Topic 4.11 Notes – Parameters for a Binomial Distribution

What a Binomial Random Variable Is

A binomial random variable $X$ counts the number of successes in a fixed number of trials.

For a setting to be binomial, it must satisfy BINS:

• Binary: Each trial has two outcomes (success/failure).
• Independent: One trial does not affect another.
• Number fixed: The number of trials $n$ is set in advance.
• Same probability: The probability of success $p$ is constant.

If any of these fail, you cannot use the binomial mean and standard deviation formulas.

The 10% Condition

When sampling without replacement, independence is reasonable if:

$$n < 0.10N$$

where $N$ is the population size.
This allows us to treat the trials as approximately independent and still use a binomial model.

The Parameters of a Binomial Distribution

If $X \sim \text{Binomial}(n, p)$, then its distribution is completely described by $n$ and $p$.

Mean (Expected Value)

$$\mu_X = E(X) = np$$

Think of this as the long-run average number of successes.

Example:
A quality inspector checks $n = 50$ items. Each item has a 0.08 probability of being defective.

$$\mu_X = 50(0.08) = 4$$

Interpretation in context:

In many samples of 50 items, we expect an average of about 4 defective items per sample.

Notice the wording. It’s about long-run average behavior, not a guarantee.

Standard Deviation

$$\sigma_X = \sqrt{np(1-p)}$$

This measures how much the number of successes typically varies from the mean.

Using the same example:

$$\sigma_X = \sqrt{50(0.08)(0.92)} = \sqrt{3.68} \approx 1.92$$

Interpretation in context:

The number of defective items in samples of 50 typically varies by about 1.92 items from the mean of 4.

Standard deviation is measured in the same units as the variable, here “number of defective items.”

What These Formulas Mean Conceptually

Why does the mean equal $np$?

• Each trial has probability $p$ of success.
• Over $n$ trials, you expect about $p$ successes per trial.
• Multiply → $np$.

Why does variability depend on $p(1-p)$?

• If $p$ is near 0 or 1, outcomes are more predictable → less variability.
• Variability is largest when $p = 0.5$.

This idea shows up in multiple-choice questions where they compare spreads.

Interpreting Probabilities and Parameters in Context

This is where points are often lost.

Interpreting a Probability

If $P(X = 6) = 0.18$, a correct interpretation looks like:

There is a 0.18 probability that exactly 6 customers out of the 20 selected prefer the new product.

You must:

• Mention the number of trials.
• State what counts as success.
• Use context words (customers, products, defects, etc.).

Interpreting the Mean

Correct style:

In many repetitions of selecting 20 customers, we expect an average of ___ customers to prefer the new product.

Never say “there will be ___.” The mean is not guaranteed.

Interpreting the Standard Deviation

Correct style:

The number of customers who prefer the new product typically varies by about ___ from the mean.

Do not treat it as a maximum or a bound.

Common Mistakes to Watch For

• Using $np$ as a probability instead of a mean.
• Forgetting to check independence in sampling without replacement.
• Using a changing probability as if it were constant.
• Giving interpretations without context (no variable, no population).
• Forgetting that mean and standard deviation are measured in number of successes, not proportions.

If a question quietly violates BINS, the formulas do not apply. That’s a common trap on quizzes.