WordGPT

t critical value, df = 18, 90% confidence

For 18 degrees of freedom the two-tailed t critical value at 90% confidence is 1.734. For a one-tailed test with α = 0.1 the critical value is 1.330. The matching z value is 1.645.

The t distribution with 18 degrees of freedom-5-4-3-2-1123450.10.20.30.4tdensityt* = 1.734t, df = 18standard normal
The t distribution with 18 degrees of freedom. Drag to move, scroll or pinch to zoom, double-click to reset.

How to work it out

  1. 1

    Degrees of freedom: for a single sample of size n they are n − 1. Here df = 18, so n = 19.

    df=n−1=18df = n - 1 = 18
  2. 2

    The confidence level 90% leaves α = 0.1 in the tails: half of it in each tail for a two-tailed interval.

    α2=0.05\frac{\alpha}{2} = 0.05
  3. 3

    Read the t-table at row df = 18 and the column for 0.05 in one tail.

    t∗=1.734t^{*} = 1.734
  4. 4

    Use it in the confidence interval for a mean when σ is unknown. For example, with x̄ = 50, s = 10 and n = 19:

    50±1.734⋅1019=50±3.978=(46.022, 53.978)50 \pm 1.734\cdot\frac{10}{\sqrt{19}} = 50 \pm 3.978 = (46.022,\ 53.978)
  5. 5

    The t value is larger than z = 1.645 because a small sample estimates the spread poorly; the difference is 5.4%, and it shrinks as df grows.

LaTeX source of the formulas on this page
\[ t^{*}_{18,\,90\%} = 1.734 \]
\[ df = n - 1 = 18 \]
\[ \frac{\alpha}{2} = 0.05 \]
\[ t^{*} = 1.734 \]
\[ 50 \pm 1.734\cdot\frac{10}{\sqrt{19}} = 50 \pm 3.978 = (46.022,\ 53.978) \]

t critical values for df = 18

ConfidenceTwo-tailed t*One-tailed t (α = 1 − C)z*
80%1.3300.8621.282
90%1.7341.3301.645
95%2.1011.7341.960
98%2.5522.2142.326
99%2.8782.5522.576

t* at 90% confidence for other degrees of freedom

dft*
16.314
22.920
52.015
101.812
201.725
301.697
601.671
1201.658

Questions

What is the t critical value for df = 18 at 90% confidence?

1.734 (two-tailed).

What is the one-tailed critical value for df = 18 and α = 0.1?

1.330.

Why is t larger than z?

With few degrees of freedom the sample standard deviation is an unreliable estimate, so the t distribution has heavier tails and needs a wider interval. At df = 18 the t value is 5.4% larger than z = 1.645.

How many degrees of freedom does a sample of 19 have?

n − 1 = 18 for a one-sample t interval or test.

Other confidence levels

Other degrees of freedom

    t critical value for 18 degrees of freedom at 90% confidence: 1.734 | WordGPT