PowerPoint Presentation
p-value
use to determine if theres a statistically significant
difference between groups of data
probability of error by convention, if p > 0.05, then there
is no effect if p < 0.05, then there is an effect
The remaining slides (optional) provide a more in-depth
description of the p-value.
Click voice clip1p-value (probability of a type I error). null
hypothesis (Ho) = there is no effect. Ho is accepted, unless there
is good evidence to reject it.accept Horeject HoHo is validcorrect
decisiontype I errorHo is invalidtype II errorcorrect decisionTable
1. Types of errors in a statistical test
2Determine the value of the p-value its a calculus / numerical
methods problemexample: Probability of the test statistic
used in the t-test , which is used to compare 2 groups of
data.
is the area under the graph that describes the probability of
getting a specific value of the test statistic.
3.Types of p-values 2-tail p-values null hypothesis: X1 =
X2there is no effect; the data is the same
alternative hypothesis: X1 X2there is an effect; the data is
different
4 1-tail p-value null hypothesis: X1 X2
alternative hypothesis: X1 > X2
5 1-tail p-value null hypothesis: X1 X2
alternative hypothesis: X1 < X2
6