abkürzungsverzeichnis XI Symbolverzeichnis XIII 1 Einleitung 1 2 Literaturübersicht 3 3 Methodik 5 3.1 Deskriptive Analyse 5 3.1.1 Lage- und Streumaße 5 3.1.2 Zusammenhangsmaße 5 3.2 Binäre logistische Regression 7 3.2.1 Allgemeines 7 3.2.2 Modellformulierung 8 3.2.3 Schätzung der logistischen Regressionsfunktion 9 3.2.4 Prüfung des Gesamtmodells 10 3.2.5 Prüfung der Merkmalsvariablen 13 3.2.6 Residuen-Analyse 14 3.2.7 Interpretation der Regressionskoeffizienten 15 4 Daten 17 4.1 Datensatz 17 4.2 Aufbereitung der Daten 17 4.2.1 Zusammenhänge in der Multilevelstruktur 18 4.2.2 Wegedaten 18 4.2.3 Personendaten 19 4.2.4 Haushaltsdatei 20 4.3 Datengrundlage 21 5 Deskriptive Analyse 23 5.1 Vorgehen 23 5.2 Streu- und Lagemaße für kardinal skalierte und klassierte Variablen 23 5.2.1 Alternativenspezifische Variablen 23 5.2.2 Sozioökonomische Variablen 27 5.3 Korrelation zwischen den metrischen Variablen 29 5.4 Relative Häufigkeiten kategorialer Variablen 29 5.4.1 Höchste Schulausbildung im Haushalt 30 5.4.2 Höchste Berufsausbildung im Haushalt 30 5.4.3 Geschlecht 30 5.4.4 Altersklassen 31 5.4.5 Erwerbstätigkeit 32 5.5 Nominale Variablen 32 5.6 Beurteilung der Variablen anhand des korrigierten Kontingenzkoeffizienten nach Pearson 34 6 Binäres Logit-Modell 35 6.1 Schätzung der Regressionskoeffizienten 35 6.2 Prüfung des Gesamtmodells 37 6.2.1 Informationskriterien und Log-Likelihood-Wert 37 6.2.2 Likelihood-Ratio-Test 37 6.2.3 Pseudo-R2-Statistiken 37 6.2.4 Klassifizierung neuer Elemente 38 6.2.5 ROC-Kurve 38 6.3 Prüfung der Merkmalsvariablen 39 6.4 Residuen-Analyse 39 6.5 Interpretation und Diskussion der Regressionskoeffizienten 40 6.5.1 Metrische Variablen 40 6.5.2 Nominale Variablen 41 6.5.3 Kategoriale Variablen 42 6.5.4 Konfidenzintervalle 44 7 Fazit 45 8 Diskussion und Literatur 47 9 Kritische Würdigung und Ausblick 49 Anhang XVII Danksagung XXXI

Movements of cyclists and m.otor vehicles at intersections involve a wide variety of potential conflicting interactions. In Australia, the high numbers of motor vehicles, particularly light passenger vehicles, mixed with cyclists results in many bicycle-light passengervehicle (LPV) crashes (3,135 crashes during 2002-2014). About 68% of cyclist deaths at Australian intersections in 2016 were due to crashes between bicycles and LPVs (DITRLDG, 2016). The high number ofLPV crashes among fatalities among cyclists is an increasing safety concem. When an LPV collides with a cyclist, the resulting impact forces in.tluence the probability of cyclist injury severity outcom.e. Therefore, the goa1 at intersections should be to understand whether and which particular crash patterns are more injurious, in order to better inform approaches to reduce the impact forces to levels that do not result in severe injury outcomes. To examine how crash pattem (or mechanism) influences the injury severity of cyclists in bicycle-motor vehicle crashes at intersections, researchers typically describe the crash mechanism in terms of crash types, such as angle crashes, head--on crashes, rear-end crashes, and sideswipe crashes (e.g., Kim et al., 2007; Pai, 2011 ). While crash types explain crash mechanisms to some extent, this study hypothesiz.es that the trajectories of the crash involved vehicles may provide additional information because they better capture the movements of the vehicles prior to collision. Furthermore, it is argued that injury pattem might be in.tluenced by vehicle travel direction and manoeuvre (Isaksson-Hellman and Wemeke, 2017). For example, when a car is moving straight ahead it is likely to have a higher speed than when it is turning, and if cyclists are struck at a higher impact speed, they tend to sustain more severe injury (Badea-Romero and Lenard, 2013). While many studies have evaluated the association between cyclist injwy severity and crash types, the factors that might influence cyclist injury severity related to trajectory types (vehicle movement and travel direction) have not yet been thoroughly investigated. This study aims to examine the factors associated with cyclists' injury severity for 'trajectory types• compared with the typically used 'crash types' at intersections.

