Developing taxonomic and trait-based approaches for assessing and predicting macroinvertebrate responses to elevated fine sediments in the Tsitsa River and its tributaries, South Africa
- Authors: Ntloko, Pindiwe
- Date: 2022-04-08
- Subjects: Water quality South Africa Mzimvubu River Watershed , Sedimentation and deposition South Africa Mzimvubu River Watershed , Aquatic invertebrates Effect of sediments on South Africa Mzimvubu River Watershed , Aquatic invertebrates Classification , Environmental monitoring South Africa Mzimvubu River Watershed , Analysis of variance , Multivariate analysis
- Language: English
- Type: Academic theses , Doctoral theses , text
- Identifier: http://hdl.handle.net/10962/294692 , vital:57245 , DOI 10.21504/10962/294692
- Description: Sedimentation of freshwater systems is one of the leading causes of water quality deterioration. The Mzimvubu River catchment, which includes the Tsitsa River and its tributaries, in the Eastern Cape is prone to elevated sediment impact due to dispersive soils that are easily erodible. In this study, taxonomy and trait-based approaches were used to assess the responses of macroinvertebrates to fine sediments in the Tsitsa River and its tributaries. Macroinvertebrates and environmental variables were sampled seasonally in winter, spring, summer and autumn of 2016 to 2018 in six selected sites, using the South African Scoring System version 5 as a collecting protocol. The sites were selected to represent a decreasing gradient of sediment influence from the highly impacted Sites 1 (Tsitsa upstream) 2 (Tsitsa downstream), and 3 (Qurana River) to moderately impacted Sites 4 (Millstream upstream) and 5 (Millstream downstream) and the least impacted Sites 6 (Pot River upstream), 7 (Little Pot River) and 8 (Pot River downstream), which were collectively referred to as the control sites. Analysis of basic physico-chemical variables, dissolved oxygen, pH, electrical conductivity, turbidity, total suspended solids, temperature and nutrients were undertaken seasonally over the study period. Sediments grain sizes were also analysed. All collected data were subjected to appropriate statistical tests – univariate and multivariate techniques. A fine-sediment-specific multimetric index was developed to monitor the impact of fine sediments on macroinvertebrate assemblages of the Tsitsa River and its tributaries. A total of 12 traits, resolved into 48 trait attributes, were selected to explore their distribution in relation to a fine-sediment stress gradient, and identify the trait-based signature of fine-sediment impact. A trait-based approach was then developed to classify South African macroinvertebrates into two groups: taxa that are potentially vulnerable to fine-sediment impact and those potentially resilient, based on the combination of traits possessed. Two-way analysis of variance (ANOVA) indicated that electrical conductivity, turbidity, embeddedness and total suspended solids were statistically significantly different between the sites. Apart from Dissolved oxygen, the remaining variables were statistically significantly lower at the control sites (P < 0.05). The two-way multivariate analysis of variance (MANOVA) indicated global significant differences between sites and seasons. The two-way MANOVA also revealed that the interaction between the sites and seasons were statistically significant. The MANOVA indicated global combined interactive effects across the sites for suspended fine-sediment grain sizes, two-way ANOVA, followed by a Tukey’s post-hoc test, was carried out to indicate where the significant differences lay. The one-way ANOVA results indicated that very fine sand, very coarse silt, medium silt, and fine silt were significantly higher at Tsitsa upstream, Tsitsa downstream, Qurana tributary that is at Millstream upstream, Millstream downstream and Control sites. The rest of the grain sizes did not differ statistically between the sites. In terms of the settled sediment grain sizes, the volumetric analysis did not show considerable differences across the sites. Settled fine-sediment grain sizes were evenly distributed across the sites. Statistically, MANOVA results indicated no significant differences across sites or across seasons. The developed Sediment Multimetric Index indicated that the sites in the Tsitsa River and those in the Qurana River were highly sedimented during the wet season, but became moderately sedimented during the dry season, indicating that the index responded to seasonality. The sediment multimetric index indicated that the control sites were less sedimented during both the wet season and dry seasons, suggesting minimal seasonal effects at the control sites. Traits such as an exposed and soft body, collector-filterers, shredding, feeding on coarse particulate organic matter and a high sensitivity to dissolved oxygen were identified as fine- sediment-sensitive indicator traits. Identified fine-sediment-tolerant traits and ecological preferences included complete sclerotisation, a cased/tubed body, a preference for fine particulate organic matter, a high tolerance to dissolved oxygen depletion, and climbing and skating behaviours. Regarding the trait-based approach followed for classifying macroinvertebrates into vulnerable taxa and resilient taxa, the