Define the population, event, or assumptions first
A z-score by itself gives no percentile unless the distribution is roughly normal. In a strongly skewed data set a score of 2 may be nowhere near the 97.7th percentile that the normal curve would suggest; without a shape assumption all you can say is that at most a quarter of values lie more than 2 standard deviations from the mean.
What each quantity represents in z-score
Name the experiment or data-generating process, distinguish sample information from population parameters, and state the event or estimate in words. Rebuild missing background in the statistics subject guide, then use statistics practice to retrieve the relationship without being told its name.
Interpret a complete z-score calculation
The numerical line is only the middle of the answer; connect it back to the event, data, or uncertainty:
Compute without losing the statistical question
- Name the unknown, and the unit the answer has to come out in.
- Match the symbols to your values. $x$ is value, $\mu$ population mean, and $\sigma$ population standard deviation.
- Check the conditions before substituting. $\sigma>0$; a negative score lies below the mean.
- Substitute, keep exact values to the last line, then test the sign, size, and unit against a rough estimate - The check that catches most statistics slips.
Check the range, direction, and interpretation
After using z-score, repeat the calculation with units attached and test the result against a boundary case or a quick numerical estimate.
An inference this formula does not justify
What goes wrong: Dividing by the variance rather than the standard deviation.
Why it fails: With a value of 85, a mean of 70 and a standard deviation of 10, the correct score is 1.5. Using the variance of 100 gives 0.15, which makes an unusual value look ordinary.
Repair: Check the units: the denominator must be in the same units as the numerator, which means the square-rooted spread, never the squared one.
Questions about interpreting z-score
Does a z-score of 2 always mean the top 2.5 percent?
Only under a normal distribution, where the area above 2 is 0.02275. For an unknown shape that figure is a guess, and for a skewed distribution it can be badly wrong.
Can a z-score be larger than 3?
Yes, and in large data sets it should occasionally happen. Under a normal model only about 0.27 percent of values fall beyond three standard deviations either way, so such a value is rare rather than impossible.
What do I use when I only have sample statistics?
Substitute the sample mean and sample standard deviation. The result is an estimate rather than an exact standardisation, which is one reason small-sample inference uses t rather than z.
What are the mean and spread of the z-scores themselves?
Standardising always produces a set with mean 0 and standard deviation 1. That is the whole point: it puts every variable on one shared ruler.
Change one input, predict whether the output should rise, fall, or stay fixed, and only then recalculate.