1 00:00:00,260 --> 00:00:04,860 One of the best ways to get insight into your data, is with data visualization. 2 00:00:04,860 --> 00:00:08,760 Data visualization takes raw data, and turns it into an image. 3 00:00:08,760 --> 00:00:13,480 This way, we can see exactly what our data looks like, without having to guess. 4 00:00:13,480 --> 00:00:16,330 There's nothing wrong with having your data in a table. 5 00:00:16,330 --> 00:00:18,250 But by using a good visualization, 6 00:00:18,250 --> 00:00:22,460 you can get people to understand the result almost 50% faster. 7 00:00:22,460 --> 00:00:25,160 And it'll be about 9% more accurate, too. 8 00:00:25,160 --> 00:00:28,970 Most often, you'll see data represented as some kind of chart. 9 00:00:28,970 --> 00:00:30,970 Let's look at a few of the more common charts, and 10 00:00:30,970 --> 00:00:32,260 talk about when you would use them. 11 00:00:33,530 --> 00:00:35,086 First up, is a Column Chart. 12 00:00:35,086 --> 00:00:38,170 Column Charts are used to compare different values. 13 00:00:38,170 --> 00:00:41,362 A good example of a Column Chart would be something like a company's 14 00:00:41,362 --> 00:00:42,651 monthly revenue numbers. 15 00:00:42,651 --> 00:00:46,081 Another example, from the perspective of a grocery store, 16 00:00:46,081 --> 00:00:49,188 could be how many of each fruit, sold in the past week. 17 00:00:49,188 --> 00:00:53,050 However, look what happens when we start to add more fruits. 18 00:00:53,050 --> 00:00:55,920 It gets harder and harder to read each one. 19 00:00:55,920 --> 00:00:59,120 At this point, it might be time to switch to a Bar Chart. 20 00:00:59,120 --> 00:01:02,570 A Bar Chart is essentially just a rotated Column Chart. 21 00:01:02,570 --> 00:01:06,000 But it does a better job of giving us room for our labels. 22 00:01:06,000 --> 00:01:10,094 Also, since we're used to seeing ranked data go from top to bottom, 23 00:01:10,094 --> 00:01:14,486 if we sort our data before creating the Bar Chart, it looks a lot cleaner. 24 00:01:14,486 --> 00:01:18,510 And paints a clear picture, that bananas and apples are the most popular. 25 00:01:19,620 --> 00:01:22,370 The next chart we need to know about is the Pie Chart. 26 00:01:22,370 --> 00:01:26,710 Pie Charts are used to show how something breaks down into its constituents. 27 00:01:26,710 --> 00:01:28,220 A good use of a Pie Chart, 28 00:01:28,220 --> 00:01:32,010 would be to compare market share of smartphone operating systems. 29 00:01:32,010 --> 00:01:34,400 An important thing to remember with Pie Charts, 30 00:01:34,400 --> 00:01:37,090 is that you don't want to have too many categories. 31 00:01:37,090 --> 00:01:39,750 It distracts the reader from the rest of the chart, and 32 00:01:39,750 --> 00:01:43,520 can be difficult to see which labels, belong to which slices. 33 00:01:43,520 --> 00:01:44,430 Once you get to six or 34 00:01:44,430 --> 00:01:49,100 so categories, it's time to start thinking about adding an other slice. 35 00:01:49,100 --> 00:01:52,010 Another frequently used chart, is the Line Chart. 36 00:01:52,010 --> 00:01:55,050 A Line Chart shows similar data to a Column Chart. 37 00:01:55,050 --> 00:01:59,210 Except typically, a Line Chart shows data that's more continuous. 38 00:01:59,210 --> 00:02:01,270 It has a lot more data points. 39 00:02:01,270 --> 00:02:05,140 For example, if a patient is wearing a heart rate monitor that reports their 40 00:02:05,140 --> 00:02:08,450 heart rate every minute, we'd probably want to use a Line Chart, 41 00:02:08,450 --> 00:02:10,000 instead of a Column Chart. 42 00:02:10,000 --> 00:02:12,280 Had to be a ton of columns. 43 00:02:12,280 --> 00:02:14,210 Next up, is the Scatter Plot. 44 00:02:14,210 --> 00:02:17,520 Scatter plots are used to show the relationship between two different 45 00:02:17,520 --> 00:02:18,675 variables. 46 00:02:18,675 --> 00:02:22,340 Here's a great example of a Scatter Plot, that shows the relationship 47 00:02:22,340 --> 00:02:26,610 between temperature and sales figures for a frozen banana stand. 48 00:02:26,610 --> 00:02:30,230 Thanks to our Scatter Plot, even know our data varies quite a bit, 49 00:02:30,230 --> 00:02:33,780 it's clear that higher temperatures means more sales. 50 00:02:33,780 --> 00:02:35,390 The cool thing about Scatter Plots, 51 00:02:35,390 --> 00:02:38,810 is that they can help us make predictions about the future. 52 00:02:38,810 --> 00:02:40,960 We can draw a line through the middle of our data, 53 00:02:40,960 --> 00:02:45,530 to help show about how many sales we should expect for every temperature. 54 00:02:45,530 --> 00:02:49,860 So even, though we haven't seen a hundred degree day yet, we have a pretty good idea 55 00:02:49,860 --> 00:02:53,612 of what kind of sales we'd see, if we were lucky enough, to hit a hundred. 56 00:02:54,740 --> 00:02:57,590 These are just a few of the data visualizations that you're likely 57 00:02:57,590 --> 00:02:58,750 to encounter. 58 00:02:58,750 --> 00:03:01,220 If you'd like to learn more about data visualization, 59 00:03:01,220 --> 00:03:03,080 we've got a whole course about it. 60 00:03:03,080 --> 00:03:06,150 Check out the teacher's notes below to learn more. 61 00:03:06,150 --> 00:03:09,970 Another thing you'll want to be careful of, is being misleading with your charts. 62 00:03:09,970 --> 00:03:14,100 It's easy to make a mistake and end up showing something you didn't intend to. 63 00:03:14,100 --> 00:03:16,690 Or maybe you did intend to be misleading. 64 00:03:16,690 --> 00:03:19,710 Take this ad for Microsoft Edge, for example. 65 00:03:19,710 --> 00:03:23,920 It looks like Edge is a lot faster than Chrome and Firefox. 66 00:03:23,920 --> 00:03:30,400 But take a second to look at that chart.It starts at 25,000 and goes to 31,000. 67 00:03:30,400 --> 00:03:34,060 So even though the difference is pretty small, it looks awfully big. 68 00:03:35,190 --> 00:03:39,630 There's a lot of different ways charts can be misleading or just plain wrong. 69 00:03:39,630 --> 00:03:43,200 In the next video, we'll get some practice with data visualization. 70 00:03:43,200 --> 00:03:44,490 We'll look at the shape of our data, 71 00:03:44,490 --> 00:03:46,530 and see if it actually looks like a bell curve.