Lesson video

Hi, my name is Chloe and I'm a geography field studies tutor.

This lesson is called Interpreting Graphs in Geography, and it's part of the geographical skills unit of work.

It's all about how we can read graphs and find all those little tricks to make us understand them in greater detail.

By the end of today's lesson, you'll be able to interpret data from a wide variety of graphs.

There's some key words to review first of all.

Bivariate data is data for two variables where one of those variables is believed to influence the other.

An anomaly is a value within the dataset that does not appear to follow the general pattern or trend.

Interpolation is estimating unknown values within a data set based on other values within the set, and extrapolation is estimating unknown values beyond the data set based on the pattern or the trend of the known values.

This lesson is in three parts.

First, how do geographers interpret a graph of data? Then how do geographers use graphs to predict data? And finally, we're going to look at how geographers interpret more complex graphs.

Let's begin with that first one about how geographers interpret a graph of data.

Interpreting data is all about recognising what that data tells geographers about the feature they're investigating.

It's likely to involve describing the data, saying what it looks like, and comparing the data, finding similarities and differences between sets of data.

There might also be correlations within the data, seeing how change in one data set affects another.

It may also involve explanation where geographers link the data to known geographical theories and ideas.

Certainly the first three of those points though are the most common.

When presented with graphical data, geographers first look at the structure of the graph.

This means identifying what type of graph has been used, and as Lucas rightly points out here, this is a line chart.

You also need to read the title and understand what the graph is trying to show.

Lucas says the line chart is showing the number of people living in rural and urban areas in the UK.

You also need to read the key.

There are two lines, blue for urban inhabitants and red for rural inhabitants.

You then look at the axis to find the variables being used.

As Lucas points out, the x axis shows time and the Y axis shows the number of inhabitants.

You could also look at the scale used or the extent of the values and their units.

Inhabitants have been measured in millions of people with the Y axis starting at zero.

Time has been measured in years and covers the time period from 1960 to 2022.

Lucas has broken down everything there is to break down about this graph, so the structure is all covered now.

Let's check our understanding of that.

Which of the following is least likely to form part of a geographer's initial interpretation of a graph? The initial interpretation.

Reading the title to understand the graph's purpose, looking at the key to understand different elements of the data.

Looking at the axes to see the variables being examined or explaining the data by linking it to known geographical theories.

Pause the video and have a think about all the things that Lucas just said about that line chart and then come back to me with the right answer.

So well done if you recognise that explaining the data would not form part of an initial interpretation of a graph, the initial interpretation is all about the graph's structure, how it's been drawn.

Geographers will then start to look more closely at the data itself.

Depending on the nature of the graph, they may try to identify general patterns in the data, maybe by drawing a line of best fit as we've got here.

They might look at the highest and lowest values in the data set.

They might look at how the data compares to the average value, or they might try to identify an outlier or data that falls outside the general trend.

GRaDE is a useful acronym that geographers use to examine graphical data more closely.

G is for general trend.

It describes the general pattern that we can see in the data.

Ra stands for the range of data describing the amount of data collected and the minimum and maximum values.

D is for a data quotation, quoting individual elements of the data, and E is for exceptions, highlighting any data that does not appear to follow the general trend.

Read Laura's initial interpretation of this graph.

What else could she discuss? Let's see what Laura says.

Generally, the noise level decreases as one moves away from the town square.

The greatest drop occurs between 100 and 200 metres away from the town square.

An exception to the trend is at 500 metres where it rises slightly from 34 decibels to 38 decibels.

What else could she discuss? Could she discuss the range of data, the ranking of the data, the rate of change in the data or the rarity of the data? Pause the video, remember the acronym.

See if you can find the right answer.

So what do you think the answer is? Hopefully yes, it's the range of data.

That's what the Ra stands for in grade.

Laura, talk about the range of the decibels or the range of the distance from the town square.

Our first task of this lesson, look carefully at the graph and write one sentence for each part of an interpretation of the graph's structure.

So very important, only the structure is what you need to do there.

Secondly, read Sam's initial interpretation of the data in the graph.

