Thesis (PhD) -- Macquarie University, School of Biological Sciences, 1990. / Thesis by publication. Bibliography: leaves 269-324.

Study project (MComm) -- University of Stellenbosch, 2001. / ENGLISH ABSTRACT: A psychometric audit of an existing selection procedure: The validity and credibility of assertions made about the effectiveness and faimess of selection procedures is dependent on the methodology with which the procedure was developed and justified. Thus the process of evaluating a selection procedure in practice entails comparing the developmental history of the procedure to the ideal approach of developing and justifying a selection procedure, derived from standard guidelines, to identify and rectify any procedural and substantial shortcomings. Such a process could be described as a psychometric audit. The actuarial approach to selection is proposed as the ideal approach, whereby a mechanical decision rule is derived from historical data and justified in terms of the faimess and utility of the decisions made. A psychometric audit on the developmental history of the selection procedure for the selection of commission advisers is undertaken and various shortcomings are identified and rectified or at least recommendations are made on rectifying them. The audit finds that the selection procedure has zero validity and negative utility and discriminates unfairly. / AFRIKAANSE OPSOMMING: Die geldigheid en geloofwaardigheid van utsprake oor die effektiwiteit en billikheid van In keuringsprosedure is 'n funksie van die metodologie waarmee die prosedure ontwikkel en regverdig is. Dus geskied die evaluasie van In keuringsprosedure in wese deur die ontwikkelingsgeskiedenis van die prosedure te vergelyk met die ideale benadering tot die ontwikkeling en regverdiging van In keuringsprosedure, afgelei uit standaard riglyne, om enige proseduriele en substantiewe tekortkominge te identifiseer en reg te stel So 'n proses sou beskryf kon word as 'n psigometriese oudit. Die aktuariele benadering tot keuring word voorgestel as die ideale benadering, waarvolgens In meganiese beslu~emingsreel afgelei word u~ historiese data en regverdig word in terme van die billikheid en nut van die besluitneming wat in gevolge die reel geskied. In Psigometriese oudit is onderneem op die ontwikkelingsgeskiedenis van die keuringsprosedure vir die keuring van kommissie adviseurs en verskeie tekortkominge is ge"identifiseer en reg gestel of ten minste aanbevelings ten opsigte van regstelling gemaak. Die oudit vind dat die keuringsprosedure oor zero geldigheid en negatiewe nutwaarde beskik en onbillik diskrimineer.

