Week 5 - Two way tables

Simpson's Paradox

2 - way tables get certain results of those results reverse when you have 3 variables.

Suppose 35% of the women in a poll support candidate A for president (and 65% do not.) In the same poll 45% of the men support candidate A for president (and 55% do not). Are gender and support (yes/no) related?

True

Which of the following questions can be answered by using a two-way table?

Whether gender of driver is related to the color being red (or not)

When comparing conditional distributions what happens if they're both the are different?

Y and X are related, there is a relationship/effect

What are the 2 ways of comparing your variables?

- Compare the conditional distribution - Compare the conditional to the marginal distribution

Suppose 35% of the women in a poll of Americans support candidate A for president (and 65% do not.) These results make up what kind of distribution?

Conditional distribution of support (yes/no) given women

What do you do when you have a joint distribution?

Divide each category by the grand total, it's the overall percentage of EACH cell. Sums to 1

What do you do when you have a conditional distribution?

Divide each category by the variable. It breaks down one variable within the values of the other.

What do you do when you have a marginal distribution?

Divide each variable by the grand total. Sums to 1

If variables A and B are related in a certain way in a two way table (with 2 variables), no matter how many other variables you look at in addition to these two, the relationship will always stay the same.

False

You can compare two marginal distributions to see if the corresponding two variables are related.

False

A __________ distribution summarizes the information from one variable ONLY, without considering ANY information from another variable.

Marginal

If 2 corresponding conditional distributions are different from each other in a two-way table, we know that the variables are related.

True

Which is the most informed data set?

The one that is further broken down

When comparing conditional distributions what happens if they're both the same?

Their is NO effect

A bar graph using StatCrunch is a good way to visualize a marginal or a conditional distribution.

True

A two-way table has this name because it contains rows going one way, and columns going the other way.

True

In the hospital example in your lecture notes, you found hospital A didn't do as well as hospital B when all the data was in one two-way table, but when an additional variable was examined, hospital A was better in all cases. What was that additional (confounding) variable that ended up reversing the results?

condition of patient (good/poor)

A two-way table allows us to examine the relationship between _____________________ variables.

two categorical