Characterization of the distribution of platinum group elements in sulphide ores within the Merensky Reef at Modikwa and Two Rivers Platinum Mines, Eastern Bushveld Complex, South Africa
- Authors: Zilibokwe, Nosibulelo Julie
- Date: 2017
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/7938 , vital:21326
- Description: The distribution of the platinum group element (PGE), in the Merensky Reef was characterized by, first determining the occurrence of the platinum group minerals (PGM), then by establishing the PGE concentration in the base metal sulphides (BMS) associated with the PGE mineralization in the Merensky Reef from selected borehole intersections, at the Two Rivers (TRP) and Modikwa Platinum Mines in the Eastern Bushveld Complex. A mineral liberation analyser (MLA) was then used to identify the PGM phases; their silicate and base metal associations; and their grain size distribution. Electron microprobe quantitative analysis and mapping were then used to determine the compositional variation of the PGM and the PGE elemental distribution in the BMS, respectively. The study showed that the BMS including pyrrhotite, pentlandite, and chalcopyrite were the principal sulphides, where pyrrhotite was most prominent with minor quantities of pyrite. Orthopyroxene, clinopyroxene and plagioclase were the most abundant primary silicate minerals identified, while secondary silicates identified included talc, serpentine and amphibole. Platinum group minerals showed three distinct groups with respect to the mineralogical association with the PGE; (i) BMS association; (ii) chromite association; and (iii) silicate association. Of the BMS, chalcopyrite showed the most dominant association with the PGMs. All samples from both mines exhibited a wide range of PGMs, including maslovite, braggite, cooperate, laurite and PGE alloys such as ferroplatinum as well as other unidentified platinum and palladium sulphides, arsenides and bismuthides, while gold was present as electrum. The PGMs ranged in size from less than a micron to about 125 microns with an average of 20 microns. The close association of PGM with BMS along the margins of sulphides indicates that the PGMs were derived from the sulphide melt. PGE distribution in the sulphides at Modikwa showed pentlandite contained the highest concentrations of palladium (up to 379 ppm) and chalcopyrite hosting the highest rhodium concentrations (up to 793 ppm). Samples from Two Rivers revealed pentlandite as the principal host to both palladium and rhodium, with concentrations reaching up to 695 and 930 ppm, respectively. Magnetite at both Modikwa and Two Rivers showed significant rhodium content, reaching up to 982 and 930 ppm, respectively. The pyrrhotite compared to other sulphides contained all the elements found in the platinum group (PPGE), namely, platinum, palladium and rhodium, with all the platinum identified found in the pyrrhotite. The concentrations for the iridium group (IPGE) namely, iridium, osmium, and ruthenium were below the detection limit. The PGE mineralization in the stratigraphy varied within each mine. The mineralization revealed top loading in the central sector (Modikwa) and bottom loading in the southern sector (Two Rivers). The sequence of the Merensky Reef at the two sectors of the Eastern Bushveld Complex showed a remarkable similarity in their mineralogy suggesting that these two sectors were formed from the same liquid or formed simultaneously within a single magma chamber; however the PGE distribution within the stratigraphy may have been controlled by the presence of cumulate sulphides.