Masters of Science / Department of Agricultural Economics / Hikaru H. Peterson / Concerns about environmental degradation due to agriculture have gained importance as it is associated with soil erosion, health hazards, and ground water pollution. Environment-friendly land use practices have been developed to gain a wide range of environmental benefits including reduced soil erosion, reduced nutrient runoff from crop and livestock facilities, increased biodiversity preservation efforts, and restoration of wetlands and other native ecosystems. No-till is one such practice where soil erosion, nutrient runoff and environmental degradation can be reduced to a certain extent. This study evaluated the factors affecting the adoption of tillage systems in Kansas. A survey was conducted with a total of 135 participants from four different locations in the state of Kansas between August 2006 and January 2007. The adoption process was modeled as a two-step econometric models consisting of perception and adoption equations to estimate the impacts of demographic variables and farmers’ familiarity with and participation in certain conservation programs. The results for the perception models showed that the farm operators’ perceptions regarding whether BPM installation and management is unfair to producers or not and whether environmental legislation is often unfair to producers do not vary systematically across farm size, producers’ familiarity and participation in conservation programs, or other demographics considered in the study. On the other hand, their perceptions regarding how polluted their water supplies varied by their thoughts on relative profitability across various tillage practices, their primary occupation, and their familiarity with conservation programs. Specifically, the results suggested that those who regarded no-till practices to be more profitable than other tillage practices or whose primary occupation was farming-related tended to believe that ground water was not polluted, and those who were less familiar with available conservation programs tended to believe that surface waters were not polluted. The adoption model results suggested that farmers with greater operating acreage, those who perceived that no-till was more profitable than other tillage systems, and those with greater familiarity with and participation in existing conservation programs were more likely to adopt more conservation tillage systems, all else equal. Further, perceptions of fairness of environmental regulations or the level of pollution did not impact the tillage choices.

本研究採用美國國家不動產經紀人協會（National Association of Realtors；NAR）編制的房屋購買力指數（Housing Affordability Index，簡稱HAI）來衡量近年來台灣地區房屋購買力變化的情況。 實證結果顯示，在各縣市中，92年度桃園區之民眾的HAI指數可望突破1，為各區購屋能力首先達到標準的縣市。但預期未來房地產回溫以及房貸利率在93年初觸底回升之後，各區將面臨到購屋能力惡化的情形。此外，在所得差距持續拉大的情況下，位於不同所得級距之民眾購屋能力的差距已拉大為四至六倍之間，其中又以高雄區與台南區貧富之間的購屋力指數差距最大。 在敏感性分析分析方面，房貸利率與房價的變動和HAI指數是呈反向變動的關係，而所得則與HAI呈正向變動，與過去各實證結果一致。而在三個變數之中，房價的下降與所得的增加，對HAI指數的提昇最為明顯。但近年來因為利率變動的幅度遠超過於房價與所得變動的幅度，因此近幾年HAI指數的提昇，主要是反映房貸利率的下降。 此外，本研究透過Logit模型來研究家戶特徵對於購屋能力的影響為何。實證結果方面，若以總括來看或分區討論之結果有相當大之差異，例如在本研究中，除就業人口數、經濟戶長性別、教育程度與從業身分以外，其餘變數像是層別、家庭組織型態、經濟戶長婚姻狀態以及所從事之行業，台灣區實證結果與各區有相當大的出入。 在各區呈現顯著正向關係的變數有戶內人口數（新竹區除外）、就業人口數、經濟戶長年齡（除新竹區、台中區近年與台南區89年以前之外）、教育程度、經濟戶長是否為雇主。值得注意的是，經濟戶長為雇主此一變數之係數在85年或86年有大幅下降的情況(台南區除外)，近年則因利率不斷調降而使雇主與非雇主間的購屋能力差距下降。 而大致呈現不顯著的變數有層別（鄉村相較於城鎮）、家庭組織型態、經濟戶長性別、經濟戶長已婚與否、經濟戶長行業（除台北市與高雄區無業之經濟戶長）。其中在家庭組織型態方面，各區近年來有單親家庭或單人家庭之購屋能力較普通家庭差的現象產生。 其餘變數則因地區的不同而有不同的實證結果。在都市化層別方面，新竹區、台中區、台南區居住在城市的家戶相較於居住在城鎮的家戶，其較有能力負擔該區中價位的房子，而台北縣、桃園區、高雄區則無差別；居住在城鎮或鄉村之家戶，除以台灣區來觀察有差別以外，各區皆家戶購屋能力皆無顯著之差別。 台北市、台北縣、高雄區有自用住宅之家戶較無自用住宅之家戶有能力負擔該區中價位之房子；桃園區、新竹區、台中區為不顯著之正相關；台南區歷年則呈現負的不顯著，有可能暗示台南區家戶即使擁有自用住宅，也面臨著房貸本息負擔過重之問題。