Thesis (MCom)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: People are often referred to in a Human Resource Development context as the organisation‟s most important resource in recognition of the important knowledge and learning they bring to the organisation (Bierema & Eraut, 2004). South African organisations experience a shortage of this valuable and important resource due to the country‟s social political past which was led by the Apartheid system. South Africa today still suffers from the consequences of the history of racial discrimination which was lead by the Apartheid system. This system was one of legal racial segregation enforced by the National Party government of South Africa between 1948 and 1993 and it deprived the majority of South Africans of the opportunity to develop and accumulate human capital. South Africa‟s past has thus left the previously disadvantaged group members with underdeveloped competency potential, as opposed to the not previously disadvantaged group members, and this has subsequently led to adverse impact in valid, fair (in the Cleary sense of the term) strict-top-down selection. This underdeveloped competency potential prohibits these individuals from succeeding in the world of work. Because of the importance of labour it is crucial that the South African labour force be developed to reach its full potential. Adverse impact in personnel selection refers to the situation where a selection strategy affords members of a specific group a lower probability of being selected compared to members of another group (Boeyens, 1989). There thus lies a vast reservoir of untapped human potential in this country, and a method to identify these individuals is required. The fact that adverse impact is created during personnel selection does not necessarily mean that selection procedures are responsible for the adverse impact. Adopting a problem orientation involves using careful analysis to identify the root causes of a problem (Bierema & Eraut, 2004). In South Africa an intellectually honest solution to the problem of adverse impact would be to provide development opportunities, rather than searching for an alternative selection instrument, to those individuals who have been denied opportunities in the past in order to develop skills, abilities and coping strategies necessary for job performance. This does not imply that affirmative action should be abolished; it rather suggests that the focus of this corrective policy should shift towards a more developmental approach. More emphasis should be placed on providing the previously disadvantaged with the necessary training and development to foster the necessary competency potential to succeed in the world of work. Affirmative developmental opportunities will entail giving previously disadvantaged individuals access to skills development and educational opportunities aimed at equipping them with the currently deficit skills and knowledge. A need exists to identify individuals who will gain maximum benefit from these developmental opportunities and who display the highest potential to learn, as resources for such developmental programmes are scarce. Attempts at accelerated affirmative development will be effective to the extent to which there exists a comprehensive understanding of the factors underlying training performance and the manner in which they combine to determine learning performance (De Goede & Theron, 2010). De Goede (2007) has already conducted a study to identify such individuals. Selection alone, although important and necessary, is not sufficient to ensure successful affirmative development interventions. Additional interventions are required, post-selection, to ensure success. The primary objectives of this study are consequently to build onto De Goede‟s (2007) foundations and it is therefore necessary to describe De Goede‟s (2007) model, explain its underlying argument, report on the fit of his proposed structural model and also to report on the findings regarding the specific causal relationships which he proposed. De Goede‟s (2007) existing learning potential structural model was expanded with the inclusion of additional non-cognitive variables in order to gain a deeper understanding of the complexity underlying learning and the determinants of learning performance. The hypothesised learning potential structural model was empirically tested and evaluated and achieved good close fit. Modification indices calculated as part of the structural equation modelling suggested a specific addition to the existing model that would improve the fit. One modification was subsequently made to the model after the consideration of the full range of fit indices, standardised residuals, modification indices and parameter estimates. No paths were removed. This decision was taken because the path-specific hypotheses that were tested referred to the specific paths when they were included in the specific model. Deleting insignificant paths from the model would therefore change the original hypotheses. The final revised structural model achieved good fit. The limitations of the research methodology, the practical implications of this study, and recommendations for future research are also discussed. / AFRIKAANSE OPSOMMING: In die konteks van Menslike Hulpbronontwikkeling word daar vele kere na mense verwys as die organisasie se belangrikste hulpbron uit erkenning vir die belangrike kennis en leer wat hulle na die organisasie bring (Bierema & Eraut, 2004). Suid-Afrikaanse organisasies ervaar „n tekort aan die waardevolle en belangrike hulpbron weens die land se verlede onder leiding van die Apartheidsisteem. Suid-Afrika ly vandag steeds onder die gevolge van die geskiedenis van rassediskriminasie onder leiding van die Apartheidstelsel. Hierdie stelsel is gebaseer op wetlike rasseskeiding, afgedwing deur die Nasionale Party regering in Suid-afrika tussen 1948 en 1993. Hierdie sisteem het die meeste Suid-Afrikaners die geleentheid op toegang tot ontwikkelingsgeleenthede ontneem. Suid-Afrika se verlede het die lede van die voorheen benadeelde groepe gelaat met onderontwikkelde bevoegdheidspotensiaal, in teenstelling met lede van bevoorregte groepe. Dit het daartoe aanleiding gegee dat geldige en regverdige (in die Cleary sin van die begrip) streng bo-tot-onder keuring „n nadelige impak teen voorheen benadeelde individue tot gevolg het. Die onderontwikkelde bevoegdheidspotensiaal verhoed die voorheen benadeelde groepe om suksesvol in die werksplek te wees. Weens die belangrikheid van arbeid is dit noodsaaklik dat die Suid-Afrikaanse arbeidsmag ontwikkel word om sy volle potensiaal te bereik. Nadelige impak in personeelkeuring verwys na die situasie waar „n keuringstrategie lede van „n spesifieke groep „n laer waarskynlikheid van keuring bied in vergelyking met lede van „n ander groep (Boeyens, 1989). Daar bestaan dus „n reuse onontginde reservoir van menslike potensiaal in hierdie land en „n metode om hierdie individue te identifiseer word benodig. Die feit dat „n nadelige impak geskep word tydens personeelkeuring beteken nie noodwendig dat die keuringsprosedures verantwoordelik is vir die nadelige impak nie. Die aanvaarding van „n probleemoriëntasie vereis die gebruik van „n versigtige analise om die grondoorsake van „n problem te identifiseer (Bierema & Eraut, 2004). In Suid-Afrika sal dit „n intellektueel eerlike oplossing ten opsigte van die probleem van nadelige impak bied om ontwikkelingsgeleenthede te voorsien aan daardie lede wat geleenthede misgun is in die verlede, om vaardighede, vermoëns en hanteringstrategieë wat benodig word vir werksprestasie te ontwikkel, eerder as om „n ander keuringsinstrument te soek. Daar word glad nie hiermee geïmpliseer dat regstellende aksie tot niet gemaak moet word nie. Daar word slegs voorgestel dat die fokus van regstellende aksie meer ontwikkelingsgerig moet wees. Groter klem moet dus daarop geplaas word om lede van voorheen benadeelde groepe die geleenthede te gee om die nodige bevoegdheidspotensiaal te ontwikkel om suksesvol in the werksplek te wees. Regstellende ontwikkelingsgeleenthede sal voorheen benadeelde individue toegang gee tot opleidings en ontwikkelingsgeleenthede wat daarop afgestem is om hulle van die nodige vaardighede en kennis te voorsien wat hulle kortkom. „n Behoefte bestaan om daardie indiwidue te identifieer wat die grootste voordeel uit hierdie ontwikkelingsgeleenthede sal trek en wat die hoogste vlak van leerpotensiaal het, aangesien hulpbronne vir die doel baie skaars is. Pogings tot versnelde regstellende ontwikkeling sal net suksesvol wees tot die mate wat daar „n omvattende begrip is van die faktore wat onderliggend is aan leerprestasie en die wyse waarop hulle kombineer om leerprestasie te bepaal (De Goede & Theron, 2010). De Goede (2007) het reeds so „n leerpotensiaalnavorsingstudie gedoen. Keuring alleen, alhoewel belangrik en noodsaaklik, is nie voldoende om suksesvolle regstellende ontwikkelingsingrypings te verseker nie. Verdere addisionele ingrypings word na keuring benodig om sukses te verseker. Die primêre doelstellings van hierdie studie is gevolglik om op De Goede (2007) se fondasies te bou. De Goede (2007) se model is beskryf, sy onderliggende argument is verduidelik, verslag is gedoen oor die pasgehalte van die voorgestelde strukturele model en ook oor sy bevindinge aangaande die spesifieke, oorsaaklike verwantskappe wat hy voorgestel het. De Goede (2007) se bestaande leerpotensiaal strukturele model is gewysig en uitgebrei deur die toevoeging van addisionele nie-kognitiewe veranderlikes om ‟n meer indringende begrip van die kompleksiteit onderliggend aan leer en die determinante van leerprestasie te verkry. Die strukturele model is empiries getoets en geëvalueer en die model het „n goeie passing getoon. Modifikasie-indekse bereken as deel van die strukturele vergelykingsmodellering het „n spesifieke baan uitgewys wat die passing van die model sou verbeter indien dit bygevoeg word tot die bestaande model. Die strukturele model is dus aangepas deur die addisionele baan by te voeg tot die bestaande model na die oorweging van die volle spektrum pasgehaltemaatstawwe, gestandaardiseerde residue, modifikasie-indekse and parameterskattings. Geen bane is verwyder nie. Die besluit is geneem omdat die baan-spesifieke hipoteses wat getoets is, verwys het na spesifieke bane toe hulle ingesluit is in die spesifieke model. Verwydering van bane wat nie statisties beduidend was nie, sou dus die oorspronklike hipoteses verander. Die bevinding was dat die finaal-gewysigde strukturele model die data goed gepas het. Die beperkinge van die navorsingsmetodiek, die praktiese implikasies van die studie en aanbevelinge vir toekomstige navorsing word ook bespreek.

