Statistics distributions
Tutorials
statistics howto (thorough until uni) https://www.statisticshowto.com/probability-and-statistics/ StatQuest Youtube Channel (percentiles, normalization, etc)
Histograms
A Bar graph to reveal the distribution of the data.
Info is clumped in equal size chunks of your data.
| | Frequency | | of | | ${X} | | | | | | | | | | | +---------------------- 1-8 9-16 17-24There are different measurements of frequency
# (ex. 2x yr1 students, 3x yr2 students) cumulative: each group includes the previous group's results. (yr1==2, yr2==5) relative: each group is mutually exclusive (yr1==2, yr2==3)
Distribution
A distribution describes the shape of a histogram.
There are various named classifications of distributions, that describe a shape in your chart: Some examples:Normal Distribution | | | | | | | | | | | +-+-+-+-+Skewed Right
| | | | | | | | | | +-+-+-+-+Bimodal
| | | | | | | | | | | +-+-+-+-+Outliers are data that is largely unrelated to the majority.
| | | | | | | | | | +----+-+-+-----+ ^ ^ +--------------+ (outliers)
Quantiles
If you arrange your data as a distribution, sorted by frequency,
Quantiles can be used to extract information based on the frequency, and eliminate anomalies from the data.| | Frequency | | of | | | ${X} | | | | | | | | | | | | | | | | | | | | | +-+-+-+-+-+-+-+ 1 4 3 5 7 6 2 ^ medianThe median is an example of a quantile.
50% of values were more frequent, and 50% were less frequent.
We can say the median (50th percentile) value was5
.Similarly, we can say the value of the 85% percentile's value is 6.