P-values & Conclusions
null hypothesis or no difference
What does the Ho mean?
n - 1
How to calculate Degrees of Freedom?
tcdf(lower,upper,df)
How to calculate P-values using the t distributions?
Less variability.
What is the benefit of increasing the sample size?
What does the Ha mean?
invT(area,df)
How to find a t-score that matches a given percent?
Lets us account for other confounding variables and helps us be more likely to find a difference.
What's the benefit of using a matched pairs in an experiment?
S.O.C.S
Shape - Outlier - Center - Spread
What is used to describe distribution of a quantitative variable?
The odds I would get my sample or more extreme if the null is really true
calculator prompt for confidence intervals for means.
What is a T-interval?
Power
Increase the null
Increase Sample Size
Matched Pairs Experiment
What are some ways to increase the probability of avoiding a Type II error?
How find interquartile range (IQR).
What is Q3 - Q1?
If the P-value is less than the null.
When to reject the Ho?
Graph sample. Look for no skew or outliers.
How to satisfy the normal (large counts) condition if it is not stated or less than 30?
Characteristics of a Paired experiment.
Always have same samples size for both, row header, "mean difference", can't scramble a row?
How to find a z-score.
What is (one data point - mean) / standard deviation?
If the P-value is greater (or) equal to the null.
When to fail to reject the Ho?
Used for estimating population parameters for small sample sizes or unknown standard deviation.
When to use T - Interval?
Characteristics of an unpaired experiment.
Sample size can be same or different; Can scramble lists without a problem; no row header, "difference in means"?
If we observe many, many trials of a random process, the proportion of times the outcome would occur is its probability.
What is the law of large numbers?