This dissertation addresses statistical issues related to endogenous peer selection in the context of social networks, social interaction models and snowball sampling methods. The first chapter studies the peer effects of friends, studymates, and seatmates on academic performance. We use data on social networks, personality traits, and cognitive ability measures collected through a unique survey conducted in three schools in Hong Kong. We estimate a social interaction model which accounts for endogenous network formation and correlation between multiple networks with a Bayesian approach. Our results show that the cognitive ability of studymates and the conscientiousness of friends positively affect a student's mathematics exam score while the conscientiousness of studymates and the cognitive ability of friends have no effect. The second chapter proposes a novel identification strategy for social interactions in a model with endogenously formed social networks. The network endogeneity arises from the correlation between the links of the network and the unobservables that determine the outcome of interest. We show that the eigenvectors of the adjacency matrix that defines the social network are control variables for network endogeneity without imposing any parametric assumption. We propose an information criterion to select the number of eigenvectors to be included as control variables. We apply the proposed method to the same empirical application as the first chapter and compare the results. The third chapter studies a snowball sampling method for social networks with endogenous peer selection. Snowball sampling is a sampling design which preserves the dependence structure of the network. It sequentially collects the information of vertices linked to the vertices collected in the previous iteration. The snowball samples suffer from a sample selection problem because of the endogenous peer selection. We propose a new estimation method that uses the relationship between samples in different iterations to correct selection. We use the snowball samples collected from Facebook to estimate the proportion of users who support the Umbrella Movement in Hong Kong.

