Bar Charts
What do the frequency counts add up to on a frequency table?
The total number of observations in the study.
What is the equation to calculate the chi square statistic?
sum of (observed-expected)^2/expected
What type of variable is chi square used to test?
Categorical
How do you determine expected counts of a category when given a proportion?
Multiply your total count by the expected proportion given.
What do the relative frequencies add up to on a relative frequency table?
100%, not the total count
What are the degrees of freedom for a chi square test?
Number of categories minus 1
What is the null hypothesis for a chi square goodness of fit test?
The observed counts in each category equal the expected counts of each category.
What is the number of categories we generally deem appropriate when displaying data using a pie chart?
3 or fewer categories
When reading relative frequencies of those eating no salad, eating 1 to 3 times per week, 4 to 6 times per week, and every day, how do you find the number of people who eat at least one salad a week?
Add the number of people who eat 1 to 3, then 4 to 6, and then everyday together.
How do you determine relative frequency?
At the end of a chi square test, if the null hypothesis states the distributions were equal with no preferences to a specific category, what can we say if we reject the null hypothesis?
We say that there is a preference for one (or more) categories, or that there is a significant difference in the expected proportions. The distributions are unequal.
What order should the categories in a bar chart be listed?
The order doesn't matter!
What is the difference between a bar chart and a histogram? List at least 2 differences.
Bar chart is for categorical variables, bars do not touch, and are in any order. Histogram is for numerical variables, bars touch, and have values in numerical order.
When comparing the chi square statistic to the critical chi square value, if your statistic is lower than the chi square value, what do you conclude?
We fail to reject the null hypothesis and conclude that there is no significant difference between the observed and expected counts.
What are the three assumptions for a chi square test?
Random samples, independent observations, and all expected frequencies should be at least 5.
Do you round to the nearest whole number when determining expected counts?
NOOOO!
Is it possible for the chi square statistic to be negative? Explain.
No, chi square is a sum of differences squared divided by counts, so it is impossible for chi square to be negative.