Chapter 9: Non-Parametric Statistics and Chi-Square Tests

0.0(0)
Studied by 0 people
call kaiCall Kai
Locked
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/51

encourage image

There's no tags or description

Looks like no tags are added yet.

Last updated 12:57 AM on 7/27/26
Name
Mastery
Learn
Test
Matching
Spaced
Call with Kai
Chat

No analytics yet

Send a link to your students to track their progress

52 Terms

1
New cards

What are the two primary requirements for using parametric tests like the t-test or F-test?

Normality (or large samples) and interval level data.

2
New cards

Why are tests like the t-test and F-test specifically categorized as 'parametric' tests?

They assume there are population parameters to be estimated and tested.

3
New cards

Statistical tests that do not require population parameter assumptions are known as __ tests.

Non-parametric

4
New cards

Which non-parametric test is used to determine if the frequency distribution of a single variable follows a specified (typically uniform) distribution?

One-Way Chi-Square Test.

5
New cards

Observed frequency (fₒ)

The actual or true frequency distribution obtained by counting.

6
New cards

Expected frequency (fₑ)

The anticipated frequency based on the assumption that the null hypothesis is true.

7
New cards

What is the most commonly used non-parametric statistic?

Chi-square (χ²).

8
New cards

The Chi-square test allows researchers to test the significance of the difference between which two values?

Observed frequency (fₒ) and expected frequency (fₑ).

9
New cards

What is the null hypothesis (H₀) for a Chi-square test regarding frequencies?

H₀: fₒ = fₑ

10
New cards

Formula: One-Way Chi-Square (χ²)

χ² = (fₒ - fₑ)² / fₑ

11
New cards

Formula: Degrees of freedom (df) for a One-Way Chi-Square test

df = k - 1

12
New cards

In the formula df = k - 1, what does the variable k represent?

The number of categories.

13
New cards

According to the Chi-square procedure, where is the critical value (χ²_c) typically found?

Table E.

14
New cards

Under what condition regarding the Chi-square statistic (χ²) and the critical value (χ²_c) should the null hypothesis be rejected?

If χ² > χ²_c

15
New cards

Under what condition regarding the Chi-square statistic (χ²) and the critical value (χ²_c) should the null hypothesis be retained?

If χ² ≤ χ²_c

16
New cards

What is the primary function of a Two-Way Chi-Square Test?

To determine whether two categorical variables are associated or not.

17
New cards

In a Two-Way Chi-Square test, what tool is used to cross-tabulate the two categorical variables?

A contingency table.

18
New cards

Formula: Degrees of freedom (df) for a Two-Way Chi-Square test

df = (r - 1)(c - 1)

19
New cards

In the Two-Way Chi-Square df formula, what do the variables r and c represent?

The number of rows (r) and the number of columns (c).

20
New cards

Formula: Expected frequency (fₑ) for a specific cell in a contingency table

fₑ = (row marginal total × column marginal total) / N

21
New cards

What is the underlying rationale for the expected frequency (fₑ) formula in a Two-Way Chi-Square test?

It assumes the two variables are not related (independent).

22
New cards

When using Chi-square, researchers should be wary of expected frequencies that fall below what value?

Five (5).

23
New cards

What is a recommended action for small samples to prevent misinterpretation due to low expected frequencies?

Collapse categories if it is logical to do so.

24
New cards

Under what specific range of expected values (fₑ) should Yate's correction be applied?

When fₑ is greater than five but less than ten.

25
New cards

Why is Yate's correction used when cell frequencies are smaller than 5?

To prevent the Chi-square test result from being erroneous.

26
New cards

How does the application of Yate's correction impact the final size of the Chi-square statistic?

It reduces the size of the statistic.

27
New cards

Yate's correction is specifically designed for use with which type of contingency table?

2 × 2 tables.

28
New cards

Formula: Chi-Square with Yate's correction

χ² = (|fₒ - fₑ| - 0.5)² / fₑ

29
New cards

What level of measurement is required for the variables in a Chi-square test of significance?

Nominal data.

30
New cards

What sampling method is a requirement for the use of the Chi-square test?

Random sampling.

31
New cards

One requirement for Chi-square is that the __ cell frequencies should not be too small.

Expected

32
New cards

Which non-parametric test is appropriate for use with ordinal data?

The Median Test.

33
New cards

Formula: Position of the Median

Position = (N + 1) / 2

34
New cards

In the Median Test, what values function as the observed frequencies (fₒ)?

The counts of numbers above and below the median.

35
New cards

Which statistical correction is typically used when calculating the Chi-square statistic for a Median Test?

Yate's correction.

36
New cards

What does the Median Test specifically compare between two or more samples?

Their medians.

37
New cards

What is the measurement level requirement for the Median Test?

Ordinal data.

38
New cards

How does the power of non-parametric procedures generally compare to parametric tests?

Non-parametric procedures are less powerful.

39
New cards

True or False: The calculation process for a Two-Way Chi-Square test remains the same even if the table has more than two columns or rows.

True.

40
New cards

The assumption that the joint frequency is the product of two marginal frequencies allows for the calculation of __.

Expected frequency (fₑ)

41
New cards

What is the first step in the general procedure for conducting a Chi-square test?

Establish the null hypothesis (H₀) and the research hypothesis (H₁).

42
New cards

Non-parametric tests are particularly useful when __ requirements, such as normality, cannot be met.

Parametric

43
New cards

Concept: fₒ

Observed frequency (the actual data counted).

44
New cards

Concept: fₑ

Expected frequency (the data predicted by the null hypothesis).

45
New cards

In a 2 × 2 table where N=40, if the Row 1 marginal is 25 and the Column 1 marginal is 20, what is the fₑ for the first cell?

12.5

46
New cards

If a researcher is comparing two samples with ordinal data and random sampling, which test should they use?

The Median Test.

47
New cards

If a researcher is comparing two nominal variables for association using a 3 × 4 table, what are the degrees of freedom?

6

48
New cards

Which test requires the calculation of the position of the median before counts can be used in a Chi-square formula?

The Median Test.

49
New cards

A One-Way Chi-Square test is used to see if a distribution follows a __ distribution.

Specified (or uniform)

50
New cards

Non-parametric tests are 'non-parametric' because they do not assume there are __ to be estimated.

Population parameters

51
New cards

If χ² = 5.2 and χ²_c = 5.2, what is the statistical conclusion?

Retain the null hypothesis.

52
New cards

In the context of the Median Test, 'random sampling' is listed as a __ for its use.

Requirement (or condition)