SYNERGIES OF CROWDSOURCING FOR THE TEACHING DESIGN OF STATISTICS COURSES IN VARIOUS SCHOOLS AND PROGRAMMES OF A HEALTH SCIENCE INCLINED UNIVERSITY
Sefako Makgatho Health Sciences University (SOUTH AFRICA)
About this paper:
Conference name: 10th annual International Conference of Education, Research and Innovation
Dates: 16-18 November, 2017
Location: Seville, Spain
Abstract:The Statistics (Stats) subject is understood well when mathematics (Math) conceptualisation is emphasised in the concepts. Basically, Math skilled statisticians are more resilient and efficient that those with no Math knowledge. Matsose and Seeletse (2016) showed that data analyses undertaken mostly by non-statisticians tended to produce unreliable results. The case of students who learn Stats from lecturers who did some Stats but being non-statisticians is also a flaw. Stats concepts taught without the Math understanding is a deficiency. For example, in the South African education, basic Stats courses do not include concepts of outliers, influential observation and robust statistics methods. As a result, some studies produce recommendations based on data with outliers and influential values. These often lead to unreliable results. Thus, use of core statisticians in teaching the Stats courses in fields of application, such as the health and business sciences, is a worthy alternative for enhancing improved analysis approaches. Also, examples from these applications are worthy in the training of core statisticians due to real life case. These opportunities lead to crowdsourcing and synergistic relationships for training in the use of Stats methods for every role player. There are challenges to address though. This paper shares experiences of these collaborations, the benefits obtained and some approaches that may lead to optimised modus operandi for joint benefiting.
Keywords: Influential observations, math concepts, outliers, robust methods.