- Full Text:
- Date Issued: 2017
- Authors: Zilibokwe, Nosibulelo Julie
- Date: 2017
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/7938 , vital:21326
- Description: The distribution of the platinum group element (PGE), in the Merensky Reef was characterized by, first determining the occurrence of the platinum group minerals (PGM), then by establishing the PGE concentration in the base metal sulphides (BMS) associated with the PGE mineralization in the Merensky Reef from selected borehole intersections, at the Two Rivers (TRP) and Modikwa Platinum Mines in the Eastern Bushveld Complex. A mineral liberation analyser (MLA) was then used to identify the PGM phases; their silicate and base metal associations; and their grain size distribution. Electron microprobe quantitative analysis and mapping were then used to determine the compositional variation of the PGM and the PGE elemental distribution in the BMS, respectively. The study showed that the BMS including pyrrhotite, pentlandite, and chalcopyrite were the principal sulphides, where pyrrhotite was most prominent with minor quantities of pyrite. Orthopyroxene, clinopyroxene and plagioclase were the most abundant primary silicate minerals identified, while secondary silicates identified included talc, serpentine and amphibole. Platinum group minerals showed three distinct groups with respect to the mineralogical association with the PGE; (i) BMS association; (ii) chromite association; and (iii) silicate association. Of the BMS, chalcopyrite showed the most dominant association with the PGMs. All samples from both mines exhibited a wide range of PGMs, including maslovite, braggite, cooperate, laurite and PGE alloys such as ferroplatinum as well as other unidentified platinum and palladium sulphides, arsenides and bismuthides, while gold was present as electrum. The PGMs ranged in size from less than a micron to about 125 microns with an average of 20 microns. The close association of PGM with BMS along the margins of sulphides indicates that the PGMs were derived from the sulphide melt. PGE distribution in the sulphides at Modikwa showed pentlandite contained the highest concentrations of palladium (up to 379 ppm) and chalcopyrite hosting the highest rhodium concentrations (up to 793 ppm). Samples from Two Rivers revealed pentlandite as the principal host to both palladium and rhodium, with concentrations reaching up to 695 and 930 ppm, respectively. Magnetite at both Modikwa and Two Rivers showed significant rhodium content, reaching up to 982 and 930 ppm, respectively. The pyrrhotite compared to other sulphides contained all the elements found in the platinum group (PPGE), namely, platinum, palladium and rhodium, with all the platinum identified found in the pyrrhotite. The concentrations for the iridium group (IPGE) namely, iridium, osmium, and ruthenium were below the detection limit. The PGE mineralization in the stratigraphy varied within each mine. The mineralization revealed top loading in the central sector (Modikwa) and bottom loading in the southern sector (Two Rivers). The sequence of the Merensky Reef at the two sectors of the Eastern Bushveld Complex showed a remarkable similarity in their mineralogy suggesting that these two sectors were formed from the same liquid or formed simultaneously within a single magma chamber; however the PGE distribution within the stratigraphy may have been controlled by the presence of cumulate sulphides.
- Full Text:
- Date Issued: 2017
Exploring employees' need for the development of an employee assistance programme at a welfare organisation
- Authors: Mncunzwa, Gcobisa
- Date: 2017
- Subjects: Employee assistance programs
- Language: English
- Type: Thesis , Masters , MSoc
- Identifier: http://hdl.handle.net/10353/5554 , vital:29325
- Description: The study was conducted in order to determine the employee needs for an employee assistance programme in a welfare organisation in East London, Eastern Cape. This study was born out of a need from the management of this welfare organisation in East London who identified a need for an employee assistance Programme (EAP) within their organisation. According to management, there was a decline in productivity, and high stress levels due to high caseloads and societal demands (Leeuw, May 15, 2015). The theoretical framework that guided this research was systems theory. The rationale for using systems theory is that this study deals with individuals within an organisation who must coexist on a daily basis. There is a general belief that people and their physical-social-cultural environment interact in processes of mutual reciprocity and complementary exchanges (Merton, 2009). A qualitative approach was used in this study. Eight (8) employees from various levels in a welfare organisation were purposively selected to participate in the study. Semi-structured interviews were conducted with the participants and the interviews took more than forty-five (45) minutes each and this enabled the acquisition of rich information which is described in the findings and is used as a basis for the recommendations in the study. After the examination of the needs of employees for an EAP from the observations of both the employer and the employee it was clear that there is a need for employee assistance programmes in welfare organisations. Employees in this organisation are experiencing stress, burnout and were highly concerned about high caseloads that makes them feel incompetent to render adequate and quality services to their employees. The study also revealed that both employees and management of this organisation have different needs, experience different types of problems and view the EAP in different contexts. Despite the differences in their outlook on the situation, the evidence indicates that both employer and employee perceive that there is a need for an EAP, especially clinical services. At the same time, the issue of funding such a programme is also a concern. Recommendations were made for the organisation to develop an EAP to address the needs identified by employees, and for a clear communication strategy be developed to make all levels of employees within the organisation aware of the existence of such a programme. Another recommendation was made for the organisation to identify specific days on which they will conduct preventative awareness and debriefing sessions for the employees. It was also recommended that the South African Council for Social Services Professionals must standardise, regulate and monitor caseloads for the social work profession.
