A column of numbers is shy. Give it the right graph and it tells you the answer at a glance.
Start hereA table makes you count and compare one number at a time. A graph turns those numbers into a shape — and your eye reads a shape in a heartbeat.
Here's the twist you'll play with on this page: the very same numbers can wear a line, a bar chart, or a scatter of dots. Each outfit shows off a different part of the story. Learning to graph well is really just learning to pick the outfit that makes the answer pop.
You planted a bean and measured its height every week. Here's your whole dataset — a small, tidy table. Look at it and try to answer: is the plant slowing down, or still speeding up? It's genuinely hard to tell from bare numbers.
| Week | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
|---|---|---|---|---|---|---|---|---|
| Height (cm) | 2 | 5 | 9 | 14 | 18 | 21 | 23 | 24 |
Nothing jumps out, right? Now let's dress these exact numbers up and watch the story appear.
Same eight numbers, three outfits. Tap Line, Bar, and Scatter and notice how each one nudges your eye toward a different question. Read the little verdict under the chart each time.
A line graph connects the dots in time order, so your eye follows the climb — and sees it flattening near the end.
Did you feel it? The line makes the slowing down obvious — the curve bends and levels off. The bars invite you to compare week-to-week heights like a row of steps. The scatter strips away the connecting line so you judge the raw relationship between week and height. Same numbers, three different answers made easy.
A line graph puts time along the bottom (the x-axis, the horizontal one) and your measurement up the side (the y-axis, the vertical one), then connects the points in order. The connecting line is a promise: these points belong to one thing changing, step by step.
That's why the line is perfect for the bean plant. Because the dots are joined, your eye reads the direction and the steepness at once. A steep part means fast growth; a flat part means it has nearly stopped. In our data the line rockets up early, then eases off around week 6 — the plant is reaching its full height.
Reach for a line when your x-axis is time (days, weeks, years) or any smooth order, and you want to see the shape of the change: rising, falling, speeding up, leveling off.
One honest warning: only connect dots with a line when the in-between actually exists. The plant really did have some height on the days between measurements, so a line is fair. Connecting unrelated groups with a line would be a costume that lies.
A bar chart gives each thing its own column, and the column's height is the number. Because your eye is brilliant at judging which of two heights is taller, bars are the champion for comparing separate groups — favorite fruits in your class, rainfall in five cities, points scored by four teams.
Here's a fresh dataset that's born for bars: a survey of how twenty classmates get to school.
| Way to school | Walk | Bus | Car | Bike |
|---|---|---|---|---|
| Number of students | 8 | 6 | 4 | 2 |
These are categories, not moments in time — "Walk" doesn't come "before" "Bus." So a line here would be nonsense; there's nothing in between to connect. Bars are exactly right: line the columns up and the tallest one, "Walk," answers what's most common? instantly. That's the whole superpower of a bar chart — ranking by eye.
Reach for bars when you're comparing distinct groups or categories and the question is which is biggest / smallest / how do they stack up?
A scatter plot is a field of unconnected dots. Each dot carries two measurements at once — one sets how far right it sits, the other how far up. You use it to ask: when one thing goes up, does the other tend to go up too? That tendency is called a correlation — a fancy word for "these two seem to travel together."
Imagine measuring, for lots of students, their hours of sleep against their reaction time in a game. Plot a dot per student. If well-rested players cluster toward faster times, the cloud of dots tilts in a clear direction — and the relationship is visible even though no single line was drawn.
Drag the slider below. It doesn't change how many dots there are — it changes how tightly they follow the pattern. Watch a shapeless cloud sharpen into an unmistakable trend.
At high strength the dots line up along a clear slope — a strong relationship. Slide left and the pattern melts into random static.
Reach for a scatter when every item has two numbers and you want to see whether they're linked. And remember the classroom rule: dots traveling together shows a relationship, not proof that one causes the other.
You don't need to memorize a rulebook. Just ask what your question is really about, and the graph almost chooses itself:
Notice the pattern: time picks the line, groups pick the bars, pairs pick the scatter. When you're unsure, describe your data out loud — the word "over time," "compare," or "linked" usually pops out and tells you which outfit to grab.
For each dataset, pick the outfit that makes the answer pop. Tap an option — you'll get instant feedback and the reason why.
Score: 0 / 3 answered.
A graph turns a shy column of numbers into a shape your eye reads instantly.
Time wants a line, groups want bars, paired numbers want a scatter.
The right graph doesn't decorate the data — it hands you the answer for free.