Thank you so much.. Iam doing master degree in community medicine in Iraq and this part of epidemiology just made me cry.. I watched your presentation and everything was so clear..
Hey Justin, your material is just absolute gold. Thank you so much for maintaining and sharing such clear tutorials. Do you use anything besides Excel? I would love to see your concepts applied in a stats workflow, say for example with Python/Pandas or R.
Great presentation. How would you go about finding, testing for possible/potential counfounding factors if, e.g., you didn't know of the effects of age in rate of disease?
a bit confused on how exactly you get the mortality rates in the standard population. it will help a newbie like me to understand exactly how you get the mortality rates in the standard population at step 1, although you mention before step 1. Nevertheless, I find this movie very helpful. Additionally, it will help emphasising on when to use the indirect standardisation vs. direct standardisation . I am saying this because indirect standardisation may apply to a dataset that does have low numbers for age group.
34 years living in Kerala and I just find out that we have a flag. Well explained video. 👍🏻 But when should we use direct standardisation and when should it be indirect?
When the age specific death rates of two or more populations are known, we will go for direct standardisation. On the other hand , when the age specific rates of the population of interest are unknown , indirect standardisation is to be applied. Even I am seeing the flag of Kerala for the very first time [mallu here :) ]
So helpful! I wonder is it possible to use standardization to standardize other rate, such as morbidity rate or hospital admission rate? or is it only specific for standardize mortality rate? and do we always need to standardize every data we read before making interpretation? I really appreciate if you could answer my questions....
I think RUclips needs to add a feature where, if you downvote a video, you have to state why! I just don't get it. I have a question, though. In the India example (direct standardization), when computing the standardized rates by age group, isn't Kerala part of the standard population for the rest of India? Wouldn't it have been better to do this computation, **excluding** Kerala?
Fantastic video! But I think the 1.2 standardized mortality at the end of the video should be compared to India's 1.7 rather than Karala's 2.2. Because you are applying Karala's death rate over to the Indian population, the 1.2 means if India had Karala's rate, India's number would have been 1.2 instead of 1.7.
What if we want to adjust for multiple factors. For example race and sex along with age. This tabular method would become complex. So how would be deal with that?
minute:8:11 is the calculation correct for pop 2.2 and 7.1? im getting a different answer: 2.2/1000*5,900,000= 12980 and 7.1/1000 *1,400,000= 9,940 . am i doing the math correctly? or am i missing something
Is it possible to estimate an age adjusted rate when you know the crude rate of death of a whole population and the age distribution of the population, but you don't know the rates of deaths for each age group?
This is exceptionally useful information. I am curious how it might relate to a mortality rates within a population. For example, in your examples related to COVID, you are evaluating mortality rates for the entire population (there are 5 million people here and 1000 died of COVID19), rather than saying there are 5 million here and 500,000 were diagnosed with COVID19 and of those, 1000 died. How might we account for age distribution with this? Would we still use the "all in" standard population or would our new standard population become just those 500,000 who had COVID19? Thanks in advance!
Yo, bruh, just got a notification about your new video. Well done. I have a question: In linear regression, have you every experienced a dramatic increase in R2 and errors after standardising the predictor variables? That is happening to me as we speak. Any explanations?
Hi Zed, how do i connect to you? mail or anything. And also in categorical x variables-advanced regression, couldn't understand the dummy variable trap thing. what exactly is it. either of agecat 1 , 2 , 3, 4 can be one. so if we are using all the agecat variables, then only all the data points will be considered in regression right. not clear, can you explain? thaNKS
The mellow music in the background makes your presentation easier and more enjoyable to listen to.
SUPER helpful for my Master's level epidemiology course! Thank you so much!
I second that!
Thank you so much.. Iam doing master degree in community medicine in Iraq and this part of epidemiology just made me cry..
I watched your presentation and everything was so clear..
You sir, are truly awesome. Far, far better explained than in my introduction to epidemiology book. Thank you.
So clear and professional...thank you! This is a terrific supplement to my epidemiology course!
You are my hero! Thank you!!! Very helpful for my health & mortality demography class. Thank you thank you thank you!!!
You saved me buddy, I cracked an offer letter today. Thanks a lot.
really helped me understand whilst trying to navigate my undergraduate Bachelor Health Science Epidemiology Unit thank you!
Excellent video, completely answered the questions i had.
Hey Justin, your material is just absolute gold. Thank you so much for maintaining and sharing such clear tutorials. Do you use anything besides Excel? I would love to see your concepts applied in a stats workflow, say for example with Python/Pandas or R.
thanks for making this so much easier to understand compared to how my lecturer explained
Exactly, similar analysis, me and my wife (statistical) have discussed. Thank you very much for detailed explanation.
That is a super clear explanation!!!!
Thanks for an amazing video. if you didn't put up this video, i would have believed what media says about kerala vs India.
Thank you sooo much for this video. It helped me understand way better for Epidemiology
Hugely helpful, the explanation and graphics made it really clear. Thank you!
