Chapter 5
What is a mean deviation score?
The difference between an individual score and the mean
What is an alpha level?
Threshold of probability below which we interpret our results as coming from a different population, an effect of our IV on our DV, or a difference in sample means instead of coming through sampling variation/probability
A confidence interval is ______.
a statistical interval around a point estimate (the mean) that provides a specified level of confidences for the true population parameter
What is sampling variation?
Differences between the sample and population that are due to chance
What do measures of dispersion tell us?
The variation, or variability in our data. How similar are our data points to each other? What are the differences among our data points? How heterogenous are our data?
What does a Z score tell us?
How many standard deviation units away from the mean a certain score is
If our level of confidence increase then what happens to the width of our confidence interval?
It gets larger
What is the difference between a one and two tailed test?
A one tailed test contains directionality (IE specifically larger or smaller/greater or less)
Which two measures of dispersion can we find using the average of the mean deviation scores?
The variance and standard deviation
What is a probability distribution?
It is the distribution of possible future outcomes based on the probability of each outcome occurring
A confidence interval is equal to point estimate ± -----
margin of error
What does this formula give us?
x̄-μ/ [σ/sqrt(n)]
A Z-score for our sample mean compared to the population distribution
You randomly select 15 numbers from 1-1000. What would you predict about the heterogeneity of this data set?
We would expect high variance and large differences between our data points. We would expect large heterogeneity.
If I have a dataset with a mean of 8 and a normal distribution, which of these two values would we be less likely to select through sampling variation/probability?: 3 or 11
3
What is the difference between a sample statistic and a population parameter?
A population parameter is a true value because we know every case. A sample statistic does not represent every case and thus may differ from our population parameter (even if it is from the same distribution)
When performing hypothesis testing regarding a sample and a population we are trying to determine what regarding their respective distributions?
We are trying to determine if our sample comes from the same population distribution as the population, or if it represents its own distribution
In words, not statistical notation, explain how you would find the standard deviation of a data set?
Find the differences between each data point and the mean (mean deviation score), square each mean deviation score and sum them all together, divide by the sample size and then take the square root.
The normal distribution is a probability distribution. Because of this, the area under the curve represents what percentage?
1 or 100% of cases will fall under the normal distribution
If I am investigating the average beaver weight in Washington and I gather a sample of 30 beavers and generate a 95% confidence interval from 28-34 lbs. interpret those results in regards to beavers in Washington
Based on our sample we are 95% confident that the TRUE average beaver weight (population parameter) is between 28 and 34 lbs
2 Part question:
If I have a population mean μ =10 and I generate two samples from that population
1:
x̄ =13
n=5
2:
x̄=13
n=50
Which would we be less likely to find through sampling variation and why?
Less likely to find the 2nd sample. As sample size increases we expect our sample to be more representative of the population and thus to become closer to the mean.