Z-Score Table: Full Negative & Positive Z Table (−3.49 to 3.49)

To read a z-score table, split the z-score in two. The row gives the whole number and first decimal, and the column gives the second decimal. For z = 1.47, go down to the row for 1.4 and across to the column for .07. The cell says 0.9292, so 92.92% of a normal distribution lies below z = 1.47.

Every value in the two tables below is that kind of number: the area under the standard normal curve to the left of z, also called the cumulative probability or Φ(z). The values are rounded to four decimal places and cover z from −3.49 to 3.49. Some textbooks print a different layout that starts at 0.0000 instead of 0.5000. The z-score calculator page explains how to tell the three common layouts apart and convert between them.

For a z-score with more decimals, an area the tables don’t list, or a value far out in the tails, use the calculator below. It is set to the worked example further down this page: the area between z = −0.75 and z = 1.30.

Z-Score Calculator

Type the score, the mean and the standard deviation in the same units. Decimals and negatives are fine.

z-score –
Normal curve (shaded area)
Step by step

    The z-Table for Negative z-Scores (−3.49 to −0.00)

    Use this half when your z-score is below zero, which means the value is below the mean. Each cell is the area to the left of z, so all of them are 0.5 or less. The row gives the sign, the whole number and the first decimal, and the column gives the second decimal: z = −1.36 is row −1.3, column .06.

    z.00.01.02.03.04.05.06.07.08.09
    −3.40.00030.00030.00030.00030.00030.00030.00030.00030.00030.0002
    −3.30.00050.00050.00050.00040.00040.00040.00040.00040.00040.0003
    −3.20.00070.00070.00060.00060.00060.00060.00060.00050.00050.0005
    −3.10.00100.00090.00090.00090.00080.00080.00080.00080.00070.0007
    −3.00.00130.00130.00130.00120.00120.00110.00110.00110.00100.0010
    −2.90.00190.00180.00180.00170.00160.00160.00150.00150.00140.0014
    −2.80.00260.00250.00240.00230.00230.00220.00210.00210.00200.0019
    −2.70.00350.00340.00330.00320.00310.00300.00290.00280.00270.0026
    −2.60.00470.00450.00440.00430.00410.00400.00390.00380.00370.0036
    −2.50.00620.00600.00590.00570.00550.00540.00520.00510.00490.0048
    −2.40.00820.00800.00780.00750.00730.00710.00690.00680.00660.0064
    −2.30.01070.01040.01020.00990.00960.00940.00910.00890.00870.0084
    −2.20.01390.01360.01320.01290.01250.01220.01190.01160.01130.0110
    −2.10.01790.01740.01700.01660.01620.01580.01540.01500.01460.0143
    −2.00.02280.02220.02170.02120.02070.02020.01970.01920.01880.0183
    −1.90.02870.02810.02740.02680.02620.02560.02500.02440.02390.0233
    −1.80.03590.03510.03440.03360.03290.03220.03140.03070.03010.0294
    −1.70.04460.04360.04270.04180.04090.04010.03920.03840.03750.0367
    −1.60.05480.05370.05260.05160.05050.04950.04850.04750.04650.0455
    −1.50.06680.06550.06430.06300.06180.06060.05940.05820.05710.0559
    −1.40.08080.07930.07780.07640.07490.07350.07210.07080.06940.0681
    −1.30.09680.09510.09340.09180.09010.08850.08690.08530.08380.0823
    −1.20.11510.11310.11120.10930.10750.10560.10380.10200.10030.0985
    −1.10.13570.13350.13140.12920.12710.12510.12300.12100.11900.1170
    −1.00.15870.15620.15390.15150.14920.14690.14460.14230.14010.1379
    −0.90.18410.18140.17880.17620.17360.17110.16850.16600.16350.1611
    −0.80.21190.20900.20610.20330.20050.19770.19490.19220.18940.1867
    −0.70.24200.23890.23580.23270.22960.22660.22360.22060.21770.2148
    −0.60.27430.27090.26760.26430.26110.25780.25460.25140.24830.2451
    −0.50.30850.30500.30150.29810.29460.29120.28770.28430.28100.2776
    −0.40.34460.34090.33720.33360.33000.32640.32280.31920.31560.3121
    −0.30.38210.37830.37450.37070.36690.36320.35940.35570.35200.3483
    −0.20.42070.41680.41290.40900.40520.40130.39740.39360.38970.3859
    −0.10.46020.45620.45220.44830.44430.44040.43640.43250.42860.4247
    −0.00.50000.49600.49200.48800.48400.48010.47610.47210.46810.4641

