00:01
Which of these statements are true about missing data? so we have four statements here.
00:05
Let's have a look at them.
00:06
A.
00:07
If a few data points are missing, not at random, they can be safely ignored.
00:12
This is not true because it is not at random.
00:16
That indicates that you may be missing a cluster of points and that's going to bias whatever remains.
00:23
So you're losing a cluster of data.
00:29
So whatever is left is likely to be no longer representative.
00:34
What about b? this is also not true.
00:37
Now there are some data sets where it might be okay.
00:40
For example, if you have very symmetric data, then it might not matter.
00:43
But if your data is not symmetric, then using the mean introduces bias because it drags your data set in the direction of your outliers.
01:05
C...