Quiz 10.2
Chapter 10 - Day 6
Students Struggled With…
-
Determining if they should be using a paired t-test or a two-sample t-test. Students often defaulted to two-sample t-test since it is what they have been working with the most.
-
Acknowledging the significance level when asked if they have convincing evidence from a confidence interval. Many students just write - yes because the value is in the interval or no because the value is not in the interval.
-
Checking the 10% condition when it is not needed. Students always need to think about if sampling without replacement is happening when deciding if this condition should be checked.
Remedies…
-
When the students are taking a quiz on this, consider scaffolding this question by asking them to make a display of the "differences". This will help them to get in the right mindset of what the problem is asking. Alternatively, have them identify how many samples there are - as noticing that there is one sample generating both results may also be helpful.
-
Throw this back to the way we write our conclusions in a significance test. We always want to say what our alpha level is when determining if we have convincing evidence, and this would be no exception.
-
We would much rather have the students always check it than not check it - but it is worthwhile to have a conversation about why this condition exists. Take time to discuss with students about how knowing it is a sampling problem vs. an experiment can help determine if this condition needs to be checked.