    The z-Table for Positive z-Scores (0.00 to 3.49)

    Use this half when your z-score is zero or above. Each cell is still the area to the left of z, so it includes the whole left half of the curve and every value is 0.5 or more. That is why the 0.00 cell is 0.5000 in both halves: half the distribution lies below the mean.

    z.00.01.02.03.04.05.06.07.08.09
    0.00.50000.50400.50800.51200.51600.51990.52390.52790.53190.5359
    0.10.53980.54380.54780.55170.55570.55960.56360.56750.57140.5753
    0.20.57930.58320.58710.59100.59480.59870.60260.60640.61030.6141
    0.30.61790.62170.62550.62930.63310.63680.64060.64430.64800.6517
    0.40.65540.65910.66280.66640.67000.67360.67720.68080.68440.6879
    0.50.69150.69500.69850.70190.70540.70880.71230.71570.71900.7224
    0.60.72570.72910.73240.73570.73890.74220.74540.74860.75170.7549
    0.70.75800.76110.76420.76730.77040.77340.77640.77940.78230.7852
    0.80.78810.79100.79390.79670.79950.80230.80510.80780.81060.8133
    0.90.81590.81860.82120.82380.82640.82890.83150.83400.83650.8389
    1.00.84130.84380.84610.84850.85080.85310.85540.85770.85990.8621
    1.10.86430.86650.86860.87080.87290.87490.87700.87900.88100.8830
    1.20.88490.88690.88880.89070.89250.89440.89620.89800.89970.9015
    1.30.90320.90490.90660.90820.90990.91150.91310.91470.91620.9177
    1.40.91920.92070.92220.92360.92510.92650.92790.92920.93060.9319
    1.50.93320.93450.93570.93700.93820.93940.94060.94180.94290.9441
    1.60.94520.94630.94740.94840.94950.95050.95150.95250.95350.9545
    1.70.95540.95640.95730.95820.95910.95990.96080.96160.96250.9633
    1.80.96410.96490.96560.96640.96710.96780.96860.96930.96990.9706
    1.90.97130.97190.97260.97320.97380.97440.97500.97560.97610.9767
    2.00.97720.97780.97830.97880.97930.97980.98030.98080.98120.9817
    2.10.98210.98260.98300.98340.98380.98420.98460.98500.98540.9857
    2.20.98610.98640.98680.98710.98750.98780.98810.98840.98870.9890
    2.30.98930.98960.98980.99010.99040.99060.99090.99110.99130.9916
    2.40.99180.99200.99220.99250.99270.99290.99310.99320.99340.9936
    2.50.99380.99400.99410.99430.99450.99460.99480.99490.99510.9952
    2.60.99530.99550.99560.99570.99590.99600.99610.99620.99630.9964
    2.70.99650.99660.99670.99680.99690.99700.99710.99720.99730.9974
    2.80.99740.99750.99760.99770.99770.99780.99790.99790.99800.9981
    2.90.99810.99820.99820.99830.99840.99840.99850.99850.99860.9986
    3.00.99870.99870.99870.99880.99880.99890.99890.99890.99900.9990
    3.10.99900.99910.99910.99910.99920.99920.99920.99920.99930.9993
    3.20.99930.99930.99940.99940.99940.99940.99940.99950.99950.9995
    3.30.99950.99950.99950.99960.99960.99960.99960.99960.99960.9997
    3.40.99970.99970.99970.99970.99970.99970.99970.99970.99970.9998

