Statistical Estimation
The formulas used in
this chapter are the same as from the normal distribution, just
re-arranged. There are two basic types
of problems, calculating a confidence interval and determining the required n
for a sample. The first set of formulas are used for calculating a confidence
interval. There are three formulas, use
the first formula if the population standard deviation is available. Use the second formula for the t distribution
if the population standard deviation is not available and the sample standard
deviation is used to estimate it. The
table for the t-statistic is included below.
Use the third formula when dealing with proportional data. Small population formulas are not included
here.
When s the population standard deviation is known use the formula
Data from the normal
table is reorganized for use in this equation as:
|
Reliability or Confidence level 1 - s |
Normal Deviate za/2 |
|
.80 |
1.28 |
|
.90 |
1.64 |
|
.95 |
1.96 |
|
.98 |
2.33 |
|
.99 |
2.57 |
|
.998 |
3.08 |
|
.999 |
3.27 |
For example, to
calculate a 95% confidence interval for a sample mean of 10 taken from a sample
of n = 25 and a population standard of 5:
m
= 10 ±
1.96(5/5) = 10 ±
1.96.
The table above also
applies for proportional data. Use the
following formula when calculating the confidence interval.
For example, to
calculate a 95% confidence interval for a sample mean of 0.4 taken from a
sample of n = 25: p = 0.4 ± 1.96(.098) = 0.4 ± 0.192
When the population
standard deviation is not available and the sample standard deviation is used
to calculate the confidence interval, the t-statistic should be used:
Use the t-table
below, Table F in the text, to find values of t. There are two calculations needed to use the
table. First, degrees of freedom is n -
1. Then take the allowable error a and divide this
value in two (since the table only provides one side of the two sided
confidence interval).
For example, to calculate
a 95% confidence interval for a sam1ple mean of 10 taken from a sample of n =
25 when the sample standard deviation is s = 5:
m
= 10 ± 2.064(5/5) = 10 ± 2.064. The value of the t-statistic (2.064) came by
getting degrees of freedom (25 - 1) and since a 95% reliability is desired the
acceptable error is .05, divide .05 in half for .025. On the t-table below find the value in the
column for a
= .025 (.05/2).
Calculating required sample size. Use the formulas
below to calculate the required sample size n to achieve some target acceptable
error e:
For example, to
calculate the required sample size for getting a 95% confidence interval when
the population standard deviation is 5 and the tolerable error is 3:
n = (1.962 (52))/32 = 10.67
for 11, always round these
solutions up to the next integer
If using proportional
data use:
For example, to
calculate the required sample size for getting a 95% confidence interval when
the population proportion is expected to be 0.5 and the tolerable error is 3%:
n = (1.962(.5)(1 - .5))/.032
= 1067.11 for 1068, always round up
This table provides
the values of the t-statistic for the area in one tail of the distribution for a
|
Degrees of Freedom |
a
= .4 |
.25 |
.1 |
.05 |
.025 |
.01 |
.005 |
.0025 |
.001 |
.0005 |
|
1 |
.325 |
1.000 |
3.078 |
6.314 |
12.71 |
31.82 |
63.65 |
127.3 |
318.3 |
636.6 |
|
2 |
.289 |
.816 |
1.886 |
2.920 |
4.303 |
6.965 |
9.925 |
14.09 |
22.32 |
31.6 |
|
3 |
.277 |
.765 |
1.638 |
2.353 |
3.182 |
4.541 |
5.841 |
7.453 |
10.21 |
12.92 |
|
4 |
.271 |
.741 |
1.533 |
2.132 |
2.776 |
3.747 |
4.604 |
5.598 |
7.173 |
8.610 |
|
5 |
.267 |
.727 |
1.476 |
2.015 |
2.571 |
3.365 |
4.032 |
4.773 |
5.893 |
6.869 |
|
6 |
.265 |
.718 |
1.440 |
1.943 |
2.447 |
3.143 |
3.707 |
4.317 |
5.208 |
5.959 |
|
7 |
.263 |
.711 |
1.415 |
1.895 |
2.365 |
2.998 |
3.499 |
4.029 |
4.785 |
5.408 |
|
8 |
.262 |
.706 |
1.397 |
1.860 |
2.306 |
2.896 |
3.355 |
3.833 |
4.501 |
5.041 |
|
9 |
.261 |
.703 |
1.383 |
1.833 |
2.262 |
2.821 |
3.25 |
3.690 |
4.297 |
4.781 |
|
10 |
.260 |
.700 |
1.372 |
1.812 |
2.228 |
2.764 |
3.169 |
3.581 |
4.144 |
4.587 |
|
11 |
.260 |
.697 |
1.363 |
1.796 |
2.201 |
2.718 |
3.106 |
3.497 |
4.025 |
4.437 |
|
12 |
.259 |
.695 |
1.356 |
1.782 |
2.179 |
2.681 |
3.055 |
3.428 |
3.93 |
4.318 |
|
13 |
.259 |
.694 |
1.350 |
1.771 |
2.160 |
2.650 |
3.012 |
3.372 |
3.852 |
4.221 |
|
14 |
.258 |
.692 |
1.345 |
1.761 |
2.145 |
2.624 |
2.977 |
3.326 |
3.787 |
4.140 |
|
15 |
.258 |
.691 |
1.341 |
1.753 |
2.131 |
2.602 |
2.947 |
3.286 |
3.733 |
4.073 |
|
16 |
.258 |
.690 |
1.337 |
1.746 |
2.120 |
2.583 |
2.921 |
3.252 |
3.686 |
4.015 |
|
17 |
.257 |
.689 |
1.333 |
1.740 |
2.110 |
2.567 |
|