1 00:00:00,120 --> 00:00:03,090 Now we want to look at another type of plot, which is the line. 2 00:00:03,090 --> 00:00:09,720 Plot and line plots are great when we want to create the sketch of a data set where there's a continuous 3 00:00:09,720 --> 00:00:12,090 relationship between the data points. 4 00:00:12,090 --> 00:00:15,630 For example, time and temperature are both great examples of this. 5 00:00:15,630 --> 00:00:22,560 So for instance, if we have the average high temperature in degrees Fahrenheit for each month in a 6 00:00:22,560 --> 00:00:29,460 city, then we can display each month along the horizontal axis of our line plot and then sketch in 7 00:00:29,460 --> 00:00:33,540 each of these average high temperatures in line with the corresponding month. 8 00:00:33,540 --> 00:00:36,120 And our line plot would look something like this. 9 00:00:36,120 --> 00:00:42,060 So we have the months along the horizontal axis and then along the vertical axis, we have this range 10 00:00:42,060 --> 00:00:48,480 of temperatures in degrees Fahrenheit and we can see that here in January, we've plotted 69 degrees 11 00:00:48,480 --> 00:00:49,260 Fahrenheit. 12 00:00:49,260 --> 00:00:55,740 In February, we've plotted 73 degrees, in March, we've plotted 78 degrees, etc. And once we plot 13 00:00:55,740 --> 00:00:58,950 all of those points, we connect them with a line. 14 00:00:59,100 --> 00:01:05,430 Now, realize here that connecting all of the data points in a scatterplot with a line like this one 15 00:01:05,430 --> 00:01:06,660 wouldn't make sense. 16 00:01:06,660 --> 00:01:10,920 Because if you remember when we looked at scatter plots, the dots were all over the place. 17 00:01:10,920 --> 00:01:14,520 They don't necessarily connect to each other in a logical way. 18 00:01:14,550 --> 00:01:21,510 The crucial thing here is that we use a line plot when the values in between each point on our line 19 00:01:21,510 --> 00:01:22,920 plot make sense. 20 00:01:22,920 --> 00:01:26,850 So, for instance, the average high in January is 69 degrees Fahrenheit. 21 00:01:26,850 --> 00:01:30,090 The average high in February is 73 degrees Fahrenheit. 22 00:01:30,120 --> 00:01:37,740 It makes sense to think about the temperature increasing from 69 degrees to 73 degrees between January 23 00:01:37,740 --> 00:01:38,490 and February. 24 00:01:38,490 --> 00:01:44,190 Those in between values of 70, 71, 72 degrees, for example, make sense. 25 00:01:44,190 --> 00:01:50,280 And so connecting these temperature values with a line gives us a logical kind of plot. 26 00:01:50,280 --> 00:01:54,300 Whereas connecting the points in a scatterplot with a line doesn't make sense. 27 00:01:54,300 --> 00:01:58,950 That's why we used the trend line to sketch a trend through the scatterplot. 28 00:01:58,950 --> 00:02:05,250 Whereas we use a line like this one to connect data points where the values in between each point make 29 00:02:05,250 --> 00:02:08,190 sense in the context of the kind of data we have. 30 00:02:08,190 --> 00:02:10,800 So we said this works well for things like temperature. 31 00:02:10,800 --> 00:02:13,500 Also for any kind of measurement of time. 32 00:02:13,500 --> 00:02:19,560 If we have some measurement of time along the vertical axis, maybe in minutes, let's say here instead 33 00:02:19,560 --> 00:02:24,480 we have 60 minutes, 70 minutes, 80 minutes, 90 minutes instead of degrees Fahrenheit. 34 00:02:24,510 --> 00:02:29,370 The values in between there make sense between 60 minutes and 70 minutes. 35 00:02:29,370 --> 00:02:31,800 We have all those continuous time values. 36 00:02:31,800 --> 00:02:39,030 So it's logical if we have data in terms of time to use a line plot for a representation of that data 37 00:02:39,030 --> 00:02:39,570 set. 38 00:02:39,570 --> 00:02:46,470 In other words, we use line plots to show changes over time or a connection between data points. 39 00:02:46,470 --> 00:02:52,410 So if showing that connection doesn't really make sense, then a line plot probably isn't the best plot 40 00:02:52,410 --> 00:02:53,070 to use. 41 00:02:53,070 --> 00:02:58,740 To give a counter example, let's say we're keeping track of the number of times that the Summer Olympics 42 00:02:58,740 --> 00:03:00,720 has been hosted on each continent. 43 00:03:00,720 --> 00:03:05,610 And so along the horizontal axis, we have continents like Europe. 44 00:03:06,410 --> 00:03:11,390 Asia, North America, etc., along the horizontal axis. 45 00:03:11,390 --> 00:03:17,060 And then we have a count along the vertical axis for the number of times that each continent has hosted 46 00:03:17,060 --> 00:03:18,380 the Summer Olympic Games. 