- Full Text:
- Date Issued: 2017
- Authors: Mncunzwa, Gcobisa
- Date: 2017
- Subjects: Employee assistance programs
- Language: English
- Type: Thesis , Masters , MSoc
- Identifier: http://hdl.handle.net/10353/5554 , vital:29325
- Description: The study was conducted in order to determine the employee needs for an employee assistance programme in a welfare organisation in East London, Eastern Cape. This study was born out of a need from the management of this welfare organisation in East London who identified a need for an employee assistance Programme (EAP) within their organisation. According to management, there was a decline in productivity, and high stress levels due to high caseloads and societal demands (Leeuw, May 15, 2015). The theoretical framework that guided this research was systems theory. The rationale for using systems theory is that this study deals with individuals within an organisation who must coexist on a daily basis. There is a general belief that people and their physical-social-cultural environment interact in processes of mutual reciprocity and complementary exchanges (Merton, 2009). A qualitative approach was used in this study. Eight (8) employees from various levels in a welfare organisation were purposively selected to participate in the study. Semi-structured interviews were conducted with the participants and the interviews took more than forty-five (45) minutes each and this enabled the acquisition of rich information which is described in the findings and is used as a basis for the recommendations in the study. After the examination of the needs of employees for an EAP from the observations of both the employer and the employee it was clear that there is a need for employee assistance programmes in welfare organisations. Employees in this organisation are experiencing stress, burnout and were highly concerned about high caseloads that makes them feel incompetent to render adequate and quality services to their employees. The study also revealed that both employees and management of this organisation have different needs, experience different types of problems and view the EAP in different contexts. Despite the differences in their outlook on the situation, the evidence indicates that both employer and employee perceive that there is a need for an EAP, especially clinical services. At the same time, the issue of funding such a programme is also a concern. Recommendations were made for the organisation to develop an EAP to address the needs identified by employees, and for a clear communication strategy be developed to make all levels of employees within the organisation aware of the existence of such a programme. Another recommendation was made for the organisation to identify specific days on which they will conduct preventative awareness and debriefing sessions for the employees. It was also recommended that the South African Council for Social Services Professionals must standardise, regulate and monitor caseloads for the social work profession.
- Full Text:
- Date Issued: 2017
Homeless men in East London : an exploration of significant life experiences to self-concepts
- Authors: Lobi, Thembelihle
- Date: 2017
- Subjects: Homeless persons Homelessness Homeless men
- Language: English
- Type: Thesis , Masters , Counselling Psychology
- Identifier: http://hdl.handle.net/10353/16069 , vital:40661
- Description: This dissertation investigated the life experiences of homeless persons and how they come to view their self-concepts as homeless men. The theoretical framework that underpinned this study was the self-concept theory and social identity theory. In this contextual descriptive study, four homeless men between ages of 28-50 participated in the study. Data was collected by means of disposable cameras and recorded interviews. Thematic analysis was employed to analyse data. The study found that when the participants first became homeless, they encountered stigmatisation and were not ready to accept the homeless identity as it conflicted with their past identity. However, with prolonged homelessness, they took on the new identity and the roles that come with it. The study found that homeless men view themselves as worthless due to their experiences on the street which include living under inhumane conditions, victimisation, and stigma associated with being homeless. The participants have internalised the negative utterances from housed people, which has negatively affected their self-image to a point where they view themselves as worthless,useless and forgetable. The study also found that homeless men use various methods to survive on the streets, which were both self-enhancing such as participating in the informal economy such as car guarding, car washing and collecting tins in exchange for money; and self-destructive such as using substances, risky sexual behaviour. The participants were found to have low self-concepts because of the tough experiences they encounter on the street.
- Full Text:
- Date Issued: 2017
- Authors: Lobi, Thembelihle
- Date: 2017
- Subjects: Homeless persons Homelessness Homeless men
- Language: English
- Type: Thesis , Masters , Counselling Psychology
- Identifier: http://hdl.handle.net/10353/16069 , vital:40661
- Description: This dissertation investigated the life experiences of homeless persons and how they come to view their self-concepts as homeless men. The theoretical framework that underpinned this study was the self-concept theory and social identity theory. In this contextual descriptive study, four homeless men between ages of 28-50 participated in the study. Data was collected by means of disposable cameras and recorded interviews. Thematic analysis was employed to analyse data. The study found that when the participants first became homeless, they encountered stigmatisation and were not ready to accept the homeless identity as it conflicted with their past identity. However, with prolonged homelessness, they took on the new identity and the roles that come with it. The study found that homeless men view themselves as worthless due to their experiences on the street which include living under inhumane conditions, victimisation, and stigma associated with being homeless. The participants have internalised the negative utterances from housed people, which has negatively affected their self-image to a point where they view themselves as worthless,useless and forgetable. The study also found that homeless men use various methods to survive on the streets, which were both self-enhancing such as participating in the informal economy such as car guarding, car washing and collecting tins in exchange for money; and self-destructive such as using substances, risky sexual behaviour. The participants were found to have low self-concepts because of the tough experiences they encounter on the street.