Use highlighter pens if you've got them, or you can underline in different colours or you could use codes to identify each stage of the GRaDE acronym.

And lemme tell you now, it might not be in the order that the GRaDE acronym is written in.

Pause the video, have a go at both tasks and then I'll show you my answers.

Right, let's go back to that first task.

So we're looking to write an interpretation of the graph, by talking about its structure.

So we need one sentence for each part of it.

Your answer may include something like this.

This is a line chart identifying the type of chart is, it shows how the gradient changes two rivers as one moves away from the source that's highlighting something from the title.

The rivers are shown by two coloured lines, purple for river A and pink for river B.

So there's an interpretation of the key there.

The x axis shows distance in kilometres from zero to 10 kilometres, and the Y axis shows gradient in degrees from three to 15 degrees.

So there's a lot in that final sentence.

It's talking about what data is on each axis, what the values are and what the units are.

Hopefully you manage to find all of those things in your answer.

Then we've got the second part where we've got Sam's interpretation of the graph using the GRaDE acronym.

How can you highlight each area here? Well, let's look at the general trend, first of all.

There it is.

In both rivers the gradient decreases as one moves away from the source.

That is the general pattern within that interpretation.

Then we've got the range of data that comes right at the start of Sam's quote.

"The gradients are compared over a 10 kilometre section of the rivers." The range is mentioned there, a quotation of data.

The lowest gradient for either river is 4.

7 degrees.

Nice little quote.

And finally an exception.

River B seems to drop gently before the 10 kilometre point, but then sees a dramatic drop in gradient.

It's talking about how it's appearing to have one pattern, but then something slightly changes.

Hopefully, he managed to get all of those four parts of the grade acronym highlighted on Sam's quote.

Let's move on to the second part of the lesson.

We're looking at how geographers use graphs to predict data now.

Now, some graphs are based on by variant data.

This is where from two variables is paired against each other.

We've got an example here.

Flow velocity of river has been paired against river channel depth.

Interpretation of bivariate data centres around understanding the nature of any relationship between the variables.

And this means really examining the scatter graphs in greater detail.

They focus on the direction of the trend line, which is also known as the line of best fit.

A trend line is a straight line that runs through most of the plots and indicates the nature of the correlation.

So here you would have a positive correlation.

If the line is going in the opposite direction, it's a negative correlation.

But if the points are so randomly spaced out that it's impossible to put a line through them, we would say it has no correlation.

The closer the pots are to the trend line, the stronger the correlation.

This means that one can predict exactly how one variable will influence the other.

A trend line also makes it possible to identify anomalies as these sit away from the trend line.

One there, for example, a piece of data can be an anomaly for a variety of reasons.

It may be that the data is an outlier due to human error in the collection of the data.

However, it is far more likely that a third variable that has not been measured has influenced this data point.

So true or false, an anomaly in a dataset is always due to human error in the collection of data.

Is that true or false? Pause video and have a think.

But hopefully you've seen that that is false.

But why is that? Well, human error may be the reason the data appears to be anomalous.

However, it's far more likely due to the influence of an unmeasured geographical variable and that's had an influence on the data and caused it to come away from the trend line.

Geographers use the trend line in a scatter graph to predict unmeasured values.

So Alex here wants to know the likely velocity of the river when the channel is at 0.

2 metres deep.

You can see we've got a lot of plots on the graph, but we don't have anything that represents 0.

2 metre.

How could he work this out? He can use the trend line to help him.

He says, I can predict that the velocity will be 0.

35 metres per second.

So how is he gonna do that? He traces up to the trend line from the 0.

2 metre mark.

Then you can use the trend line to read the corresponding velocity.

You can see how he's predicted it across to the Y axis there to about 0.

35.

This type of prediction is known as interpolation.

If geographers want to predict a value beyond the current range of data, they use a technique called extrapolation.

The trend line is extended using a dotted line.

Remember, a dotted line is all about projected data or estimated data.

It's not data that is actually known.

Additional values can then be read using the extended line.

So Sofia would like to know the likely velocity of the river when the channel is 0.

6 metres deep.

