Calculating PCI2 in SPSS

General Overview

All files should be stored in the directory C:\PCI
If this directory does not exist, you will need to create it.
File names should not contain any spaces.

To Calculate PCI2 in SPSS:

Download and unzip the SPSS PCI2 macro: SPSS_macro
Compile the macro that generates the PCI2
Highlight all code between
“DEFINE” (line 30)
and “!ENDDEFINE” (line 284)
Click the “Run” button

The following code generates the PCI2

SPSS command Comment
PCI2 infile=|’C:\PCI\Example.sav’| Original input filename
Replace “Example” with the name of your file
outfile=|’C:\PCI\Example.CSV’| Output filename
Replace “Example” with the name of your file
PCIOUT=|’C:\PCI\Sim_Data.sav’| Output filename Do not change the name of this file
WORKDIR=|C:\PCI\| Working directory (must be C:\PCI)
Subsample(s) calculate PCI for each group
(e.g., males and females in this example)
Up to 4 Subsample variables can be specified.
If the analysis does not have any Subsamples these lines can be eliminated
Scale_width=7 Scale width (e.g., 2, 3, 4, 5, 6, 7, 8, 9)
This example is based on a 7-point scale
Replace the 7 with the width of your scale
var={VARIABLE NAME} Variable to be used in PCI calculation
Distance_function=D1 Distance Function:
D1 (or D2) for bipolar scales
D1 currently recommended for bipolar
D3 for unipolar scales
Power_of_distance=|1.5 Power:
P1 = 1 = Unsquared difference scores
P2 = 2 = Squared difference scores
Power can be any value > 0 and < 5

The following files are examples of SPSS data, macro syntax, and output

File type File name
Data Galapagos.sav
Syntax Galapagos_macro.sps
Output Galapagos_Frequencies_Crosstabs_output.spv
Output Galapagos_PCI_output.spv

The file contains the above four files.
Down load the ZIP file and unzip the files to a directory/folder where you can run the SPSS application.

In the Galapagos example:

The SubSample1 variable is “Island” (the respondent’s island of residence).
The Island variable has 3 responses:
1 = San Cristobal
2 = Santa Cruz
3 = Isabela

The macro calculates the PCI for the “normfish” variable (Norms for illegal fishing).
Because a SubSample variable was specified, 3 actual and 3 simulated PCI values are generated;
one for each of the islands.

The scale for “normfish” is 4-point unipolar (i.e., scale_width = 4):
0 = Do nothing if the fisherman violates the rules for illegal fishing
1 = Fine the fisherman
2 = Take away the fisherman’s permit for 15 days
3 = Take away the fisherman’s permit for the rest of the year
Note: The minimum value for a unipolar scale should always be 0.

Because the scale is unipolar, the appropriate distance function is D3.

The power of distance function is currently set to 1 (i.e., Power_of_distance = 1).