No description available.

No description available.

Simulated evolution is used as a tool for investigating the selective pressures that have influenced the design of animal signalling systems. The biological literature on communication is first reviewed: central concepts such as the handicap principle and the view of signalling as manipulation are discussed. The equation of “biological function” with “adaptive value” is then defended, along with a workable definition of communication. Evolutionary simulation models are advocated as a way of testing the coherence of a given theory. Contra some ALife enthusiasts, simulations are not alternate worlds worthy of independent study; in fact they fit naturally into a Quinean picture of scientific knowledge as a web of modifiable propositions. Existing simulation work on the evolution of communication is reviewed: much of it consists of simple proofs of concept that fail to make connections with existing theory. A particular model (MacLennan & Burghardt, 1994) of the evolution of referential communication in a co-operative context is replicated and critiqued in detail. Evolutionary simulations are then presented that cover a range of ecological scenarios; the first is a general model of food- and alarm-calling. In such situations signallers and receivers can have common or conflicting interests; the model allows us to test the idea that a conflict of interests will lead to an arms race of ever more costly signals, whereas common interests will result in signals that are as cheap as possible. The second model is concerned with communication during aggressive interactions. Many animals use signals to settle contests, thus avoiding the costs associated with fighting. Conventional game-theoretic results suggest that the signalling of aggression or of strength will not be evolutionarily stable unless it is physically unfakeable, but some recent models imply that cost-free, arbitrary signals can be reliable indicators of both intent and ability. The simulation, which features continuous-time perception of the opponent’s strategy, is an attempt to settle the question. The third model deals with sexual signalling, i.e., elaborate displays that are designed to persuade members of the opposite sex to mate. The results clarify the question of whether such displays are the pointless result of runaway sexual selection, or whether they function as honest and costly indicators of genetic quality. The models predict the evolution of reliable communication in a surprisingly narrow range of circumstances; a serious gap remains between these predictions and the ethological data. Future directions for simulation work are discussed.

Understanding the dynamics of functional genetic variability in small populations can have important implications in their conservation. I screened the variation of MHC II gene in Galapagos mockingbirds to evaluate the evolutionary forces that shaped the genetic variation. I found out that genetic drift affected the MHC variation together with a specific form of natural selection. Although the MHC is supposed to be under a pathogen-mediated selection I found no evidence for this theory in the mockingbird study system.