results revealed that the relative abundance and richness of the vulnerable taxa decreased predictably along the increasing gradient of sediment impact. However, the relative abundance and richness of resilient taxa showed no marked response to the impact of an increasing gradient of fine sediments. Overall, the present study makes a contribution to the complementary application of trait-and taxonomy-based approaches to freshwater biomonitoring. The trait-based approach enables predictions to be made and tested based on the mechanistic understanding of the mediating roles of traits in organism- environment interaction. A fundamental challenge, which showcases the limitation of the current study, is the sparse trait data on Afrotropical macroinvertebrates at the species or generic levels. In this regard, the iv trait-based approaches developed here were the family level instead of species or genus. This is the first study in South Africa to develop explicit trait-based indicators of elevated fine sediments as well as an approach for predicting macroinvertebrate vulnerability and resilience to fine-sediment effects, thus advancing the science and practice of freshwater biomonitoring. , Thesis (PhD) -- Faculty of Science, Institute for Water Research, 2022
- Full Text:
A survey of organizational culture and organizational performance in a manufacturing company
- Authors: Dom, Veliswa Virginia
- Date: 2018
- Subjects: Corporate culture -- South Africa -- Eastern Cape , Organizational behavior -- South Africa -- Eastern Cape , Factories -- South Africa -- Eastern Cape , Analysis of variance , Shift systems -- South Africa -- Eastern Cape , Employees -- Rating of -- South Africa -- Eastern Cape , Competing Values Framework (CVF)
- Language: English
- Type: text , Thesis , Masters , MBA
- Identifier: http://hdl.handle.net/10962/61861 , vital:28069
- Description: The purpose of this research was to explore the relationship between organizational culture and organizational performance in a manufacturing company. This company supplies metal products to the motor vehicle manufacturing industry, where the quality and quantity of parts produced is of paramount importance, since the buyers of these products are quality conscious. The organization has three production shifts that rotate each week. Anecdotal observations are that irrespective of the time that a shift operates over the course of the month, the different shifts tend to perform at different levels in terms of quality and quantity of output. This study therefore sought to investigate if these differences between shifts are statistically significant, and if so, whether these shifts also have differences in organizational culture. Components of organizational culture include values, norms, beliefs, attitudes, principles and expectations that give the organization a unique personality and differentiate it from other organizations. While the literature indicates that qualitative or quantitative approaches can be used in organizational culture research, this research adopted the quantitative approach, making use of the Competing Values Framework (CVF). The CVF is a four-category organizational culture typology established by Cameron and Quinn (2006). The framework is based on two dimensions: stability or flexibility of the organization, and external or internal focus. In this research, the CVF was used as a measurement tool to evaluate organizational culture. In order to determine differences in organizational culture between shifts, a survey was undertaken (N=138) which measured employee perceptions pertaining to the existing organizational culture of each of the three production shifts at the company. Secondly, differences in performance between the shifts were examined by using the performance data for a three-month period for each shift in terms of quantity and quality. This data was obtained from the management of the production process at the company. Statistical analysis was done using ANOVA to analyse the differences between the shifts. The findings indicated that the dominant existing organizational culture at the company under investigation is a clan culture. Furthermore, there is a statistically significant difference between the cultures of all the three shifts at the manufacturing company. The results also indicated that there is a statistically significant difference in the quantity and quality of production between the three shifts. In conclusion, the research indicated there are differences in culture and in performance, but given the nature of the data, it was not possible to statistically analyse the relationship between shift culture and performance. However, it is conceivable that cultural differences between shifts may be contributing to performance differences. With regards to further research, it is recommended that this research be extended to other branches of the manufacturing company in other regions, in order to determine whether there are any significant differences in culture and performance between these branches and their shifts. Research could also be extended to other South African organizations to create a sufficiently large sample of shift and/or business units, so as to be able to do statistical analysis of the relationship between culture and performance.
