Careful with The Statistics and Graphs

Statistics and graphs can be misleading, unfortunately, the person making the graphs or writing the figures can convey more misinformation than information depending on the intent. We as readers of such data therefore must be careful in how we interpret the information presented to us.

Broadly, it is always good to check out some aspects before believing any inferences presented with “supporting statistics”:-

1. If two sets of data are presented as two different graphs, the first thing one should check is whether the axis of both the graphs is same or not. Comparing two graphs with two different axes values is like comparing apples to oranges.
2. One should also check whether the two sets of data are from the same time period or are in context, that is if we are comparing the milk production of two different countries, we need to see same time periods.
3. If one wants to interpret the results of a survey one should find out the questions or the set of questions asked from the target population, the composition of the target population, number of samples collected, the probability, the confidence interval of the conclusion. All good surveys will mention these figures in the end of the survey results.
4. Lastly, one must understand that if two things are correlated, this does not in any way mean that one is causing the other. There are so many correlations in daily life, cold weather and incidents of common cold have a high correlation, but cold weather does not cause common cold, which is caused by failure to protect oneself adequately during cold weather.

One should therefore be careful while drawing conclusions after seeing statistics and graphs. Sanne Blauw, who calls herself a number nerd, talks about how to be defend oneself from misleading statistics in the news in her youtube video.

My Random Thoughts