Chi Square ( χ2 test of independence)

a

A study is designed to test whether there is a relationship between natural hair color and eye color. If a significant Chi-Square value is obtained, this suggests that: Select one: a. there is a significant association between natural hair color and eye color. b. a person's hair color is determined by his or her eye color. c. there is a significant difference between average eye color and average hair color. d. there is insufficient Information to suggest an association

c

Suppose you run a chi-square test and determine that there is a significant relationship between Age group (18-24, 25-34, 35-49, 50+) and Political Affiliation (Democratic, Republican, Independent). What is an appropriate next step? Select one: a. Examine the Pearson Residuals to see which age groups have significantly higher/lower counts than would be expected if there was no relationship. b. Run several additional chi-square tests by looking at two levels of the explanatory variable and two levels of the response variable at a time and use the Bonferonni adjusted alpha level. c. Either (a) or (b) is appropriate. However, option (b) would be tedious since there are more than 2 levels of the response variable. d. Neither (a) or (b) is appropriate.

d

Which of the following scenarios would best be analyzed using a chi-square test of independence? Select one: a. Determining if a smoking cessation program reduces smoking behavior in a sample of patients. b. Examining the distribution of grades in a class. c. Determining whether males or females have a higher mean score on a test of interpersonal sensitivity. d. Testing whether more males or females are coffee drinkers versus non-coffee drinkers.

sample, whole population

as in testing men and women and nonbinary on income, in statistical analysis, you cant test every man and every woman and every nonbinary person, so we take a __________ and try to see if it represents the ___________ of men and women and nonbinary folk

hypothesis, null, P-value, conclusion in context

four steps of chi squared an I think other stat tests... 1. say H0... gender and drunk status are not related, and HA.... stating the ______________ 2. seeing how far the numbers in terms of males and females and their drunk status, compares to the ______ hypothesis that these two categorical variables are independent of one antother 3. to see how likely our results would be if H0 were true so if gender and drunk status were independent and not related, we calculate the __________________- 4. finally, last step is to ___________ for example the SCOTUS said that since the P value was 0.201, and based on this test, the law making it legal for women but not for men doesn't make sense.

Bonferroni adjustment

if we are chi squaring color of cigarette used (blue, orange, black, or yellow) to see if it is the explanatory variable for a TWO LEVELED response variable of if the person is addicted to nicotine (yes or no), then we will use the post hoc test called the _____________________. to see which of the four leveled explanatory variables of cigarette color, we will compare blue to orange, blue to black, orange to black, orange to black, so we will compare all of the different explanatory variable categories. so 0.05/number of comparisons we are making.

expected count, multiply, p(drunk), 409

in order to find the ___________ of how many men would be drunk, and how many women would be drunk, if H0 were true, we ______________ P(men) and _______ and same for women So if 93 men and women were drunk, and there were 619 people in all, and there were 409 men in all, to find expected count we'd do 93/619 * _______/619

Pearson Residuals

post hoc test we'd use when we chi squaring data that has explanatory variable of pants type (jean, khaki, shorts, long underwear) and a response variable of MORE THAN TWO LEVELS of what kind of chafing in the leg they experience (blue chafing, red chafing, orange chafing) pearson residuals greater than 2 represent relatively high counts and values less than -2 represent relatively low counts.

independent

the null hypothesis (H0), saying that there prolly not a relationship between gender and whether you are a bike owner, means that these two CATEGORICascladla variables are _______________ of one another

explanatory, conditional percentages

the two steps in explanatory data analysis... 1. i.e., finding out which of gender and drunk driving is the __________ variable 2. i.e. finding that this % of males were drunk driving, this $ of females were drunk driving, so finding ___________, which are the % of the response variable for each value of the explanatory variable