The aim of this research was to construct a freight mode choice model, from the perspective of New Zealand freight shippers, identifying the possibility of mode substitution effects. Shipper’s freight modal choice depends on freight demand and infrastructure as well as the quality of service characteristics of alternative modes, such as transport cost, delivery time, reliability, damage and loss and frequency of service. Freight logistics characteristics, such as the attributes of the shipper, the attributes of the commodities to be transported, and the spatial attributes of shipments, strongly influence modal choice. In New Zealand, due to the heterogeneity of firms and issues of confidentiality and reliability of data, relatively little research has been done on modelling freight mode choice. This research involved revealed preference (RP) and stated preference (SP) surveys of representative freight shippers and agents. User-specific data make it possible to better identify the dependence between shipper’s mode shift behaviour and freight logistics in New Zealand circumstances. Moreover, by applying a discrete choice approach, the possibility of mode substitution effects was investigated. This research approach was prompted by substantial changes in New Zealand’s freight transport patterns due to the increasing use of logistic processes, and previously developed models using a four-stage approach fail to model elements of firms’ characteristics (i.e. size of shipments, delivery distance, export volume, product shelf-life, size and location of firm, number of road fleets, and relationship with contracted carriers). The outcomes of this research have shown that many of the operational and logistical influences that affect mode choice vary with the shipper and the industry. As a result, public policy makers should recognize that effective policy must consider both the needs of the transportation service provider and user. In particular, the public policy maker should recognize that freight transport mode choice results from an array of interactions among transportation characteristics, logistics characteristics and product characteristics.

Energy is the driver of activity in every economy and, therefore, its importance cannot be overemphasised. However, Sub-Saharan Africa (SSA) faces general problems of access to modern energy. Most households and industry in SSA use traditional and unclean energy resources for activities such as cooking, lighting and drying of farm produce. Many households in less developed countries have very limited choices with regard to alternatives to traditional energy supplies. Energy poverty is overt in many poor countries, particularly in the Sub-Saharan region where 700 million people are deprived of access to modern energy facilities. In Malawi, less than 6 percent of the population have access to electricity. There are, therefore, many questions regarding the state of energy poverty still to be answered, not only in Malawi, but also the entirety of the SSA region. Questions such as what is the level of energy poverty in these regions? What determines this level of poverty? Why are people not adopting renewable energy facilities for their household needs? Are some energy facilities inferior to others? Such questions were the centre of the present study. These questions are important because, with energy poverty, nearly all the Millennium Development Goals are unachievable and sustainable development could not be a success story where the dominant source of energy for both households and industry is biomass. This study was based on a survey conducted in South Lunzu Township (SLT), which is a low income area to the east of Ndirande Mountain in the city of Blantyre, Malawi. The survey administered a standard questionnaire through face-to-face interviews with heads of households. Data was collected from 319 respondents who were selected through random sampling techniques. The descriptive statistics suggest that the average household size for South Lunzu Township is 5 people. The average age of the sampled respondents was about 38. Energy Poverty and Sustainable Development The findings of the thesis suggest that over 90 percent of the households sampled were energy poor with energy expenditure exceeding 10 percent of total household expenditure. In terms of energy resources used in SLT, 2.9 percent used electricity for cooking meals. Only 2 households, representing 0.63 percent, use liquefied petroleum Gas (LPG) and just 1 household, representing 0.31 percent, depended mostly on solar power. On the other hand, energy facilities that are considered dirty, inefficient and a danger to the heath of people seem to be popular. For instance charcoal and firewood were used by 25 percent and 4.7 percent of the total sample respectively. Most households use a combination of energy facilities; however, those that are considered inferior are preferred. Of the sample, 42 percent use both charcoal and firewood to cook their meals. Further, the results of the Engel functions suggest that charcoal and wood were not regarded as inferior products for the cooking needs of households despite improvements in income. Electricity, which was also regarded as a normal energy resource, had positive income elasticity. To improve access to modern energy facilities at the household level, the thesis recommends that a flexible trade and tax regime, one that will improve the availability and affordability of renewable energy to the majority, should be adopted. The Logit model of energy poverty reveals that household expenditure on transport, income level, age, and education level of the head of household; household size; and home size, are important factors in explaining the level of energy poverty in South Lunzu Township. Further, the results revealed that expenditure on housing and marital status could not be relied upon as important predictors of the probability of energy poverty in South Lunzu. Expenditure on education was associated with lower levels of energy poverty. Households who spent more on schooling also spent more on food items and their expenditure on energy resources was less than 10 percent of the total expenditure per month. In addition, those households that spent more on food were also likely to be energy well-off. Energy Poverty and Sustainable Development Results of the multinomial logit (MNL) model suggest that most socioeconomic variables under study were inelastic in influencing the probability for the outcomes, at the household level, to be used for the purposes of cooking. Statistically, age, income and education level of the head of household, together with household size, were important factors that influenced the choice of most of the outcomes for cooking purposes, including electricity, charcoal, firewood and LP gas. The major recommendation of this study is that campaigns emphasising the abilities of renewable energy be developed and disseminated. That renewable energy is relegated to poor and uncivilised societies is a notion that must be rooted out of the mindset of the average, civilised urban dweller. Also, the use of LP gas for cooking purposes must be encouraged. Import tax regimes that discourage international trade of renewable energy resources must be removed to encourage lower prices on such facilities. These policies would ensure sustainable development by reducing reliance on biomass, which is depleting at a faster rate than it is regenerating. / Thesis (PhD (Economics))--North-West University, Vaal Triangle Campus, 2013