Table of Contents
Meta-Analysis Workshop for ESMARConf2023
|Workshop Dates||24 March, 2023|
|Workshop Times||17:00 - 00:00 CET (check local time)|
|Workshop Location||online / live stream|
|Workshop Fee||none (this workshop is free!)|
|Materials||can be found here or here|
This workshop was held prior to the Evidence Synthesis & Meta-Analysis in R Conference (which took place on 27-31 March 2023) as a pre-conference workshop. For more information about the conference itself, check out the conference website.
Meta-analysis denotes a particular approach to research synthesis that makes use of quantitative methods for aggregating and comparing the results from related studies in a systematic manner and has become the method of choice for summarizing the constantly growing body of research in the social, health, and natural sciences. The focus of this workshop is on current methods and techniques for conducting a meta-analysis, using the statistical software package R for the analyses.
We will start by looking at methods for quantifying the results from individual studies included in a meta-analysis in terms of various effect size or outcome measures (e.g., raw or standardized mean differences, ratios of means, risk/odds ratios, risk differences, correlation coefficients). We will then delve into methods for combining the observed outcomes (i.e., via equal- and random-effects models) and for examining whether the outcomes depend on the characteristics of the studies from which they were derived (i.e., via meta-regression and subgrouping). A major problem that may distort the results of a meta-analysis is publication bias (i.e., when the studies included in a meta-analysis are not representative of all the relevant research that has been conducted on a particular topic). Therefore, current methods for detecting and dealing with publication bias will be discussed next. Finally, time permitting, we will look at some advanced methods for meta-analysis to handle more complex data structures that frequently arise in practice, namely when studies contribute multiple effect sizes to the same analysis, leading to dependencies in the data that need to be accounted for (via multilevel/multivariate models and robust variance estimation).
Emphasis throughout the workshop is on the application of the methods and the interpretation of the results (i.e., theory will be covered only as necessary to properly interpret the results).
The workshop will be taught online as a live stream via the streaming platform Twitch. Once the stream goes live (around 5-10 minutes before the actual start time), simply go to this link to start watching. Note that the starting time of the workshop is fixed, but the ending time might vary a bit.
As the workshop is specifically focused on the statistical methods for conducting a meta-analysis, a basic understanding of the research synthesis process as a whole is useful. Reading a few meta-analyses from one's field of interest is a good starting point. Siddaway et al. (2019) also provide a nice introduction. Chapters 1-5 from Cooper (2017) provide a more thorough introduction to the steps prior to the analysis.
Second, basic knowledge of statistical methods (e.g., regression, analysis of variance, hypothesis testing) is assumed. If you passed the intro to stats course for your discipline, you should be fine.
Finally, the primary software package to be used for the analyses during the workshop will be R (see below for more details). Although no prior experience with R will be assumed, this will not be a comprehensive R programming course. Therefore, if you are new to R, it would be useful to familiarize yourself with R a little bit ahead of time (see also my notes on preparing to use R).
How to Prepare for the Workshop
You will need a computer with the current version of R installed. You can download R from the Comprehensive R Archive Network (CRAN). Follow the appropriate "Download R" link depending on your operating system and follow the instructions for downloading and installing R. If you already have R installed, please check that it is the current version (you can check what the 'latest release' of R is by going to CRAN and then compare this with the version shown when you start R). If not, please update.
Although not strictly necessary, it will also be useful to install an integrated development environment (IDE) for R. A popular choice these days is RStudio. So, unless you already have a different setup, download the appropriate installer of RStudio from here and install in the usual manner.
Once R and RStudio are installed, please also install the metafor package (an add-on for conducting meta-analyses with R). More details about the package can be found on the metafor package website. You should be able to install the package by starting R/RStudio and then typing
install.packages("metafor") into the 'Console'.
You do not need a Twitch account to watch the stream, but if you would like to post comments or questions via the chat (see below), then you do need to be signed in. To create an account, go to Twitch, click on "Sign Up", and follow the instructions. Make sure you also verify your email address as part of the registration procedure (otherwise you cannot use the chat).
Chat (Purpose and Rules)
You can ask questions and comments via the chat. If something is unclear, please let me know! You are also welcome to answer questions from other workshop participants if you know the answer.
If possible, I will also be happy to answer 'but how do I do this or that?' type questions (assuming they relate to what is being covered in the workshop at that moment). However, if there are too many questions of this type at the same time, the chat will become unusable, so please consider carefully whether it is appropriate to ask such questions at particular moments. Also, it will generally not be possible to answer specific questions about the meta-analysis you are currently working (but you can try to abstract your question away from the specifics of your own meta-analysis to make it of interest to the general audience).
Note that there is a slight delay when live streaming via Twitch (between a few and sometimes up to 10-20 seconds), so keep that in mind when asking questions. This can also make 'back-and-forth' questioning difficult (e.g., if I have to ask for some clarification about your question).
Any type of harassment or hateful conduct, inappropriate commenting, or disruptive behavior will not be tolerated and will lead to temporary chat timeouts or to being banned from the chat permanently.
Other Things / Notes / FAQs
Some other notes and frequently asked questions that may come up:
- I will not be able to troubleshoot general computer problems during this workshop (e.g., with installing R, RStudio, packages, and so on).
- I reserve the option to end the workshop at any point (e.g., due to low attendance, technical problems). Also, depending on how things go (e.g., if there are technical problems, lots of questions), I might not be able to cover all topics planned.
- If, for whatever reason, the stream goes down (and I cannot reconnect and get it running again), I will post a note at the top of this page.
- I cannot provide 'course certificates' or some other form of certification that you have participated in the workshop.
- The workshop will not be recorded.
- If the stream lags/stutters, try reloading the stream (just refresh the page) or reduce the stream resolution (click on the little gear icon at the bottom right of the stream window, select 'Quality', and try 720p or 420p; you could also try disabling 'Low Latency' under 'Advanced'). Note that the option to reduce the stream resolution might not be available (which is not something I have control over).