 here we're using a whole lot of information so if you're looking at something that would conform to a bell curve you would think that if you had the more data that you had the more smooth looking the shape would look and the more you can be kind of certain that it looks like it follows a bell curve type of distribution so we're going to go okay it looks like a bell curve kind of does the thing I'm going to move this off to the side and let's start to build an actual bell curve so I'm going to say let's say that this is going to be x this is our x this is going to be p of x let's say I'm going to make this black and white so we're going to go hometown font group making this black and this white and then I'll center it and then let's make a skinny x skinny up the x it's on the lighter side of our bell curve it weighs less because it's skinny it's a skinny x but it could be skinny with big bones you know it just has but any case so now we're going to say okay well what am I going to start because I'm measuring in pounds I could start from like one pound and go up to like I don't know 500 pounds or something but it's I don't probably need to go that far because it's not likely that someone weighs one pound so what's the range that I need on my x's that are going to be on the x parameter so I could say well everything let's do that for standard deviations again that should pick up almost everything and for standard deviation so let's go that we're going to say the numbers of of standard deviations which I'm just going to say sd's in our chart the number of x's that we want are going to be equal to uh well let's four let's just say four of them and so that means that the lower x and then the upper x we can calculate so the lower x is going to be the mean the middle point minus the standard deviation 1166 times four of them to go to go four standard deviations and we're going to say okay boom so we've got 80 so I so people don't weigh less than 80 oftentimes right and then we're going to say pounds we're talking pounds I know if you're not in the United States you're like what are you talking about or what a pound what in the world but whatever that's how we do it man so now we're going to say the upper one is going to be 127 uh plus a standard deviation times four and enter so now uh 173 pounds is going to be our upper end okay and so now let's let's just build this I'm going to say this is going to be an x of let's go from starting point is just going to be one less than this one or or 80 you know at the lower end I'm going to get rid of the decimals because I'm not going to do the decimals here I'm just going to take it to the whole pounds and I'm going to drag it well let's do 81 so we can see the pattern get rid of the decimals select those two I'm going to drag it down to 174 174 and that should be good so I'm going to drag it down until I get the pattern down to 174 that doesn't even fit me man I'm because I'm yoked up what about my someone like me that's like huge like Arnold Swartz snager but not now because he's old but like when he was I'm not you know when he was not old no anyways uh so now we're going to say p of x we're going to say this equals then norm dot disc so we've seen this in prior presentation so we're just going to take the norm dot disc I won't do the spill function this time I'll just do the normal process I'm going to take the x comma the mean the mean is outside the data we're working in so I'm going to f for it making it absolute so it will copy down uh and not move that cell down comma the standard deviation is that one again it's outside the data I'm working with I don't want it to move down so I'm going to select f for dollar sign before the f and the three making it absolute comma cumulative not cumulative false or zero and then enter I'm going to percentify it home home tab number group percentify you better percentify if you want to recognize I still don't recognize it's still zero but if I double click on the fill handle copies it down and there we have it now if I if I was to total this up total and say alt equals to some trusty some formula we get one or 100 because basically all the data is in there all right so that looks good and so so now I'm going to say okay uh so so now let's ask well let's see how closely that fits to our actual data now now remember our actual data is actual samples not given to us in percents so I want to either adjust my actual data to percents or adjust this curve to actual data right so first I've got to take all my information here and group it together and that's going to be our frequency so this is going to be actual frequency actual frequency and then I'll wrap that home tab alignment wrap it center it black white so so this is going to be the one where we're going to say the frequency is going to tell us everything for example this one that is above 80 up to and including 81 is going to go into that bucket when we count our actual data all right so we're going to use our frequency which is a spill array function frequency tab I'm just going to select all of our data with the little arrow selecting all of the data comma and then I'm going to select all of my x's by putting my cursor on the top x control shift and down arrow don't want to pick the total so I'm just going to say shift up and then control backspace to get up to the top without the dancing frequency ants uh stopping their frequency dance and then I'm going to say enter and boom spills it on down so these are the number of of occurrences of these datas that are are in between 100 in this case or above 100 up to and including 101 I don't want it to it spilled down a little one more than I wanted it to go so I'm going to double click here I'm going to get rid of that last one so it spills out right to there now if I sum this up how do I know it picked up all my numbers well if I say alt equals down here summing it up comes out to 25 000 and I could say does that equal my count of data right I can check my actual count say how many data points do I have over here imagining these are our bunch of data that we actually counted right so we're going to say this is going to be count I'm just going to say count all of my data how many are there there's 25 000 we picked up all 25 000 are represented in one of these buckets so that looks good check confirmed so now I can either convert my percentages to counts by multiplying all the percentages by 25 000 then I can then I can compare my actual to my to my uh the smooth curve or I can convert which is probably the better thing to do the actual frequency to the percent by taking the percent of the total and I'm going to say let's go home tab numbers black white wrap it center it and I'm going to do