Z Score of 1.866: Meaning, Percentile, and Probability

Home » Blog » Z Score of 1.866: Meaning, Percentile, and Probability

A z score of 1.866 means a value is 1.866 standard deviations above the mean of its reference distribution. If the distribution is standard normal, that score corresponds to a cumulative probability of about 0.9690: roughly 96.9% of values are below it and 3.1% are above it.

What a z score of 1.866 tells you

A z score expresses a value’s distance from a mean in standard-deviation units. Its sign gives the direction: a positive score is above the mean, while a negative score is below it. The size tells how far away it is relative to the spread of the data.

For example, if test scores have a mean of 70 and a standard deviation of 10, a score with z = 1.866 is 18.66 points above the mean, or 88.66. That conversion works when the stated mean and standard deviation apply to the score being evaluated. The z score itself does not reveal the original measurement or tell you whether the result is important in practice.

A score of 1.866 is therefore above average, but not unusually extreme by common statistical conventions. It is below the frequently used threshold of 1.96 standard deviations, which corresponds approximately to the upper 2.5% of a standard normal distribution in a two-sided comparison.

Percentile and probability

When a z score is interpreted using the standard normal distribution, its cumulative distribution value is the area under the normal curve to the left of the score. For z = 1.866, this value is approximately 0.9690. In practical terms:

  • Below the score: About 96.9% of observations in a standard normal population fall below it.
  • Above the score: About 3.1% fall above it.
  • Between the mean and the score: About 46.9% fall between z = 0 and z = 1.866.

These percentages describe areas under a theoretical normal curve. They are not automatically the percentage of people in a particular study who scored lower or higher. That interpretation is appropriate when the normal model is relevant and the score has been standardized against the right population.

For a two-tailed probability—the chance of a standard normal value being at least this far from zero in either direction—double the upper-tail area: about 2 × 0.0310 = 0.0620, or 6.2%. This is different from the one-tailed upper probability of 3.1%.

How the score and lookup are calculated

For an observation x, population mean μ, and population standard deviation σ, the z score is calculated as:

z = (x − μ) / σ

If you have a sample mean and sample standard deviation rather than population values, those estimates are used for descriptive standardization. In inferential tests, the appropriate standard error or test-specific formula may be needed instead; a raw observation’s z score is not interchangeable with every statistical test statistic.

To find the probability for 1.866, use a standard normal cumulative distribution function, a z table, or statistical software. A calculator or software command that returns the cumulative normal probability gives approximately 0.9690. Some traditional printed z tables list areas from zero to z rather than cumulative areas from the far left; for this score, the table entry is approximately 0.4690, which must be added to 0.5000 to get 0.9690.

Values can differ slightly across tables because of rounding. A table may show z only to two decimal places, requiring you to use 1.87 or interpolate between nearby entries. Reporting 0.9690 is a useful approximation, not a claim that the probability is exact to four decimal places in every application.

One-tailed, two-tailed, and significance

The phrase “probability for a z score” can mean different things. If the question is how much of the normal curve lies below 1.866, the answer is about 0.9690. If it asks for the upper-tail probability, use 1 − 0.9690 = 0.0310. If it asks for a two-tailed probability, use about 0.0620.

In a hypothesis test, which tail applies depends on the research question and the alternative hypothesis—not on which result gives a smaller p-value. Under a standard normal null model, a one-sided test in the direction of a positive z statistic has a p-value near 0.031. A two-sided test has a p-value near 0.062. Thus, at a conventional 5% significance level, the one-sided result would be below 0.05, while the two-sided result would not.

That comparison alone does not establish that an effect is meaningful or that a hypothesis is true. Statistical significance depends on the test assumptions and design; practical importance also depends on the size and context of the effect, measurement quality, and uncertainty.

What the number does—and does not—mean

A z score is a relative position, not a direct measure of rarity in every dataset. The normal-curve percentages apply only if the reference distribution is standard normal or is reasonably modeled by one. If the original data are skewed, have outliers, or come from a small or unusual sample, converting a value to a z score does not make the underlying distribution normal.

It is also a misconception that z = 1.866 means there is a 96.9% chance the observation is correct, or a 96.9% chance that a person will achieve a particular outcome. The 96.9% figure is a cumulative proportion under a specified distribution. It does not measure the probability that an individual claim or prediction is true.

For a normally distributed variable, a value near z = 1.866 is in the upper portion of the distribution, but it is not beyond the common 1.96 cutoff. That comparison is only a quick reference: 1.96 is associated with a two-sided 5% standard-normal threshold, while actual confidence intervals and tests may use other distributions, degrees of freedom, or assumptions.

Frequently Asked Questions

What percentile is a z score of 1.866?
It is approximately the 96.9th percentile under the standard normal distribution. In other words, about 96.9% of values are lower and about 3.1% are higher.

What is the p-value for z = 1.866?
For an upper-tailed standard normal test, it is approximately 0.031. For a two-tailed test, it is approximately 0.062. The correct choice depends on the test’s prespecified alternative hypothesis.

Is a z score of 1.866 statistically significant?
It is significant at the 5% level for an upper-tailed standard-normal test, but not for a two-tailed test, whose p-value is about 0.062. Significance should be judged using the test design and assumptions, not the score alone.

How many standard deviations above the mean is 1.866?
It is 1.866 standard deviations above the mean. The corresponding original-unit distance is 1.866 multiplied by the standard deviation.

How do I find the area to the right of 1.866?
Subtract its cumulative left-tail probability from 1: 1 − 0.9690 ≈ 0.0310, or about 3.1%.

Quick interpretation

For a standard normal distribution, z = 1.866 is above the mean, near the 96.9th percentile, with about 3.1% of the distribution to its right. Before applying that result, check whether the normal model and the chosen one- or two-tailed interpretation match the question you are trying to answer.

Related

Scroll to Top