analysing categorical data Flashcards
(21 cards)
What is the purpose of a Chi-Square (χ²) test?
To test whether the observed frequencies in categories differ from what we would expect by chance.
What is the null hypothesis (H₀) in a Chi-Square test?
That the observed frequencies are due to chance.
What is the alternative hypothesis (H₁) in a Chi-Square test?
That the observed frequencies reflect real differences between categories.
What are the two key assumptions of a Chi-Square test?
1) Independence of observations; 2) Expected frequencies must mostly be ≥ 5 and all > 1.
What happens if the assumptions of the Chi-Square test are violated?
The power of the test is reduced.
How can you address violations of expected frequencies in a Chi-Square test?
Use an exact test (e.g., Fisher’s), collapse categories, remove sparse levels, or collect more data.
What is a contingency table?
A table showing the frequency of occurrences for combinations of categories.
How do you calculate degrees of freedom for a Chi-Square test with one variable?
Degrees of freedom = number of categories − 1.
What does it mean if χ² observed > χ² critical?
The null hypothesis is rejected; the result is statistically significant.
How do you calculate expected frequency for two variables in a contingency table?
(Row total × Column total) ÷ Grand total.
What SPSS menu path is used for a Chi-Square test with one variable?
Analyze → Nonparametric Tests → Legacy Dialogs → Chi-Square.
What SPSS menu path is used for a Chi-Square test with two variables?
Analyze → Descriptive Statistics → Crosstabs.
What does the “Observed Count” in a Chi-Square output represent?
The actual number of observations in a category.
What does the “Expected Count” in a Chi-Square output represent?
The number of observations expected under the null hypothesis.
What is Cramer’s V or Phi used for in Chi-Square tests?
Measuring effect size or strength of association between variables.
How would you report a Chi-Square test with one IV in APA style?
“χ²(df) = value, p = value” (e.g., χ²(1) = 5.4, p = .02).
How would you report a Chi-Square test with two IVs and effect size in APA style?
“χ²(df) = value, p = value, Cramer’s V = effect size” (e.g., χ²(1) = 5.44, p = .02, Cramer’s V = .301).
When should you use a binomial test instead of a Chi-Square test?
When your variable has only two categories.
What SPSS path is used for a binomial test?
Analyze → Nonparametric Tests → Legacy Dialogs → Binomial.
How do you report a binomial test result in APA style?
“A binomial test indicated no significant difference (n₁ = x, n₂ = y, p = value).”
What is the key difference between a Chi-Square and Binomial test?
Chi-Square is used for variables with more than two levels or two variables; Binomial is for single variables with exactly two levels.
