• Weitkunat Rolf

The principles of quantitative evidence aggregation will be introduced. A background will be provided on the history, basic concepts, terminology, and methods of systematic and comprehensive information retrieval, data extraction, data pre-processing and management, derivation of different effect measures, and meta-analysis. The role of study designs, sample representativeness, data description and inferential analysis will be addressed. General graphical and numerical methods will be covered related to fixed- and random effect(s) estimation, meta-regression, heterogeneity, and subgroup- and sensitivity analysis. Additional topics include bias adjustment, complex data structures, overlapping evidence, and study quality assessment.

The course aims at providing insights into the contexts, principles, and methods of meta-analysis as well as at promoting hands-on experience. Participants will learn how to assess as well as how to plan and to conduct standard meta-analyses with Excel and R, including the tabulation of results and the generation of funnel and forest plots. Preparation is required, as detailed in the Course Information Sheet, available to participants on “Gestens” upon mandatory pre-registration. 

I. Requirements and Preparation

-­‐ You must be admissible to participate in accordance with the program criteria, and the program administration must have accepted your registration; participation is only possible for pre-­‐registered students

-­‐ You have to bring a laptop, with Excel and RStudio (available for free) installed. Verify that both programs work. Trouble-­‐shooting at the basic level with multiple participants means trouble for the ordered course of events; thus, if possible, familiarize yourself to some degree with the software

-­‐ Wireless will be available to install R packages. Make sure, if ever possible, that you have already accessed the WLAN at the premises and have obtained the password from the program administration. Ceterum censeo, trouble-­‐shooting at the basic level with multiple participants means trouble for the ordered course of events.

-­‐ Obtain the preparatory literature, print it (all course literature is open access), read it upfront (!), and don’t forget bringing it along to the course:

1. Lakens, D., Hilgard, J. & Staaks J. (2016). On the reproducibility of meta-­‐analyses: six practical recommendations. BMC Psychology, 4:24. https://bmcpsychology.biomedcentral.com/articles/10.1186/s40359-­‐016-­‐0126-­‐3
2. Tan, L., Wang, M.J., Modini, M., Joyce, S., Mykletun, A., Christensen, H., & Harvey, S.B. (2014). Preventing the development of depression at work: a systematic review and meta-­‐analysis of universal interventions in the workplace. BMC Medicine, 12:74. http://www.biomedcentral.com/1741-­‐7015/12/74 [There is an erratum published for this paper, which you should read, as it addresses a common problem (duplicate publications on identical data sets), but which we will quite nonchalantly neglect for the purposes of the course]
3. Kitchener, B.A. & Jorm, A.F. (2004). Mental health first aid training in a workplace setting: A randomized controlled trial. BMC Psychiatry, 4:23. http://bmcpsychiatry.biomedcentral.com/articles/10.1186/1471-­‐244X-­‐4-­‐23

Borenstein M.,  Hedges L.V.,  Higgins J.P.T., & Rothstein, 

     H.R. (2009). Introduction to meta-analysis. Chichester: 

    Wiley.

Cooper, H., Hedges, V., & Valentine, J.C. (2009, eds.). 

    The handbook of research synthesis and meta-

    analysis. New York: Russell Sage Foundation. 

Higgins, J.P.T. & Green, S. (2011, eds.). Cochrane

    handbook for systematic reviews of interventions. 

    Chichester: Wiley.

Normand, S.L. (1999). Meta-analysis: formulating, 

    evaluating, combining, and reporting. Statistics in 

    Medicine, 18(3), 321-359.

Ressing, M., Blettner, M., & J. Klug, S.J. (2009). 

    Systematic literature reviews and meta-analyses. 

    Deutsches Ärzteblatt International, 106(27), 456-

    463.

Written report based on an assigned meta-analysis homework, due within 4 weeks

Written report based on an assigned meta-analysis homework, due within 4 weeks

Written report based on an assigned meta-analysis homework, due within 4 weeks

Written report based on an assigned meta-analysis homework, due within 4 weeks