Inferential statistics is a way of guessing about things based on what we observe. Imagine you and your friend are trying to guess how many blueberries are in a jar. You can look and make an estimation based on what you see, but you can't know for sure unless you count them all. This is like descriptive statistics, where we describe what we see in the world around us.
But sometimes we don't have the ability or resources to count everything, especially if we're dealing with very large or complex systems. So what inferential statistics does is take a small sample of what we're interested in - like counting only some of the blueberries in the jar - and then make an educated guess about the rest.
It's kind of like playing a guessing game. Let's say you counted 10 blueberries in the jar and your friend counted 12. You both have different numbers, but you can use those numbers to make a guess about how many blueberries are really in the jar.
Now, people have come up with really special ways to do this guessing accurately. They use fancy math and formulas to make sure that their guesses are as close to the real answer as possible. They also try to make sure their guesses are not just lucky guesses, but are based on scientifically sound methods. That way, if they can't count everything, they can still make pretty good guesses.
And that's inferential statistics - it's a way of making educated guesses about things based on a small sample, using math and science to help us make good guesses that are as close to the real answer as possible.