This week I’m learning from a book called “Hypothesis Testing, an Intuitive Guide for Making Data Driven Decisions” by Jim Frost, MS.
While learning through this book, I figure I will do a couple of posts with some basic concepts I find useful to know at a fundamental level.
one of which is….
♦ What is a Hypothesis Test?
✨ It is an inferential statistical procedure that evaluates (ie. tests) for two mutually exclusive statements about a population.
These are the 👉 null hypothesis and 👉the alternative hypothesis.
✨ And determines which statement the data supports while using the evidence in samples to make inferences about the characteristics of the population.
To take it further, rejecting the null hypothesis means the results are statistically significant for the data to support the theory proposed is supported by the data towards the greater population level.
*** 👆 that is the simplest way to boil it down from the book
Now… since Hypothesis Testing is a type of Inferential Statistics….
♦ What is inferential Statistics?
This type of statistics becomes important where we need to draw conclusion about a population from a randomized sample.
In other words, it’s about generalizing the sample results to people outside the sample.
✨ And with this there are some considerations to be aware of:
🔘 the confidence that the sample gets as close as possible to reflecting the population, with the awareness there is always a level of sampling error (which is the difference between the sample statistic and the population value)
“Sampling error blurs the line between real effects and random variations caused by sampling.”
🔘 population parameters are most often unknown, a sampling error is almost never known in such cases, unless you can determine the population parameters
🔘 hypothesis tests make assumptions about the data collection process (ie. That the method used tends to produce representative samples), therefore choosing the sample method is important.
🔘 it’s most often wanted to have unbiased estimates of parameters (ie. Not too high or too low) because they are correct on average for a larger population)…
Check out what others may have said about this subject on Linkedin