P-value
By: Pruthvi
244007
II Msc Statistics
Contents:
Introduction to Statistical Significance
What is a P-value ?
How to calculate P-value?
Example
Limitations of the P-Value
Conclusion
Statistical Significance
Statistical significance is a measure used in
hypothesis testing to determine whether the
observed result in a data sample is likely due to a
real effect rather than random chance.
Statistical significance is usually decided
by a number called a p-value. If this p-value is
small enough—often less than 0.05 or 5%—it
shows the result is probably real and not random.
So, if the p-value is 0.05 or less, the findings are
called statistically significant, meaning there's
strong evidence that something meaningful is
going on.
What is P-value?
A p-value is a number that tells you how likely it is to get
the results you observed (or something even more
extreme) if the null hypothesis is true. The null hypothesis
usually means there is no real effect or difference.
Smaller P-value: Stronger evidence
against the null hypothesis.
Larger P-value: Weaker evidence
against the null hypothesis.
First introduced by Karl Pearson in
his Pearson’s chi-squared test.
What P-value Tells You!
The p-value is the level of marginal significance in a
statistical hypothesis test, representing the
probability of an event occurring.
The vertical coordinate in the distribution plot is the
probability density for each outcome calculated under
the null hypothesis, and the p-value represents the
area under the curve in these regions.
If the p-value is less than or equal to α (the
significance level), we reject the null hypothesis. If the
p-value is greater than α, we do not reject the null
hypothesis.
P≤α: Reject null hypothesis
P>α.: Fail to reject null hypothesis
α is the level of significance.
How to calculate P-value?
First, a null hypothesis (H0) and an alternative hypothesis (H1) are formed based
on the research question.
Then, a test statistic is calculated from the sample data, such as a t-statistic or
z-statistic, which measures the deviation of the sample from the null hypothesis.
General Formula for Test Statistic (e.g., Z-test or T-test):
For a sample mean test (t-test or z-test), the test statistic is calculated as:
Where:
−
x = sample mean
μ0= hypothesized population mean under the null hypothesis
s= sample standard deviation
n = sample size
^
p= sample proportion (for proportion tests)
p0= hypothesized population proportion
Calculating P-value Using the Test Statistic
For a left-tailed test, p-value = P(TS ≤ observed value)
For a right-tailed test, p-value = P(TS ≥ observed value)
∣
For a two-tailed test, p-value = 2×P(TS≥ observed value ) ∣
Here, TS is the test statistic, and the probabilities are obtained from the
appropriate probability distribution (Normal, t-distribution, etc.).
Example:
Testing Average Weight of a Product
Problem
A company claims that the average weight of their packaged cereal bags is
500 grams. A quality control analyst wants to test if the true average weight
differs from this claim. A sample of 30 bags is taken, and the sample average
weight is found to be 495 grams with a sample standard deviation of 10
grams.
Step 1: State Hypotheses
Null hypothesis H0:μ=500 grams (the average weight is 500 grams)
Alternative hypothesis H1:μ≠500 grams (the average weight is not equal to
500 grams)
Step 2: Calculate Test Statistic (t-statistic)
Step 3: Find p-value
Using a t-distribution with n−1=29 degrees of freedom, find the probability
of observing a t-value as extreme as −2.74 or +2.74 (two-tailed).
From t-tables, P(T≤−2.74)≈0.0055.
Because it is two-tailed, multiply by 2:p≈2×0.0055=0.011.
Step 4: Conclusion
Since p=0.011<0.05, reject the null hypothesis.
Interpretation: There is strong evidence that the average weight of the
cereal bags differs from 500 grams. The company's claim is unlikely to be
true based on this sample.
Limitations of the P-Value
Sample size sensitivity: It is No measure of effect size: A
affected by sample size. With a small p-value does not mean the
big enough sample, even tiny effect is important or large; it just
and unimportant differences means the data are unlikely under
can look significant. the null hypothesis.
.
P-values alone cannot
confirm scientific findings;
they need to be interpreted
with measures of effect size,
study quality, and context.
Conclusion:
The p-value is a fundamental tool in statistical hypothesis
testing that helps quantify the evidence against the null
hypothesis. A smaller p-value indicates stronger evidence to
reject the null hypothesis, suggesting the observed data are
unlikely under the null assumption. However, it is important to
remember that the p-value does not measure the size or
importance of an effect, nor does it give the probability that the
null hypothesis is true.
Interpreting p-values requires careful consideration of the
context, study design, and complementary statistics such as
effect size and confidence intervals. With these in mind, p-
values provide valuable guidance in scientific decision-making
but should not be used in isolation. Understanding their
limitations and proper usage enhances the rigor and reliability of
statistical conclusions.
Thank you!
