this is an empty file
this is an empty file
this is an empty file
this is an empty file
this is an empty file
this is an empty file
this is an empty file
this is an empty file
this is an empty file
10) When you just start the program there are only two menu items to
consider:
s -- sample size, if you want to see how sample size and sample
accuracy interact.
d -- describe a sample, if you want to be able to pick a sample.
You will want to check the settings in:
c -- set the choices, before you you do any serious work.
To get more specific help, select an item and then ask for help.
20)Precision reflects the understanding that our sample value may not
be exactly the value for the population as a whole.
For example:
A value of 20% which we observed for our sample might be reported
as a population value of between 16% and 24%. A shorthand way of
writing this is 20 plus or minus 4%. This is read as twenty plus or minus
four percent. In this case the precision value is 4%.
Many studies use precision values of 3 to 5 percent.
Remember that smaller precision requires a larger sample size.
30)Confidence is the possibility that our findings are outside of the range
set by our precision figure. For example:
A confidence of 90% means that there is one chance in ten that our
result plus or minus the precision is wrong. The 90% is the nine chances
out of ten that our sample value is within the stated precision.
Confidence reports our certainty that we have avoided a large error.
We usually set the confidence value at 90% or 95%.
40)We estimate the variance by estimating the percent of the population
that will be in the most important category. For example:
In a close presidential election we might estimate that the potential
vote was evenly split with 50% going to each candidate.
If we were studying undiagnosed reading disorders in children, we
might guess that 6% of all grade school students suffered from such
problems.
If we don't have any idea about what the split would be, the most
conservative course is to use an estimate of 50%.
50)This menu lets you enter the values necessary to calculate a sample size.
Before you enter a value, the text indicates that no value has been entered.
Any value, once entered, can be changed -- simply select the appropriate
letter and enter the new value.
If you need more help, first enter the letter
and then when prompted for a value type h for help
60)If you are doing a simple sample like picking from a numbered list of
500 students you would have only one category:
Category 1 student numbered from 1 to 500
A more complicated sample might be times throughout the week
Category 1 day named Mon Tues Wed Thur Fri
Category 2 hour named 9am 10am 11am 12noon 1pm 2pm
3pm 4pm
Category 3 minute numbered from 0 to 59
70)Sometimes the items in our population are designated by numbers:
A class list numbered from 1 to 28,
Accession numbers from 20660 to 23999.
Other times the values which we wish to select are not numeric.
Names of faculty members; Ross, Penchkoski, and Smith
The hours of the day from 9 am to 4 pm. are numbers but we don't
want them sorted in number order. Entering them as names solves
this problem.
You can enter na for named or nu for numbered.
80)Almost all of the time we sort the sample because it makes it easier
when we actually go to collect the data.
A problem with sorting the sample occurs if the research has to be
suspended before every individual selected in the sample is examined. The
partial data is somewhat suspect since only part of the population had a
chance to be selected.
90)We have assumed in drawing our sample that the population is
known so that we can set the range of random numbers to be drawn.
If we do not know this size we have to draw numbers that allow for the
biggest possible population and then throw away values that are too big.
Discarding values gives us problems when we are using an ordered list
because we will not get to use some numbers. This would be acceptable as
long as we remember that when we are drawing from an ordered list we
cannot stop just because we have as many data points as we would like To
100)Editing is a two step process:
First you indicate which editing command (change, add, delete)
that you wish to do,
Then you will be prompted to provide a category number
on which to perform the operation.
110)If you are really doing a sampling study it is always safest to save
the raw file. This uses some disk space but that is the only negative
Saving a raw file saves both the sample design and the values drawn.
You need a raw file if, after drawing a sample, you need more items. This
happens when some of the cases you selected in your sample are
unavailable.
The raw file lets the program draw the extra items without
repeating any of the items already drawn.
120)Changing a category requires retyping all of the category information.
There is no provision for editing only a part of a category.
130)Saving a file lets you use some other program, such as a word processor
or a spreadsheet to read, print, or manipulate the sample that is drawn.
Saving a file is so important that you should always do it.
If you are drawing a supplemental file, be sure that you use a new file name
to save your result. The program will erase everything in the file you name
before starting to save your result.
140)Help is available for any item on this menu. Enter the appropriate
letter and then type h in response to the question. When it initially
appears, this menu is set to the normal settings for running the
program.
150) You can add a category to any place in the sampling design.
First you will be prompted to indicate where to add the category
and then you will be asked to describe the new category.
If you are replacing a category it does not matter whether you add first
and then delete or delete first and then add.
160)If you are doing a simple sample like picking from a numbered list
of 500 students you would have only one category:
Category 1 student numbered from 1 to 500
You start by entering the category name, this is usually one or two
words like "student number" or "hour." Since these names are printed with
the sample we usually keep them short.
If you don't like a name after you have chosen it, you can use "edit" to
edit the sample description.
170) When the sample size gets close to the size of the entire population
the size of the sample you need to achieve a certain accuracy diminishes.
This item lets you tell the computer what your estimate of the population
size is. The program then calculates a revised sample size.
If you have the program calculate sample size again it does not continue
to correct for population size.
Most studies do not correct for sample size.
180) Population size should be entered as a whole number without
commas or decimal point.
If you don't know population size use a reasonable estimate.
If you experiment, you will see that, in most cases, correcting for the
population size has very little effect on sample size. So it is not
worth complicating the study and its explanation by doing it.
190) If you want to try out different values of confidence, precision, and
variance, you will want to record this information.
When you record information a file is created that shows all the sample
size calculations that you have done.
You can later use any word processor to view this file or to print it out.
You don't need to shut off the recording. Quitting the program does this
automatically.
200) What sample size did you want to consider?
By entering a value of zero for precision you indicated that you want
the program to calculate the precision for a sample size that you select.
If this is not what you want to do, just select precision again and put in
a non zero value.
210) When categories are numbered, sample designs look like this:
Category 1 student numbered from 1 to 500
Category 1 accession numbered from 65371 to 84843
Category 1 something numbered from -300 to 700
Do not use commas when entering the low and high numbers. The
program will not let the second number be lower than the first.
220) You must always enter a sample size number even if you want
the computed value. Enter the number without commas.
Remember that a sample design implies a maximum sample size:
Category 1 week numbered from 1 to 3
Category 2 day named Mon Tues Wed Thur Fri
Implies a maximum size of 15. That is five days each week in any of
three weeks. Since this is a program to draw samples, the largest sample
that could be drawn is 14. (Since 15 would be drawing everything which
is not a sample.)
230) This option appears when precision is set to zero.
This option is used to investigate what level of precision can be
achieved for a specific sample size.
The precision will be calculated after values for Size, Confidence, and
Variability have been set. The calculated value is entered just as if
you had entered a value for precision.
The calculated sample size reported may not be exactly the one
you entered because of rounding off.
240) Enter the size sample you want drawn.
You must enter a size even if you agree with the calculated size.
Remember that if you want the sample saved you must use c--Choices
from the Main Menu BEFORE you pick the sample. You always
should save the sample except when you are just playing with the
program.
For your protection, once a sample is saved, the save feature is
turned off until you use Choices again.
250) Special categories are frequently used named categories.
This feature can save typing time. If the special category you
want is not available, it can be created by the normal category
creation process.
The program tells you that special categories are available only
before the first category but you can use special categories at
any time by entering a @ when prompted for the category name.
this is an empty file
this is an empty file
this is an empty file
this is an empty file
this is an empty file
this is an empty file
this is an empty file
this is an empty file
this is an empty file