Chapter 4 Flashcards
It is used to determine if sample results are convincing enough to infer something about the population
Statistical test
A claim that there is no effect or difference
Null hypothesis
A claim for which we seek evidence
Alternative hypothesis
It is created by collecting the values of a sample statistic for each simulated sample
Randomization distribution
the proportion of samples, when the null hypothesis is true, that would give a statistic as extreme as (or more extreme than) the originally observed sample
P-value
The hypothesis that is only written with equality
Null hypothesis
The hypothesis that is only written with inequalities
Alternative hypothesis
It signifies the need for a right-tail test
Greater than
It signifies the need for a left-tail test
Less than
It signifies the need for a two-tail test
Not equal to
The threshold below which the p-value is deemed small enough to reject the null hypothesis
Significance level
Written conclusion when p-value <a
“We have (strong) evidence that [fill with the alternative hypothesis]”
Written conclusion when p-value >a
“We do not have enough evidence to conclude that [fill with the alternative hypothesis]”
T or F: We never accept the null hypothesis, only reject it
True
When the p-value is less than the significance level
Statistically significant
When you reject a true null hypothesis. It can also be considered a “false positive”
Type I error
When you fail to reject a false null hypothesis. It can also be considered a “false negative”
Type II error
Only the significant results are published, concealing all of the insignificant results and causing very misleading results
Publication bias
If assuming that the null hypothesis is true, the significance level will show the number of results that will display significant results by chance
The problem of multiple testing
Why is replication so important?
Replication with similar conclusions increases credibility. Replication also helps guard against Type I error
How does the significance level influence the errors?
Lower a = Lower Type I error
Higher a = Lower Type II error
How much is the significance level, unless otherwise specified?
a = 0.05