Chapter 7 Flashcards
the second main application of inferential statistics
hypothesis testing
2 things hypothesis testing does
-measures the likelihood a relationship exists between variables in a population
-probability that a relationship is due to random chance such as a sampling error
5 step process for hypothesis testing
1.)Make assumptions and meet test requirements.
2.)State the null hypothesis.
3.)Select the sampling distribution and establish the critical region.
4.)Compute the test statistic.
5.)Make a decision and interpret the results.
what is a hypothesis
a specific statement about the relationship between variables that is logically derived from theory
Null hypothesis
H0 meaning no relationship
research hypotheisis
H1 contradicts null hypothesis
sampling ditribution
theoretical understanding that endless samples will result in a bell-curve
Critical region
area under the sampling distribution encompassing sample outcomes defined as “unlikely” by the researcher
If our sample falls into the critical region
we reject the null hypothesis of no relations and we accept that there is a relationship
What is the test statistic used for
evaluate the probability of the sample outcome (assuming the null hypothesis is true)
p-value
probability located in the tails of the sampling distribution beyond the value of the test statitic
what doesn’t hold in the alpha region and what is the alpha region also known as
the null hypothesis doesn’t hold in the critical region
the test statistic is in the critical region what is the decision and interpretation
we reject the null hypothesis and say the difference is statistically signiicant
the test statistic is not in the critical region what is the decision and interpretation
fail to reject the null hypothesis and the difference is not statistically significant
failing to reject the null hypothesis means
there is a relationship between the varibales
Type I or alpha error
probability that we reject the null-hypothesis in an error of 0.05 or whatever level we set alpha
Type II or beta error
the risk of failing to reject the null hypothesis increases
Which type error do we try to control and how
a type I error by setting the alpha level
put in chart
what levels is a chi square test good for
nominal or ordinal
the independent variables go into
columns (up and down)
the dependent variables go into
rows (side to side)
When are 2 variables considered independent
when the classification of one case has no effect the the case will also fall into the other catergory
2 types of frequency’s
observed and expected
how to calculate expected frequencys
How to find chi squared
Fo-Fe
square it
divide by Fe
what does the chi square being nonparametric mean
makes no assumption the population has a normal-curve
How to state H0
Fo=Fe
How to state H1
Fo=/ Fe
How to find degrees of freedom for chi square
(r-1) x (c-1)
X2 critical= 3.84
X2 obtained= 10.78
Null hypothesis?
we reject the null hypothesis because it is unlikely to happen by chance
Chi square should be used with (2)
4 or fewer categories
small sample
