Setting
up your data sheet
For
EZAnalyze to work smoothly, the Excel file that you use
to provide the data for analysis needs to be set up with
a few simple rules in mind. While there may be exceptions
to these guidelines, following them will have you spending
less time preparing your data, and more time analyzing
it!
| ROWS |
Each
row in your Excel data sheet (rows are numbered down
the left side) should contain one "case." Usually,
each row will contain data from one student or person.
However, if your data is from schools in a district,
and you are looking at how schools are different
from one another, your "cases" would then
be each individual school, and you would put each
schools data into the rows.
THE
FIRST ROW IN YOUR DATA SHEET MUST CONTAIN VARIABLE
NAMES! (see COLUMNS below)
|
| COLUMNS |
Each
column in your Excel data sheet (columns are lettered
across the top) should contain one "variable." EZAnalyze
reads the variable names it will use for your analyses
from the first row of data. The ability of EZAnalyze
to read these variable names from the first row of
data is what makes it special - and what makes it easy
for you to analyze your data. |
Sample
of what your data should look like
Notice how the first row contains the variable
names, and rows 2-11 contain the values for each
case
|
| MISSING
VALUES |
Sometimes,
you might have some data that is "missing." While
you should make every effort to make sure that you
have all of the data for each case, sometimes that
is not possible. When this happens, you should leave
the cell blank - do not put a number, letter, or
anything in the cell where you have missing data.
EZAnalyze will tell you how many missing values you
have if you use the DESCRIBE
- PERCENTAGES function. You can still calculate
descriptive statistics using the DESCRIBE
- DESCRIPTIVE STATISTICS function - the mean,
median, and mode can still be calculated using all
of the data that is valid (not missing).
For
specific information on how missing values affect
the creation of new variables, see the help topic
associated with NEW VARIABLES
|
| SIZE |
EZAnalyze
can work with Excel data sheets that are 256 columns
wide and over 65000 rows long. If you are looking at
data that contains more than 256 variables and 65000
cases, you should probably look into buying more powerful
statistical analysis software such as SPSS.
NOTE: Even though Excel 2007 supports more than 256 columns
and 65000 rows, EZAnalyze does not at this time. |
| DATA |
EZAnalyze
works best when all of the data (values contained in
each cell) are numbers - HOWEVER, there are some exceptions.
For example, if you have a variable named "gender" (see
example above), you could represent "males" with
a 1 and "females" with a 2. Since "gender" is
what is known as a categorical variable (a
variable that creates categories for naming purposes
only - glossary), you will
not be computing descriptive statistics such as means
and standard deviations. In this case, then you can
type "male" or "female" into your
data sheet instead of numbers. TO BE CLEAR, however,
you will always be safe if you use numbers. |
|