- Full Text:
- Date Issued: 2017
Medication errors in a private hospital closed intensive care unit: a retrospective analysis of process change
- Authors: Cruickshank, Deborah Claire
- Date: 2017
- Subjects: Medication errors
- Language: English
- Type: Thesis , Masters , MPharm
- Identifier: http://hdl.handle.net/10948/15501 , vital:28259
- Description: Healthcare professionals should be concerned about the safety of the patients in their care and the references to patient safety go back as far as the Hippocratic Oath. Study of literature however shows that medical errors are still of concern and the majority of medical errors are medication errors. The aim of the study was to determine whether process changes introduced reduced both the medication prescribing and medication administration errors in the intensive care unit, thereby contributing to an increase in overall patient safety in the intensive care unit. This study retrospectively analysed the effect of the process changes on medication prescribing and administration errors. The research objectives were to: Identify the number and type of prescribing medication errors prior to the implementation of process changes; Identify the number and type of administration medication errors prior to the implementation of process changes; Identify the process changes implemented; o Determine number and type of prescribing medication errors post the implementation of process changes; Determine number and type of administration medication errors post the implementation of process changes; Assess whether the process changes affected the number and type of prescribing medication errors; and o Assess whether the process changes affected the number and type of administration medication errors. In the Phase One (1 November 2013 to 31 January 2014) 534 patient prescription charts for 172 patients were examined. Medication error rates of 57.6% (n=99) of individual patients reviewed and 18.4% of total patient prescription charts reviewed were found. A total of 69 medication prescribing errors were found in Phase One. This equates to an error percentage of 12.9% per patient chart reviewed and 40.1% per patient reviewed. Thirty medication administration errors were identified in Phase One of the study representing 17.4% of patients reviewed and 5.6% of patient prescription charts reviewed. Medication administration errors included both errors of commission, incorrect doses administered, (n=19) and omission, dose missed, (n=11). Process changes were then introduced and the results of these changes analysed in Phase Two (1 April 2014 to 31 December 2014) show an overall reduction in total medication errors with relation to number of patients reviewed from 57.6% in Phase One to 40.5% in Phase Two. In relation to number of prescription charts reviewed the medication error rate in Phase One was 18.4% and in Phase Two 14.4%. Prescribing errors in relation to number of patients reviewed reduced from 40.1% in Phase One to 26.19% in Phase Two. Overall reductions in percentage of errors were seen in all categories of prescribing errors except duplication of therapy which showed a slight increase. Based on the number of patients reviewed a reduction of prescribing errors was seen in the following categories: transcription errors (13.3% to 7.6%), anticoagulant not prescribed when indicated (3.5% to 2.4%), medication safety (5.2% to 2.9%), dose errors (9.3% to 6.6%) and duration of therapy (6.3% to 3.6%). An increase in the duplication of therapy error rate was seen (2.3% to 3.1%). There was also a reduction in administration errors in relation to number of patients reviewed with a total number of errors of 17.4% in Phase One and 15.8% in Phase Two. The number of prescribing errors per medication chart in Phase Two showed a statistically significant reduction (p=.002). A statistically significant reduction was also seen when the number of errors per patient was reviewed (p=.008). The total number of medication administration errors per medication chart showed a significant reduction (p=.042) as did the number of administration errors per patient (p=.003). When combining the total number of medication errors (both prescriobing and administration) a significant reduction was seen for both the number of charts reviewed (p=.001) and the number of patients reviewed (p=.002). These results indicate that the desired goal of increasing patient safety with regard to medication errors has been achieved but ongoing study is required to ensure the sustainability of the process changes.
