Exporting data to SPSS
Last updated
Last updated
Only certain older versions of the 3di contain SPSS export functionality. This feature was removed in versions from 2021 onwards.
The contents of all or of selected patient records in the currently linked backend can be exported to an SPSS file. Whichever backend is involved (even if it resides on a removable drive) the file is saved automatically in the SPSS folder which sits in the same folder as your 3di frontend – the 3di installation folder. The file is not opened automatically for inspection; indeed, the file can be created on a machine which does not have the SPSS application installed.
The file is given a name on the pattern
3di_abcdef_031105_085710
in which:
abcdef is the name of the backend from which the data come 031105 is the date in the form yymmdd 085710 is the time in the form hhmmss
You are free to rename (and delete) SPSS export files.
click Export data to SPSS on the Main Menu
or click on the toolbar
or press <Ctrl + Shift + S> (think of S-elector)
select the patients whose data you wish to export by setting the check box in the right-hand column
if you click a button at a column head (eg DOB) the patients will be sorted on that value
a clicked button has a pressed appearance; a second click will release the button, so that patients return to their default order by name
clicking another button will replace one sort order with another (and release the button previously pressed)
if you click a button at a column foot, it will set or clear the check box for all the related records in the right-hand column (eg clicking Set f will ensure that all and only females are selected)
click Export selected to Spss or Export ASD subscales to Spss – the latter is much faster, but exports only details from the Case Manager together with ASD scores (both summative eg B-scale, and contributory down to the lowest level eg B1a, B1b, and so on).
remember that the exported file will not be opened and displayed
With the exception of Repeating Answer questions (which are not currently exported), every question in the interview is represented by one or more variables in the Spss datafile. All questions of a given type (Text, Numeric, Combo and so on) share a particular pattern of Spss variables.
Also exported are all the computed outcomes of the 3di – for example, the Children’s Communication Checklist scores, the Autistic Spectrum Disorders scores, and the results of the Comorbidities Tool. The variables relating to computed outcomes are described in the next subsection.
Each exported question has an Spss variable, here termed the question’s base variable, whose name may be derived from the question’s ID within the 3di system as follows:
determine the question’s ID – one way is to hover the mouse pointer over a question’s interview number (eg 5.2.8) on the interview form;
write the ID as a four-digit number by prefixing zeroes as necessary
prefix the result with q (for q-uestion)
For example:
The further details of this variable, and of any further variables associated with a question, depend upon the question’s type, as follows:
The base variable:
Questions of type Text have no additional Spss variables.
The base variable:
Questions of type Numeric have no additional Spss variables.
The base variable:
Questions of type Age have no additional Spss variables.
The base variable:
Questions of type Date have no additional Spss variables.
The base variable:
Questions of type Time have no additional Spss variables.
The base variable:
Questions of type Combo or OptionGroup have no additional Spss variables.
The base variable:
In addition to the base variable for the ListMulti question, one variable for each item in the list box, built on the following pattern:
As well as the responses to individual questions, many computed outcomes of the 3di are exported to SPSS. These include the Children’s Communication Checklist scores, the Autistic Spectrum Disorders scores, and some results from the Comorbidities Tool.
The SPSS variables which receive the exported outcomes are fully described by their SPSS labels: hover the pointer over the variable name at the head of a column in the SPSS data file to see the label (or switch to variable view in SPSS).
3di system ID (Qid) | Name of Spss variable |
---|---|
property of variable | property value/s | note |
---|---|---|
property of variable | property value/s | note |
---|---|---|
property of variable | property value/s | note |
---|---|---|
property of variable | property value/s | note |
---|---|---|
property of variable | property value/s | note |
---|---|---|
property of variable | property value/s | note |
---|---|---|
property of variable | property value/s | note |
---|---|---|
property of variable | property value/s | note |
---|---|---|
6
q0006
25
q0025
564
q0564
name
qnnnn (eg q0178)
see Spss base variable on the naming of a question’s base variable
label
text of question
tokens (such as %name for the patient’s name) remain unsubstituted
value
if text has been entered
text of response
otherwise, whichever one of the following is appropriate:
-ToDo-
-NotApp-
-Dk-
Spss does not provide a missing values device for String variables; the export system provides a work-around by inserting special text values for null text boxes.
