When should I use CONFIDENCE.NORM instead of other confidence functions?

Use CONFIDENCE.NORM for large samples (n ≥ 30) or when you know the true population standard deviation. If you have a small sample or only a sample standard deviation, use CONFIDENCE.T with the t-distribution instead for more conservative (wider) confidence intervals.

How do I convert the margin of error into an upper and lower confidence bound?

Subtract the margin of error from your sample mean for the lower bound, and add it for the upper bound. For example, if CONFIDENCE.NORM returns 2000 and your average is 50000, the interval is [48000, 52000].

What if I have raw salary data in a spreadsheet column instead of pre-calculated inputs?

Nest functions to calculate standard deviation and sample size on the fly. Use the formula =CONFIDENCE.NORM(0.05, STDEV(A2:A20), COUNT(A2:A20)) to compute all three inputs directly from your salary column.

What does alpha=0.05 mean, and why do people use it?

Alpha is the significance level, representing the probability of error. An alpha of 0.05 corresponds to 95% confidence, meaning there is a 5% chance the true mean falls outside your calculated interval. The 95% standard is a convention balancing statistical certainty with practical margins.

CONFIDENCE.NORM function

Returns the margin of error for a population mean estimate using the standard normal distribution with a specified confidence level and sample size.

=CONFIDENCE.NORM(alpha, standard_dev, size)

Generate a CONFIDENCE.NORM formula

Describe what you need. The generator will reach for CONFIDENCE.NORM where CONFIDENCE.NORM is the right tool, and tell you when it is not.

Arguments

How CONFIDENCE.NORM reads its arguments

Argument	Required	Description
alpha	Required	The significance level, a decimal strictly between 0 and 1 (e.g., 0.05 for 95% confidence). Values outside this range return #NUM!.
standard_dev	Required	The standard deviation of the population or sample, in the same units as the data. Must be non-negative; negative values return #NUM!.
size	Required	The sample size (count of observations), must be a positive integer ≥ 1. Values less than 1 return #NUM! or #DIV/0!.

Returns

Returns a single number representing the margin of error (half-width of the confidence interval).

Availability

Excel: All · Google Sheets: Supported

Worked examples

1. Calculate margin of error for sales salary analysis

Name	Department	Salary
Sarah Chen	Sales	65000
Robert Wilson	Sales	72000
Jennifer Lee	Sales	68000

=CONFIDENCE.NORM(0.05, 3500, 3)

Result: 3960.08

With 95% confidence (alpha=0.05), a standard deviation of $3,500, and 3 sales staff, the margin of error is $3,960.08. The true average sales salary is estimated to lie within the sample mean ± $3,960.08.

2. Determine confidence width for engineering department

Name	Department	Salary
Marcus Johnson	Engineering	95000
James Park	Engineering	85000
Emma White	Engineering	80000

=CONFIDENCE.NORM(0.05, 7500, 3)

Result: 8485.46

The engineering department's higher salary variation (std dev $7,500) produces a larger margin of error of $8,485.46. This reflects greater uncertainty in the true average, even with the same confidence level and sample size.

3. Find confidence margin for company-wide salary estimate

ID	Name	Department	Salary
1	Sarah Chen	Sales	65000
2	Marcus Johnson	Engineering	95000
3	Lisa Rodriguez	Marketing	58000
4	James Park	Engineering	85000
5	Angela Davis	HR	62000
6	Robert Wilson	Sales	72000
7	Emma White	Engineering	80000
8	David Brown	Finance	78000
9	Jennifer Lee	Sales	68000
10	Michael Torres	Marketing	61000

=CONFIDENCE.NORM(0.10, 12500, 10)

Result: 6501.61

Using 90% confidence (alpha=0.10) for all 10 employees with standard deviation $12,500, the margin of error is $6,501.61. The lower confidence threshold produces a narrower margin than 95% confidence, accepting more risk for a tighter estimate range.

Common errors

Which CONFIDENCE.NORM error are you seeing?

Error	Why it happens	How to fix it
#NUM!	alpha is not strictly between 0 and 1 (e.g., 0, 1, −0.05); or standard_dev is negative; or size is less than 1.	Ensure 0 < alpha < 1, standard_dev ≥ 0, and size ≥ 1. For 95% confidence use 0.05; for 90% use 0.10. Verify cell references contain valid numbers.
#VALUE!	One or more arguments contain text, dates, or other non-numeric data instead of numbers.	Verify all three arguments are pure numbers. Convert text to numbers with VALUE() if needed, or reference numeric cells directly.
#DIV/0!	size equals 0, causing division by zero in the √size denominator of the margin-of-error formula.	Ensure size ≥ 1. If using a COUNT formula, verify your data range is not empty.

Tips and when to use something else

CONFIDENCE.NORM returns only the margin of error, not the full confidence interval. Add it to the sample mean for the upper bound and subtract for the lower bound. For example, if the sample mean is $70,000 and the margin is $4,000, the 95% confidence interval is $66,000–$74,000.
This function assumes the data is normally distributed. For small samples (n < 30) or when the population standard deviation is unknown, use CONFIDENCE.T instead, which employs the t-distribution for more accurate results.
Pre-calculate standard deviation and sample size from your raw data using helper columns. Filter by department with COUNTIF and AVERAGEIF before passing aggregated values to CONFIDENCE.NORM.
Smaller alpha values (e.g., 0.01 for 99% confidence) produce wider confidence intervals; larger values (e.g., 0.10 for 90% confidence) produce narrower intervals. Choose alpha based on your decision's stakes.

Related functions

COUNTCOUNT counts the number of cells containing numeric values and dates in one or more ranges, ignoring text and blank cells.AVERAGEAVERAGE calculates the arithmetic mean of numbers, ignoring text and logical values; returns a single numeric value representing the average.AVERAGEIFReturns the average of cells matching a single criterion, useful for conditional averages like mean cost by warehouse or mean inventory per SKU.COUNTIFCounts cells in a range that meet a specific criterion, returning a number — use it to tally matching rows like employee counts by department.

Frequently asked questions

When should I use CONFIDENCE.NORM instead of other confidence functions?: Use CONFIDENCE.NORM for large samples (n ≥ 30) or when you know the true population standard deviation. If you have a small sample or only a sample standard deviation, use CONFIDENCE.T with the t-distribution instead for more conservative (wider) confidence intervals.
How do I convert the margin of error into an upper and lower confidence bound?: Subtract the margin of error from your sample mean for the lower bound, and add it for the upper bound. For example, if CONFIDENCE.NORM returns 2000 and your average is 50000, the interval is [48000, 52000].
What if I have raw salary data in a spreadsheet column instead of pre-calculated inputs?: Nest functions to calculate standard deviation and sample size on the fly. Use the formula =CONFIDENCE.NORM(0.05, STDEV(A2:A20), COUNT(A2:A20)) to compute all three inputs directly from your salary column.
What does alpha=0.05 mean, and why do people use it?: Alpha is the significance level, representing the probability of error. An alpha of 0.05 corresponds to 95% confidence, meaning there is a 5% chance the true mean falls outside your calculated interval. The 95% standard is a convention balancing statistical certainty with practical margins.

Need a different formula?

The full generator is not scoped to one function — describe any spreadsheet problem and it will pick.

Open the formula generator

Reviewed 2026-09-17