Statistics are often used to support arguments, but how can you tell if they are accurate and relevant? In this article, you will learn some tips and techniques to evaluate the quality and credibility of statistics in an argument. You will also discover some common pitfalls and errors to avoid when using or interpreting statistics.
The first step to evaluate the accuracy of statistics in an argument is to check the source. Who collected and reported the data? What is their expertise and reputation? What is their purpose and motivation? How did they obtain and analyze the data? You should look for sources that are reliable, unbiased, transparent, and up-to-date. You should also verify if the source has been peer-reviewed, cited, or endorsed by other experts or organizations.
I would want to see other measurements of the data and plots before I get statistical results. Or at least commented on. Why do you want to do that test what are you trying to show etc.?
The second step to evaluate the accuracy of statistics in an argument is to check the sample. The sample is the group of individuals or items that the data is based on. How large and representative is the sample? How was the sample selected and divided? How was the data collected and measured? You should look for samples that are sufficient, random, and representative of the population of interest. You should also avoid samples that are biased, skewed, or cherry-picked.
I would still ask what are they trying to show before I questioned the sample. There are also lots of difference between the laboratory where the environment can be controlled and taking measurements outside and in the field.
The third step to evaluate the accuracy of statistics in an argument is to check the method. The method is the way the data is processed and presented. How was the data summarized and organized? What kind of statistics and graphs were used? How was the data interpreted and explained? You should look for methods that are appropriate, clear, and consistent. You should also avoid methods that are misleading, confusing, or inconsistent.
Do they understand any biases and assumptions from the methods, do they really understand what the result means? Are they using parametric data with a parametric test?
The fourth step to evaluate the accuracy of statistics in an argument is to check the context. The context is the background and situation that the data is related to. What is the scope and scale of the data? What are the assumptions and limitations of the data? What are the comparisons and contrasts of the data? You should look for context that is relevant, accurate, and complete. You should also avoid context that is irrelevant, inaccurate, or incomplete.
I would look at those things before this point. It really is too late now to decide on these problems. The problem is not in really as linearised as it is shown here going from box to box.
The fifth step to evaluate the accuracy of statistics in an argument is to check the logic. The logic is the reasoning and inference that the data is used to support. What is the main claim and evidence of the argument? What are the premises and conclusions of the argument? What are the warrants and qualifiers of the argument? You should look for logic that is valid, sound, and strong. You should also avoid logic that is invalid, unsound, or weak.
The sixth and final step to evaluate the accuracy of statistics in an argument is to check the counterarguments. The counterarguments are the alternative or opposing views that challenge the argument. What are the possible objections and refutations of the argument? What are the strengths and weaknesses of the counterarguments? How does the argument address and respond to the counterarguments? You should look for counterarguments that are fair, respectful, and constructive. You should also avoid counterarguments that are unfair, disrespectful, or destructive.