Z Score of 1.65: Meaning, Probability, and Uses

Home » Blog » Z Score of 1.65: Meaning, Probability, and Uses

A z score of 1.65 means a value is 1.65 standard deviations above the mean of a distribution. In a standard normal distribution, that corresponds to a cumulative probability of about 0.9505, or the 95th percentile; the exact interpretation depends on whether you are looking at a one-tailed or two-tailed probability.

What a z score measures

A z score expresses how far a value is from a distribution’s mean in units of standard deviation. Its formula is z = (x − μ) / σ, where x is the value, μ is the mean, and σ is the standard deviation. A positive score is above the mean; a negative score is below it.

For example, suppose test scores have a mean of 70 and a standard deviation of 10. A score with z = 1.65 is 70 + (1.65 × 10) = 86.5. This calculation describes position relative to the mean and spread, not whether the score is good or bad.

  • z = 0: exactly at the mean.
  • z = 1: one standard deviation above the mean.
  • z = −1.65: 1.65 standard deviations below the mean.

Probability associated with 1.65

If observations follow a standard normal distribution, the probability of a value at or below z = 1.65 is approximately 0.9505. In other words, about 95.05% of the distribution lies to its left, and about 4.95% lies to its right. Tables and calculators may round this to 0.950 or 0.9505.

The area between the mean and 1.65 is about 0.4505, because half the distribution, 0.5000, lies below the mean. These are different quantities: a z table may report cumulative area, area from zero, or an upper-tail area. Check the table’s heading before interpreting its number.

A z score alone does not guarantee a probability statement about real-world data. The normal-curve percentages apply when the normal model is appropriate. For a skewed distribution, or data with unusual outliers, the same standardized score may not correspond to the same percentile.

Why 1.65 appears in statistical tests

The value 1.645, usually rounded to 1.65, is a common critical value for a one-sided test at a 5% significance level under the standard normal model. For an upper-tailed test, the area above 1.645 is approximately 0.05. A result beyond that cutoff is often called statistically significant at the 5% level, given the test’s assumptions and setup.

For a lower-tailed test, the corresponding cutoff is −1.645. For a two-sided test with 5% significance, the 5% is split between both tails: the familiar standard normal cutoffs are approximately −1.96 and +1.96, not ±1.65. Thus, saying “1.65 is the 5% critical value” is incomplete unless the test is one-sided.

Critical values are decision thresholds, not measures of effect size or practical importance. A statistically significant result can be small in real-world terms, and a result that misses a cutoff does not prove that there is no effect.

Using 1.65 for a confidence interval

For a normally distributed estimate with a known standard error, a one-sided 95% bound uses approximately 1.645 standard errors. For example, an upper one-sided bound is commonly calculated as the estimate plus 1.645 times its standard error. A lower one-sided bound uses subtraction.

A conventional two-sided 95% confidence interval instead uses approximately 1.96 standard errors on each side under the standard normal method. The distinction is about the question being asked: a one-sided interval sets a bound in one direction, while a two-sided interval gives lower and upper bounds. In many practical analyses, especially when estimating a mean with a small sample and unknown population standard deviation, a t distribution is more appropriate than the standard normal distribution. Its critical value depends on degrees of freedom.

How to calculate or find it

To find a z score for a raw observation, subtract the mean and divide by the standard deviation. To find the raw value from a z score, rearrange the formula: x = μ + zσ. If you already have z = 1.65, the raw value is therefore 1.65 standard deviations above the mean.

To look up its probability, use a standard normal table or a statistics calculator and identify the kind of area requested:

  • Left-tail probability: P(Z ≤ 1.65) ≈ 0.9505.
  • Right-tail probability: P(Z > 1.65) ≈ 0.0495.
  • Two-tailed probability beyond ±1.65: about 2 × 0.0495 = 0.099.

For a measured dataset, a z score is meaningful only in relation to the mean and standard deviation used to calculate it. Whether those are population values or estimates from a sample matters, particularly in formal inference.

Common interpretation mistakes

  • Confusing 1.65 with 1.65%: It is a standardized distance, not a percentage. The associated probability depends on the area being considered.
  • Treating 95% as the area above the score: About 95.05% is below 1.65; only about 4.95% is above it.
  • Using a one-tailed cutoff for a two-tailed question: At a 5% two-sided significance level, use approximately ±1.96 for a standard normal test.
  • Assuming every dataset is normal: A z score can be calculated for many datasets, but normal-curve probabilities require a suitable distributional model.
  • Equating a percentile with a probability about a hypothesis: A percentile describes location under a distribution. A p-value is calculated under a specified null hypothesis and test procedure; they are related through tail areas in some tests but are not interchangeable ideas.

Frequently Asked Questions

Is a z score of 1.65 good?
A score of 1.65 is above the mean by 1.65 standard deviations. Whether that is good depends on what is being measured and whether higher values are desirable. In a normally distributed population, it is around the 95th percentile.

What percentile is z = 1.65?
It is approximately the 95.05th percentile under the standard normal distribution. The percentile is the share of the modeled distribution at or below that score.

What is the p-value for z = 1.65?
For an upper-tailed test, the p-value is about 0.0495. For a lower-tailed test, it is about 0.9505; for a two-sided test based on the magnitude 1.65, it is about 0.099. The test’s direction must be specified to give the correct p-value.

Is 1.65 statistically significant?
It meets the approximate 5% cutoff for an upper-tailed standard normal test, but not the usual 5% cutoff for a two-tailed test. Significance also depends on the test assumptions, the chosen threshold, and whether the analysis was planned as one- or two-sided.

What does z = −1.65 mean?
It means the value is 1.65 standard deviations below the mean. Under the standard normal model, about 4.95% of observations lie below it and about 95.05% lie above it.

Related

Scroll to Top