- Full Text:
Development and validation of a health literacy measure for limited literacy public sector patients in South Africa
- Authors: Marimwe, Chipiwa
- Date: 2018
- Subjects: Health literacy -- South Africa , Patient education -- South Africa , Communication in medicine -- South Africa , Health literacy -- Social aspects -- South Africa , Poor -- Medical care -- South Africa , Analysis of variance , Multidimensional Screener of Functional Health Literacy (MSFHL)
- Language: English
- Type: text , Thesis , Doctoral , PhD
- Identifier: http://hdl.handle.net/10962/62661 , vital:28227
- Description: The growing complexity of healthcare demands greater patient involvement and skills to navigate this complex system. It has therefore become increasingly important to identify individuals with inadequate health literacy, by using efficient, short and reliable measures for doing so. Most research on the development and validation of health literacy tests has been conducted in high-income countries, with very little reported from low-and middle-income countries (LMICs). Existing health literacy measures have come under scrutiny for their lack of cultural sensitivity, bias towards certain population groups and failure to acknowledge health literacy as a multidimensional concept. These measures usually have limited application in LMICs due to the significantly different structuring of healthcare systems, they overlook the extreme discrepancies in educational levels, and rely too heavily on the ability to read health information. No health literacy data for South Africa are available, and only a few health literacy-based research papers have been published in this country. The aim of the study was to develop and validate a health literacy measure that is contextually and culturally appropriate to measure health literacy in limited literacy public sector patients in South Africa. An Item Bank of 30 questions was developed with the input of a diverse expert consultant panel, and included skills-based and self-reported questions which ensured cultural, contextual and educational level appropriateness. The Information and Support for Health Actions Questionnaire (ISHA-Q) is a health literacy measure developed to assess health literacy for LMICs which includes 14 core scales. These were useful in ensuring coverage of a range of health literacy constructs within the Item Bank. The 30 questions were then allocated to one of three health literacy domains: Procedural knowledge, Factual knowledge and Access to healthcare, health services and social support. Ethical approval for the study was obtained. The questions were translated into isiXhosa and underwent pilot testing. Following pilot testing, 120 isiXhosa first-language speakers, at least 18 years old, who attended public sector facilities and had a maximum 12 years of education were recruited from a primary healthcare clinic in Grahamstown. An interpreter was trained and he participated in all interviews. A questionnaire was used to collect data on the 30-question Item Bank. The Multidimensional Screener of Functional Health Literacy (MSFHL) was used as the primary comparator.The second phase of the study involved the refinement of the 30 questions in the Item Bank, which involved a multi-stage process. Data were analysed statistically using t-test, correlations, chi-square and ANOVA tests at a 5% level of significance, in order to identify problematic questions. Item Response Theory was used to ascertain difficulty and discriminatory ability of the questions. Each question was further subjected to in-depth interrogation by a panel of healthcare professionals to ensure that questions were supported by the conceptual framework and the definitions of health literacy adopted for this study. The number of questions was reduced from 30 to 12, and formed the new Health Literacy Test - Limited Literacy (HELT-LL). To validate the HELT-LL, 210 patients with the same inclusion criteria as previously noted, were recruited from four primary healthcare clinics in the Eastern Cape Province. Individual interviews were conducted with the assistance of the interpreter to collect sociodemographic data as well as data from the HELT-LL, the primary comparator (MSFHL), and a secondary comparator which was a South African modified version of the Newest Vital Sign (NVS-SA). The HELT-LL was re-administered to 40 patients in a follow-up interview two weeks later. The HELT-LL categorised only 17.6% of the patients as having adequate health literacy, just over a third with inadequate health literacy, and the majority with marginal health literacy. Questions in the cognitively demanding Procedural knowledge domain were the most poorly answered, with a mean score of 48.6±24.9%. Patients had great difficulty performing the basic numeric tasks in this domain. The overall mean score for the HELT-LL was 52.8±18.4%, compared with the more cognitively demanding NVS-SA with a mean of 28.6±21.1%, and clearly illustrated the impact of the strategy to include in the HELT-LL a variety of questions with differing cognitive load. The MSFHL, which is based on demographic characteristics and perceived difficulties with reading and writing, had an overall mean score of 44.4±26.2%. Demographic characteristics including age, education and English literacy, were found to be good predictors of limited health literacy, with significant correlations being found between these variables and the mean HELT-LL score. An acceptable value for Cronbach’s alpha, excellent test-retest reliability and excellent concurrent validity show that the HELT-LL is a valid and reliable measure of health literacy in our target population. As there is a paucity of health literacy research emanating from developing countries, this study presents a significant contribution to literature. It is the first study to report the development and validation of a health literacy measure to address the dearth of available health literacy measures applicable for South Africa. If implemented for use in clinical settings and for research purposes, it could provide valuable South African health literacy data which could inform the development of interventions focusing on improving health literacy and health outcomes.
