ScholaFly

GE04-05 Geography Watch

Spotting weaknesses in how data is presented

Watch on YouTube

In this lesson

Spotting weaknesses: how to spot and explain the weaknesses in how data are presented, with worked examples and the mistakes examiners report. AQA 8035, Edexcel 1GA0, Edexcel 1GB0, Eduqas C111, Eduqas C112, OCR J383 and OCR J384 GCSE. Graphs 5 of 5. More at https://scholafly.com

What it covers

  1. 0:00 Intro: spotting weaknesses: graphs
  2. 0:50 Spotting weaknesses: graphs: the cut axis
  3. 4:06 Other selective tricks
  4. 6:13 How a graph is built
  5. 8:37 Scatter graph limits

About this video

Spotting weaknesses in how data is presented | GCSE Geography | GE04-05

Spotting weaknesses: how to spot and explain the weaknesses in how data are presented, with worked examples and the mistakes examiners report.
AQA 8035, Edexcel 1GA0, Edexcel 1GB0, Eduqas C111, Eduqas C112, OCR J383 and OCR J384 GCSE. Graphs 5 of 5.
More at https://scholafly.com

Watch first: GE04-02 - https://youtu.be/pPgYdrEWZ0M, GE04-04 - https://youtu.be/x5eCb37nQRk
Video code: GE04-05 (search "ScholaFly GE04-05")

#GCSEGeography #Graphs #SpottingWeaknessesGraphs

For teachers
This GCSE Geography lesson teaches spotting weaknesses in how data is presented. By the end, students should be able to spot and explain the weaknesses in how data are presented - a cut axis, cherry-picked years, a small or unequal sample, an average that hides the spread, uneven or overlarge classes, missing labels or units - develop each weakness with its consequence, suggest a specific improvement and say how it helps (OCR A and B print this), and give the limits of a scatter graph (Eduqas A and B name it). It works through three worked examples and the mistakes examiners report.

Exam board specification references:
AQA 8035
- 3.4.4
Edexcel 1GA0
- MS:Statistical:4
Edexcel 1GB0
- MS-Stat.4
Eduqas C111
- AppA:2.4
Eduqas C112
- AppA:2.4
OCR J383
- 3.3(12)
- 3.4(3)
- 5d[statistical]-b4
OCR J384
- 5d[Statistical skills][4]
- 9.3.12
- 9.4.3

Read the transcript

Can a graph made of true numbers still tell a false story? Between the twenty eleven and twenty twenty-one censuses, the population of England and Wales grew by six point three per cent, says the Office for National Statistics. It went from just over fifty-six million people to nearly fifty-nine point six million. Draw those two numbers as bars, but start the side axis at fifty-five million instead of nought. The twenty twenty-one bar comes out more than four times as tall as the twenty eleven bar. The numbers are true, and the picture shouts population explosion. Catching a graph doing that, and saying what it does to the reader, is a skill you can learn.

Selective presentation means showing true data in a way that bends the story it tells. A weakness in a presentation is anything that makes it mislead, hide something, or be hard to read accurately. The first trick is a cut axis, one that starts above nought. The twenty eleven figure is about fifty-six point one million, so above an axis starting at fifty-five million, its bar shows only about one point one million. The twenty twenty-one figure is about fifty-nine point six million, so its bar shows about four point six million. Four point six million against one point one: about how many times as tall? Just over four times, because four lots of one point one is four point four, and four point six is a little more. Start the axis at nought, and the two bars differ by only six point three per cent, so they look nearly the same height. The numbers are true either way. The cut axis bends the picture, which is why the first look at any graph goes to where its axis starts. Naming a weakness is the first step. Developing it means adding its consequence for the reader, joined with so. Three students criticise the census graph. Answer A says the graph is not clear. Answer B says the axis starts at fifty-five million, and the axis does not begin at nought. Answer C says the axis starts at fifty-five million, so the newer bar looks more than four times as tall, though the rise was only six point three per cent. Which answer develops its weakness: A, B or C? C. It names the weakness, then spells out its effect on the reader: a small increase appears huge. A names no weakness at all. Now, B does name the cut axis. Why does it still fall short? Because B says the same thing twice in different words. It never says what the cut axis does to the reader. One examiner's report records many generic answers, such as it's not clear, or not easy to read, and says these are not specific enough to gain marks. The same report adds that some students are becoming increasingly confident in criticising statistical techniques. Another report names a common error: repeating the first point in slightly different wording, rather than extending it. The fix is to give each weakness one so, and make the so say what changes for the reader.

A cut axis is one trick, and true data can mislead in other ways too. Start and end years can be picked to flatter the story, and a very short period can hide the longer trend. A small sample can mislead, and so can an unequal one, such as many rich countries and only a few poor ones. An average can hide the spread, so the extremes vanish. A percentage shown without its total can hide how few people it describes. A total shown without a percentage can hide how small a share it is. In the fictional land of Owlobia, Featherby council's leaflet shows its recycling rate as two bars, with invented figures. It was forty-two per cent in twenty nineteen and forty-eight per cent in twenty twenty-four. The side axis runs from forty to fifty per cent, with one small square for each per cent. The task is to find two weaknesses in how this is presented, and to explain each one. What two weaknesses can you find? Pause it there and work it out for yourself. The first weakness is that the axis starts at forty per cent, not nought. The twenty twenty-four bar stands eight squares above forty, and the twenty nineteen bar only two. That makes the newer bar four times as tall, though the rate rose by only six percentage points. Second, only two years are shown, chosen by the council. That means the reader can't tell whether twenty nineteen was an unusually low year, or whether the rate rose steadily. The trend in between is hidden. The improvement is to start the axis at nought, and to show every year from twenty nineteen to twenty twenty-four.

