Hypothesis Testing Null and Alternative Hypotheses (One‑tailed, Two‑tailed)

Concept Summary

• A null hypothesis is a default statement that there is no significant difference or relationship between variables, often denoted as H0.
• An alternative hypothesis is a statement that there is a significant difference or relationship between variables, often denoted as H1.
• Null and alternative hypotheses are used in hypothesis testing to determine whether a sample data supports or rejects a specific claim.
• One-tailed hypotheses are used when the direction of the effect is predicted, while two-tailed hypotheses are used when the direction of the effect is not predicted.
• The choice between one-tailed and two-tailed hypotheses depends on the research question and the expected direction of the effect.

Questions

WHAT (definitional)

1. What is a null hypothesis?
2. Answer: A null hypothesis is a default statement that there is no significant difference or relationship between variables.
3. Real-world example: In a study on the effect of exercise on blood pressure, the null hypothesis might be that exercise has no effect on blood pressure.

4. Misconception cleared: The null hypothesis is not a statement of what we expect to find, but rather a statement of what we assume to be true before collecting data.

5. What is an alternative hypothesis?

6. Answer: An alternative hypothesis is a statement that there is a significant difference or relationship between variables.
7. Real-world example: In a study on the effect of exercise on blood pressure, the alternative hypothesis might be that exercise has a significant effect on blood pressure.

8. Misconception cleared: The alternative hypothesis is not a statement of what we expect to find, but rather a statement of what we hope to find if our research is correct.

9. What is the difference between one-tailed and two-tailed hypotheses?

10. Answer: One-tailed hypotheses are used when the direction of the effect is predicted, while two-tailed hypotheses are used when the direction of the effect is not predicted.
11. Real-world example: In a study on the effect of a new medication on blood pressure, a one-tailed hypothesis might be that the medication lowers blood pressure, while a two-tailed hypothesis might be that the medication has an effect on blood pressure.
12. Misconception cleared: One-tailed hypotheses are not more powerful than two-tailed hypotheses, but rather they are used when the direction of the effect is predicted.

WHY (causal reasoning)

1. Why do we need to specify a null hypothesis?
2. Answer: We need to specify a null hypothesis to provide a clear statement of what we assume to be true before collecting data, and to determine whether our sample data supports or rejects this statement.
3. Real-world example: In a study on the effect of exercise on blood pressure, specifying a null hypothesis allows us to determine whether our sample data supports or rejects the statement that exercise has no effect on blood pressure.

4. Misconception cleared: The null hypothesis is not a statement of what we expect to find, but rather a statement of what we assume to be true before collecting data.

5. Why do we need to specify an alternative hypothesis?

6. Answer: We need to specify an alternative hypothesis to provide a clear statement of what we hope to find if our research is correct, and to determine whether our sample data supports or rejects this statement.
7. Real-world example: In a study on the effect of exercise on blood pressure, specifying an alternative hypothesis allows us to determine whether our sample data supports or rejects the statement that exercise has a significant effect on blood pressure.

8. Misconception cleared: The alternative hypothesis is not a statement of what we expect to find, but rather a statement of what we hope to find if our research is correct.

9. Why do we need to choose between one-tailed and two-tailed hypotheses?

10. Answer: We need to choose between one-tailed and two-tailed hypotheses based on the research question and the expected direction of the effect.
11. Real-world example: In a study on the effect of a new medication on blood pressure, choosing a one-tailed hypothesis allows us to determine whether the medication lowers blood pressure, while choosing a two-tailed hypothesis allows us to determine whether the medication has any effect on blood pressure.
12. Misconception cleared: One-tailed hypotheses are not more powerful than two-tailed hypotheses, but rather they are used when the direction of the effect is predicted.

HOW (process/application)

1. How do we specify a null hypothesis?
2. Answer: We specify a null hypothesis by stating what we assume to be true before collecting data, often in the form of a statement that there is no significant difference or relationship between variables.
3. Real-world example: In a study on the effect of exercise on blood pressure, the null hypothesis might be that exercise has no effect on blood pressure.