And you can see that goes beyond the range of data that we currently have.

What is her predicted value? Can she work it out the same way that Alex worked out his? So she's gonna come up from the 0.

6 metre line, go across to the Y axis, and you can see that the predicted value is 0.

77 metres per second or thereabouts.

Let's check her understanding of that.

Complete the sentences with the missing words.

Pause the videos.

You can have a look through the paragraph and come back to me shortly, right? Let's see what you got.

Geographers use a trend line to predict values that have not been measured.

A value can be interpolated when it sits between two known values.

It can be extrapolated when it sits beyond the range of values in the dataset.

So our second practise task of the lesson, use the graph to complete the values in the table.

You can see this is going to be about interpolation and extrapolation.

We've got four values in the table already.

You might need to work very closely with the graph to work out the other four.

So pause the video so you can really examine the graph and then come back to me with your answers.

Let's take a look at your answers.

Now you can see I've drawn a dotted line on my graph so I can work with the extrapolated data.

My two kilometres is paired with 120 millimetre bed load size by 4.

5 kilometres is with a hundred millimetres, 7.

5 kilometres pairs with 75 millimetres and 10 kilometres is with 55 millimetres.

I hope you're pretty close to those values.

We're onto our third and final part of the lesson.

How do geographers interpret more complex graphs? More complex graphs may show more than two variables.

This means that the graphs are unlikely to be drawn using a typical X and Y axis structure that we're very used to.

They're gonna look slightly different.

A complex graph may also show more than one aspect of the data on the same axis, and this means that the plots on the graph may look a little different.

So let's look at some examples here.

Now, climate graph is an example of a complex graph.

You might have come across these before.

It shows three variables.

The first is time, and this is shown as months of the year, and it's read off the x-axis.

In most cases, it will simply be written as the initial letter of each of the months.

So you can see them there ranging from January to December.

Temperature is shown as a line.

It's often red and it's read off the left hand y axis.

And you can see there I've got my temperature labelled on that axis in degrees Celsius.

I then have precipitation and this has got a second y-axis on the right hand side with a different scale.

And you can see it's been labelled up clearly with millimetres, but my precipitation is not with a line this time, it's with bars and it's very particularly with histogram bars and that's important because we're dealing with continuous data there.

A population pyramid also shows three variables.

First variable is age, and this is shown in normally year groups.

It's read off the Y axis that runs through the centre of the pyramid.

The percentage of males in the population is shown as histogram bars and it's read off the left hand axis and the percentage of females in the population is shown equally, its histogram bars and that's read off the right hand axis.

In this case, it's used percentages.

It might be that you have actual real data as your Y axis values.

Either way it's going to be that way round.

A triangular graph also uses three variables, but it presents 'em in a very different way.

And these are read off three different axis lines.

So here we've got an X, a Y and a Z, and these lines form together a triangle.

So you've got an X axis and you can see it's running from zero to 100 here.

So it's almost always going to be used with percentages.

Your Y axis runs along the base and your Z axis runs along the left hand side of the triangle.

Employment structured data commonly uses triangular graphs.

Locations are plotted onto the graph according to their structure.

So you see where the blue dot is there.

We've got, let's look at our three axes.

We've got extractive industries, so things like mining and agriculture is on our X axis, manufacturing industries on our Y axis and the services industries on our Z axis.

So this location has 35% employment in the extractive industries.

Look where the dotted line is running through the axes to the extractive industries.

50% employment in manufacturing industries.

Again, look at the exact position of that dotted line and 15% in the service industries.

So the position of that dot tells us all that information by using the triangular graph.

Let's check our understanding of triangular graphs now.

What percentage employment in the services industries is represented by the plot in this triangular graph? So same graph set up, but the dot is in a different place now.

Is it 62%, 80% or 18%? You might want to go back to the previous slide so you can actually check again which axes you are reading in which direction, but pause the video and then come back to me with your answer.

Right, this is not easy.

Very common that you might get this wrong the first time.

The actual answer is 62%.

Look at how the line is going straight across from the services industry.

And so, you are reading it off that left hand axis, 62%.

A radial graph can show many variables.