<i>Leptophyes punctatissima</i> is unusual in that both sexes call. The male calls, the female replies and the male performs phonotaxis to the stationary female. Consequently mate choice could occur at either of two stages: first, during the interchange of calls and second, on the basis of proximate criteria once the male has approached. There is no evidence that females choose their mates on the basis of calling behaviour or call characteristics, though males that call more may achieve more matings. There is no evidence that body asymmetry has any effect on mating success for either sex. Males on a protein-supplemented diet do not produce larger spermatophores than males whose diet is not supplemented, but they do mate more often, possibly as a result of female choice but more likely because diet affects the rate at which males can produce spermatophores. Unsupplemented females mate more often than supplemented females, possibly as a result of male male choice or because they are seeking matings so that they can supplement their diet with spermatophores. Males give larger spermatophores to unsupplemented females. Larger males produce larger spermatophores. They also mate more often than smaller males, possibly as a consequence of female choice, success in male-male contests, or because larger males have larger energy reserves and can produce spermatophores more quickly. Larger females mate more often than smaller females but only when their diet was supplemented. Females lay more eggs the more times they mate. Females lay heavier eggs after their first mating than they do in later batches, and unsupplemented females lay more eggs after their first mating than supplemented females do, but otherwise female size, diet or level of asymmetry has no effect on the size or weight of eggs, or the number of eggs laid. The total weight of spermatophores females receive does not affect any measure of female reproductive success: neither fecundity, egg size or egg weight is affected by the weight of spermatophores females consume, irrespective of the diet the females were maintained on. Diet, size or number of matings does not affect female longevity.

The use of consumer credit background checks in employee selection has been increasing and, in recent years, has been the topic of heated debate. Supporters and opponents contest the benefits and consequences of the use of credit background checks for personnel selection, with arguments on both sides predominantly based on anecdotal evidence; empirical research is missing from the debate. The lack of research to support these arguments is concerning due to the historical, evidence-based, relationship between employee selection and organizational justice. Job applicants pursue employment to fulfill economic and socio-economic needs and expect fair processes and outcomes. Imbalances in the input-to-output ratio have been suggested to result in behavioral outcomes intent to restore balance. Two experimental studies examined justice-related consequences of the use of ECCs in personnel selection. Study 1 examined potential applicants' perceptions of organizational justice as well as their engagement in both pro-social (organizational citizenship behaviors-OCB) and anti-social (counterproductive workplace behavior-CWB) behavior as a result of failing a job selection hurdle on the basis of a poor ECC outcome (in comparison to a standard personnel selection criteria- job qualifications and work experience). A sample of adults (N = 171) was recruited from Amazon MTurk to ostensibly pilot test an online employee selection battery. They were randomly assigned to either pass both the ECC and Job Qualifications/Experience tests or to fail one or the other (thus being dropped from further consideration). Applicants denied employment based on their consumer credit experienced significantly lower distributive and procedural justice. They were also more likely than those denied employment on the basis of qualifications and experience and those passing both assessments to engage in a CWB. There were no effects on OCB. The effect of failing on the basis of ECC on CWB engagement was mediated by justice perceptions. Study 2 examined how applicants with weak credit, in comparison to applicants with weak qualifications/experience are perceived by raters. Study 2 also examined the potential for disparate treatment against minority applicants on the basis of ECCs - an issue of distributive justice. A similar sample of (N = 155) working adults recruited from Amazon MTurk were asked to make personnel selection judgments of applicants who varied by type of Applicant Credential (weak consumer credit history but strong job qualifications and experience; or weak job qualifications and experience but strong consumer credit history) and race (White/Black). Type of Applicant Credential significantly affected employability ratings such that those with weak qualifications and experience but strong credit were rated as less employable than those with strong qualifications/experience but weak credit. Also, applicants with weak credit (but strong qualifications/experience) were perceived as more likely to exhibit behavioral indicators of fraud than applicants with weak qualifications/experience (but strong credit). Race of the applicant did not moderate these effects. These studies provide evidence of both individual, and organizational, level outcomes associated with the use of ECCs as well as potential retaliatory behavior (CWB) directed at the organization from applicants denied employment based on credit. However, the findings also suggest that ECCs are not prone to race discrimination effects. The findings fill a necessary gap in the research literature by providing empirical evidence directly related to the use of consumer credit in selection.