How to read scientific research

Tim Huntley who sometimes writes for Robb Wolf passed on a link to a great piece has has done on how to read scientific research: it is written for a person who is new to science but would like to be able to understand the structure and primary concepts in a research article.

It is a great resource - thanks Tim!

Scientific Research 101

You might be a student with a homework assignment, someone who is searching for answers to a health question, or someone who is intrigued by a research study mentioned in the newspaper or on television.    Honestly there could be a thousand reasons why you want to lean how to read scientific research.

Independent of your age or education, you have the capability to understand scientific research studies.   I’m not suggesting that everyone should become a research scientist, but with a reasonable amount of effort, you can read, evaluate, and utilize the researcher’s work.  You simply need a roadmap to guide you along the way, and that is the purpose of this article.

The section on Bad Science is really useful!  For example:

Absolute vs. Relative Percentages
Suppose that there was a medical problem that caused 2 people  in 1,000,000 to have a stroke, and suppose there was a treatment that would reduce the problem to only 1 person per 1,000,000.  This would be an improvement of 0.0001% in an absolute sense or NO BIG DEAL.  However had I reported the results using relative percentages, I could have stated:  ”New medical treatment yields a 50% reduction in risk of stroke.”  This would obviously be quite misleading, but it is a common practice.

Why Most Published Research Findings Are False.

Following on from The future: I just saw it, I thought you might find this 2005 study by John P. A. Ioannidis of interest.

"Summary

There is increasing concern that most current published research findings are false. The probability that a research claim is true may depend on study power and bias, the number of other studies on the same question, and, importantly, the ratio of true to no relationships among the relationships probed in each scientific field. In this framework, a research finding is less likely to be true when the studies conducted in a field are smaller; when effect sizes are smaller; when there is a greater number and lesser preselection of tested relationships; where there is greater flexibility in designs, definitions, outcomes, and analytical modes; when there is greater financial and other interest and prejudice; and when more teams are involved in a scientific field in chase of statistical significance. Simulations show that for most study designs and settings, it is more likely for a research claim to be false than true. Moreover, for many current scientific fields, claimed research findings may often be simply accurate measures of the prevailing bias. In this essay, I discuss the implications of these problems for the conduct and interpretation of research."

For more on this, see Who pays the piper and Who pays the piper part 2.

Research shows.....

....that people who eat a low-fat breakfast like Special K are more likely to be slimmer than those who don't.

So states the Special K advert. I wonder whether the researchers used a control group who ate a high-fat breakfast? Somehow, I doubt it. I eat a high-fat breakfast, if you look at the Blog post immediately below this one and I am slimmer than when I didn't eat breakfast. Skipping breakfast causes low blood glucose, which encourages over-eating later-on. Therefore, people who eat breakfast are more likely to be slimmer than those who don't.

That's the problem with "Research". It usually shows what people want it to show. By omitting control groups, such research is utterly worthless. While I'm on a mini-rant, "Clinically proven" is another phrase often bandied about in advertisements. I have news for you. Clinical studies can only disprove something. The best that a clinical study can do is provide evidence that something is beneficial to some or most of the people in the study only. It cannot be extrapolated to the rest of the population.