Each variable is plotted on its own Y-axis, so there as many Y-axis drawn as are needed for the number of variables.

There's probably going to be a limit.

If you have too many, it's going to be quite difficult to read, but as it is normally, yeah, you can have as many as you'd like.

Environmental scores could be plotted onto the actual axes for two different locations.

You can see here I've got a pink line and I've got a blue line that's representing two different places.

And the scores that they got in these different categories, either a one, a two or a three.

With the points joined, these create comparable polygons.

Now you can see in general, the blue polygon is much larger than the pink one.

And this shows in general, whatever the place is that is represented by the blue line, it is scoring high more highly on those environmental scores.

Our pink line, you can see some of the values where it's not scoring very well.

Tranquillity, it only gets a value of one.

Naturalness, it only gets a value of one as well.

But equally, it scores well in greenery and cleanliness, both scoring three in those categories.

In this example, there are eight variables, one for each of the cardinal and in cardinal directions.

So there are eight y-axis radiating out from the centre you can see here.

The number of days in a year when wind blows from each of the directions has been plotted on the axis.

And this type of radial graph is known as a wind rose.

So let's just take a look at the graph here.

You can see that we've got 20, 40, 60, 80, 100.

That's our 100 is our top score.

The number of days where the wind is blowing from that direction.

So we can see that mostly the wind is coming from the southwest.

It's probably coming least from the east.

And then the others, it's roughly unequal.

So as Andy suggests, on 90 days of the year, the wind blew from the southwest.

He's using the radial graft, read some quite precise data.

A box and whisker plot does not show multiple variables, unlike all the ones we've seen so far.

But it does show different aspects of a whole dataset all at the same time on one set of axis.

So the top and bottom of what we call the whiskers show the minimum at the bottom and the maximum values in the dataset.

The median value of the dataset is also given its own line.

The top of the box sits on the value of the upper quartile, the median of the upper half of the values in the dataset.

And the bottom of the box sits on the value of the lower quartile, and that's the median of the lower half of values in the dataset.

Align joins the values together simply so it all looks nice and neat.

The box represents the inter quartile range for a dataset.

So the smaller the box, the more clustered the data is around the median value when one considers the dataset as a whole.

With that in mind, what is the lower quartile value of this dataset? Let's take a look at our graph here.

Is it 2.

5, 5, 10 or 15? Pause the video and have a think.

Maybe have a chat with somebody nearby as well to see what their ideas are and then come back to me.

It's not easy this, but well done if you've got that, it's five.

So it's the lower quartile value is the represented by the bottom of the box, not the bottom whisker.

That's the minimum value.

So the bottom of the box is what we're looking for there.

And if you look on the graph, you can see that has a value of five.

Our final task, Aisha has drawn two box and whisker plots to show the average daily wind speed data for February and August.

Which month has the greater range of daily wind speed values? And how do you know this? So the box in whisker needs to be looked at quite carefully there.

Secondly, in which month were the values in the data set closest to the median.

And how do you know this? Again, look carefully.

I'm gonna let you pause the video here because it's gonna take a little bit of time to have a think about this then come back to me.

Right, let's look at that first idea.

Which month has the greatest range of daily wind speed values? Well done If you worked out that it was February.

Now, how did you work this out? It's because of the whiskers.

So the whiskers on the February graph are at a greater distance apart.

The minimum and maximum values show the greatest range, far larger than it is on the August graph.

But in which month were the values in the dataset closest to the median? Yes, August.

But how do you know this? It's because the box on the August graph is much smaller than the box on the February graph.

Let's summarise our ideas.

Geographers initially interpret graphs by looking at the structure of the graph.

And for simple patterns in the data, bivariate data graphs require geographers to understand how trend lines show the direction and strength of correlation as well as predict unknown values within and beyond the dataset.

More complex graphs require geographers to read the values of multiple variables from more than two axes, or to interpret different aspects of data from a single data set.

Some of those concepts were quite complex, weren't they? It would definitely be worth looking at examples of each of the graphs that we had to in today's lesson to really get a feel for how they are used by geographers.

Best of luck.