Professional Documents
Culture Documents
Introduction On Chi Square Distribution
Introduction On Chi Square Distribution
Significance Test
Just how low is sufficiently low? The choice is somewhat arbitrary but by convention
levels of 0.05 and 0.01 are most commonly used.
For instance, an experimenter may hypothesize that the size of a food reward does
not affect the speed a rat runs down an alley. One group of rats receives a large
reward and another receives a small reward for running the alley. Suppose the mean
running time for the large reward were 1.5 seconds and the mean running time for
the small reward were 2.1 seconds.
The difference between means is thus 2.1 - 1.5 = 0.6 seconds. The test of whether
this difference is significant consists of determining the probability of obtaining a
difference as large or larger than 0.6 seconds given there is really no effect of
magnitude of reward. If the probability is low (below the significance level) then the
null hypothesis that magnitude of reward makes no difference is rejected in favor of
the alternate hypothesis that it does make a difference. The null hypothesis is not
accepted just because it is not rejected.
Degrees of Freedom
Significance Level
In hypothesis testing, the significance level is the criterion used for rejecting the null
hypothesis. The significance level is used in hypothesis testing as follows: First, the
difference between the results of the experiment and the null hypothesis is
determined. Then, assuming the null hypothesis is true, the probability of a
difference that large or larger is computed . Finally, this probability is compared to
the significance level. If the probability is less than or equal to the significance level,
then the null hypothesis is rejected and the outcome is said to be statistically
significant. Traditionally, experimenters have used either the 0.05 level (sometimes
called the 5% level) or the 0.01 level (1% level), although the choice of levels is largely
subjective. The lower the significance level, the more the data must diverge from the
null hypothesis to be significant. Therefore, the 0.01 level is more conservative than
the 0.05 level. The Greek letter alpha (α) is sometimes used to indicate the
significance level. See also: Type I error and significance test.
There are two kinds of errors that can be made in significance testing: (1) a true null
hypothesis can be incorrectly rejected and (2) a false null hypothesis can fail to be
rejected. The former error is called a Type I error and the latter error is called a Type
II error. These two types of errors are defined in the table.
The probability of a Type I error is designated by the Greek letter alpha (a) and is
called the Type I error rate; the probability of a Type II error (the Type II error rate) is
designated by the Greek letter beta (ß) . A Type II error is only an error in the sense
that an opportunity to reject the null hypothesis correctly was lost. It is not an error
in the sense that an incorrect conclusion was drawn since no conclusion is drawn
when the null hypothesis is not rejected
A Type I error, on the other hand, is an error in every sense of the word. A conclusion
is drawn that the null hypothesis is false when, in fact, it is true. Therefore, Type I
errors are generally considered more serious than Type II errors. The probability of a
Type I error (α) is called the significance level and is set by the experimenter. There is
a tradeoff between Type I and Type II errors. The more an experimenter protects
himself or herself against Type I errors by choosing a low level, the greater the
chance of a Type II error. Requiring very strong evidence to reject the null hypothesis
makes it very unlikely that a true null hypothesis will be rejected. However, it
increases the chance that a false null hypothesis will not be rejected, thus lowering
power. The Type I error rate is almost always set at .05 or at .01, the latter being
more conservative since it requires stronger evidence to reject the null hypothesis at
the .01 level then at the .05 level.
Applications:-
To test if, more than two population proportions can be considered equal.
Characteristics:-
Continuous distribution.
Chi square distribution curve starts at the origin and lies entirely to the right
of Y axis.
5.Shape of Chi square distribution curve is skewed for very small d.f and
changes drastically as d.f. increases . For large d.f, Chi-square distribution
looks like a normal distribution curve.
All expected frequency > 10 for good accuracy, and should be minimum 5.
Take Ho: There is no significant difference between the sample proportions or
between the observed and the corresponding expected values.
If the degrees of freedom and area required in the right tail (ie, significance
level of test ) are given, the critical values of chi-square can be found from the
table.
The c2 test is a statistical method that tests whether a given set of data fit a hypothesis.
Can the test accept the hypothesis and it’s wrong? Can we reject a correct
hypothesis? You bet! That’s why we call it probability and not absolutility!
But in the following pages you will see how statistics can gauge the confidence
of our answer.
Predict the outcome of the data if your null hypothesis were correct
Draw a conclusion
These 7 steps will be explained in detail in the following example. :-
Example: If we observed 99 purple and 45 white flowers, is this a 3:1 ratio? Use
c2 to test this hypothesis.
Pp X Pp -> 3 P_ : 1 pp
Pp X Pp
gives
3 Purple : 1 White
- The given data are the observed values (ie, in this case 99 purple and 45
white)
108 - 99 = 9 purple
36 - 45 = -9 white
(9)2 = 81 purple
(-9)2 = 81 white
expected
To answer that question we have to remember that c 2 is determined from the eviations
from the expected values. Therefore the smaller the number is the smaller the
deviation is from the expected values:
If c2 = 0
Conversely, the larger the number is the larger the deviation is from the expected
values.
One phenotype is not variable with itself and so would have 0 degrees of freedom.
Two phenotypes have one variable and therefore would have 1 degree of freedom, etc.
We can therefore summarize this by saying that:
We will take a little detour to demonstrate that probability depends on sample size:
Let’s look at 3 sample sizes below, 4, 8, and 40. Here we are looking at tall vs. dwarf
trees. Now let’s look at the predicted distribution of trees having the indicated
number of tall trees on the left for each category and the probability for each outcome
on the right. Since tall is dominant, it is not surprising to see that the probability is
higher for higher numbers of trees to be tall. You will notice though that the larger the
number the more accurate the prediction of ¾ should be tall.
If we look at the graph, 95% of the area is under the curve and 5% is in the shoulder
regions. This 5% is the data that is most in doubt – do probabilities that fall into this
region mean that the hypothesis is incorrect? No. Do probabilities that fall under in
the 95% area mean that the hypothesis is correct? No. But depending upon sample
size if we are within those 95% confidence limits we accept the hypothesis.
• Degrees of freedom (df) are listed in the outer columns
If we go from the graph to the c2 table below we see columns with probabilities as
headings.
Within the body of the table are the c2 values that correspond to the probability at the
top.
Locate the 2 columns that span the c2 value from your analysis
Since we have 1 df and our c2 is 3.0 our probability is between 0.1 and 0.05.
These data for purple and white are indicative of a 3:1 ratio
The c2 test can be used to test any type of genetic hypothesis using these 7 steps: