Chi-Squared Test Flashcards
What is the chi-squared test?
It is a statistical test used to see if the results of an experiment support a theory.
What is done first?
The theory is used to predict a result - this is called the expected.
Then, experiment is carried out and actual result is recorded - this is called the observed result.
What is a null hypothesis?
It says there’s no significant difference between the observed and expected result.
Why is the chi -squared carried out?
To compare how well the observed results matches the expected result.
Outcome either supports or rejects null hypothesis.
When would you use a chi-squared test?
Whenever there is categorical data and you want to compare observed and predicted results.
EXAMPLE - INHERITANCE OF WIN LENGTH
What is the theory?
Wing length in fruit flies is controlled by a single gene with 2 alleles (monohybrid inheritance).
Dominant allele N - normal wings.
Recessive allele n - vestigial wings.
EXAMPLE - INHERITANCE OF WIN LENGTH
What are the expected results?
With mono-hybrid inheritance, if you cross homozygous dominant parent with homozygous recessive you’d expect 3:1 ratio.
Normal: vestigial wings in F2 generation.
EXAMPLE - INHERITANCE OF WIN LENGTH
What are the observed results?
The experiment of crossing a homozygous dominant with a homozygous recessive parent, is carried out on fruit flies and number of offspring with normal wings and vestigial wings is counted.
EXAMPLE - INHERITANCE OF WIN LENGTH
What is the null hypothesis?
There is no significant difference between the observed and expected results.
EXAMPLE - INHERITANCE OF WIN LENGTH
What happens in the chi-squared test?
Chi-squared value is compared to critical value.
If there is no significant difference between predicted and observed they are not significantly different they we can reject null hypothesis.
The data therefore supports the theory that wing length is controlled by monohybrid inheritance.
Chi-squared formula.
X2 = SUM OF (observed - expected)2 divided by expected.
What is the critical value?
It is the value of chi-squared that corresponds to a 0.05% (5%) level of probability that the difference between the observed and expected is due to chance.
What happens if your chi-squared value is larger (or equal) than your critical value?
There is a significant difference between the observed and expected results.
(something other than chance is causing the difference eg epistatsis or sex linkage).
Null hypothesis can be rejected.
What happens if your chi squared value is smaller than your critical value?
There is no significant difference between the observed and expected results.
The null hypothesis can’t be rejected.
