Deck 11: Testing for Significance: the Chi-Square Test

Question 1

Nonparametric statistics are called that because they

A) are free of assumptions about table parameters
B) don't make assumptions requiring a specific data distribution
C) are distribution-free statistics
D) b and c above

Answer 1

D

Question 2

The larger the value of Chi-square value, the smaller the probability.

Answer 2

True

Question 3

Larber values of Chi-square indicate bigger differences.

Answer 3

True

Question 4

The larger the value of chi-square:

A) the smaller the difference between the observed and expected frequencies
B) the larger the difference between the observed and expected frequencies
C) the less the distribution approaches normality
D) none of the above

Question 5

Large sample sizes, all other things being equal, generate

A) smaller probability values for a Chi-square value
B) larger Chi-square values
C) smaller marginal differences in tables
D) larger marginal differences in tables

Question 6

If a null relationship exists in a crosstabulation table

A) there are no differences in the expected values
B) there are no differences in the observed values
C) the expected values and observed values are substantially different
D) the expected values and observed values are essentially equal

Question 7

Degrees of freedom in a Chi-square test refers to the total number of cases divided by the number of cells in a table.

Question 8

The Chi-square test is designed to be used with

A) any level of data
B) nominal-level data
C) ordinal-level data
D) interval-level data

Question 9

Interval-level data is particularly appropriate for Chi-square because it is less likely to violate assumptions.

Question 10

Chi-square gets its name because of the fact that the O - E differences are squared.

Question 11

Because sample size affects probability levels, never use a measure of significance to mean strength of relationship.

Question 12

Which of the following can cause problems in the accurate interpretation of Chi-square?

A) a low probability level
B) a small chi-square value
C) related samples
D) nominal level data

Question 13

If you have a 2x2 table and a small number of cases, the best choice of statistic below is:

A) Chi-Square Test
B) Fisher's Exact Test
C) 2x2 Combinatorial Test
D) Modified Chi-square Test

Question 14

Which of the following statistics can be used to determine whether or not there is a statistically significant relationship between two variables in a 2x3 crosstabulation table?

A) Chi-square
B) t-test
C) ANOVA
D) Fisher's Exact Test

Question 15

If you have a 2x2 table, the number of degrees of freedom is

A) 1
B) 2
C) 3
D) 4

Question 16

Fisher's Exact Test is expressly designed for large samples that Chi-square cannot handle.

Question 17

Observed frequencies in Chi-square refers to

A) the observed computed value of Chi-square as opposed to the value present in tables
B) the total number of cases observed in each category of the independent variable
C) the actual number of cases in each table cell
D) the frequency of occurrence of cells in each of the table variables

Question 18

Expected values are used in cells because they

A) are always different from observed values
B) are easier to calculate than unexpected values
C) standardize the cell sizes for the total number of cases in the sample
D) compensate for unequal row and column marginals

Question 19

Assumptions of the Chi-square test include

A) small sample sizes
B) equal number of cells in the rows and columns
C) equal marginals
D) b and c above

Question 20

Chi-square is a measure of the cumulative differences between what we expect to get and what we really get.

Question 21

Chi-square is very sensitive to violations of its assumptions.