- Full Text:
- Date Issued: 2017
- Authors: Cruickshank, Deborah Claire
- Date: 2017
- Subjects: Medication errors
- Language: English
- Type: Thesis , Masters , MPharm
- Identifier: http://hdl.handle.net/10948/15501 , vital:28259
- Description: Healthcare professionals should be concerned about the safety of the patients in their care and the references to patient safety go back as far as the Hippocratic Oath. Study of literature however shows that medical errors are still of concern and the majority of medical errors are medication errors. The aim of the study was to determine whether process changes introduced reduced both the medication prescribing and medication administration errors in the intensive care unit, thereby contributing to an increase in overall patient safety in the intensive care unit. This study retrospectively analysed the effect of the process changes on medication prescribing and administration errors. The research objectives were to: Identify the number and type of prescribing medication errors prior to the implementation of process changes; Identify the number and type of administration medication errors prior to the implementation of process changes; Identify the process changes implemented; o Determine number and type of prescribing medication errors post the implementation of process changes; Determine number and type of administration medication errors post the implementation of process changes; Assess whether the process changes affected the number and type of prescribing medication errors; and o Assess whether the process changes affected the number and type of administration medication errors. In the Phase One (1 November 2013 to 31 January 2014) 534 patient prescription charts for 172 patients were examined. Medication error rates of 57.6% (n=99) of individual patients reviewed and 18.4% of total patient prescription charts reviewed were found. A total of 69 medication prescribing errors were found in Phase One. This equates to an error percentage of 12.9% per patient chart reviewed and 40.1% per patient reviewed. Thirty medication administration errors were identified in Phase One of the study representing 17.4% of patients reviewed and 5.6% of patient prescription charts reviewed. Medication administration errors included both errors of commission, incorrect doses administered, (n=19) and omission, dose missed, (n=11). Process changes were then introduced and the results of these changes analysed in Phase Two (1 April 2014 to 31 December 2014) show an overall reduction in total medication errors with relation to number of patients reviewed from 57.6% in Phase One to 40.5% in Phase Two. In relation to number of prescription charts reviewed the medication error rate in Phase One was 18.4% and in Phase Two 14.4%. Prescribing errors in relation to number of patients reviewed reduced from 40.1% in Phase One to 26.19% in Phase Two. Overall reductions in percentage of errors were seen in all categories of prescribing errors except duplication of therapy which showed a slight increase. Based on the number of patients reviewed a reduction of prescribing errors was seen in the following categories: transcription errors (13.3% to 7.6%), anticoagulant not prescribed when indicated (3.5% to 2.4%), medication safety (5.2% to 2.9%), dose errors (9.3% to 6.6%) and duration of therapy (6.3% to 3.6%). An increase in the duplication of therapy error rate was seen (2.3% to 3.1%). There was also a reduction in administration errors in relation to number of patients reviewed with a total number of errors of 17.4% in Phase One and 15.8% in Phase Two. The number of prescribing errors per medication chart in Phase Two showed a statistically significant reduction (p=.002). A statistically significant reduction was also seen when the number of errors per patient was reviewed (p=.008). The total number of medication administration errors per medication chart showed a significant reduction (p=.042) as did the number of administration errors per patient (p=.003). When combining the total number of medication errors (both prescriobing and administration) a significant reduction was seen for both the number of charts reviewed (p=.001) and the number of patients reviewed (p=.002). These results indicate that the desired goal of increasing patient safety with regard to medication errors has been achieved but ongoing study is required to ensure the sustainability of the process changes.
