Lesson video

Welcome to this Oak National Academy lesson, which is called Data Science, and it's taken from the unit Computer Systems and Data Science.

Thank you for joining me for this lesson today, and I look forward to learning along with you.

The outcome of today's lesson is, "I can describe data science and explain how data is used to make decisions." We have four keywords or phrases in this lesson.

So the first one is analyse, and analyse means to look at something carefully to understand it better.

Then we have data pattern.

A data pattern is a recurring trend, relationship or structure found within a set of data.

Next, we have data cleaning.

And data cleaning means fixing or removing mistakes in data to make it correct and ready to use.

And finally, we have data bias.

So data bias is when data is unfair or unbalanced, which can lead to wrong or unfair results.

So there's our four keywords or phrases.

Analyse, data pattern, data cleaning, and data bias.

There's two learning cycles in this lesson.

So the first learning cycle is to describe data science, and we'll start that in a minute.

And the second learning cycle is explain how data is used to make decisions.

So let's begin that first learning cycle, describe data science.

Sam has a question for Jun.

"Have you ever wondered how video streaming services seem to know what you like to watch?" And Jun says, "I think that's data science in action, Sam." So let's see what he means.

Data science is all about using data like numbers and facts, to solve problems and make smart decisions.

And it's a bit like being a detective.

So just like a detective will use data like numbers and facts to solve a case, data scientists collect, prepare, and analyse data to discover data patterns and answers.

Data science is used in the real world every day in lots of different ways.

Let's have a look at some examples.

So if Sam has an example, companies use data science to understand their customers, make smart business choices and save time and money.

So there's lots of uses of data science in that statement there.

And we'll look at some examples of those as we go through this lesson.

So here's the first example.

Professional sports teams use data science a lot.

So coaches and managers, as Sam points out, can track player statistics to improve player performance and choose the best team.

And they can do that across a whole multitude of sports.

So here we have basketball, football, and tennis.

So in team sports, they can monitor players to make sure they're contributing to the team and performing their role correctly.

In individual sports, they can look at technique, fitness, and all sorts of things.

There's so many different ways sports teams can use data science to improve performance.

Doctors and healthcare workers also use data science.

Doctors can diagnose and treat patients more accurately.

They can also track, predict, and prevent health problems. So this is a real way in which data science is making it a proper impact on people's lives and wellbeing.

There are many other examples of how data science is used in modern life.

And these include weather forecastings predict storms, rain, sunshine, using data from satellites.

Traffic apps use data to show the fastest route.

Social media uses data to organise your feed.

And online shopping uses data to recommend products and show you what's popular.

These are all really good examples of how data science is used in the real world.

Data science isn't magic.

It follows a clear process like solving a puzzle.

So let's have a look at that process now.

First of all, you ask a question, then you collect the data, then you clean the data, then you look for patterns, and finally you share what you find.

So we'll have a look at all those stages as we go through this lesson.

Before we do that, Laura has a question, but what exactly is data? Data is facts, numbers or details we can measure, count or observe.

You might not see it, but data is all around us in daily life.

Can you think of any examples? Here's just a few.

You could have come up with.

So, and there are so many more than this.

Here's some examples of some data.

So it could be something like favourite songs or favourite foods or steps per day or time exercised, shoe sizes or test scores.

So a real variety of different ways that data is around us every day.

And you probably came up with some different examples yourself.

So Laura has a question, "Is data the same as information?" And it's not.

They are closely related, but data and information are different.

Data is raw, unorganised facts and figures.

Information is processed, organised, and structured data that provides context and meaning as shown in the table below.

So we've got some data and some information side by side.

So on the left we've got the data, the raw facts of some test scores.

So we've got 85, 90, 72, and 88.

And the information meaning behind that is, could be something like this.

The average score is 83.

75, and most students did well on the test.

So we've got some survey results there, which says pizza 10, tacos 5, and salad 3.

We don't really know what that means until it's interpreted as information.

So this time it means that pizza is the most popular lunch choice on the menu.

Data is really perfect and can be messy, wrong, or missing important pieces.

So data cleaning is about fixing or removing mistakes in the data to make it correct and ready to use.

You might also remove unneeded data or fill in any missing pieces.

These are all parts of data cleaning.

When data scientists analyse data, they look for data patterns.

And a data pattern is a repeating trend or structure found within a set of data.

Here's an example, a data scientist may find a data pattern that more people buy ice cream in the summer months of Jun, July, and August.

Laura has a question, but what if the data is biassed and isn't fair? Data bias can be a problem.

Data bias is data that's unfair or unbalanced, which can lead to wrong or unfair results.

And it happens when the data we collect doesn't show the full picture.

So the results aren't fair or accurate.

Time for some questions.

So the first question, data is, is it A, never biassed, B, provides context and meaning, or C, raw unorganised facts and figures.

Well done.

That's right.

Data is raw, unorganised facts and figures.

Next question, which of the following as an example of data? Is it A, a news article, B, a list of temperatures over a week, or C, a social media video clip.

Well done.

It's B, a list of temperatures over a week.

That's an example of raw unorganised data.

And our next question information is, is it A, structured data that provides context and meaning? B, recording the maximum daily temperature for a week? Or C, a list of vehicle speeds passing a speed camera? And well done, information is A, structured data that provides context and meaning.

Time for your task.

Andeep has a question.

What's data science? So two parts to this task.

First, in your own words, describe what data science is to Andeep and then describe the difference between data and information.