Some weaknesses sit in how a graph or table is built, rather than in which data it shows. Class intervals can be uneven, or far too wide, and scales can jump in uneven steps. A title, units, labels or a key can be missing. Shades can be too alike to tell apart, and a pie can have too many thin slices. On a map, one shade over a whole area hides the differences inside it. A map of only some places can't support a full comparison. Back in Owlobia, students measured the depth of the River Tawny at thirty points across it, with invented results. Nought to ten centimetres has six points. Ten point one to twenty has nine. Twenty point one to thirty has eight. And thirty point one to one hundred and fifty centimetres has seven. What is the weakness in how these depths are grouped? The weakness is that the groups are uneven. The deepest group spans a far wider range than the rest, about twelve times as wide. The last class covers one hundred and twenty centimetres, while each of the others covers only ten. Some courses ask you to suggest an improvement and say how it helps. Now, what would you change, and how would that help the reader? Split the last class into equal ten centimetre classes, thirty point one to forty, forty point one to fifty, and so on. The reader can then see how the seven deepest points are spread out, and every class can be compared fairly. One examiner's report notes students doing less well when they wrote that more colours should be used to make the map clearer. A good improvement names the exact change and says how it helps the reader. And improve the method you're shown, unless the question asks for an alternative graph.

A scatter graph has limits of its own, and some courses name them. The first is that a relationship does not prove that one thing causes the other. A scatter graph of ten districts in Owlobia, with invented figures, shows that districts with more doctors per thousand people have a higher life expectancy, meaning how long people can expect to live. Why can't this graph alone prove that more doctors cause longer lives? Because a third factor could drive both. Richer districts tend to have more doctors, and also better food, clean water and housing. Wealth may be causing both. Now, what about the graph itself limits it? Ten cases is a small sample. And one unusual year, such as a flood in one place, could pull a point far from the others and change the pattern. Points bunched together in one corner, or points with no labels, limit a scatter graph too. Estimates and projections have a limit of their own. Saying it is an estimate only rewords the question. Give the reason it could be wrong: birth rates, migration or a crisis can change. A photograph or diagram has limits as well. It shows one place, at one moment, from one viewpoint, so say what else would help, such as a second photo from another date.

Last run through the ideas, and you take each question first.

Before reading any bar chart, where do you look first?

Where the side axis starts. If it starts above nought, every difference looks bigger than it is.

Now, what turns a named weakness into a developed one?

Its consequence for the reader, joined with so: what the weakness makes them see wrongly, or miss.

Now, a graph shows a town's crime for only two picked months. What's the weakness?

The period is too short and the months are hand-picked, so the longer trend is hidden, and nobody can tell whether those months were unusual.

Now, ice cream sales and sunburn cases rise together. Why isn't that proof?

It isn't proof because a third factor, hot sunny weather, drives both. A relationship never proves cause on its own.

Now back to the census bars for England and Wales: how do you fix them?

Start the axis at nought. The twenty eleven and twenty twenty-one bars then differ by six point three per cent, the true rise, and the false story of an explosion disappears.

If you can name a weakness and say what it does to the reader, give this a thumbs-up, because you own this one. If it hasn't landed, come back to it later in the week.

That wraps up the chapter on drawing and reading graphs. The next chapter is Population pyramids, climate graphs and wind roses. Hopefully that's next in your feed, but if not, search for ScholaFly G E oh five, oh oh.

Every other video is on scholafly.com too, easy to search by video code or by keyword. Whatever you're revising, go there and have a look, or, watch the next video.

Related terms

For: AQA GCSE 8035, Edexcel GCSE 1GA0, Edexcel GCSE 1GB0, Eduqas GCSE C111, Eduqas GCSE C112, OCR GCSE J383, OCR GCSE J384

On the specification

BoardSpecStatement
AQA GCSE 80353.4.4
Edexcel GCSE 1GA0MS:Statistical:4
Edexcel GCSE 1GB0MS-Stat.4
Eduqas GCSE C111AppA:2.4
Eduqas GCSE C112AppA:2.4
OCR GCSE J3833.3(12)
OCR GCSE J3833.4(3)
OCR GCSE J3835d[statistical]-b4
OCR GCSE J3845d[Statistical skills][4]
OCR GCSE J3849.3.12
OCR GCSE J3849.4.3
For teachers

This GCSE Geography lesson teaches spotting weaknesses in how data is presented. By the end, students should be able to spot and explain the weaknesses in how data are presented - a cut axis, cherry-picked years, a small or unequal sample, an average that hides the spread, uneven or overlarge classes, missing labels or units - develop each weakness with its consequence, suggest a specific improvement and say how it helps (OCR A and B print this), and give the limits of a scatter graph (Eduqas A and B name it). It works through three worked examples and the mistakes examiners report.