    Left, Right, Between or Outside: Four Areas From the Same Cells

    The table only ever gives the area to the left. Every other area comes from one or two cells with a subtraction:

    The area you needHow to get it from the tableExample
    Left of z (below)Read the cell for zLeft of 1.30 = 0.9032
    Right of z (above)1 − the cell for zRight of 1.30 = 1 − 0.9032 = 0.0968
    Between z₁ and z₂Cell for the larger z − cell for the smaller zBetween −0.75 and 1.30 = 0.9032 − 0.2266 = 0.6766
    Outside −z and z (two tails)2 × the cell for −|z|Outside ±1.30 = 2 × 0.0968 = 0.1936

    The between rule works whatever the signs are. For −0.75 to 1.30, the cell for 1.30 holds everything below 1.30, and the cell for −0.75 holds everything below −0.75. Take the second away from the first and only the strip between them is left: 0.6766, or 67.66% of the distribution.

    A right-hand area for a negative z goes above 0.5. To the right of −0.75 is 1 − 0.2266 = 0.7734, because that area includes the whole upper half of the curve. If you get 0.2266 for “above −0.75”, you have read the left side by mistake.

    The outside rule gives a two-tailed p-value. For z = 1.30, the chance of landing at least that far from the mean in either direction is 0.1936. A one-tailed test would use only one of the two tails, 0.0968.

    Two smaller points. First, “less than” and “less than or equal to” give the same area. A single exact point has no width, so P(Z = 1.30) is 0 and P(Z < 1.30) = P(Z ≤ 1.30) = 0.9032. Second, the between rule still works when both z-scores are negative. Between −2.10 and −0.40, the area is 0.3446 − 0.0179 = 0.3267. Subtract the smaller cell from the larger and the answer can’t come out negative.

    One Table Covers Both Signs

    The standard normal curve is symmetric around zero, so the area left of −z equals the area right of +z. In the table, every pair of cells for z and −z adds up to 1:

    • Left of −1.02 = 0.1539
    • Left of 1.02 = 0.8461
    • 0.1539 + 0.8461 = 1.0000

    That is why some books print only the positive half. With a positive-only table, the area left of a negative z is 1 minus the cell for the matching positive z. For z = −1.02, that is 1 − 0.8461 = 0.1539. Calling 0.8461 the left area of −1.02 is a slip: it is the area to the right of −1.02.

    Splitting a z-table into a “negative table” and a “positive table” is only a page layout. Both halves are cut from the same curve, and either half can be rebuilt from the other.

    Symmetry also gives a quick way to catch a misread cell. A left area below 0.5 means a negative z, and a left area above 0.5 means a positive z. The cells for ±1 should give the 68% of the 68–95–99.7 rule: 0.8413 − 0.1587 = 0.6826. The exact figure is 0.6827. The last digit differs because each cell was rounded on its own.

    From Your Own Measurement to the Table

    The table works in z-scores, so a raw value has to be converted first with z = (x − mean) ÷ standard deviation.

    Say a machine makes bolts with a mean length of 50.0 mm and a standard deviation of 0.2 mm, and lengths follow a normal curve. The specification accepts bolts from 49.7 mm to 50.4 mm. What share passes?

    1. Lower limit: z = (49.7 − 50.0) ÷ 0.2 = −1.5. The cell for −1.50 is 0.0668.
    2. Upper limit: z = (50.4 − 50.0) ÷ 0.2 = 2.0. The cell for 2.00 is 0.9772.
    3. Between them: 0.9772 − 0.0668 = 0.9104.

