MAIN MENU

type s or size to determine a sample size

type d or describe to describe a random sample

type c or choice to set the choices

type e or edit to edit a sample description

type p or pick to actually pick the sample

If you need help type h or help

To quit the program just press return.

The rest of these menu items refer to later steps in the sampling process. They will be described below. For now we will proceed as if we were creating a new sample from scratch type s and then press the return key. This brings up the sample size menu which initially gives you three choices:

SAMPLE SIZE

Type the letter to set or change the values used in calculating a sample size

p -- Precision--how close the sample value should be to the true value

you have not yet chosen a value for precision.

c -- Confidence--how certain that the observation is within the

given precision.

you have not yet chosen a value for confidence.

v -- Variability--what percent of the sample will be in the most

important category?

you have not yet chosen a value for variability.

To record this display type r.

To quit this menu just press return.

As we noted above, all three of these values are needed to calculate a sample size. To enter any value simply type the appropriate letter and the program will ask for a value. The values can be entered in any order and can be changed at any time. You can get help by typing the letter h or the word "help" or you can go to the previous menu by not typing anything but just pressing the return key. For example if we want to select a 90% confidence value we first type c and then press return. The machine will ask:

Please enter a value for confidence.

Enter the figure as a whole number with no decimal point or percent sign.

In response we would type 90 and again press return. If we went through a similar process for precision and variability, the sample size menu would look like this:

SAMPLE SIZE

Type the letter to set or change the values used in calculating a sample size

p -- Precision--how close the sample value should be to the true value

you have chosen plus or minus 5%

c -- Confidence--how certain that the observation is within the

given precision.

you have chosen 90%

v -- Variability--what percent of the sample will be in the most

important category?

you have chosen 50%

e -- Explain -- To explain the sample size based on the values above.

CALCULATED size is 272

This display is being recorded in Size Results. To change this type r.

To quit this menu just type return.

To quit this menu just type return.

Each time you make a choice it is recorded in the sample size menu. The program remembers these values until you change them by selecting the item letter and then entering a new value or until you quit the program. The program recognizes that we have entered the values necessary for sample size calculation there is a new item in the Sample Size menu that shows us the calculated sample size. If we type e we get the sample size report.

for a confidence level of 90%

and a precision of plus or minus 5%

the sample size is 272

This means that if in your study you observe a characteristic in

50% of the sample you will know that there is

only one chance in 10 that the value for

the population is outside the range of

45% to 55%

Do you want to correct the sample size for the population size?

Enter yes or no

The sample size report repeats the confidence and the precision values that were chosen. These are then repeated in terms of the value selected for variation. In the report above 50% was the value given for the variation. One chance in ten is another way of stating the confidence level of ninety percent. The range of 45% to 55% is the variation number minus and plus the precision value of five percent.

[Back]

[Next]

[Contents}

this page is at http://testbed.cis.drexel.edu/sample/setting.html