47 00:03:18,410 --> 00:03:24,290 Well, in theory, we could plot the value for each continent, plot the value for Europe, for Asia, 48 00:03:24,290 --> 00:03:26,930 for North America, and connect them with the line. 49 00:03:26,930 --> 00:03:33,230 But it doesn't really make sense to connect those discrete data points with a line because they're not 50 00:03:33,230 --> 00:03:35,300 really related to each other. 51 00:03:35,300 --> 00:03:41,060 Or another way to put it, there's no in between values between Europe and Asia or between Asia and 52 00:03:41,060 --> 00:03:43,100 North America that make sense. 53 00:03:43,100 --> 00:03:47,720 These are just discrete buckets or discrete categories that are separate from one another. 54 00:03:47,720 --> 00:03:54,020 And so we wouldn't connect those data points with a line graph, whereas something like these temperature 55 00:03:54,020 --> 00:03:59,540 values have those in between values, we can connect them with a line plot and it makes a lot more sense. 56 00:03:59,570 --> 00:04:05,060 Now, whenever we have a line plot, we want to be able to interpret what we're seeing. 57 00:04:05,060 --> 00:04:12,710 Of course, each point along the line graph gives us a particular value, but we can also look at the 58 00:04:12,710 --> 00:04:16,190 slope or the change between certain points. 59 00:04:16,190 --> 00:04:21,800 For instance, let's look here at the change between August and September. 60 00:04:21,800 --> 00:04:29,240 So here we're at August and then here is September and then this is October. 61 00:04:29,240 --> 00:04:34,820 What we can see is that the change between August and September is much less than the change between 62 00:04:34,820 --> 00:04:36,350 September and October. 63 00:04:36,350 --> 00:04:39,590 And we see that reflected in the slope of each line. 64 00:04:39,620 --> 00:04:45,590 The slope between August and September is much shallower, whereas the slope between September and October 65 00:04:45,590 --> 00:04:46,790 is much steeper. 66 00:04:46,790 --> 00:04:53,840 So the steeper each segment is, the bigger the change in this case in temperature, the more shallow 67 00:04:53,870 --> 00:04:58,250 the particular segment, the less change in temperature we have. 68 00:04:58,250 --> 00:05:05,030 So a line plot can help us see where the greatest rate of change occurs and where the least amount of 69 00:05:05,030 --> 00:05:06,050 change occurs. 70 00:05:06,050 --> 00:05:11,330 For instance, the change up here between June, July, August, September is pretty shallow. 71 00:05:11,330 --> 00:05:12,500 There's not a lot changing. 72 00:05:12,500 --> 00:05:14,180 The temperature is pretty consistent. 73 00:05:14,180 --> 00:05:20,450 But to go, for instance, from March to June, the temperature is changing significantly every month. 74 00:05:20,450 --> 00:05:25,790 To go from September to December, the temperature's falling off significantly every month. 75 00:05:25,790 --> 00:05:32,030 So we want to pay attention both to the values along the line plot, but also to the change in between 76 00:05:32,030 --> 00:05:37,310 each value, because that slope, depending on how shallow or how steep it is, gives us a good idea 77 00:05:37,310 --> 00:05:40,880 of the rate of change in that part of the data. 78 00:05:40,880 --> 00:05:45,860 The last thing we want to say about line plots is that there's another type of line plot we want to 79 00:05:45,860 --> 00:05:49,910 be aware of, and that specific kind of plot is called an O Jive. 80 00:05:49,910 --> 00:05:54,620 And it looks like this basically, it's just an accumulating line plot. 81 00:05:54,620 --> 00:06:01,070 So for this particular example, we again have months along the horizontal axis. 82 00:06:01,810 --> 00:06:05,120 And then we have revenue along the vertical axis. 83 00:06:05,140 --> 00:06:11,980 Remember that all of these graphs and plots we're looking at are all about displaying information in 84 00:06:11,980 --> 00:06:16,870 the way that makes the most sense or communicates the data most clearly. 85 00:06:16,900 --> 00:06:23,560 So to take this example, if we have revenue data, in other words, the revenue that our company earns 86 00:06:23,560 --> 00:06:29,290 each month of the year, we could certainly take that data and create a line plot and that would be 87 00:06:29,290 --> 00:06:33,280 useful because we'd be able to see the revenue earned each month. 88 00:06:33,280 --> 00:06:39,670 But maybe instead what we're trying to communicate is more about the total revenue for the year and 89 00:06:39,670 --> 00:06:42,130 less about our month to month revenue. 