    About 91.04% of bolts meet the specification, and 8.96% don’t. Of those, 6.68% are too short and 2.28% too long. The two limits are not the same distance from the mean, so the rejects are not split evenly.

    The answer is only as good as the normal assumption. If the lengths are skewed, the table can be well off in the tails, where the rejects are. If you have the measurements themselves, the standard deviation calculator gives the mean and SD to start from, and its dot plot shows whether the values look roughly bell-shaped.

    Reading the Table Backwards: From an Area to z

    Sometimes you know the area and need the z-score: the cut-off for the top 3%, or the z below which 15% of values fall. Then you search the body of the table for the area and read the row and column at the edges.

    1. Turn the area into a left area. The top 3% is 3% to the right, so 97% (0.9700) is to the left.
    2. Find the cell closest to it. In the positive half, 0.9699 sits at z = 1.88 and 0.9706 at z = 1.89. 0.9699 is closer.
    3. Read off z = 1.88.

    For the bolts, the longest 3% are those over 50.0 + 1.88 × 0.2 = 50.376 mm.

    For a small area, search the negative half. The bottom 15% needs a left area of 0.1500. The cells nearby are 0.1492 at z = −1.04 and 0.1515 at z = −1.03, so z ≈ −1.04.

    Some cut-offs you’ll need again and again:

    Left areaNearest cellsTable answerExact z
    0.80000.7995 (0.84), 0.8023 (0.85)0.840.8416
    0.90000.8997 (1.28), 0.9015 (1.29)1.281.2816
    0.95000.9495 (1.64), 0.9505 (1.65)1.6451.6449
    0.97500.9750 (1.96)1.961.9600
    0.99000.9898 (2.32), 0.9901 (2.33)2.332.3263

    0.95 is the awkward one. It sits exactly halfway between two cells, 0.0005 from each, so neither 1.64 nor 1.65 is closer. Taking the midpoint gives 1.645, the value printed in tables of critical values. 0.975 is the easy one: it appears in the table exactly, at z = 1.96, which is where the ±1.96 for 95% confidence comes from.

    For a cut-off on a real scale, such as a test score, the Percentile → z tab of the z-score calculator does both steps and turns the z back into a score.

    Three Decimals in z: Interpolating Between Cells

    The table steps in hundredths, so a z of 1.237 falls between two columns. You can round it to 1.24, or you can interpolate, which means going the matching fraction of the way from one cell to the next:

    1. The cell for 1.23 is 0.8907 and the cell for 1.24 is 0.8925. The gap is 0.0018.
    2. 1.237 is 0.7 of the way from 1.23 to 1.24.
    3. 0.8907 + 0.7 × 0.0018 = 0.89196, which rounds to 0.8920.

    The exact value is 0.89196, so interpolation gets the fourth decimal right here. Rounding z to 1.24 would have given 0.8925.

    The same idea works backwards. For a left area of 0.9500, the cells 0.9495 and 0.9505 are 0.0010 apart, and 0.9500 is halfway, so z = 1.64 + 0.5 × 0.01 = 1.645. The exact z is 1.6449.

    Interpolating along a straight line between cells is a small approximation, because the curve bends. Between two cells only 0.01 apart, the bend is too small to change the fourth decimal in these examples. It matters more when a table only lists z to one decimal place.

    How Far Off a Rounded z Can Be

    Rounding z to two decimals before looking it up costs some accuracy. Take a score of 62 on a test with a mean of 55 and a standard deviation of 12:

    • z = (62 − 55) ÷ 12 = 0.58333…
    • The cell for 0.58 is 0.7190.
    • The exact area is 0.7202.

    The table answer is 0.12 percentage points low. That is too small to matter for a rough percentile, but enough to change the last digit of a reported probability.

    The error is largest near the middle of the curve, where the curve is tallest and the area changes fastest. Rounding z by up to 0.005 there moves the area by up to about 0.002. In the tails the absolute error shrinks, but so does the area itself. Near z = 3, rounding 2.995 up to 3.00 gives a tail area of 0.00135 instead of 0.00137, which is 1.6% too small.

