lesson

Updated 6 days ago Β· 1 view
Imagine two gaming companies both advertise that their "average" player spends 5 hours a day playing.
At Company A, almost everyone plays around 5 hours; at Company B, most people play for 1 hour while a few streamers play for 20 hours straight.
Single numbers can hide the truth, so how do we inspect the full picture of our data?
Distribution Shapes
The shape of a distribution tells us how data points are spread out across a number line.
πInteractive diagram
A distribution is skewed in the direction of its long tail, not where its tall peak sits.
In a symmetrical distribution (such as bell-shaped or uniform), data is balanced evenly on both sides, meaning meanβmedian.
A positively skewed (skewed right) distribution has a long tail stretching to the right, which pulls the mean>median.
A negatively skewed (skewed left) distribution has a long tail stretching to the left, dragging the mean<median.
In a symmetrical distribution (such as bell-shaped or uniform data), both halves mirror each other, making meanβmedian.
A positively skewed (skewed right) distribution has a long tail to the right, pulling the average upward so that mean>median.
A negatively skewed (skewed left) distribution has a long tail to the left, dragging the average downward so that mean<median.