Test #1
As evidence against the null hypothesis being true increases, what happens to the test statistic?
It gets larger. The larger the test statistic the further the result is from the null hypothesis.
Liz gets a sample and calculates a 95% confidence interval for the proportion of movie-goers who purchase from the concession stand. It is 30%-45%. Steven is going to replicate Liz’s study (he is going to take a sample of the same size from the same population. True or False: There is a 95% chance that Steven’s sample average will be in the interval 30%-45%.
False
If you have a choice between using an independent samples t-test or related samples t-test where there is a large carry-over effect, which test is best to use?
Use the independent samples. Vaguely correct is better than consistently wrong.
Why is it impossible for independent tests to have order effects?
If it is different people, they can’t experience order in different treatments.
What is a disadvantage of Gamma compared to Somers’ d?
It inflates the association by not considering when Y doesn’t change.
Explain what determines the probability of type 1 error.
The Power
Why are the tails of a t-distribution “fatter” than the tails of a normal curve?
The standard deviation of the population is often underestimated which “inflates” it.
True or False: Independent t-test have order and carry-over effects.
When is it appropriate to use a Wilcoxon test?
When it is an interval/ratio variable with two variables and two categories. It is a related sample and violates our assumption of normality.
What is meant by an asymmetric measure of association?
Using X to predict Y is different from using Y to predict X. (X and Y can be independent/dependent variables too!)
In a test to see if a certain fifth-grade class can read faster than average, the teacher concludes that the student don’t do bettern than average when they actually do. What kind of error is this?
Type II error
Why are the critical values for a t-test pushed farther out in the tails (when compared to a z-test)?
Because we want to keep the probability of type I error equal to the alpha significance level.
What problem occurs if you assume that you have homogeneity of variance, when you in fact don’t have it?
We are increasing the chance error. We must determine if we are using heterogeneity of variance or if we violate our assumptions of normality and use a Mann-Whitney test instead.
Why are independent tests less powerful than related samples tests? Specifically, what causes the more uncertain estimates?
There can be a variability due to the fact that the samples are different. Any observed difference could be due to differences in the samples.
Brian is measuring the association between stress (on a scale of 1-5) and Happiness (on a scale 1-5). What measure of association should he use?
Gamma or Somers’ d
A researcher is interested in the relationship between pet ownership and recovery from illness. The researcher sampled 50 pet owners among hospital patients diagnosed with the flu. He records what breed of pet they own, whether the pet cuddles with its owner, and how many days their flu lasts.
none of the above
Tony is testing to see if people who go to a lot of movies are smarter concerning film history than the average person. She asks them how many actors they can list who starred in the film “Avengers: Endgame”. This is her only question. She anticipates that, on average, people will know about ten actors. She has a sample of two thousand people. Most people can name only one, two, or three actors, but quite a few could name 25 or more actors. Not many actually named ten. What kind of hypothesis testing do we use?
t-test for means
Cheryl is testing to see if attending LARC affects test scores. The data is below. The numbers are test score. [two independent samples t-test without homogeneity of variance]
Attends LARC: 120 137 132 127 140 130 145 133 134
Doesn’t attend LARC: 120 60 99 81 90 102 96 111 135
an independent samples t-test with heterogeneity of variance
A psychologist want to know if hand dominance (left-handed v right-handed) is related to preferred learning style (visual, auditory, kinesthetic). What test should be used?
Chi-square test for independence (and a z-test is not appropriate)
The UCI Study Abroad Center is interested in whether UCI students are different from US college students in where they study abroad. They thus take a sample of 500 UCI students. It is known that among US college students, 60% study abroad in Europe, 20% in Asia & Oceania, 10% in the Americas, and 10% in the Middle East and Africa. Among the sample of UCI students, 60% study abroad in Europe, 33% in Asia & Oceania, 5% in the Americas and 2% in the Middle East and Africa.
Chi-square goodness of fit (and a z-test is not appropriate)