The document discusses calculating the average and standard deviation of experimental results. It explains that taking the average of multiple experimental runs can provide a single representative value for the results. However, the standard deviation must also be considered to determine how well the average represents the data, with a smaller standard deviation indicating the average is more representative. Excel functions for calculating average and standard deviation are demonstrated, as well as graphing data with error bars to visually represent the standard deviation.
(Mathematics in Industry 13) Wil Schilders (auth.), Wilhelmus H. A. Schilders, Henk A. van der Vorst, Joost Rommes (eds.) - Model order reduction_ theory, research aspects and applications-Springer-Ve
The document discusses calculating the average and standard deviation of experimental results. It explains that taking the average of multiple experimental runs can provide a single representative value for the results. However, the standard deviation must also be considered to determine how well the average represents the data, with a smaller standard deviation indicating the average is more representative. Excel functions for calculating average and standard deviation are demonstrated, as well as graphing data with error bars to visually represent the standard deviation.
The document discusses calculating the average and standard deviation of experimental results. It explains that taking the average of multiple experimental runs can provide a single representative value for the results. However, the standard deviation must also be considered to determine how well the average represents the data, with a smaller standard deviation indicating the average is more representative. Excel functions for calculating average and standard deviation are demonstrated, as well as graphing data with error bars to visually represent the standard deviation.
The document discusses calculating the average and standard deviation of experimental results. It explains that taking the average of multiple experimental runs can provide a single representative value for the results. However, the standard deviation must also be considered to determine how well the average represents the data, with a smaller standard deviation indicating the average is more representative. Excel functions for calculating average and standard deviation are demonstrated, as well as graphing data with error bars to visually represent the standard deviation.
experiments and it's kind of when you do like an experiment many times just to see if your results are reproducible or just the experiment doesn't work well so usually you get the like different results from the same experiment and actually this is not something wrong just because you have some experimental error and you'll have some differences in the values so actually to get a result that represents all these results that you got from the experiment like to get one number to say that this is the result of my experiment then you need to do something it's called getting the average of these numbers and by doing this you can say that this experiment is gonna give this results at this condition so we have here two experiments this is experiment number one and this is experiment number two and you have five runs each one gave different result here and there five results here and actually what we're gonna do now is to get the average and average as we see from this equation it is the summation of the values that you have divided by the number of experiments that you did so if you wanna calculate and we can use one function here it's called count so count just it counts the number of cells that you choose so in this case it automatically counts it you can you can just put a 5 if you want just if you in case you have a number a bigger number of experiments it's gonna be easy instead of writing something wrong and then the average it's gonna be the submission of these results divided by the number of runs and then the average is 25 we're gonna do the same here just copy and paste that we'll do the same thing and you'll find that this is the account and this is the Sun and you found that this too so the average is now done and you find that the average of these two experiments is gonna be is the same storified and now we can say that okay the 2x friends are gonna give the same result and this is the average end and stop but actually if you take a closer look at these results here you'll find that this is 21 24 25 26 and 29 which are kind of close or close to this 25 the average that you got so you can say that it it's kinda represents these results well however if you look here you'll find that there is 4 and 48 and it's pretty difficult to say that this 25 represents 4 and 48 you can you can find it not logical and that's why there is another parameter that you calculate and this parameter is the major that you can tell is this average representative or natural represented and this parameter is called the standard deviation which is calculated from this equation so actually it will tell you like you can you can see from in different papers of scientific articles it says that your average Ori or the experiment is giving let's say 25 plus or minus 5 and this 5 after the plus or minus is the standard deviation it it shows you how for the standard deviation or they or they the numbers from the average rating so if the standard deviation is high it means that your your average is not really representing your results but if it's small it means that your average is representing so let's see what we have here it says that X minus X bar experts the average that we just calculated and then it's the number of of cells sorry and this is the the s square and we'll get the square root of this to get the standard deviation so the standard deviation in this case is gonna be the summation of or we can we can do this on a separate column it is this minus X bar which is the average and we will press a for just to freeze this and and this is to the power of 2 and here we go and we'll do the same for the other guy just to be done with this okay so now have this X minus X bar power two and we will just get the sum of this divided by n minus one and this is the standard deviation but we need to get the square root of this because this is the square so we will use the square root function and now the standard deviation is two point nine one five let's do it here for this other experiment it's now the square root is calculated from these and from this so we are now good and you see that this is giving standard deviation of nineteen point twenty seven so in the first experiment our our average or the result is twenty five plus or minus two point nine which is pretty good but in the in the other case it's twenty five plus or minus nineteen point three which is like doesn't make any sense so in this case you can tell that the second experiment is not good and the results are not consistent but in the first case the results are really consistent so this is how you can do the calculations using these functions but there is something that Excel can can make your life much easier if you if you know you can do the average and the standard deviation using one function without doing all these calculations if you just press average or use the the function average average and then you just choose these cells it's gonna give you the average and the standard deviation its stdev and you just highlight the same cells and this is what we got if you if you repeat this for the second experiment is gonna give you the same this is the average and this is the standard division so in this case you don't need to go through all these columns and all these calculations you just get the function and it will take care of everything so this is something that's pretty cool what Excel you don�t� that I just showed this to let you know how how these are calculated but you don't need to do all this I can go use the functions and then you're you're fine another thing I gonna show you is related to the average and the standard deviation we have here an experiment of reaction and you have the the reaction rate at different temperatures for different catalysts and you have these results you have six runs for seven temperatures and you now need to represent this graphically and show your results and actually it's not gonna be good to plot seven lots on the same on just on the same graph so what we're gonna do is exactly what we did we will calculate the average and the standard deviation and then see how we can put this on a graph and it's pretty pretty cool so the averages we know is this function you're gonna get the average and standard deviation and we repeat this for the sense sometimes it shows this green color at the corner just to tell you that there's some may be something wrong because it doesn't understand that this is something different from those so it just wants to take this hundred into consideration just to let you know that maybe you forgot this but we know what we're doing so actually what we know you're gonna plot now is the temperature and the average so the average is kind of showing that your results are or this is the average of your of your results are so let's go back and go do this color thing and and now you have your results it's the average and actually I'm not gonna do all these excess titles and all this stuff because it's just waste of time so we have here this axis is the rate and this is the e temperature but now what we show is just the average we didn't show the standard deviation and the way you can show the standard deviation graphically is using what so called error bars and error bars are just some bars that you have at each dot just to show you how for this standard deviation are from the IDI average and it's pretty easy you can you can go to layout here when you choose or highlight the table I mean data graph and then you go to a bar there are a couple of options but I prefer to go to more error bar options and then you have this window and it asks you what is the edge what's the error amount you'll have directions in both and you can have cap or no cap you can change the line color shadow and all these stuff you don't want to play with this stuff now but if you want to go to custom because do you have your values you don't need Excel to do anything and you can choose this for the positive and negative and then press close I'm sorry okay and then close and what you see now is at each point it has these error bars which are actually the standard deviation it shows you have these values varying up and down so you can know that in this case the error bars are small Services has turned a small standard deviation which is this point and this point has a lower standard the maximum a standard deviation is at this 350 which is this so it is just a way of showing graphically your data the average and the standard deviation so it's kind of showing all your data just in one single graph so that's all for today things
(Mathematics in Industry 13) Wil Schilders (auth.), Wilhelmus H. A. Schilders, Henk A. van der Vorst, Joost Rommes (eds.) - Model order reduction_ theory, research aspects and applications-Springer-Ve