I dont know what test is appropriate for the any problems
Stay tuned for later in the stream when we go over the inference portion of statistics. This should help you understand when tests are appropriate.
If the problem talks about means then use means. (T test)
If it talks about percentages use proportion. (Z test)
UNITS CAN HELP.
If you see just 1 sample or just 1 proportion/mean use a 1 sample test.
If you see 2 samples or something about the difference between 2 numbers use 2 sample.
Caution: check for independence so you don’t confuse matched pairs with a 2 sample t- test.
Here’s a helpful image.
When you have proportions, we use a z-test with one sample or two samples based on the situation. If you have means, the story is a little more complex. If the population standard deviation is known, we will use a z-test, but usually, we don’t know this, so we will use a t-test. When there is 1 sample, we use a one-sample t-test. If we have two-samples, there are 2 types of t-tests. If the data is in a matched-pair style in which the data shows a change over time, we use a matched-pair t-test which is a one-sample t-test in which the data is the difference between the two samples. Otherwise with 2 sample means, we use a 2-sample t-test. In a one-tailed t-test, the alternate hypothesis is that the parameter is less than or less than a null hypothesis value, and in a two-tailed t-test, the alternate hypothesis is that the parameter is not equal to a null hypothesis value. In a two-sample test, the null hypothesis value is usually 0.