Working with Data

You can enter data straight into Data Editor window or recover from a file. You can as well perform a broad range of alterations on your data. Remember, any changes that you make to the data it will not exist next time when you run SPSS if you do not save your data.

Data Entry

Data Editor gives two views of your data first is Variable View that allows you to describe your variables like data type, value labels, missing values, variable label, column format). Data can be entered in Data View.

SPSS Data Files

Data stored as SPSS data file (*.SAV) can be opened straight in the Data Editor. This is one of the most common types of data file that is used with SPSS.

Other Types of Data Files

SPSS can understand many types of data files that include text files, spreadsheets as well as databases.

Transformations

You can make new data values that are based on numeric changes of the existing variables. Of the several transformations accessible, recoding values plus numeric computations are the common.

To modify variable codes choose Recode from Transform menu. The typical uses for recoding can be to reverse code 5-point Likert scale when 1 is recoded to 5 and 2 is recoded to 4 or 4 to 2, plus 5 to 1; or to end various values in a single value to make a grouping variable. Recoding the same variable you will put back the original values with recoded values. If you desire to keep the original values just recode in a different variable.

New variables can be computed from the old ones by choosing Compute from Transform menu. You can also use relational operators, arithmetic expressions, and different kinds of predefined utilities in your computations.

Missing Data

Observations that are unknown or not specific are allocated the system missing value. SPSS utilizes a period to point out a missing data for the numeric variables and blank for string variables. In example given, a third case is missing and subject label and a subject A37 is missing at the second test score.

At times it is useful to identify why the data is missing. You can classify an observation as missing through the Variable View in the Data Editor and these values are named as user missing. By stating subject H19 third test score like user missing, 9 is flagged for unique treatment and is expelled from further calculations.

Care must be taken while working with a missing data as missing data is broadcasted through arithmetic terms. For instance, if you have calculated average test score for the subject A37 by adding three test scores and after that dividing by 3, SPSS will produce a system-missing value.

All of the SPSS statistical functions consist of a strategy where only valid values are been used in the computation. MEAN function will produce the value 78, average of two non-missing test scores for the subject A37. Many statistical analyses provide substitute methods to deal with the missing data. Keep in mind you can see an explanation of any part in a dialog box just by right clicking on it.