-Dk- can appear only where the question is provided with a ‘?’ option button as well as a text box
-NotApp- can appear only where the question may be disabled as the result of responses to other questions
value labels
Spss does not provide value labels for String variables
missing values
Spss does not provide a missing values device for String variables; but see note above on the value property
name
qnnnn (eg q0615)
see Spss base variable on the naming of a question’s base variable
label
text of question
tokens (such as %name for the patient’s name) remain unsubstituted
value
if a date has been entered: date entered, in the format dd-MMM-yy otherwise no value is exported
eg 04-JUN-97
value labels
No practical use can be made of value labels for dates
missing values
Spss does not provide a missing values device for Date variables
name
qnnnn (eg q0024)
see Spss base variable on the naming of a question’s base variable
label
text of question
tokens (such as %name for the patient’s name) remain unsubstituted
value
if a time has been entered time entered, in the format hh:mm:ss otherwise no value is exported
eg 21:30:00
value labels
No practical use can be made of value labels for times
missing values
Spss does not provide a missing values device for Time variables
name
qnnnn (eg q0019)
see Spss base variable on the naming of a question’s base variable
value
the numeric value entered
label
text of question
tokens (such as %name for the patient’s name) remain unsubstituted
value labels
ToDo
NotApp
Dk
values are mapped to labels as follows:
Note that Dk is included only where the question is provided with a ‘?’ option button as well as a box for numeric input
value
label
-1
ToDo
-2
NotApp
-3
Dk
missing values
-1
-2
-3 (but see note on value labels)
ie ToDo, NotApp and Dk are treated as missing values
name
qnnnn (eg q0113)
see Spss base variable on the naming of a question’s base variable
value
the age entered in the separate dropdowns for years and months converted into an age in months
for example, an age entered as 3 years 5 months would be exported to Spss as the value 41 (ie 3 x 12 plus 5)
label
text of question
tokens (such as %name for the patient’s name) remain unsubstituted
value labels
ToDo
NotApp
Dk
values are mapped to labels as follows:
Note that Dk is included only where the question is provided with a ‘?’ option button as well as dropdowns for years and months
value
label
-1
ToDo
-2
NotApp
-3
Dk
missing values
-1
-2
-3 (but see note on value labels)
ie ToDo, NotApp and Dk are treated as missing values
name
qnnnn (eg q0248)
see Spss base variable on the naming of a question’s base variable
value
the 3di’s numeric encoding for the selected item
label
text of question
tokens (such as %name for the patient’s name) remain unsubstituted
value labels
the precise text of the items from which a single selection is made (combo box dropdown text items or option button labels)
option_1
option_2
…
option_n
and
ToDo
NotApp
for example:
and
value
label
0
often
1
sometimes
2
never
value
label
-1
ToDo
-2
NotApp
missing values
-1
-2
ie ToDo and NotApp are treated as missing values
name
qnnnn (eg q0853)
see Spss base variable on the naming of a question’s base variable
value
the count of items selected from the list box except that if ‘none applies’ has been selected, then a count of 0 is exported
thus any value greater than 0 means a ‘positive’ response to the question; and the larger the value the ‘stronger’ the response
label
text of question
tokens (such as %name for the patient’s name) remain unsubstituted
value labels
ToDo
NotApp
value
label
-1
ToDo
-2
NotApp
missing values
-1
-2
ie ToDo and NotApp are treated as missing values
name
qnnnn#NN (eg q0853#01)
NN runs from ‘01’ to the number of items in the list box; the variable qnnnn#01 always corresponds to the ‘none applies’ item which appears first in every list box
value
if item #NN has not been selected
0
otherwise
1
label
text of question
tokens (such as %name for the patient’s name) remain unsubstituted
value labels
two labels, as follows:
(ie the dash character)
text of item #NN in list box
for example:
value
label
0
-
1
second trimester
missing values
none assigned
ToDo and NotApp are conveyed by the base variable for the question; where either of these applies, nothing is exported to the ‘item’ variables (which therefore appear as ‘system missing’)
value
label
-1
ToDo
-2
NotApp
-3
Dk
value
label
-1
ToDo
-2
NotApp
-3
Dk
value
label
0
often
1
sometimes
2
never
value
label
-1
ToDo
-2
NotApp
value
label
-1
ToDo
-2
NotApp
value
label
0
-
1
second trimester