1 00:00:00,006 --> 00:00:03,333 [SOUND] Hi I'm Ben and 2 00:00:03,333 --> 00:00:09,013 welcome to data analysis basics. 3 00:00:09,013 --> 00:00:11,050 Making decisions with data. 4 00:00:11,050 --> 00:00:12,650 We all make decisions. 5 00:00:12,650 --> 00:00:14,140 We decide when to get up and 6 00:00:14,140 --> 00:00:18,130 go to sleep, what to eat, where to go to school or work. 7 00:00:18,130 --> 00:00:20,850 And that's only scratching the surface. 8 00:00:20,850 --> 00:00:24,380 Data analysis is the process we use to examine information and 9 00:00:24,380 --> 00:00:26,020 make better positions. 10 00:00:26,020 --> 00:00:29,040 Sometimes, we'll see this is beautiful charts and graphs. 11 00:00:29,040 --> 00:00:32,180 But it doesn't have to be fancy to be data analysis. 12 00:00:32,180 --> 00:00:36,680 In fact, we all do this naturally every time we go to the grocery store. 13 00:00:36,680 --> 00:00:39,410 Imagine you're looking for some peanut butter. 14 00:00:39,410 --> 00:00:42,670 It's a small store and they only have two options. 15 00:00:42,670 --> 00:00:47,810 A 10 ounce container that costs \$5 and a 20 ounce container that costs \$8. 16 00:00:47,810 --> 00:00:51,720 Since you get twice as much peanut butter with the 20-ounce option and 17 00:00:51,720 --> 00:00:55,520 the price is less than doubled, the larger jar is the better deal. 18 00:00:56,690 --> 00:00:59,840 That's a pretty simple example of data analysis. 19 00:00:59,840 --> 00:01:03,110 We took the prices and sizes of two different products and 20 00:01:03,110 --> 00:01:07,160 made a decision which product to buy based on that information. 21 00:01:07,160 --> 00:01:09,500 Though it's not always so simple. 22 00:01:09,500 --> 00:01:13,730 What if you are the CEO of McDonald and you had to pick between two cities for 23 00:01:13,730 --> 00:01:15,660 a new location, what would you do? 24 00:01:16,770 --> 00:01:20,040 Our brains can only process a limited amount of input, so 25 00:01:20,040 --> 00:01:24,750 we take shortcuts to determine what's important, and sometimes we make mistakes. 26 00:01:24,750 --> 00:01:28,570 These kinds of errors in perception are called cognitive biases. 27 00:01:28,570 --> 00:01:31,490 One of these that affects our ability to analyze data 28 00:01:31,490 --> 00:01:35,910 is that humans tend to find patterns in completely random data. 29 00:01:35,910 --> 00:01:38,990 These can cause us to identify trends which are really just 30 00:01:38,990 --> 00:01:41,270 random variations overtime. 31 00:01:41,270 --> 00:01:45,460 If I flip a coin ten times and get ten heads in a row, 32 00:01:45,460 --> 00:01:49,740 its tempting to say that the next flip will almost and definitely be heads. 33 00:01:49,740 --> 00:01:52,110 But really, it's just about 50% chance. 34 00:01:53,430 --> 00:01:58,260 Other cognitive biases lead us to put more weight on some observations than others. 35 00:01:58,260 --> 00:02:01,600 If we already have an idea of what the data will say, 36 00:02:01,600 --> 00:02:04,880 then we'll tend to notice examples that confirm our theory. 37 00:02:04,880 --> 00:02:09,520 We'll also give more weight to more recent data than we give to older data. 38 00:02:09,520 --> 00:02:13,150 Data analysis gives us methods to see beyond our biases, 39 00:02:13,150 --> 00:02:15,470 to get at the truth of what the data shows. 40 00:02:16,490 --> 00:02:18,430 Let's go back to the grocery store. 41 00:02:18,430 --> 00:02:22,030 This time, it's a large supermarket with dozens of options and 42 00:02:22,030 --> 00:02:23,690 several different brands. 43 00:02:23,690 --> 00:02:29,200 We've got chunky, smooth, natural, there's even one with a chocolate swirl. 44 00:02:29,200 --> 00:02:32,480 If you're on a budget you could be here forever trying to figure out 45 00:02:32,480 --> 00:02:35,550 which choice is the most cost effective. 46 00:02:35,550 --> 00:02:40,610 But at some stores they help you out by showing you each jar's price per ounce. 47 00:02:40,610 --> 00:02:43,910 They do the math for you, reducing the amount of information that 48 00:02:43,910 --> 00:02:48,670 you have to process and making it easier to compare different options. 49 00:02:48,670 --> 00:02:52,790 If you wanted to use data analysis to help pick the new McDonald's location, 50 00:02:52,790 --> 00:02:55,420 maybe you'd look at traffic data for each city and 51 00:02:55,420 --> 00:02:58,260 compare that to the available properties to help narrow it down. 52 00:02:59,530 --> 00:03:03,740 Just like the price per ounce label, data analysis reduces complex sets of 53 00:03:03,740 --> 00:03:07,560 information into insights that are easy to understand. 54 00:03:07,560 --> 00:03:12,350 And these insights can help us make difficult decisions much simpler. 55 00:03:12,350 --> 00:03:15,925 Coming up, we'll look at where where we can find some data to help us make those 56 00:03:15,925 --> 00:03:17,020 difficult decisions.