A z score of 2.05 means a value is 2.05 standard deviations above the mean of its reference distribution. If the data follow a standard normal distribution, it corresponds to a cumulative probability of about 0.9798: roughly the 98th percentile, with about 2.02% of values above it.
What a z score tells you
A z score expresses how far an observation is from a mean in units of standard deviation. It makes values from distributions with different means or scales easier to compare, provided the reference distributions and calculations are appropriate.
The standard formula is z = (x − μ) / σ, where x is the observation, μ is the population mean, and σ is the population standard deviation. A positive z score indicates a value above the mean; a negative one indicates a value below it. A score of zero is exactly at the mean.
For example, if a test has a mean of 70 and a standard deviation of 10, a score of 90 has z = (90 − 70) / 10 = 2. If a score is 2.05 standard deviations above the mean, its raw value is μ + 2.05σ. The z score alone does not reveal the raw value unless the mean and standard deviation are known.
Percentile and probability for 2.05
For a standard normal distribution, the cumulative distribution function at 2.05 is approximately 0.9798. This is often described as the 97.98th percentile: about 97.98% of observations lie below the score, and about 2.02% lie above it.
These figures depend on the normal-distribution assumption. A z score is a standardized distance, but it does not automatically guarantee that real data are normally distributed. If a distribution is skewed, has heavy tails, or contains unusual clusters, the normal-table percentile may not describe the actual share of observations below the value.
- Below 2.05: about 97.98% under the standard normal model.
- Above 2.05: about 2.02% in the upper tail.
- Between −2.05 and 2.05: about 95.96% in a standard normal distribution.
The last figure is a two-sided area, not the same as the one-sided upper-tail probability. That distinction matters when interpreting test results.
How to calculate and interpret it
To calculate a z score from a raw value, subtract the relevant mean and divide by the relevant standard deviation. For a sample, analysts commonly use the sample mean and sample standard deviation when describing how an observation compares with that sample. In inferential statistics, the appropriate standard error—not necessarily the standard deviation of individual observations—may be needed instead.
- Identify the observation and the correct comparison group or reference distribution.
- Find the mean and the standard deviation, or standard error if the statistical procedure calls for it.
- Subtract the mean from the observation, then divide by the chosen scale.
- Use a normal table or calculator only if a normal-model probability is relevant and justified.
Interpret the result in context. A 2.05 might be high for a normally distributed exam score, but the practical meaning depends on the measure, population, and consequences. A high score is not inherently good or bad: it could represent a strong test result, an unusually long wait, or a measurement that merits checking.
One-tailed and two-tailed interpretations
A z score of 2.05 has an upper-tail probability of approximately 0.0202 under the standard normal distribution. For a two-tailed test that treats equally extreme positive and negative values as evidence, the probability of a result at least 2.05 standard deviations from zero in either direction is about 0.0404.
Which probability applies depends on the question and the test’s design. A one-tailed test asks about an effect in a specified direction; a two-tailed test considers departures in either direction. The choice should be based on the research question and made before looking at the result, not selected afterward to obtain a smaller p-value.
These probabilities are not the chance that a particular hypothesis is true. A p-value describes how unusual a result at least this extreme would be under a specified null model and its assumptions. It does not measure effect size, practical importance, or the probability that the finding will replicate.
Common misconceptions and limits
A z score of 2.05 is not the same as a 2.05% increase, nor does it mean that an observation is “2.05 times the average.” It measures a difference in standard-deviation units. The underlying units disappear when the value is standardized, but the result still depends on which mean and standard deviation were used.
Another common mistake is treating the 98th-percentile interpretation as universal. It is accurate for a standard normal model, and can be a useful approximation when data are reasonably normal. For a real dataset, calculate the empirical percentile directly if the distribution’s shape matters. Also check whether extreme values have distorted the mean or standard deviation.
When a z score is used in a hypothesis test, the test statistic may be calculated with a standard error. For example, a sample mean is compared with a hypothesized population mean using the standard error of the mean, often σ divided by the square root of the sample size when σ is known. That statistic answers a different question from standardizing one individual observation.
Frequently Asked Questions
Is a z score of 2.05 statistically significant?
It depends on the test and significance threshold. Under a standard normal model, its one-tailed p-value is about 0.0202 and its two-tailed p-value is about 0.0404. Both are below 0.05, but significance also depends on the chosen test, assumptions, and whether the analysis is one- or two-tailed.
What percentile is a z score of 2.05?
It is approximately the 97.98th percentile in a standard normal distribution. In other words, about 97.98% of values are lower and about 2.02% are higher under that model.
Is 2.05 a high z score?
It is relatively high in a standard normal distribution, sitting just above the 98th percentile. Whether it is unusual or important in practice depends on the distribution and the setting.
What is the p-value for z = 2.05?
The upper-tail probability is about 0.0202. For a symmetric, two-tailed test, the p-value is about 0.0404. A test’s direction and assumptions determine which result is appropriate.
Can I convert a z score of 2.05 into a raw score?
Yes, if you know the relevant mean and standard deviation: raw score = mean + 2.05 × standard deviation. For instance, with a mean of 70 and standard deviation of 10, the corresponding value is 90.5.
