0:03 hi welcome back in this section we will continue
0:04 continue
0:06 studying important characteristics of data
0:07 data
0:09 using statistical concepts such as accuracy
0:10 accuracy
0:12 and precision and why is it important
0:14 for us to know these concepts
0:17 as we the lean six sigma before we go to
0:18 the details
0:21 let's focus first on a recap of what are
0:22 measures of central tendency
0:26 in statistics central tendency measures
0:27 are measures that are used to compare
0:29 the center of a data set
0:32 as against to a reference point measures
0:34 of central tendency include
0:38 mean median and mode min talks about the
0:39 arithmetic average
0:42 of all the data values as we know it the average
0:43 average
0:46 however median this talks about the
0:48 midpoint of the data values
0:51 while mode it talks about the most
0:53 frequently according data value
0:56 now let's take an example to put context
0:58 say we have a data set
1:01 with data values 1 1
1:05 1 3 5 and 9. let's take the average
1:07 so let's take the sum of all the data
1:10 values and divide it to the number of
1:12 values that we have summed
1:15 that gives us a value of 3.33
1:19 the mean of this data set is 3.33
1:21 now let's go to the median for us to
1:22 calculate median
1:24 we have to arrange the data set from
1:26 lowest to highest
1:28 and take the midpoint because the total
1:30 number of data
1:33 in our set is even which is 6
1:35 therefore the midpoint is located in the
1:36 two middle values
1:38 now we have to take the average of the
1:40 two values which is 1
1:43 plus 3 all over 2 which give us a median
1:47 of 2.0 in data sets where the total
1:49 number of data values
1:52 is add the midpoint is certain now let's
1:53 take the mode
1:55 because the value one is the most
1:57 frequently occurring value
1:59 therefore mode is one now these are the
2:01 three central tendency measures that we
2:02 are using
2:05 in statistics most of the time we use
2:06 the average
2:09 but in cases that there are outliers we
2:10 rather use the median
2:12 this is because median is not
2:14 susceptible to bias
2:17 due to outliers or unusual observations
2:19 if you're using continuous data most of
2:20 the time
2:22 you will be using mean and regen and if
2:24 you're using discrete data
2:26 mode is more appropriate so this is the
2:28 central tendency measure
2:29 now let's talk about measures of
2:32 dispersion in statistics
2:33 this person describes the
2:36 characteristics of data and how far they
2:38 are from each other
2:40 common measures are range standard division
2:41 division
2:44 and variance range is the difference
2:45 between the maximum value and the
2:46 minimum value
2:48 standard deviation is the distance of
2:50 the values from the mean
2:52 or the average distance of each values
2:54 from the mean
2:56 and variance is the squared value of the
2:57 standard deviation
2:59 now let's put context by giving an example
3:00 example
3:02 let's use the same data set from the
3:04 measures of central tendency
3:08 range maximum minus minimum therefore
3:12 9 minus 1 which is 8. standard deviation
3:15 using our formula or using excel it will
3:16 give us
3:19 3.2 while variance as the square of the
3:21 standard deviation
3:24 we can get 10.24 now this will give us
3:25 an idea
3:28 of how disperse our data values are from
3:29 each other
3:31 now let's talk about accuracy and precision
3:33 precision
3:35 accuracy is based on the measures of
3:36 central tendency
3:39 it gives us an idea of how far is the center
3:40 center
3:42 of the data set from the target while precision
3:43 precision
3:46 it is guided by the concepts of measures
3:47 of dispersion
3:50 how dispersed are the data values from
3:52 each other
3:54 in linsix sigma what we want to see is a
3:56 process that can give us an output that
3:58 is accurate
4:01 and precise take note that we use
4:05 and not or the picture on the upper left
4:07 of this slide gives us the ideal state
4:09 where there is high accuracy
4:12 and high precision while on the upper right
4:12 right
4:15 we can see that there is high precision
4:16 but there is low accuracy
4:19 the data values are not dispersed they
4:21 are close to each other
4:24 but they are far away from the target
4:25 the lower part of the picture
4:29 on the left side it gives us the idea of
4:32 high accuracy but low precision
4:35 data values are around the target but
4:36 they are
4:38 a bit dispersed compared from the
4:41 previous exercises or examples
4:44 therefore there is low precision and the
4:46 last picture on the lower right side
4:48 gives us the worst case scenario wherein
4:50 there is low accuracy
4:52 and there is low precision in lean six
4:54 sigma what we want to achieve
4:56 is a process that can give us an
4:58 accurate and precise output
5:02 remember as accuracy increases
5:03 the more conforming products we can
5:06 produce but not only accuracy is important
5:07 important
5:09 as well as precision as we may know quality
5:10 quality
5:13 is inversely proportional to variability
5:16 and dispersion and precision is the same
5:17 with variability
5:20 so as we have more variability in the process
5:21 process
5:23 the lower the potential quality level
5:25 will be these are the important concepts
5:27 of accuracy and precision
5:29 as we take them into our lean six sigma journey
Chrome Extension