90 00:06:42,160 --> 00:06:47,530 As you can imagine, the line plot would do a better job communicating the month to month revenue, 91 00:06:47,620 --> 00:06:53,230 whereas if we use an O five where the values in our data set accumulate, then we'll get a much better 92 00:06:53,230 --> 00:06:55,780 picture of revenue for the entire year. 93 00:06:55,780 --> 00:07:02,500 So what the OGI shows here for December is not just the revenue we earned in December, it's the revenue 94 00:07:02,500 --> 00:07:04,600 we earned over the course of the entire year. 95 00:07:04,810 --> 00:07:11,020 This value here representing November is the revenue we've earned up and through November, this data 96 00:07:11,020 --> 00:07:15,130 point above October is the revenue we've earned up and through October. 97 00:07:15,130 --> 00:07:21,400 So every month that goes by as we earn more and more revenue, this line is always going to grow, assuming 98 00:07:21,400 --> 00:07:22,750 our revenue is always positive. 99 00:07:22,750 --> 00:07:28,630 And you can imagine how this jive of revenue gives us a different picture of what's happening than we 100 00:07:28,630 --> 00:07:34,090 get from a line plot of revenue and which one we should use really just depends on what we're trying 101 00:07:34,090 --> 00:07:37,960 to communicate and which of these does a better job of communicating. 102 00:07:37,960 --> 00:07:38,890 Our point. 103 00:07:38,920 --> 00:07:45,010 Keep in mind here also that we're showing two different kinds of line plots and lives here in this line 104 00:07:45,010 --> 00:07:45,550 plot. 105 00:07:45,550 --> 00:07:52,930 We've sketched in these points indicated by these dots along the way where is this has no dots. 106 00:07:52,930 --> 00:07:55,360 It's just a collection of line segments. 107 00:07:55,360 --> 00:07:58,390 We can sketch a line, plot or an octave. 108 00:07:58,390 --> 00:08:04,900 Either way, we can include these little dots as points for the line plot or for the five, or we can 109 00:08:04,900 --> 00:08:09,820 exclude them and have a line that looks like this, just a collection of a bunch of segments for both 110 00:08:09,820 --> 00:08:12,970 a line plot and an F both are acceptable. 111 00:08:12,970 --> 00:08:18,430 And then finally, we just want to say that in the same way that a line plot is only appropriate for 112 00:08:18,430 --> 00:08:23,080 certain types of data, Ogi lives are also only appropriate for certain types of data. 113 00:08:23,080 --> 00:08:28,300 For instance, even though a line plot works really well for temperature data, an O jive doesn't really 114 00:08:28,300 --> 00:08:30,130 make sense at all for temperature data. 115 00:08:30,130 --> 00:08:36,039 If we're working here with this temperature data, we would show a temperature of 69 for January, but 116 00:08:36,039 --> 00:08:42,309 then a temperature of 69 plus 73 or 142 degrees for February. 117 00:08:42,309 --> 00:08:49,270 And in the context of temperature, it makes no sense to show an accumulated 142 degrees for February. 118 00:08:49,270 --> 00:08:55,630 So this kind of temperature data is nonsensical in an OGI, but revenue data works really well. 119 00:08:55,630 --> 00:09:00,400 So again, just paying attention to which kind of plot makes sense for the kind of data we have. 120 00:09:00,400 --> 00:09:06,430 And then with both line plots and drives, the last thing we want to say is that we can also plot multiple 121 00:09:06,430 --> 00:09:08,650 series on the same graph. 122 00:09:08,650 --> 00:09:16,270 So for example, with this temperature data, this blue line plot could represent one particular city 123 00:09:16,270 --> 00:09:19,570 and the red line plot could represent another city. 124 00:09:19,570 --> 00:09:25,930 Plotting them together on the same set of axes gives us a great way to compare average high temperatures 125 00:09:25,930 --> 00:09:27,460 across multiple cities. 126 00:09:27,460 --> 00:09:31,720 So line plots are particularly useful for seeing the comparison that way. 127 00:09:31,720 --> 00:09:34,660 And of course, we can do the same thing with an five. 128 00:09:34,690 --> 00:09:40,720 Maybe instead of plotting revenue accumulation over the course of a year for our entire business, we 129 00:09:40,720 --> 00:09:45,550 want to divide the business up into different departments or different lines of business, different 130 00:09:45,550 --> 00:09:46,990 segments of our business. 131 00:09:46,990 --> 00:09:53,470 For instance, maybe we have a business where we sell products online and also in person at conferences 132 00:09:53,470 --> 00:09:59,770 all over the country where we set up a booth and sell product and so we could plot in an off here revenue 133 00:09:59,770 --> 00:10:05,860 accumulation for our online business in blue over the course of the year, and then revenue accumulation 134 00:10:05,860 --> 00:10:10,480 for our in-person business at conferences in red over the course of the year. 135 00:10:10,480 --> 00:10:17,050 And that would allow us to see the comparison between those two parts of our business in the same graph, 136 00:10:17,050 --> 00:10:18,790 in the same five.