Deck 16: Cross Tabulation and Chi-Square

Question 1

The table value for chi-square at the .05 level of significance is 3.84 with one degree of freedom and 5.99 with two degrees of freedom.Our calculated chi-square value in a table with 2 rows and 3 columns is 4.50.Our chosen level of significance is .05.What does this mean?

A)Our results are statistically significant. *
B)Our results are NOT statistically significant.
C)We had one degree of freedom.
D)Our results are statistically significant and we had one degree of freedom.

Answer 1

Our results are NOT statistically significant.

Question 2

The correction for continuity is used whenever our cross tabulation table has more than two rows and two columns and has any expected frequencies more than 20.

Answer 2

False

Question 3

When applying the chi-square formula without the correction for continuity you should:

A)Add the squared differences between each observed and expected frequency.
B)Add the squared differences between each observed and expected frequency and then divide by N.
C)Add the differences between each observed and expected frequency.
D)Add the differences between each observed and expected frequency and then divide by N.

Answer 3

Add the squared differences between each observed and expected frequency and then divide by N.

Question 4

Two cells in a 2 by 2 table contain expected frequencies o4.The other two cells contain expected frequencies of 6.Which of the following statements is/are true about this table?

A)The Correction for Continuity should be used in testing the statistical significance of the results.
B)The Fisher's Exact Test should be used in testing the statistical significance of the results.
C)Pearson's chi-square statistic should be used in testing the statistical significance of the results.
D)None of these.

Question 5

The Fisher's Exact Test is used whenever any cells in our 2 by 2 cross-tabulation table have an expected count of less than 5.

Question 6

The chi-square formula involves adding in every cell the expected frequency plus the observed frequency and then dividing the sum by the observed frequency.

Question 7

Each cell in a 2 by 2 table contains 20 observed cases.Which of the following statements is/are true about these results?

A)The results are statistically significant.
B)There are 19 degrees of freedom
C)The chi-square statistic equals zero.
D)The Fisher's Exact Test should be used.

Question 8

Cross tabulation tables show whether and how two or more nominal-level variables are related.

Question 9

To obtain the expected frequency for a cell in a two-by-two table, we:

A)Multiply the column total by the row total and then divide the product by N.
B)Divide the column total by the row total and then multiply by N.
C)Divide the row total by the column total and then multiply by N.
D)Multiply the column total by the row total and then divide the square root of N.

Question 10

It is appropriate to use parametric tests of statistical significance when:

A)At least one variable in the test has an interval or ratio level of measurement.
B)The tested parameters of the interval or ratio variables are distributed normally in the population.
C)The groups being compared are independent of each other.
D)All of these.

Question 11

The goodness of fit test should be used when:

A)We have more than two variables.
B)We have exactly two variables.
C)We want to compare sample proportions on a single variable to proportions known to exist in the broader population.
D)When we have 10 or more degrees of freedom.

Question 12

Chi-square is a nonparametric test.

Question 13

The number of degrees of freedom in a cross tabulation table is one less than the number of rows times one less than the number of columns.

Question 14

McNemar's test should be used when:

A)We have more than two variables.
B)We have a nominal dichotomous independent variable and a severely skewed dependent variable that is at the ordinal, interval or ratio level of measurement.
C)We want to examine change from pretest to posttest within one group on a nominal variable.
D)We want to compare sample proportions on a single variable to proportions known to exist in the broader population.

Question 15

Two cells in a 2 by 2 table contain expected frequencies of 8.The other two cells contain expected frequencies of 12.Which of the following statements is/are true about this table?

A)The Correction for Continuity should be used in testing the statistical significance of the results.
B)The Fisher's Exact Test should be used in testing the statistical significance of the results.
C)Pearson's chi-square statistic should be used in testing the statistical significance of the results.
D)None of these.

Question 16

Which of the following statements is/are true about nonparametric tests of statistical significance?

A)They are preferable to parametric tests.
B)They require the same assumptions as parametric tests.
C)They require variables at the interval or ratio level of measurement.
D)None of these.

Question 17

Nonparametric tests of statistical significance assume that at least one variable in the test has an interval or ratio level of measurement.

Question 18

The median test should be used when:

A)We have more than two variables.
B)We have a nominal dichotomous independent variable and a severely skewed dependent variable that is at the ordinal, interval or ratio level of measurement.
C)We have a nominal dichotomous independent variable and a normally distributed dependent variable that is at the ordinal, interval or ratio level of measurement.
D)We want to compare sample proportions on a single variable to proportions known to exist in the broader population.

Question 19

How many degrees of freedom are there in a table with 4 rows and 3 columns?

A)6
B)7
C)8
D)12

Question 20

Nonparametric tests require all of the same assumptions as parametric tests.

Question 21

A common use of the goodness of fit test is when we want to compare our sample proportions to proportions known to exist in the broader population.

Question 22

To use the median test, we first calculate the median for the overall sample.Next, we collapse the interval or ratio data in each group into two (dichotomous) categories: 1) the scores that fall above the median, and 2) the scores that fall below the median.Then we analyze those categories using the chi-square formula.