I Know this is very late to the party, but felt I had to point this out. At the beginning H₀ :μ₁ = μ ₂, with Hₐ :μ₁ < μ ₂, The nul and alt hypothesis by definition have to be complementary events. if H₀ :μ₁ = μ ₂, then Hₐ :μ₁ ≠ μ ₂ I do realise that by taking the modulus of the test statistic value you can only have positive numbers indeed a difference can only be positive, still that part was noticeable. The video itself was great, we need more people explaining stuff, as everyone presents the same topics in slightly different manner, even if only by mannerisms or turn of phrase.
Please please please show what you are doing with the T-table. Why would you explain everything so clearly and then at the most important part casually talk it out loud as if its common knowledge. This is unbelievably frustrating when im trying to study for an exam
This was incredibly helpful. I watched a couple videos before this and when it got to the T-Table part none of it made sense to me. This explained a majority of that section very simply. The only thing I'm slightly struggling with is the cut off and rejecting the null hypotheses. Do you just reject anything that ends up being more than .05 unless otherwise specified? Nor sure, but I guess that's what I'm sticking with for now.
For a two samples t test if it is not given whether the population variances are not equal or not. Then which formula do I have to use to find degrees of freedom
It’s sample size minus 1. So for example if one of the samples is 15 and the other sample is 13, you take the smallest sample which would be 13 and subtract by 1. So it would be 12df
The formula for degrees of freedom in a two-sample t-test (df=n1+n2-2) assumes that the variances of the two groups are equal. In some situations, particularly when the sample sizes are unequal and/or the assumption of equal variances is questionable, researchers may opt to use a modification of the formula that adjusts the degrees of freedom based on the sample variances. One common modification is the Welch's t-test, which uses a more complex formula for degrees of freedom (you can search it up, im not typing all that). The Welch's t-test provides a more accurate assessment when the assumption of equal variances is in doubt or when dealing with unequal sample sizes. Subtracting one from the smaller group (often seen in textbooks or older statistical software) might be an attempt to approximate the Welch's degrees of freedom when the sample sizes are unequal. However, it is a simplification and doesn't fully account for the complexities introduced by unequal variances and sample sizes. In modern statistical software and practices, it's generally recommended to use the Welch's t-test when the assumption of equal variances is in question or when dealing with unequal sample sizes. @@marlonbrando6826
I Know this is very late to the party, but felt I had to point this out.
At the beginning H₀ :μ₁ = μ ₂, with Hₐ :μ₁ < μ ₂, The nul and alt hypothesis by definition have to be complementary events.
if H₀ :μ₁ = μ ₂, then Hₐ :μ₁ ≠ μ ₂
I do realise that by taking the modulus of the test statistic value you can only have positive numbers indeed a difference can only be positive, still that part was noticeable.
The video itself was great, we need more people explaining stuff, as everyone presents the same topics in slightly different manner, even if only by mannerisms or turn of phrase.
Please please please show what you are doing with the T-table. Why would you explain everything so clearly and then at the most important part casually talk it out loud as if its common knowledge. This is unbelievably frustrating when im trying to study for an exam
Is degree of freedom not n1+n2 -2?
This was incredibly helpful. I watched a couple videos before this and when it got to the T-Table part none of it made sense to me. This explained a majority of that section very simply. The only thing I'm slightly struggling with is the cut off and rejecting the null hypotheses. Do you just reject anything that ends up being more than .05 unless otherwise specified? Nor sure, but I guess that's what I'm sticking with for now.
For a two samples t test if it is not given whether the population variances are not equal or not. Then which formula do I have to use to find degrees of freedom
Please explain what the variables for the variance formulae are.
Thank you so much that was helpful!
How u have calculated the degree of freedom here ?
It’s sample size minus 1. So for example if one of the samples is 15 and the other sample is 13, you take the smallest sample which would be 13 and subtract by 1. So it would be 12df
@@marlonbrando6826 No, it's the smallest sample size minus 1. Weird, but it's just how it is.
The formula for degrees of freedom in a two-sample t-test (df=n1+n2-2) assumes that the variances of the two groups are equal. In some situations, particularly when the sample sizes are unequal and/or the assumption of equal variances is questionable, researchers may opt to use a modification of the formula that adjusts the degrees of freedom based on the sample variances.
One common modification is the Welch's t-test, which uses a more complex formula for degrees of freedom (you can search it up, im not typing all that). The Welch's t-test provides a more accurate assessment when the assumption of equal variances is in doubt or when dealing with unequal sample sizes.
Subtracting one from the smaller group (often seen in textbooks or older statistical software) might be an attempt to approximate the Welch's degrees of freedom when the sample sizes are unequal. However, it is a simplification and doesn't fully account for the complexities introduced by unequal variances and sample sizes.
In modern statistical software and practices, it's generally recommended to use the Welch's t-test when the assumption of equal variances is in question or when dealing with unequal sample sizes.
@@marlonbrando6826
i love you clutch
You didn't explain it clearly