Please enable JavaScript.
Coggle requires JavaScript to display documents.
Hypothesis Testing, Type errors, H1 can be either..., Greater test…
Hypothesis Testing
Foundation
Hypotheses
Significance Testing
Decisions and Errors
Test Statistics
Type errors
A Type II error means failing to reject a false H₀.
A Type I error means rejecting a true H₀.
Lowering α decreases Type I error risk but increases Type II error risk
H1 can be either...
Directional (one-tailed)
Nondirectional (two-tailed)
Greater test statistics are produced by higher deviations from H₀
A test statistic calculates the sample result's deviation from H₀
Standard error measures expected sampling fluctuation
The foundation for assessing significance is provided by sampling distributions
If H₀ were true, the result would be unlikely due to statistical significance
The goal is to determine whether a treatment effect exists
Probability and sampling distributions are the foundation of the process
The null hypothesis (H₀) states no effect/no difference
According to the alternative hypothesis (H₁) there is an effect
Hypothesis is directly tested is H₀
Only when H₀ is rejected is H₁ supported
Alpha = level of significance
Alpha levels of .05 and .01 are typical
The likelihood of a Type I error is determined by alpha
Unlikely outcomes under H₀ are present in the critical region
Critical region's boundaries are indicated by critical values
Reject H₀ when the test statistic falls in the critical region
Fail to reject H₀ when it does not fall in the critical region
One method of inferential statistics is hypothesis testing
It draws judgements about a population based on sample drama
NO hypothesis is ever proven to be entirely correct by researchers