Credit points: 15
The literature abounds with findings that collectively may offer important new insights for the betterment of the medical, psychological and life sciences, to name just a few. This subject is designed to provide students with the ability to combine estimated measures of evidence, known as effects, from comparable studies to increase power. Estimators are introduced which are commonly found in meta-analytic research and pitfalls are discussed. On completion of the subject, the student will have an understanding of the different effects that can be collected from the literature as well as an appreciation of how effect sizes arising from data measured on different scales can be combined. Importantly, this subject also shows students how meta-regression can be used to account for study-specific covariates that cannot be adequately accounted for using random-effects models. The freely available software packages R and RevMan are used throughout the subject.
SchoolSchool Engineering&Mathematical Sciences
Subject Co-ordinatorLuke Prendergast
Available to Study Abroad StudentsYes
Subject year levelYear Level 5 - Masters
Prerequisites Must be admitted in the Master of Data Science (SMDS) and have passed STM4PSD or both STA4SS and STM4PM. Other students require Coordinators Approval.
|Resource Type||Title||Resource Requirement||Author and Year||Publisher|
|Readings||Online learning materials||Prescribed||Prendergast 2017||La Trobe University|
Graduate capabilities & intended learning outcomes
01. Present clear, well-structured arguments that validate the combination of effect sizes from several studies that may differ in several aspects, including with respect to moderators and data measurement scales.
- Modeled via example in online lectures and readings. Fortnightly computer laboratories require students to work on real data problems under guidance of the demonstrator.
02. Describe summary output from meta-analyses in a manner which can be understood by scientists with weak statistical backgrounds, but with knowledge from the discipline for which the research question is of interest.
- Modeled via example in online lectures and readings. Fortnightly computer laboratories require students to work on real data under guidance of the demonstrator.
03. Demonstrate an expert understanding of the use of statistical software packages by carrying out valid analyses of real meta-data and providing clear interpretations of all computer output.
- Online videos provide details of how to use statistical software packages for a wide range of applications. Fortnightly computer laboratories require students to work on real data under guidance of the demonstrator.
04. Critique and summarise published meta-analyses, clearly highlighting weaknesses and flaws that undermine the key findings.
- Modeled via example in online lectures and readings. In some computer laboratory classes, students will be required to highlight problems with published findings and re-analyse the data correct to observe differences. This is under the guidance of the demonstrator.
Select to view your study options…
Melbourne, 2019, Semester 2, Blended
Maximum enrolment sizeN/A
Subject Instance Co-ordinatorLuke Prendergast
Unscheduled Online Class
One 2.0 hours unscheduled online class per week on any day including weekend from week 31 to week 43 and delivered via online.
"Approximately two hours per week of videos and readings."
One 2.0 hours computer laboratory every two weeks on weekdays during the day from week 31 to week 43 and delivered via face-to-face.
|Online quizzes (750-word equivalent)||There are 5 short quizzes throughout the semester. Students can attempt each quiz a maximum of three times and the best mark for that quiz taken. Randomly assigned questions for each quiz instance.||10||02, 03|
|Two written assignments submitted online (each 750-word equivalent)||20||01, 02, 03|
|Written Project (2250-word equivalent)||Students must choose a recently published meta-analysis for appraisal/critique. The chosen paper must be pre-approved by the Subject Coordinator. Project must include replication of the presented results which will form part of the assessment.||20||02, 03, 04|
|one 2-hr final examination||50||01, 02, 04|