91Ó°ÊÓ

The normal approximation is a useful method for simplifying problems that deal with binomial distributions, especially when the number of trials is large. In essence, when you have a situation like our McDonald's scenario where 1,000 people are tested, calculating probabilities directly using the binomial formula can become cumbersome.

The key principle behind this method is that a binomial distribution can be approximated by a normal distribution if the number of trials, denoted as \( n \), is large enough, and both \( np \) and \( n(1-p) \) are greater than 5. In our problem, both these conditions hold true, making the approximation applicable.

• The mean \( (\mu) \) of the distribution is calculated as \( np \). In our case, \( \mu = 1000 \times 0.03 = 30 \).
• The variance \( (\sigma^2) \) is obtained by \( np(1-p) \). For us, it is \( 29.1 \).
• The standard deviation \( (\sigma) \) then is the square root of \( 29.1 \), approximately \( 5.39 \).

This method is particularly helpful in scenarios with more trials, helping us to estimate probabilities more efficiently by using a normal distribution.

In probability problems, calculating the probability of a certain number of "successes" out of many trials can sometimes be complex. Let's break down how we tackled this in the specific context of our McDonald's fries taste test.

For part (a) of our exercise, we needed the probability that at least 40 people can taste a difference. By using continuity correction, we first adjusted our target number to 39.5. Then, we computed the Z-score using the formula:

• \( Z = \frac{X - \mu}{\sigma} \), where \( X \) is the adjusted number of successes.

So, substituting in values, we have \( Z = \frac{39.5 - 30}{5.39} \approx 1.76 \). This Z-score helps us find the cumulative probability from standard normal distribution tables: \( P(Z < 1.76) \approx 0.9608 \).

For part (b), the process was similarly followed for 50.5 corresponding to at most 5% of people (or 50 people). After finding \( Z \), we got \( P(Z < 3.81) \approx 0.9999 \). These Z-scores give us the cumulative probability for events occurring up to a certain point. Subtracting from 1 guides us to the chance of our event of interest happening.

The standard normal distribution is a cornerstone concept in statistics and probability, allowing us to easily find probabilities for normally distributed variables. It is a special type of normal distribution that has a mean of 0 and a standard deviation of 1.

When we calculated Z-scores in our McDonald's exercise, we transformed our binomial distribution into this standard normal form. But why?

• Converting our distribution to this standard form allows us to use extensive, pre-calculated tables of probabilities.
• These tables tell us the probability that a normally distributed random variable is less than or equal to a given value.

For instance, with a Z-score of 1.76, we identified that \( P(Z < 1.76) \approx 0.9608 \), meaning about 96.08% of the distribution lies below this point. Such standardized values make probability calculations accessible and efficient for real-life problems through a straightforward, consistent process.