| Misconception . | Adequate interpretation . |
|---|---|
| The P-value represents the probability of H0 being true | As the P-value is calculated under the assumption that H0 is true, it cannot represent a probability of a hypothesis. Instead, it represents the probability of finding similar—or more extreme—data in future similar experiments, under the assumption that H0 is true. |
| The P-value represents the 1-probability of H1 being true | Similarly, as the P-value is calculated under the assumption that H0 is true, it cannot simultaneously represent the probability that H1 is true. |
| A P-value >0.05 implies that H0 is true | As the P-value is calculated under the assumption H0 is true, a P-value above 0.05 merely implies that similar or more extreme data would be observed in >5% of similar future experiments. |
| A P-value <0.05 implies that H1 is true | Similarly to the previous two misconceptions, a P-value below 0.05 only represents the <5% probability that similar or more extreme data would be observed in such future experiments. |
| A P-value <0.05 implies that there is an important difference between groups | Smaller P-values represent the probability of the observed data under the null hypothesis, which can particularly occur in very large trials, although a treatment effect may be very small. |
| A P-value >0.05 implies that there is no difference between groups | Larger P-values simply represent that the observed data are still not that unusual under the null hypothesis, which may be particularly evident in smaller trials. |
| Misconception . | Adequate interpretation . |
|---|---|
| The P-value represents the probability of H0 being true | As the P-value is calculated under the assumption that H0 is true, it cannot represent a probability of a hypothesis. Instead, it represents the probability of finding similar—or more extreme—data in future similar experiments, under the assumption that H0 is true. |
| The P-value represents the 1-probability of H1 being true | Similarly, as the P-value is calculated under the assumption that H0 is true, it cannot simultaneously represent the probability that H1 is true. |
| A P-value >0.05 implies that H0 is true | As the P-value is calculated under the assumption H0 is true, a P-value above 0.05 merely implies that similar or more extreme data would be observed in >5% of similar future experiments. |
| A P-value <0.05 implies that H1 is true | Similarly to the previous two misconceptions, a P-value below 0.05 only represents the <5% probability that similar or more extreme data would be observed in such future experiments. |
| A P-value <0.05 implies that there is an important difference between groups | Smaller P-values represent the probability of the observed data under the null hypothesis, which can particularly occur in very large trials, although a treatment effect may be very small. |
| A P-value >0.05 implies that there is no difference between groups | Larger P-values simply represent that the observed data are still not that unusual under the null hypothesis, which may be particularly evident in smaller trials. |
Partly based on Goodman et al. [5] and Greenland et al. [6].
H0: null hypothesis, H1: alternative hypothesis.
| Misconception . | Adequate interpretation . |
|---|---|
| The P-value represents the probability of H0 being true | As the P-value is calculated under the assumption that H0 is true, it cannot represent a probability of a hypothesis. Instead, it represents the probability of finding similar—or more extreme—data in future similar experiments, under the assumption that H0 is true. |
| The P-value represents the 1-probability of H1 being true | Similarly, as the P-value is calculated under the assumption that H0 is true, it cannot simultaneously represent the probability that H1 is true. |
| A P-value >0.05 implies that H0 is true | As the P-value is calculated under the assumption H0 is true, a P-value above 0.05 merely implies that similar or more extreme data would be observed in >5% of similar future experiments. |
| A P-value <0.05 implies that H1 is true | Similarly to the previous two misconceptions, a P-value below 0.05 only represents the <5% probability that similar or more extreme data would be observed in such future experiments. |
| A P-value <0.05 implies that there is an important difference between groups | Smaller P-values represent the probability of the observed data under the null hypothesis, which can particularly occur in very large trials, although a treatment effect may be very small. |
| A P-value >0.05 implies that there is no difference between groups | Larger P-values simply represent that the observed data are still not that unusual under the null hypothesis, which may be particularly evident in smaller trials. |
| Misconception . | Adequate interpretation . |
|---|---|
| The P-value represents the probability of H0 being true | As the P-value is calculated under the assumption that H0 is true, it cannot represent a probability of a hypothesis. Instead, it represents the probability of finding similar—or more extreme—data in future similar experiments, under the assumption that H0 is true. |
| The P-value represents the 1-probability of H1 being true | Similarly, as the P-value is calculated under the assumption that H0 is true, it cannot simultaneously represent the probability that H1 is true. |
| A P-value >0.05 implies that H0 is true | As the P-value is calculated under the assumption H0 is true, a P-value above 0.05 merely implies that similar or more extreme data would be observed in >5% of similar future experiments. |
| A P-value <0.05 implies that H1 is true | Similarly to the previous two misconceptions, a P-value below 0.05 only represents the <5% probability that similar or more extreme data would be observed in such future experiments. |
| A P-value <0.05 implies that there is an important difference between groups | Smaller P-values represent the probability of the observed data under the null hypothesis, which can particularly occur in very large trials, although a treatment effect may be very small. |
| A P-value >0.05 implies that there is no difference between groups | Larger P-values simply represent that the observed data are still not that unusual under the null hypothesis, which may be particularly evident in smaller trials. |
Partly based on Goodman et al. [5] and Greenland et al. [6].
H0: null hypothesis, H1: alternative hypothesis.
This PDF is available to Subscribers Only
View Article Abstract & Purchase OptionsFor full access to this pdf, sign in to an existing account, or purchase an annual subscription.
