Tip:
Highlight text to annotate it
X
The study of statistics is commonly divided into two categories. We have what are known
as descriptive statistics and also inferential statistics.
Lets start first by discussing descriptive statistics. A good way of describing them
is that they're really just a method of organizing, analyzing, as well as presenting data in some
useful way. So let me give you an information on how descriptive statistics works and how
it would actually play out practically. One of the commons themes that we see is when
businesses issue financial reports. And so financial reports are a great example of descriptive
statistics because they're merely describing something. They're taking information, so
a company's financial performance, they're organizing it and of course they're presenting
it in a way that, ideally, the reader would be able to understand it. And they can draw
their own conclusions. But really its just designed to present the information in its
own right.
So something that might be describing a company's financial information could be showing income
from a couple periods of time. So maybe we would have income for the past three years.
So maybe, you known, 2011, 2012, and of course 2013. And then of course we have our individual
projections on how much money was generated in that particular period of time. So I'm
going to go ahead and fill in certain numbers here in thousandths. This would be an example
of descriptive statistics because we're merely organizing all of our financial information.
We're obviously presenting it in a way that is somewhat useful. And that it was much better
than if it was all expressed individually. So this is something that we would use for
descriptive statistics.
You can also do such things as determining the number of people with certain characteristics
or certain traits. That would be describing something. The key thing to think about when
looking at the difference between the two is one are we merely describing something,
or are we trying to generate assumptions based on information. And that's really where the
difference between the two lies.
Is with inferential statistics we're really trying to draw conclusions, we're trying to
make assumptions based on how a smaller group of individuals or objects behaves we're trying
to infer that the larger group would essentially behave the same way. So in order for this
to work, we need to have a smaller subset of individuals that ideally and hopefully
reflect the characteristics of a larger group. And so some terms that you've probably heard
of that are very important to inferential statistics include is known as a sample, and
also what is known as a population.
Now a sample is merely a subset. It's a smaller group of a larger group that is designed to
reflect those same characteristics. Population is the entire group in itself. So that the
idea here is that we would survey a smaller group of people and then basically draw conclusions
and assume that well if a smaller group of people behaves this way then probably the
population will behave in a similar fashion. So that really depends on the accuracy of
our sampling right. If we're going to select the right people because if you select the
wrong people than it's difficult to draw conclusions from.
So the way to look at this though is that the sample is really that smaller group of
the population. And we want to try to select a sample that mirrors the characteristics
if you will, like demographics and those things, of our population so that we can draw those
meaningful conclusions.
So the first thing that you're probably asking is there's going to be kind of these issues
with obtaining an accurate sample, which we'll talk more about over the course of this series,
then why don't we just continue to survey the population? Why don't we do something
like the U.S. census, which in idea is kind of a study of the population, not everybody
submits it so it's maybe more of a sample you might say. But why don't we do something
like that? The big determinant of not doing those types of things is of course the cost
of it. And you think about if you wanted to survey a population of consumers, like american
consumers, how costly would it be to try and track all of them down to complete this survey
and it would be very very time consuming. Not only that but very very costly from a
labor as well as a explicit cost standpoint. So that is one of the biggest reasons. It's
very hard, especially if you're a business trying to survey every one of your target
customers, to find all of them and to get them to interact with you in such a way that
they can complete the survey. Whereas if you use inferential statistics and you simply
kind of list out the traits and the commonalities and the characteristics of the people that
you want to find then you can go out and find a good amount of those. Enough so that you
can do those conclusions and you can draw those inferences.
So that's a little bit about descriptive statistics and kind of how they compare to inferential
statistics. We're going to talk more about those over the course of this series. So it's
important that we get kind of the purpose of both of them, but also how they are different
and how they can be used to present information in a useful way.
Thank you for watching this video on descriptive and inferential statistics. In the next video,
we're going to explore the types of variables that we commonly see in statistics. For questions
please leave them in the comment box below and I'll do my best to get back to them in
a timely fashion. If you want more from Alanis Business Academy, you can subscribe to our
channel, or go to alanisbusinessacademy.com where you can find additional content, quizzes,
and more. Thanks for watching.