    To keep a table answer close, round z once at the end, not at each step, and interpolate when the third decimal matters. To skip the error altogether, type the full z into the calculator at the top of this page. When you start from a raw score, its Score → z tab also names the row and column to use, so you can see where the table answer would land.

    Where the Numbers in the Table Come From

    Each cell is the area under the bell curve y = e^(−z²/2) ÷ √(2π) from far left up to z. That area has no formula in ordinary functions, so it is computed numerically, and in terms of the error function it is Φ(z) = ½ × [1 + erf(z ÷ √2)]. The tables on this page were generated that way and each value was rounded half up to four decimals.

    You can check any cell yourself. In Excel or Google Sheets, =NORM.S.DIST(1.47, TRUE) returns 0.929219…, which rounds to the 0.9292 in row 1.4, column .07. With FALSE in place of TRUE, the same function returns the height of the curve at 1.47 (0.1354), not the area, so a check with the wrong last argument will never match the table.

    Because every cell is rounded on its own, sums of cells can be off by 0.0001. That is the 0.6826 against 0.6827 in the symmetry check above, and it is why two books can differ in the last digit of a calculation without either being wrong.

    Eight Lookups to Practise On

    Cover the answers and work each one from the two tables.

    1. The area to the left of z = 0.67.
    2. The area to the right of z = 2.14.
    3. The area to the left of z = −1.88.
    4. The area to the right of z = −0.43.
    5. The area between z = −2.10 and z = −0.40.
    6. The two-tailed area outside z = ±2.58.
    7. The z-score with a left area of 0.6000.
    8. The z-score that cuts off the top 25%.

    Answers:

    1. 0.7486, read straight from row 0.6, column .07.
    2. 0.0162: 1 − 0.9838.
    3. 0.0301, read straight from row −1.8, column .08.
    4. 0.6664: 1 − 0.3336. It is above 0.5 because the area includes the whole upper half.
    5. 0.3267: 0.3446 − 0.0179.
    6. 0.0098: 2 × 0.0049. The exact value is 0.00988, so doubling a rounded cell lands 0.0001 low.
    7. 0.25: the nearest cells are 0.5987 at 0.25 and 0.6026 at 0.26. The exact z is 0.2533.
    8. 0.67: the top 25% means a left area of 0.7500, and the nearest cell is 0.7486 at 0.67 (0.7517 at 0.68 is further). The exact z is 0.6745.

    Past the Edge: What the Table Can’t Show Beyond ±3.49

    A z-score can be any number. Values outside ±3 are rare under a normal curve, but they are not impossible. A printed table has to stop somewhere, and this one stops at ±3.49.

    Near the edge, four decimal places stop telling the cells apart. Every cell from 3.40 to 3.48 shows 0.9997, and 3.49 shows 0.9998. A table that went on to 3.90 would show 1.0000, but the area above z = 3.9 is not zero. It is 0.0000481, about 1 in 20,800. Rounding has simply pushed it out of sight.

    Further out, the tail areas are:

    zArea to the rightAbout 1 in
    3.50.0002334,300
    4.00.000031731,600
    4.50.0000034294,000
    6.00.000000000991.01 billion

    The 4.5 row is the source of the Six Sigma quality target of 3.4 defects per million. Six Sigma programs allow for the process mean drifting 1.5 standard deviations over time, so a limit set six standard deviations from the mean ends up 4.5 standard deviations away. The one-sided area beyond 4.5 is 3.4 in a million.

    These tail figures are exactly where the normal model is weakest. Real data don’t have to follow a bell curve that far out, so a z of 5 in a real data set is more likely a measurement error, a mix of two groups, or a skewed distribution than a 1-in-3.5-million event. The calculator gives tail areas this small in scientific notation, but treat them as properties of the normal curve, not as proof about your data.