Murray Mike
University of KwaZulu-Natal, School of Mathematics, Statistics and Computer ScienceWestville Campus, Durban, KwaZulu-Natal, South Africa.
PLoS One. 2016 Apr 12;11(4):e0153091. doi: 10.1371/journal.pone.0153091. eCollection 2016.
Wealthy schools appoint better qualified teachers, less wealthy schools under qualified teachers. Added to this mix is a powerful teacher's union whose policies attempt to entrench the job security of teachers in the less wealthy schools irrespective of whether they can teach their subjects or not. Can one isolate these effects from that of other socio-demographic factors that may also be affecting the performance of students when they enrol for a degree at the University of KwaZulu-Natal (UKZN)? An outcome variable that subtracts the number of courses that have been failed from the number of courses that have been passed, dividing this by the total number of years that they have spent studying for a particular degree will be used as a response variable for this paper.
The system of secondary education in South Africa is highly polarized. On the one hand, we have a group of mainly Black African students, forming about 80% of the total student population, that come from a vastly under-resourced rural or township based community. On the other hand, we have a group of predominantly White and Indian students who are able to attend a far better resourced set of private schools. Added to this mix, we have 240,000 of South Africa's total number of 390,000 primary and secondary school teachers who belong to a powerful teacher's union which enjoys a strong political alliance with the ruling party in South Africa. With most of their union members teaching in the less wealthy schools in South Africa, `school background' now includes a politically motivated component that focuses on teacher self-interest rather than the education of the child. What sort of effect does school background have on the performance of students when they enter an institution of higher learning? More importantly, can one isolate the effect of school background from that of other possibly confounding factors such as gender, financial aid and the receipt of some form of residence based accommodation that will also impact on their performance while at university?
A total of 6,183 students enrolling for a degree at the University of KwaZulu-Natal (UKZN) over the period 2008 to 2012 were used a dataset for this study. Permission to use this dataset was given by the Teaching and Learning Office at UKZN. The database that was used for this study was obtained from the Division of Management Information (DMI) office at UKZN. The percentage based marks that students have managed to record for Mathematics, English, Biology and Accounting in their school leaving exams together with some other important but observable socio-economic factors were included in a regression model to determine how students will perform at UKZN. Socio-economic variables relating to gender, race and whether they have receivd some form of financial aid or residence based accommodation while studying at university were also included as predictor variables in our regression based model structure.
An interaction effect associated with being a Black African student who has been privileged enough to attend a quintile five school was found to be significant. A main effect associated with being able to attend a more privileged quintile 5 school however was found to be nonsignificant even after an adjustment has been made for gender, race, the receipt of some form of financial aid and residence based accommodation. Given that UKZN already has a number of bridging programs in place that target students who have come from a less privileged background, for university based policymakers, this result may help to justify the targeted selection of Black African students from the less privileged schools that is taking place. Because some of the disparity in matric performance that we are observing may also be associated with teacher competency and the protective influence of a powerful teacher's union, this paper may also help to highlight some of the economic costs related with having under-prepared students. "A mind is a terrible thing to waste"--United Negro College Fund.
富裕学校聘请资质更优的教师，不太富裕的学校则聘请资质不足的教师。此外，还有一个势力强大的教师工会，其政策试图巩固不太富裕学校教师的工作保障，而不论他们是否能教授所授科目。在夸祖鲁 - 纳塔尔大学（UKZN）攻读学位时，能否将这些影响与其他可能也在影响学生表现的社会人口因素的影响区分开来呢？本文将使用一个结果变量，即通过用学生通过的课程数量减去未通过的课程数量，再除以他们攻读特定学位所花费的总年数，以此作为响应变量。
南非的中等教育体系高度两极分化。一方面，我们有一群主要是非洲黑人学生，约占学生总数的80%，他们来自资源极度匮乏的农村或城镇社区。另一方面，我们有一群主要是白人和印度学生，他们能够进入资源好得多的私立学校。此外，在南非总共39万中小学教师中，有24万教师属于一个势力强大的教师工会，该工会与南非执政党有着强大的政治联盟。由于其大多数工会成员在南非不太富裕的学校任教，“学校背景”现在包含了一个出于政治动机的因素，该因素关注教师自身利益而非孩子的教育。当学生进入高等学府时，学校背景对他们的表现有何种影响？更重要的是，能否将学校背景的影响与其他可能混淆的因素，如性别、经济援助以及是否获得某种形式的校内住宿等因素的影响区分开来，这些因素在学生大学期间也会对他们的表现产生影响？
本研究使用了2008年至2012年期间在夸祖鲁 - 纳塔尔大学（UKZN）攻读学位的6183名学生作为数据集。使用该数据集的许可由UKZN的教学办公室提供。本研究使用的数据库来自UKZN的管理信息部门（DMI）办公室。学生在离校考试中数学、英语、生物和会计取得的基于百分比的成绩，以及一些其他重要但可观察的社会经济因素，被纳入一个回归模型，以确定学生在UKZN的表现。与性别、种族以及他们在大学学习期间是否获得某种形式的经济援助或校内住宿相关的社会经济变量，也作为预测变量纳入我们基于回归的模型结构。
发现一个与有幸就读于五分位5级学校的非洲黑人学生相关的交互效应是显著的。然而，即使在对性别、种族、获得某种形式的经济援助和校内住宿进行调整之后，与能够进入更优质的五分位5级学校相关的主效应仍被发现不显著。鉴于UKZN已经有一些针对来自不太优越背景学生的衔接课程，对于大学政策制定者来说，这一结果可能有助于证明目前从不太优越学校有针对性地选拔非洲黑人学生的合理性。因为我们观察到的高考成绩差异中的一些可能也与教师能力以及一个势力强大的教师工会的保护影响有关，本文可能也有助于凸显因学生准备不足而产生的一些经济成本。“浪费头脑是一件可怕的事情”——联合黑人大学基金。