- Full Text:
Generalized linear models, with applications in fisheries research
- Authors: Sidumo, Bonelwa
- Date: 2018
- Subjects: Western mosquitofish , Analysis of variance , Fisheries Catch effort South Africa Sundays River (Eastern Cape) , Linear models (Statistics) , Multilevel models (Statistics) , Experimental design
- Language: English
- Type: text , Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/61102 , vital:27975
- Description: Gambusia affinis (G. affinis) is an invasive fish species found in the Sundays River Valley of the Eastern Cape, South Africa, The relative abundance and population dynamics of G. affinis were quantified in five interconnected impoundments within the Sundays River Valley, This study utilised a G. affinis data set to demonstrate various, classical ANOVA models. Generalized linear models were used to standardize catch per unit effort (CPUE) estimates and to determine environmental variables which influenced the CPUE, Based on the generalized linear model results dam age, mean temperature, Oreochromis mossambicus abundance and Glossogobius callidus abundance had a significant effect on the G. affinis CPUE. The Albany Angling Association collected data during fishing tag and release events. These data were utilized to demonstrate repeated measures designs. Mixed-effects models provided a powerful and flexible tool for analyzing clustered data such as repeated measures data and nested data, lienee it has become tremendously popular as a framework for the analysis of bio-behavioral experiments. The results show that the mixed-effects methods proposed in this study are more efficient than those based on generalized linear models. These data were better modeled with mixed-effects models due to their flexibility in handling missing data.
- Full Text:
Improved tree species discrimination at leaf level with hyperspectral data combining binary classifiers
- Authors: Dastile, Xolani Collen
- Date: 2011
- Subjects: Mathematical statistics , Analysis of variance , Nearest neighbor analysis (Statistics) , Trees--Classification
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: vital:5567 , http://hdl.handle.net/10962/d1002807 , Mathematical statistics , Analysis of variance , Nearest neighbor analysis (Statistics) , Trees--Classification
- Description: The purpose of the present thesis is to show that hyperspectral data can be used for discrimination between different tree species. The data set used in this study contains the hyperspectral measurements of leaves of seven savannah tree species. The data is high-dimensional and shows large within-class variability combined with small between-class variability which makes discrimination between the classes challenging. We employ two classification methods: G-nearest neighbour and feed-forward neural networks. For both methods, direct 7-class prediction results in high misclassification rates. However, binary classification works better. We constructed binary classifiers for all possible binary classification problems and combine them with Error Correcting Output Codes. We show especially that the use of 1-nearest neighbour binary classifiers results in no improvement compared to a direct 1-nearest neighbour 7-class predictor. In contrast to this negative result, the use of neural networks binary classifiers improves accuracy by 10% compared to a direct neural networks 7-class predictor, and error rates become acceptable. This can be further improved by choosing only suitable binary classifiers for combination.
- Full Text:
The comparative bioavailability and in vitro assessment of solid oral dosage forms of paracetamol
- Authors: Braae, Karen
- Date: 1981 , 2013-04-02
- Subjects: Acetaminophen , Bioavailability , Drugs -- Bioavailability , Drugs -- Dosage forms , Analysis of variance
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: vital:3827 , http://hdl.handle.net/10962/d1006288 , Acetaminophen , Bioavailability , Drugs -- Bioavailability , Drugs -- Dosage forms , Analysis of variance
- Description: The dissolution profiles of eight lots of paracetamol tablets representing seven different tablet brands are determined in a USP rotating basket assembly and a stationary basket-rotating paddle apparatus. The in vitro data are expressed in terms of dissolution parameters and inter-tablet differences are assessed statistically using analysis of variance (ANOVA) and the Scheffe test. Highly significant differences are observed between a number of the tablets at the 95% confidence level. Representative tablets from the dissolution rate study and a control dose of paracetamol dissolved in water are subsequently investigated in a 4 x 4 latin square design bioavailability trial. Serum and urine samples are collected and assayed for paracetamol alone (serum) and together with its metabolites (urine) by means of high pressure liquid chromatography. The in vivo data are expressed in terms of bioavailability parameters and differences between the test doses are assessed by means of ANOVA. No significant differences are observed between the dosage forms at the 95% confidence level.
- Full Text: