SPSS software
Two main windows:
- Data editor = enter data and carry out statistical functions
o Data view = show, enter data
o Variable view = defining characteristics of variables within data editor
- Viewer = results
x-axis = independent variable
y-axis = dependent variable
Other windows:
- Syntax editor = enter SPSS commands manually
Top bar:
- File = saving, opening, printing, export etc.
- Edit = cut, paste, options etc.
- View = gridlines, value labels
- Data = make changes to data editor
o Insert variable (add/subtract columns)
o Insert cases (add/subtract row)
o Split file (split up a grouped variable)
o Select cases (run analysis on a screen of cases)
- Transform = manipulate variables in some way; “recode” to change values of certain
variables; “compute” for transforming data
- Analyze = all statistical procedures; “descriptive statistics” for mean, mode, median
etc., general data exploration
- Correlate = correlation techniques
- Regression = linear, multiple linear, logistic
- Dimension reduction = factor analysis
- Scale = reliability analysis
- Graphs = histograms, scatterplots, box-whiskers etc.
- Utilities = comment on dataset
Start with empty dataset: “untitled”
- Row = entity (element) data
- Column = variable (no distinction between independent and dependent variables)
Variable can be “between-group” = subjects belong to either one or another test group, no
overlap
Use numbers to represent categories (Likert: 1 = bad, 2 = below average, 3 = average etc.)
Create variables:
1) Use “variable view” at bottom (Tab)
2) Rows in “variable view” = variable
Name = variable name (short, no spaces)
Type = numeric (numbers), string (names, lettercodes etc.), currency, date (longitudinal)
Label = longer, more intricate description of variable
Values = assign numbers to represent groups (1 = cat, 2 = dog, 3 = gerbil etc.)
Missing = assign numbers to missing data
Measure = nominal, ordinal, scale
Role = is a variable a “predictor” (input), an “outcome” (target) or both
“Split” = divides variable into sub-groups (age: 0-5, 6-10, 11-15 etc.)
“Partition” = selects part of the data
“None” = a variable without a pre-defined role
Two main windows:
- Data editor = enter data and carry out statistical functions
o Data view = show, enter data
o Variable view = defining characteristics of variables within data editor
- Viewer = results
x-axis = independent variable
y-axis = dependent variable
Other windows:
- Syntax editor = enter SPSS commands manually
Top bar:
- File = saving, opening, printing, export etc.
- Edit = cut, paste, options etc.
- View = gridlines, value labels
- Data = make changes to data editor
o Insert variable (add/subtract columns)
o Insert cases (add/subtract row)
o Split file (split up a grouped variable)
o Select cases (run analysis on a screen of cases)
- Transform = manipulate variables in some way; “recode” to change values of certain
variables; “compute” for transforming data
- Analyze = all statistical procedures; “descriptive statistics” for mean, mode, median
etc., general data exploration
- Correlate = correlation techniques
- Regression = linear, multiple linear, logistic
- Dimension reduction = factor analysis
- Scale = reliability analysis
- Graphs = histograms, scatterplots, box-whiskers etc.
- Utilities = comment on dataset
Start with empty dataset: “untitled”
- Row = entity (element) data
- Column = variable (no distinction between independent and dependent variables)
Variable can be “between-group” = subjects belong to either one or another test group, no
overlap
Use numbers to represent categories (Likert: 1 = bad, 2 = below average, 3 = average etc.)
Create variables:
1) Use “variable view” at bottom (Tab)
2) Rows in “variable view” = variable
Name = variable name (short, no spaces)
Type = numeric (numbers), string (names, lettercodes etc.), currency, date (longitudinal)
Label = longer, more intricate description of variable
Values = assign numbers to represent groups (1 = cat, 2 = dog, 3 = gerbil etc.)
Missing = assign numbers to missing data
Measure = nominal, ordinal, scale
Role = is a variable a “predictor” (input), an “outcome” (target) or both
“Split” = divides variable into sub-groups (age: 0-5, 6-10, 11-15 etc.)
“Partition” = selects part of the data
“None” = a variable without a pre-defined role
