Chapter 6 Flashcards
Why do you use samples? And why do you use Populations?
Samples: Samples are easier to collect data but are more biased
Populations: Populations are less biased but harder to collect data
What is the central limit theorem?
it makes the probability calculations behind most inferential tests possible. The Central Limit Theorem addresses three aspects of sampling distributions: their shapes, their means, and their standard deviations.
What are the 6 steps of making a sample distribution?
1.Choose your population.
2.Choose a sample size n.
3.Randomly sample with replacement. That is, one score is chosen at random – which is actually a very complicated process – and then put back into the sample, which means that it could be sampled again. This may sound strange but it keeps the sample size constant and makes probability calculations much easier.
4.Calculate the statistic of interest for your sample. This is often the mean but it can be anything at all, such as the median or the interquartile range or whatever.
5.Do it again an infinite number of times. That’s how you know this is really a theoretical process and not one that is carried out empirically in standard statistical practice.
6.Create a distribution of your sample statistics. This is called a sampling distribution because it was developed through a sampling process. Also, it is called a sampling distribution to distinguish from the raw score distribution from whence it came.
What statistics can have sampling distributions?
ANY SAMPLE STATISTIC
What are the 3 different names for sampling distributions?
*The sampling distribution of means (which is the most common)
*The sampling distribution of standard deviations
*The sampling distribution of kurtosis
What is the typical shape for a sampling distribution?
The shape of a sampling distribution is typically bell-shaped, or normal, regardless of the shape of the population distribution.
If the original distribution is a perfect bell curve What is the value of n (the sample size)?
n =1
Does the raw scores shape effect the shape of the sampling distribution as the value of n is increasing to make the shape more normal (bell curved)?
NO THEY DO NOT
What needs to happen to get the sample distribution and the population distribution to be the same shape?
n=1
If the sample distribution is super skewed (negatively or positively) then what needs to happen to get the sample distribution to become normal (more bell curved)?
n (the sample size) NEEDS TO GET BIGGER AND INCREASE
How does the mean of the sampling distribution relate to stats being sampled? And name some examples
Whatever you are sampling the sampling distribution will equal the original value
Examples: If you sample the mean the mean of the sampling distribution equals the original populations mean
The mean of the sampling distribution equals the original populations variance (if you are sampling the variance)
What happens if moral or ethical responsibilities fail to represent the population while gathering data?
It will lead to an error NOT A DEVIATION
What happens to the standard error, sampling distribution, and the shape of the sampling distribution as n increases? (the sample size)
Standard error: gets smaller
Sampling distribution: gets narrower
Shape: gets more like a bell curve
What is the standard error?
The standard deviation of the sampling distribution, how much variation is likely to be between different samples drawn from the same population
How do you calculate the standard error?
standard deviation of the raw score population divided by the sample sized square rooted (SD/Sample size square rooted)
