Name : Rabbiya Shabbir

Roll no : 1939
Course code: BOT-403
Subject : Biostatistics
Semester : 3rd
Topic : Testing of Hypothesis and
its importance in various
Biological experiments.
Abstract.
In this essay we come to know the meaning of hypothesis testing and its types including
the problem of introducing divergence-based statistics to test composite
hypotheses related to p-value was still open when sample sizes were not equal.
Predictions for any statement must have statistically significant so that one could
proved , with the help of hypothesis testing where we can gathered the data and
through procedure analysed which one hypothesis was true also putting a
biological example in this we concluded the results showing valid approach(two
tailed test , Z-Test ) and significance of testing of hypothesis was very helpful in
biological experiments too , so that one could assume and then verified its own
theory about the populations or sampling or even the drugs used in
pharmaceutical composition.

Essay.
Hypothesis testing was a conventional plan of action for investigating your ideas about
any statement. It was widely used by scientists to test particular predictions which was
called hypotheses, that were imported from theories.
There are two types of Hypothesis when we want to explain the testing of hypothesis.
These are as follows.
 Null Hypothesis: A unconfirmed aspect was formed about the framework.  This
assumption was called as the Null hypothesis ,Designated by H0
 Alternative hypothesis. An opposite statement or fact to the Null hypothesis.
We would Divide it also on the basis of variables .

 Simple Hypothesis. Which commonly called Basic Hypothesis, that an

independent variable was responsible for a corresponding dependent variable.
Based on only two variables.
 Composite hypothesis . More than two independent variables are present which results
the more dependent variables.

Procedure.
The purpose of hypothesis testing was to examined and analysed in between the null
hypothesis and alternative hypothesis which one was the most acceptable. Only one
can be true. Once you know the variables for the null hypothesis, the next step is to
examined the alternative hypothesis. The alternative hypothesis could be overcome the
null assumption by suggesting the statement was true. Now you can make the Null
hypothesis by yourself or by some other research . Then make alternative hypothesis,
collection of data make tests . Which should be of 99% of confidence. Then this will
decide whether the Hypothesis was true or not and hence we can reject the false one .
o P-value. A P-value  is used to estimate the importance of the statistical test. If the null
hypothesis was rejected, then the alternative one was accepted. To determine whether a
result is statistically significant, a researcher calculates a p-value, which is the probability
of observing an effect of the same magnitude or more extreme given that the null
hypothesis is true
o Specify significant value (Threshold) . 0.5
o Z-Test . It can be one tailed test(left or right) or two tailed test (both sides), also called
test statistic as stated in the formula.

Where the X=Sample mean

U= population mean
SD=standard deviation

N=sample size

Example No 1:
A neurologist is testing the effect of a drug on response time by injecting 100 cate with a
section dose of the drug, treated each to external stimulus, and recording its response time.
The neurologist knows that the mean response time for cats not injected with the drug is 1.2
seconds. The mean of the 100 injected rats' response times is 1.05 seconds where the standard
deviation is 0.5 seconds. Do you think that the drug has an effect on response time?
Here we know that two hypothesis are given.

Ho = drug has no effect . Mean μ = 1.2 even with the drug

Ha = drug has an effect. Mean μ # does not equal to 1.2 when the drug is given.

If the Null hypothesis was true than what was the probability.

By calculating the Z-score as formula. 1.2-1.05/0.05 = 3 ,

where the 0.05 comes when SD / square root of n

The result is from 3 standard deviation far from mean .. we Draw this on two tailed test
If the Z value is 3 .. then the movement from mean from left side is 3 Standard deviation also to left thus
the value is more unlikely and more extreme to this result of μ =1.2 .

If we assume that the drug has no effect and null hypothesis is true then the probability is only 0.3% ,So
the result seem to be in favour of alternative hypothesis. We reject the null hypothesis .After draw the
results we get p-value which is in extreme = .003 and much less probability this value is also less then
the threshold value which is 0.5 so rejection accepted.

Otherwise there are two types of errors could be overcome the results. E.g If the null hypothesis tell
that there is no difference between two variables but we find out the difference but actually there is no
difference so tha null hypothesis was true but we declined it wrong we caused Type I error and if we
rejected the alternative hypothesis but actually it was true then we caused Type ll error .

Chi-Square test also can perform . Formula is

Where the O = observed frequencies ,E= expected frequencies

T-test can perform

Importance in biological experiments

For many scientific experiments computing a p-value is the standard method for
reporting the outcome. It is a simple way of summarizing the information in the
data.

To provide different interpretations for a parameter.

These properties make our statement more general than the usual statement

Current scientific techniques in genomics and image processing routinely produce

hypothesis testing problems with hundreds or thousands of cases to consider
simultaneously. This poses new difficulties for the statistician, but also opens new
opportunities
Conclusion.
By the tremendous use of this statistics work best and gave solutions to many
problems creating in scientific researches and publishing projects however there is
some causes of errors now a days scientists use ways and methodologies to
remove type 1 and type 2 errors regarding to their interest. But for the huge
populations this might not be worked due the largely handed data and to compute
this might be difficult so there should be some other methods for this solution.