This guy is the best better than my teacher🔥😂
Cameroon ECID, Hello,
Clear and concise... Thank you
Good video, good diagrams. Appreciate your effort ZED. Will be going through the rest of the series.
Nicely explained . Thanks
Great presentation. How would you go about finding, testing for possible/potential counfounding factors if, e.g., you didn't know of the effects of age in rate of disease?
Excellent! Very clear explanation with COVID-19 as an example made the talk very contemporary and understandable to all. Thank you.
Thank you so much, it's very easy and understandable once i watch this video and I love how you explain it. Big thanks!
How would the calculation look like if I wanted to standardise for another variable e.g. male-female-distribution in the comparing countries?
Sooo helpful! Love how you explained it! Thank you!!
Stunning stuff!
The best explanation ever!!
Nice explanation, this will help my test.
a bit confused on how exactly you get the mortality rates in the standard population. it will help a newbie like me to understand exactly how you get the mortality rates in the standard population at step 1, although you mention before step 1. Nevertheless, I find this movie very helpful. Additionally, it will help emphasising on when to use the indirect standardisation vs. direct standardisation . I am saying this because indirect standardisation may apply to a dataset that does have low numbers for age group.
34 years living in Kerala and I just find out that we have a flag. Well explained video. 👍🏻
But when should we use direct standardisation and when should it be indirect?
When the age specific death rates of two or more populations are known, we will go for direct standardisation.
On the other hand , when the age specific rates of the population of interest are unknown , indirect standardisation is to be applied.
Even I am seeing the flag of Kerala for the very first time [mallu here :) ]
Justin I wish you could start a series about time series ARIMA model!!!
So helpful! I wonder is it possible to use standardization to standardize other rate, such as morbidity rate or hospital admission rate? or is it only specific for standardize mortality rate? and do we always need to standardize every data we read before making interpretation? I really appreciate if you could answer my questions....
Yes possible for any numeric variable
I think RUclips needs to add a feature where, if you downvote a video, you have to state why! I just don't get it.
I have a question, though. In the India example (direct standardization), when computing the standardized rates by age group, isn't Kerala part of the standard population for the rest of India? Wouldn't it have been better to do this computation, **excluding** Kerala?
YOU'RE A GENIUS! THANK YOU!!!
Recommended book for demography and population studies?
Fantastic video! But I think the 1.2 standardized mortality at the end of the video should be compared to India's 1.7 rather than Karala's 2.2. Because you are applying Karala's death rate over to the Indian population, the 1.2 means if India had Karala's rate, India's number would have been 1.2 instead of 1.7.
I think so too.
What if we want to adjust for multiple factors. For example race and sex along with age. This tabular method would become complex. So how would be deal with that?
Hi Zed,
Thank you for sharing. What do you infer, If the crude mortality rate is less than the Standardised mortality rate?
minute:8:11 is the calculation correct for pop 2.2 and 7.1? im getting a different answer: 2.2/1000*5,900,000= 12980 and 7.1/1000 *1,400,000= 9,940 . am i doing the math correctly? or am i missing something
Is it possible to estimate an age adjusted rate when you know the crude rate of death of a whole population and the age distribution of the population, but you don't know the rates of deaths for each age group?
This is exceptionally useful information. I am curious how it might relate to a mortality rates within a population. For example, in your examples related to COVID, you are evaluating mortality rates for the entire population (there are 5 million people here and 1000 died of COVID19), rather than saying there are 5 million here and 500,000 were diagnosed with COVID19 and of those, 1000 died. How might we account for age distribution with this? Would we still use the "all in" standard population or would our new standard population become just those 500,000 who had COVID19? Thanks in advance!
Yo, bruh, just got a notification about your new video. Well done. I have a question: In linear regression, have you every experienced a dramatic increase in R2 and errors after standardising the predictor variables? That is happening to me as we speak. Any explanations?
How would I compare deaths by age from one year to the next? The age groups always stay the same, but my populations grow and deaths grow as well.
Nicely explained!!!!
Wonder how to identify the cause(s) of the significant difference between groups?
Hi Zed, how do i connect to you? mail or anything. And also in categorical x variables-advanced regression, couldn't understand the dummy variable trap thing. what exactly is it. either of agecat 1 , 2 , 3, 4 can be one. so if we are using all the agecat variables, then only all the data points will be considered in regression right. not clear, can you explain? thaNKS
Clear explanation! Thanks!
Super helpful, thanks 😊
nice explanation
Hi I'm a demography student from KERALA, India
This is a good tutorial. Please dont post subtitle above the calculations. It's a little uncomfortable. Thanks for the explanation. Tc
Where'd you get the intro music from?
Why is indirect method not preferred? i.e. comparison after indirect method is not valid
what is the general population? white people? and what are the confounding factors?
I think the general population is everyone, including the test population.
You're amazing!
It totally clear my confusion.
You are my savior
Nice example !
Is this the same as normalization?
your disclaimer at 13:20 didn't age very well... lol
Yikes. Yeah, they were innocent times.
#am_from #kerala
great video .....
Thank you!
Thank you!!
Thanks
Much appreciated
Very helpful :')
Tx sir