logo Epi Result

Good research. Valid results. Better decisions.


Articles published in the category: Methods

Conducting a multi-centre analysis on migration and health.

Migration has positive and negative consequences for the health of the migrant. Being involved in the production of a book reporting on this in different countries, we realized that in order to get a clear picture of the impacts of migration on health, a multi-site study should be conducted, with all sites using similarly defined variables and the same analyses techniques. continue reading

Reliability and measurement error

When teaching students about evaluating tests, I always said that 'Reliability means that repeating the same measurement in the same circumstances should result in the same findings (the findings are reproducible). Reliability is related to the precision of the instrument used for scientific observations.' However, how reliability and precision (or measurement error) actually different I was not totally sure about, until I read about this in the book Measurement in Medicine, authored by Riekie de Vet, my former PhD co-supervisor. In this article she explains this to all of you. continue reading

Challenges with using estimates when calculating ART need among adults in South Africa

In October 2012 an article with this title was published in the South African Medical Journal (1). I wrote this together with colleagues from the Foundation for Professional Development (FPD) and SACEMA, both clients of Epi Result. It describes a problem I experienced when using data published by Statistics South Africa (Stats SA) to calculate need for ART provision among adults in the City of Tshwane Metropolitan Municipality (CTMM). continue reading

Strategies for Choosing and Planning a Statistical Analysis

Recently I read a blog article entitled 'Strategies for Choosing and Planning a Statistical Analysis' which highlights a message that I always give to researchers that are writing their research proposal: Develop your analysis plan at this stage, otherwise you might encounter problems when you are starting the actual data analysis. However, that does not mean that it is easy to do that. So therefore I hope that this article, written by Karen Grace-Martin a colleague consultant who focuses on statistical consultancy, will be helpful to researchers in my network. continue reading

Unique approach to conveying research results

Recently I was alerted on the website "They go to die" featuring a film documentary on the phenomenon that gold miners in South Africa contract diseases (especially HIV TB co-infection) at their workplace. When a worker becomes sick at the mine, their illness deems them "unfit for work" and subsequently they are sent home to the rural areas. Since these areas often have little or no access to medication/care, this process is termed, "sending them home to die" I was taken by this project for two reasons. continue reading

Sample size calculation in cross-sectional studies

I often see people being a bit anxious when it comes to sample size calculations: I recently had a client having this problem when setting up a cross-sectional study. I here explain the solutions found to assist the client with this issue. Two different tools that are freely online available were used. continue reading

I publish articles in the following categories

Cases Epidemiology in the news Methods News