4. Misconception cleared: The null hypothesis is not a statement of what we expect to find, but rather a statement of what we assume to be true before collecting data.

5. How do we specify an alternative hypothesis?

6. Answer: We specify an alternative hypothesis by stating what we hope to find if our research is correct, often in the form of a statement that there is a significant difference or relationship between variables.
7. Real-world example: In a study on the effect of exercise on blood pressure, the alternative hypothesis might be that exercise has a significant effect on blood pressure.

8. Misconception cleared: The alternative hypothesis is not a statement of what we expect to find, but rather a statement of what we hope to find if our research is correct.

9. How do we choose between one-tailed and two-tailed hypotheses?

10. Answer: We choose between one-tailed and two-tailed hypotheses based on the research question and the expected direction of the effect.
11. Real-world example: In a study on the effect of a new medication on blood pressure, choosing a one-tailed hypothesis allows us to determine whether the medication lowers blood pressure, while choosing a two-tailed hypothesis allows us to determine whether the medication has any effect on blood pressure.
12. Misconception cleared: One-tailed hypotheses are not more powerful than two-tailed hypotheses, but rather they are used when the direction of the effect is predicted.

CAN (possibility/conditions)

1. Can we specify a null hypothesis for any research question?
2. Answer: No, we can only specify a null hypothesis for research questions that involve a comparison or a relationship between variables.
3. Real-world example: In a study on the effect of exercise on blood pressure, we can specify a null hypothesis that exercise has no effect on blood pressure.

4. Misconception cleared: The null hypothesis is not a statement of what we expect to find, but rather a statement of what we assume to be true before collecting data.

5. Can we specify an alternative hypothesis for any research question?

6. Answer: No, we can only specify an alternative hypothesis for research questions that involve a comparison or a relationship between variables.
7. Real-world example: In a study on the effect of exercise on blood pressure, we can specify an alternative hypothesis that exercise has a significant effect on blood pressure.

8. Misconception cleared: The alternative hypothesis is not a statement of what we expect to find, but rather a statement of what we hope to find if our research is correct.

9. Can we choose between one-tailed and two-tailed hypotheses for any research question?

10. Answer: No, we can only choose between one-tailed and two-tailed hypotheses based on the research question and the expected direction of the effect.
11. Real-world example: In a study on the effect of a new medication on blood pressure, choosing a one-tailed hypothesis allows us to determine whether the medication lowers blood pressure, while choosing a two-tailed hypothesis allows us to determine whether the medication has any effect on blood pressure.
12. Misconception cleared: One-tailed hypotheses are not more powerful than two-tailed hypotheses, but rather they are used when the direction of the effect is predicted.

TRUE/FALSE (misconception testing)

1. The null hypothesis is a statement of what we expect to find.
2. Answer: FALSE
3. Real-world example: In a study on the effect of exercise on blood pressure, the null hypothesis is that exercise has no effect on blood pressure, not what we expect to find.

4. Misconception cleared: The null hypothesis is a statement of what we assume to be true before collecting data, not what we expect to find.

5. The alternative hypothesis is a statement of what we hope to find if our research is correct.

6. Answer: TRUE
7. Real-world example: In a study on the effect of exercise on blood pressure, the alternative hypothesis is that exercise has a significant effect on blood pressure, which is what we hope to find if our research is correct.

8. Misconception cleared: The alternative hypothesis is a statement of what we hope to find if our research is correct, not what we expect to find.

9. One-tailed hypotheses are more powerful than two-tailed hypotheses.

10. Answer: FALSE
11. Real-world example: In a study on the effect of a new medication on blood pressure, choosing a one-tailed hypothesis allows us to determine whether the medication lowers blood pressure, but it is not more powerful than choosing a two-tailed hypothesis.
12. Misconception cleared: One-tailed hypotheses are not more powerful than two-tailed hypotheses, but rather they are used when the direction of the effect is predicted.