Creating and Editing Data Files in SPSS

Creating and Editing a Data File

• The presentation discusses creating and editing data files using IBM SPSS Statistics.
• The software has a data view and a variable view.
• The data view shows the raw data in rows and columns.
• The variable view shows details of each variable (e.g., name, type, width, decimals, labels).

1.1 Name

• Variable names must be unique.
• Names can be any length but should be shorter than 10 characters.
• Names must begin with a letter, followed by letters, numbers, @, #, _, or $.
• Names cannot end with a period.

1.2 Type

• The Type window allows you to specify the variables' data types (numeric, comma, dot, scientific notation, date, dollar, custom currency, string, restricted numeric).
• Numeric variables usually honor digit grouping.
• Strings are nonnumeric and have limited operations.
• Numeric and string are the most common choices.

1.3 Decimals

• The Decimals column sets the number of decimal places for each variable.
• The default in SPSS is 0 for string variables and 2 places for numeric variables.
• The GPA variable in the given dataset requires decimals.

1.4 Width

• Variable width is determined by the largest number or longest string value for each variable.
• The width of variable "id" is 6 digits.
• The width for "lastname" is 10 to accommodate longer names.
• The width of GPA is 4 (one digit left, two on the right, and one for the decimal point).

1.5 Label

• The Label column labels variables with longer or more descriptive meanings.
• Longer labels (max 256 chars) are cumbersome.
• Concise labels (e.g., 20-30 characters) are enough for most cases.

1.6 Values

• Value labels are used for multiple choices such as gender, marital status, or ethnicity.
• Value labels make the output more understandable.
• The software provides up to 60 characters for value labels; however, shorter ones are more practical.

1.7 Missing

• The Missing column is rarely/not used unless a specific missing value needs a particular designation.
• Subjects refusing to answer a question, belonging to an unspecified group can be coded.
• Values entered in Missing are ignored in analyses.

1.8 Columns

• The column identifies how much room is allocated for the data and labels.
• Wide columns improve the user experience for large datasets.
• Narrow columns allow for more columns on the screen, but names might be truncated.
• Default number of columns is 8.

1.9 Align

• The data in each cell can be aligned left, right, or center.
• Numeric variables align to the right by default
• String variables align to the left by default.

1.10 Measure

• Measures (scale, ordinal, nominal) determine the type of data (categorical, ranked, continuous).
• Scale measures have inherent numeric meaning (like age).
• Ordinal measures have inherent order, but mathematical manipulations aren't always useful (e.g., service ratings).
• Nominal measures are used for identification but lack order (e.g., gender, ethnicity).

1.11 Role

• The Role column tracks variables in large datasets, which are essential for experimental projects (e.g., target, input)
• This column is typically ignored for most studies.

1.12 Entering Data

• Save the file (e.g., grades.sav) to start entering data.
• Subsequent saves require only one or two clicks to save changes.

1.13 Search for Data

• The search tool enables quick identification of values within a dataset.
• The search function helps find errors in a dataset for correction.

2. The Compute Procedure: Creating Variables

• The "Compute Variable" procedure creates new variables.
• It computes "total" (sum of quizzes and final) and "percent" (percentage).

This quiz covers the essential processes of creating and editing data files using IBM SPSS Statistics. It focuses on understanding the data and variable views, as well as the rules for naming variables and specifying data types. Test your knowledge on managing variables and their attributes in SPSS.