- Full Text:
- Date Issued: 2017
Missing values: a closer look
- Authors: Thorpe, Kerri
- Date: 2017
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/d1017827 , vital:20798
- Description: Problem: In today’s world, missing values are more present than ever. Due to the ever-changing and fast paced global society in which we live, most business and research data produced around the world contain missing data. This means that locating data which is meticulously precise can be a hard task in itself, but at times may prove essential as the consequences of making use of incomplete data could be disastrous. The reasons for missing data cropping up in almost all forms of work are numerous and shall be discussed in this dissertation. For example, those being interviewed or polled may choose to simply ignore questions which are posed to them, recording equipment may malfunction or be misplaced, or organisers may not be able to locate the respondent in order to rectify the missing data. Whatever the reasons for data being incomplete, it is necessary to avoid having to use inefficient and incomplete data as a result from the above problems. Therefore, various strategies or methods have been developed in order to handle these missing values. It is important, however, that these strategies or methods are utilised effectively as missing data treatment can introduce bias into the analysis. This dissertation shall look at these and other problems in more detail by using a data set which consists of records for 581 children who were interviewed in 1990 as part of the National Longitudinal Survey of Youth (NLSY). Approach: As mentioned above, many strategies or methods have been developed in order to deal with missing values. More specifically, traditional methods such as complete case analysis, available case analysis or single imputation are widely used by researchers and shall be discussed herein. Although these methods are simple and easy to implement, they require assumptions about the data that are not often satisfied in practice. Over the years, more up to date and relevant methods, such as multiple imputation and maximum likelihood have been developed. These methods rely on weaker assumptions and contain superior statistical properties when compared to the traditional techniques. In this dissertation, these traditional methods shall be reviewed and assessed in SAS and shall be compared to the more modern techniques. Results: The ad hoc techniques for handling missing data such as complete case and available case methods produce biased parameter estimates when the data is not missing completely at random (MCAR). Single imputation techniques likewise produce biased estimates as well as result in the underestimation of standard errors. Although the expectation maximisation (EM) algorithm yields unbiased parameter estimates, the lack of convenient standard errors suggests that using this algorithm for hypothesis testing is not a good idea. Multiple imputation, however, yields unbiased parameter estimates and correctly estimates standard errors. Conclusion: Ignoring missing data in any analysis produces biased parameter estimates. Using single imputation to handle missing values is not recommended, as using a single value to replace missing values does not account for the variation that would have been present if the variables were observed. As a result, the variance will be greatly underestimated. The more modern missing data methods such as the EM algorithm and multiple imputation are preferred over the traditional techniques as they require less stringent assumptions and they also mitigate the downsides of the older methods.
- Full Text:
- Date Issued: 2017
- Authors: Thorpe, Kerri
- Date: 2017
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: http://hdl.handle.net/10962/d1017827 , vital:20798
- Description: Problem: In today’s world, missing values are more present than ever. Due to the ever-changing and fast paced global society in which we live, most business and research data produced around the world contain missing data. This means that locating data which is meticulously precise can be a hard task in itself, but at times may prove essential as the consequences of making use of incomplete data could be disastrous. The reasons for missing data cropping up in almost all forms of work are numerous and shall be discussed in this dissertation. For example, those being interviewed or polled may choose to simply ignore questions which are posed to them, recording equipment may malfunction or be misplaced, or organisers may not be able to locate the respondent in order to rectify the missing data. Whatever the reasons for data being incomplete, it is necessary to avoid having to use inefficient and incomplete data as a result from the above problems. Therefore, various strategies or methods have been developed in order to handle these missing values. It is important, however, that these strategies or methods are utilised effectively as missing data treatment can introduce bias into the analysis. This dissertation shall look at these and other problems in more detail by using a data set which consists of records for 581 children who were interviewed in 1990 as part of the National Longitudinal Survey of Youth (NLSY). Approach: As mentioned above, many strategies or methods have been developed in order to deal with missing values. More specifically, traditional methods such as complete case analysis, available case analysis or single imputation are widely used by researchers and shall be discussed herein. Although these methods are simple and easy to implement, they require assumptions about the data that are not often satisfied in practice. Over the years, more up to date and relevant methods, such as multiple imputation and maximum likelihood have been developed. These methods rely on weaker assumptions and contain superior statistical properties when compared to the traditional techniques. In this dissertation, these traditional methods shall be reviewed and assessed in SAS and shall be compared to the more modern techniques. Results: The ad hoc techniques for handling missing data such as complete case and available case methods produce biased parameter estimates when the data is not missing completely at random (MCAR). Single imputation techniques likewise produce biased estimates as well as result in the underestimation of standard errors. Although the expectation maximisation (EM) algorithm yields unbiased parameter estimates, the lack of convenient standard errors suggests that using this algorithm for hypothesis testing is not a good idea. Multiple imputation, however, yields unbiased parameter estimates and correctly estimates standard errors. Conclusion: Ignoring missing data in any analysis produces biased parameter estimates. Using single imputation to handle missing values is not recommended, as using a single value to replace missing values does not account for the variation that would have been present if the variables were observed. As a result, the variance will be greatly underestimated. The more modern missing data methods such as the EM algorithm and multiple imputation are preferred over the traditional techniques as they require less stringent assumptions and they also mitigate the downsides of the older methods.
- Full Text:
- Date Issued: 2017
- «
- ‹
- 1
- ›
- »