Question 1

If H 0 is validly rejected, H 1 must be true.

Accepted Answer

If H 0 is rejected, it only means that there is sufficient evidence to suggest that H 0 is not true. It does not necessarily mean that H 1 is true. Other explanations or factors may also be at play, and further research or analysis may be needed to confirm the validity of H 1.

Question 2

If H 0 is true, beta equals 0.00.

Accepted Answer

If H 0 is true, beta (Type II error rate) equals 0.00, which means there is no risk of failing to reject a false null hypothesis.

Question 3

It is impossible to get 20 pluses out of 20 pairs of scores in a replicated measures design due to chance alone.

Accepted Answer

In a replicated measures design, it is possible, though highly unlikely, to get 20 pluses out of 20 pairs of scores due to chance alone. The probability of such an event depends on the specific circumstances and assumptions of the experiment, but it is not zero.

Question 4

It is permissible to use a directional H 1 when there are good theoretical as well as strong supporting data to justify the predicted direction.

Accepted Answer

When there are good theoretical and strong supporting data to support a predicted direction, it is permissible to use a directional H1. However, it is important to ensure that the hypothesis is clearly stated and justified in the research.

Question 5

H₀ always asserts that the dependent variable has no effect on the independent variable.

Accepted Answer

H₀ always asserts that the independent variable has no effect on the dependent variable, not the other way around.

Question 6

It is always appropriate to use a directional H 1 .

Accepted Answer

It is not always appropriate to use a directional H1. The decision to use a directional or non-directional H1 depends on the research question and the hypothesis being tested. When the hypothesis is directional, a directional H1 is appropriate, but when the hypothesis is non-directional, a non-directional H1 is appropriate.

Question 7

Unless the very same subject is used in the control and experimental condition the design cannot be a replicated measures design.

Accepted Answer

Replicated measures design involves using the same subjects in both the control and experimental conditions. Therefore, different subjects are used in the control and experimental conditions in a non-replicated measures design.

Question 8

It is technically correct to conclude by "accepting" rather than "failing to reject" H 0 .

Accepted Answer

In hypothesis testing, we never "accept" the null hypothesis (H0); we only "fail to reject" it, indicating insufficient evidence against it.

Question 9

The sign test analyzes raw scores.

Accepted Answer

The sign test analyzes the direction of differences between pairs of observations, not the raw scores themselves. It is a non-parametric test used for assessing the median of differences between paired samples.

Question 10

A replicated measures design is the same thing as a correlated groups design.

Accepted Answer

A replicated measures design is a type of experimental design where each participant is tested multiple times under different conditions, and the results are compared. Correlated groups design is another name for this type of design, as the measures or conditions are correlated within each participant.

Question 11

H 1 and H 0, taken together, cover the entire continuum with regard to possible effects of the independent variable.

Accepted Answer

Both H1 and H0 account for all possible effects of the independent variable, with H1 suggesting a specific effect and H0 suggesting no effect at all.

Question 12

The statement "Drug X has no effect on Y and any observed effect is due to chance alone" is an example of a nondirectional alternative hypothesis.

Accepted Answer

The statement provided is actually an example of a null hypothesis, not an alternative hypothesis. It states that there is no effect and any observed effect is due to chance, which is the opposite of an alternative hypothesis, which proposes a specific relationship or effect between variables. Additionally, since the statement suggests that any observed effect is due to chance, it is not a directional alternative hypothesis, which would propose a specific direction for the effect (e.g. Drug X increases Y).

Question 13

If alpha is made more stringent, beta increases.

Accepted Answer

When alpha is made more stringent (i.e. the level of significance is decreased), it means that the researcher is less willing to accept a type I error (rejecting the null hypothesis when it is actually true). This increases the chances of a type II error (failing to reject the null hypothesis when it is false), which is represented by beta. Therefore, as alpha becomes more stringent, beta is likely to increase.

Question 14

The sign test is used with the replicated measures design.

Accepted Answer

The sign test is a non-parametric test used with the replicated measures design. It is used to compare two related samples or measures to determine whether there is a significant difference between them.

Question 15

It is impossible to prove with certainty the truth of H 1 when using sample data.

Accepted Answer

Hypothesis testing involves making a decision based on sample data, but there is always a chance that the conclusion reached is incorrect due to sampling variability. Therefore, it is impossible to prove with certainty the truth of H1 (alternative hypothesis) when using sample data.

Question 16

If the result turns out to be in the direction opposite to a directional H 1 , we must conclude by retaining H 0 .

Accepted Answer

That's correct. When the result goes against the directional alternative hypothesis (H1), we retain the null hypothesis (H0) as the result does not provide evidence in favor of the alternative hypothesis.

Question 17

H₀ and H₁ must be mutually exclusive and exhaustive.

Accepted Answer

Mutually exclusive means that the two options cannot occur at the same time, while exhaustive means that together they cover all possible options. Given that H₀ and H₁ differ only in the last two digits, and these are binary digits, it means that they represent two possible outcomes of the same variable. Therefore, they cannot occur at the same time (mutually exclusive) and cover all possible options (exhaustive).

Question 18

The sign test ignores the magnitude of the difference scores.

Accepted Answer

The sign test only considers whether the difference scores are positive or negative, not how large the differences are.

Question 19

It is not possible to analyze the probability of the alternative hypothesis because the laws of probability are derived for chance events. This is why we test H₀ instead of H₁ .

Accepted Answer

This statement is true. The laws of probability are based on chance events, and the alternative hypothesis is not a chance event that can be analyzed using these laws. Instead, we must test the null hypothesis against the alternative hypothesis using statistical tests.

Question 20

If results are statistically significant, the independent variable must have had a large effect.

Accepted Answer

Statistical significance only indicates that there is a low probability that the results occurred due to chance. It does not necessarily indicate the size or magnitude of the effect of the independent variable. Other factors such as the sample size and measurement precision can also influence statistical significance.

Quiz 10: Introduction to Hypothesis Testing: Using the Sign Test