Measures of Spread & Variability: Range, Variance, SD, etc| Statistics Tutorial | MarinStatsLectures

[@marinstatlectures]

This statistics video introduces the different measures of spread/variability for a numeric variable. In this video, we introduce the Interquartile Range, Variance, Standard Deviation, and other measures of "spread" for a numeric variable. One measure is not necessarily 'better' than the other, they simply try to convey similar information in slightly different ways, each with their respective pros and cons. 👍🏼Best Statistics & R Programming Language Tutorials: ( goo.gl/4vDQzT ) 🎗 Like to support us? You can Donate (bit.ly/2CWxnP2), Share our Videos, Leave us a Comment, Give us a Like or Write us a Review! Either way, We Thank You!

[@mariairudayaraj6069]

This video completely made me feel better about statistics and concepts. Your approach is commendable and really explains a lot for people who are scared of formulas and concepts. Way to go !

[@varshajoyprakash7404]

Hi Mike, the tutorial is the best what is available online. Your daughter has perfectly summed up. You have made statistics so much fun and easy to learn. Kudos.

[@yimbanji]

Your Son is right, You are a ninja, I'm a self-taught data analyst, Excel and Power BI Mostly, and I started using R about a year ago for some more heavier work, now I'm certain I'll be a Data Consultant for big projects soon because you are sorting so many scattered pieces of the practice in my mind into proper order,

[@patrickcordero6673]

i like how you can discuss everything in just a single board. Also it's very easy to understand and not boring.

[@sureshgeddada6166]

Sir you made statistics very easy for me, very usefull course, Even in paid courses we cannot see such kind of explanation,

[@buihung3704]

Nice, after learning so many methods of analysis, I'm starting to lose myself about the basic concepts :))))) thanks Professor Marin for bringing this backed up.

[@nwakerenduike8151]

It is really wonderful to have come across your videos. Your style makes one more comfortable learning statistics. Well done

[@jivankushwah4566]

You deserve milions subscribers

[@arjunmohan6210]

Thank you so much for these wonderful series of videos, as it was the last video for the beginners playlist for me , It was an amazing experience and will recommend it to others who are interested in data science. Good job!

[@DanRebCadavedo]

Hi Mike, been watching since your first video on this course, I was wondering why didn't you get the mean first before calculating squared deviations? Just a question, I love your videos and I'm going to finished these videos btw. Thank you IN ADVANCE

[@teelakkhananukun283]

These materials are gold. Please keep delivering high quality videos/lectures like these- so people like me who are really interested in learning can learn from pros like you. Now from a business perspective, I think you can spend more time on marketing yourself in this market. I know you are doing this for free but at some point you gotta make a living out of it too. ( I know the more people watch your video, you get paid more but you can definitely expand this a lot more). Consider put your course on Udemy or some MOOC platforms to get some incentives on your lectures (if you haven't done so). I really think people who are aspired to be data scientists will really appreciate your videos. Keep up the good work! I am on the 12th video now and I am planning to finish the whole series.

[@marinstatlectures]

we don't make much money with this approach, but we dont want to put our content on something like Udemy or something like that as we believe that education should be free for all. we know our content is better than 99% of any paid content, but we can't in good conscience charge people money for their desire to learn and better themselves. instead we will rely on the small ad revenue, and many donations we receive to contribute toward our cause.

[@-lll7585]

Great videos! Thank you

[@sandeepchotalia1131]

Awesome precise explanation...loving your videos..please upload some videos on data analysis, especially taking some datasets from kaggle and explaining with Python. Keep up the good work..!!

[@jivankushwah4566]

from heart

[@محمدعليمهدلي-ح6ق]

Thanks so much for the video what is the type of microphone you are using?

[@mistrzwewszystkim]

thank you for the great material! Could somebody enlighten me why the dots on the chart are spread this way? Lowering dot positions towards higher values on X-axis?

[@samuraisom636]

Its just convenient to show the variances graphically if the values are shown that way...

[@marcelaonofre8412]

❤love u

[@jeffreylin235]

I've seen IQRs presented as Q1 and Q3 in some papers. For example, "IQR: 64 to 89" instead of "IQR 25". Strictly speaking, this is wrong. Many journals and reviewers will accept that. An IQR of 25 could be "64 to 89" or "65 to 90" or "60 to 85". Presenting as Q1 to Q3 provides more information than a single number. What do you think is a better way to present the spread of non-normally distributed data?

[@marinstatlectures]

Hi Jeffrey, sure i think it's fine to report IQR as the two values...it does provide a bit more info. I guess it depends a bit on what you want to use it for, but i do think that sometime people get too caught up in what is technically right or wrong, instead of thinking about what is the best way to convey the info you are hoping to show. regarding non-normally distributed data, i think the measure of spread you use is a separate issue from the shape of the distribution....the SD is not only relevant to the Normal distribution. for instance, both the Poisson distribution and the Exponential distribution are defined by a rate (or 1/rate in case of exponential), and this rate provides the mean as well as the SD for the distribution...so knowing the SD is useful here. same with a Binomial distribution, it is not a Normal distribution, but the SD is helpful here. so id say that it is not about normal/non-normal, but rather what measure of spread captures what you are interested in (similar to the idea of a mean vs a median..) i think that because the SD is usually presented together with the Normal distribution, they tend to get stuck together in people's minds.... but too me they are separate issues.

[@deborahrippe5641]

How are you getting your numbers?

[@hossammorsi482]

Sir, Q1 in the first example is 58 not 64 and Q3 is 90 not 89. plz clarify this so if i wrong i can know

[@marinstatlectures]

Hi, there are many accepted ways of calculating quartile. I what I’ve used here is that the 58 and 72 are “sharing” the space that divides the data into 1/4 below and 3/4 above, and averaged these. BUT there are other accepted ways to calculate quartile...I believe R has 9 different methods. So the concept is what’s most important...you will pretty much always use software for calculations, and really it’s most important to understand the idea of what they are, and software can do the calculation. Also worth mentioning that the difference between the different methods becomes smaller and smaller as the data set is larger.

[@rachitmakhija9703]

where can i find the detailed video about variance ?

[@marinstatlectures]

In this video I go over the concept of standard deviation (which is the square root of the variance). I focus on the concept, not the calculation: kzbin.info/www/bejne/pJ3QapqcqNRno9k

[@funkyfemme]

Watching it for psychology research statistics, and all I kept thinking is that your were writing everything backwards. Top notch recording

[@mariamkhalid4075]

no one explained SD better