So here's what you might have written for the first part, data science is about using data like numbers and facts to understand things and solve problems. It's kind of like being a detective because you collect data, clean it up and look for patterns to find answers and solve problems. Analysing data can lead to new and useful information that can be used by lots of different industries such as healthcare, business, and sports.

And for the second part, describe the difference between data and information.

So data and information are closely related, but they are different.

Data is raw, unorganised facts and figures, whereas information is processed, organised, and structured data that provides context and meaning.

Information is what we learn after we look at the data and understand what it means.

We can now move on to the second learning cycle, which is to explain how data is used to make decisions.

Sam has a question for Jun.

How did the school decide to change the lunch menu? And Jun says they looked at the data from the survey the students did last week.

90% said they wanted more wraps instead of sandwiches.

In this way, data science can help you make smart informed decisions.

Every day we make decisions using data from what to eat, what to wear, and what to watch.

Here's a really simple example.

You might check the weather before choosing your clothes.

If it's gonna be sunny, you'll wear something different to if it's gonna be rainy.

You might look at reviews or ratings to decide where to go on holiday.

In this way, data helps us choose wisely, not just randomly.

And data-driven decision making means using facts and data to guide the choices we make.

And it means you don't have to guess.

You can look at meaningful information before deciding what you're going to do.

Data-driven decision making helps us make smarter, fairer, and more accurate decisions.

When we use data to make decisions, we're more likely to make the best choice.

The data can help us compare options, spot patterns, and make decisions that are fair and based on facts.

Even with data, it might not be possible to make a perfect decision, but using data can help us feel more confident about the choices we make.

Time for a true or false question.

Data-driven decision making means using facts and data to guide the choices we make.

Is that true or false? Well done.

That's true.

So Jun has an idea of how we could use data to help us plan the end-of-term party at school.

And Sam agrees.

"That's a great idea!" So let's have a look at how they might do that.

What data do you think Jun and Sam should collect to help them make their decisions? So here's an idea from Jun.

We could collect data about what people's favourite songs are to help us choose the playlist for the party.

Sam says, "We could collect data about what food people would prefer to eat at the party." And then Jun says, "We might also need data to decide what time people would like the party to start." And Jun adds, "We need to make sure our party data is not biassed." How would you try to reduce the bias in the data? So Sam says, "We could make sure we collect the data from all students who will go to the party." That's a great way of reducing bias.

If you are asking all the students, you're gonna get a fair representative view of everybody who's going to be at the party.

So Jun and Sam collected data from all 25 students who are going to the party, and we're gonna have a look at the results in the table below.

So first of all, for music for the playlist, five people said they liked rock.

15 said pop, and five said dance.

For the food, it was a little bit clearer.

So 20 people said burger was their favourite.

Two people said hot dogs and three people said pizza.

And then for the start time, two people wanted an early start time at 3:30, six people said 4:30 PM and most people 17 said 5:00 PM.

So from that data, they've made some decisions.

Jun said 60% of students going to the party preferred pop music.

So Sam thinks we should mostly play pop songs with a few rock and dance songs to suit most people, which sounds like a pretty good compromise.

Sam says, "80% of all students going to the party said they would prefer to have burgers." Jun said, "That's really useful information and it's helped us decide which type of food to order." This is a little example where the data might hide a few things.

So for example, if the people that didn't say they wanted burgers may be vegetarians, you may want to think about framing that question slightly differently or using the results slightly differently.

So it doesn't always tell us everything.

So there's our results; 80% of burgers, 12% of pizza, and 8% of the hot dogs.

And finally, the data shows that 68% of people would prefer a 5:00 PM start time.

Sam says, "If that's the most popular choice, we should probably decide to start the party at 5:00 PM" which seems like a good choice.

And finally, Jun says, "I hope everyone enjoys the party." And Sam agreed.

Making decisions with the data doesn't mean you will always get it exactly right, but it does give you a better chance.

Even when choices are difficult, using data makes the process fairer and more balanced.

Let's have a look at a question.

Using data to help make decisions helps because A, you make random guesses.

B, you make smart and fair choices, or C, you ignore facts.

Well done.

That's right.

Using data to help make decisions helps because you make smart and fair choices.

Next question, bias data can be a problem because A, it includes too much information.

B, it always gives the right answer, or C can lead to unfair or wrong decisions.

That's right.

Well done.

It can lead to unfair or wrong decisions.

And now time for your task.

Choose one real life situation from the list below and explain how data could help someone make a smart decision in that situation.

And we've got three example situations.

You just need to pick one of them.

So they are, a local library wants to buy more books.

A family is choosing where to go on a weekend trip or a student wants to improve their study habits.

Choose one of those situations and explain how data could help someone make a smart decision.

If you chose the local library scenario, here's what you may have written.

The library could analyse borrowing records about which genre of books have been borrowed the most, like mystery, fantasy, or graphic novels.

They could also collect some new data by completing a short survey and asking people what kinds of books they want more of.

Analysing the data will give the library useful information that would help them make smarter decisions on what books might be popular if they bought them.

Here's a summary of this lesson, data science.

Data science is about using data like numbers and facts to solve problems and make smart decisions.

Data scientists analyse data to find data patterns and discover useful information.

Data cleaning is used to fix or remove mistakes in data to make it correct and ready to use.

Data can be biassed if it doesn't represent everyone fairly.

Thank you very much for joining me for this Oak National Academy lesson.

I hope you enjoyed learning along with me today, and I look forward to seeing you again for another lesson in the future.