1/15
Looks like no tags are added yet.
For what kinds of tests can chi-square analysis be used?
(1) whether a sample could have come from a given type of population distribution (e.g., the normal distribution),
(2) whether two nominal variables could be independent of each other, and
(3) whether two or more independent samples could have the same population proportion.
Is the chi-square distribution a continuous distribution or a discrete distribution? Explain.
The chi-square distribution is a family of distributions which take on different shapes according to the associated degrees of freedom. A given chi-square distribution is a continuous curve and chi-square is always positive.
In what way are the chi-square and normal distributions related?
The chi-square distribution is continuous and approaches the normal distribution as d.f. increases.
Sketch the approximate shape of the chi-square curve when df 5 2 and when df 5 100
A chi-square distribution with a small d.f. will be very positively skewed.
The distribution with 100 d.f. will approach a normal distribution

p( chi > A ) = 0.90
use .90 and df
p( chi < A ) = 0.90
subtract 1-0.90 for alpha
In carrying out a chi-square goodness-of-fit test, what are the “k” and “m” terms in the “df=k-1-m" expression and why is each term present?
Given k is the number of categories or groups in the test,
and m is the number of population parameters that must be estimated from sample data in order to carry out the test. "k-1" is present because, if we know the total number of observations, we need only know the contents of "k-1" cells to determine the count in the kth cell. In addition, each population parameter that must be estimated also reduces our degrees of freedom
test normal distribution chi square what parameters
Sample mean and standard deviation
poisson distribution chi square what parameters
sample mean
Chi square tests Eij expected value is ..
at least 5
This is because the chi-square distribution is continuous and is being used to analyze discrete (counts) data.
If one or more expected frequencies do not meet this minimum requirement, we should either
increase the total sample size or combine adjacent rows or columns in the table.
test of variable independence hypothesis

Chi-square test for the independence of two variables df
(r-1)(k-1)
r=rows
k=column
goodness of fit hypothesis

Expected frequencies when testing for the independence of variables
sum of column / total x total row

test for equality of proportions hypothesis

test for equality of proportions, each table (observed and expected) will always have r =
2