Okay, let's tackle this question about the chi-square test. The question is asking which statement is not true about the chi-square test. Since the options aren't provided, I need to recall common misconceptions or incorrect statements associated with the chi-square test.

First, the core concept here is understanding the chi-square test's purpose and limitations. The chi-square test is a statistical method used to determine if there's a significant association between categorical variables. It's often used in contingency tables and for goodness-of-fit tests. The test assumes that the data is nominal, the observations are independent, and the expected frequencies are sufficiently large (usually at least 5 in each category).

Now, thinking about common incorrect statements. One frequent mistake is that the chi-square test can be used on continuous data, but it's actually for categorical data. Another is that it can handle small sample sizes well, but in reality, small expected frequencies can invalidate the test. Also, some might think it's appropriate when the sample size is too small, which isn't the case. Another misconception is that it can test for causation, but it only tests for association.

So, the correct answer would be an option that states something like "Chi-square test is suitable for small sample sizes with expected counts below 5" or "Chi-square test can establish causation between variables." These are incorrect because the test requires adequate expected frequencies and only shows association, not causation.

The other options (the incorrect ones) would be distractors that correctly describe the chi-square test. For example, an option stating that it's used for categorical data is correct. Another correct statement is that it's non-parametric. So, the incorrect options would be the ones that contradict these points.

Clinical pearl: Remember that the chi-square test requires that no more than 20% of the expected counts are less than 5, and all individual expected counts are at least 1. Also, when expected counts are low, Fisher's exact test is preferred.

**Core Concept**
The chi-square test is a statistical method used to assess associations between categorical variables or goodness-of-fit. It relies on comparing observed frequencies to expected frequencies under the null hypothesis, assuming independence or a specified distribution.

**Why the Correct Answer is Right**
The chi-square test **cannot** be used when expected frequencies in any cell of a contingency table are less than 5 (as per the rule of thumb), as this violates the test's assumptions. Small sample sizes or sparse data lead to inaccurate results, necessitating alternatives like Fisher’s exact test. Additionally, the chi-square test detects associations but **does not imply causation**.

**Why Each Wrong Option is Incorrect**
**Option A:** *"Chi-square test is non-parametric."* – This is correct; it does not assume a specific distribution for the data.
**Option B:** *"Chi-square test is used for categorical data."* – Correct; it analyzes nominal or ordinal variables.
**Option C:** *"Chi-square test measures strength of association."* – Incorrect; it only tests for statistical significance, not strength (Cramer’s V or odds ratio quantify strength).

**Clinical Pearl / High-Yield Fact**
Always check expected frequencies before using the chi-square test. If more than 20% of cells have expected counts <5, apply Fisher’s exact test instead. Never infer causation from a chi-square result—association

✓ Correct Answer: C. Directly measures the strength of association