random selection from multiple measures

[tino]

Dear All,

I have repeated measures of the same observational units; specifically, for each observational unit, there are 6 measurements.

From these six measurements, I need to randomly select one measurement for each person.

For example, if there were 100 persons measured six times, there would be 600 measurements. From these 600 measurement, I would need 100 measurements, but each person should only be sampled once.

Say "person" is numbered from 1 to 100, and "measurements" is numbered from 1 to 6. Which sytnax could I use to make the kind of selection outlined above? Many thanks for your help!

Tino

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Bruce Weaver]

Here are some functions you may find useful.

RV.UNIFORM(min, max). Numeric. Returns a random value from a uniform distribution with specified minimum and maximum. See also the UNIFORM function.

RND(numexpr[,mult,fuzzbits]). Numeric. With a single argument, returns the integer nearest to that argument. Numbers ending in .5 exactly are rounded away from 0. For example, RND(-4.5) rounds to -5. The optional second argument, mult, specifies that the result is an integer multiple of this value—for example, RND(-4.57,0.1) = -4.6. The value must be numeric but cannot be 0. The default is 1.

tino wrote

Dear All,

I have repeated measures of the same observational units; specifically, for each observational unit, there are 6 measurements.

From these six measurements, I need to randomly select one measurement for each person.

For example, if there were 100 persons measured six times, there would be 600 measurements. From these 600 measurement, I would need 100 measurements, but each person should only be sampled once.

Say "person" is numbered from 1 to 100, and "measurements" is numbered from 1 to 6. Which sytnax could I use to make the kind of selection outlined above? Many thanks for your help!

Tino

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[David Marso]

Presuming there is some sensible reason for tossing 84% of the data ;-(
Study this, fill in the <blanks> and ...< be well and prosper or as Dr. Who would quip RUN>!

GET FILE <yourdata> /KEEP <your6vars>.
COMPUTE ID=$CASENUM.
DATASET NAME rawdata.
*------*.

DATASET DECLARE picker.
SET MXLOOPS=200.
MATRIX.
LOOP #=1 TO 100.
SAVE GRADE(UNIFORM(1,6)) /OUTFILE picker/VARIABLES rank1 TO rank6.
END LOOP.
END MATRIX.

DATASET ACTIVATE picker.
COMPUTE ID=$CASENUM.
VARSTOCASES MAKE rank FROM rank1 TO rank6/ INDEX=index.
SELECT IF rank=1.
EXECUTE.

MATCH FILES FILE */FILE rawdata / BY ID.
VECTOR vars=<youristvar TO your6thvar>.
COMPUTE Keep=vars(rank).

tino wrote

Dear All,

I have repeated measures of the same observational units; specifically, for each observational unit, there are 6 measurements.

From these six measurements, I need to randomly select one measurement for each person.

For example, if there were 100 persons measured six times, there would be 600 measurements. From these 600 measurement, I would need 100 measurements, but each person should only be sampled once.

Say "person" is numbered from 1 to 100, and "measurements" is numbered from 1 to 6. Which sytnax could I use to make the kind of selection outlined above? Many thanks for your help!

Tino

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[David Marso]

That last line should read
COMPUTE Keep=vars(index).
---

David Marso wrote

Presuming there is some sensible reason for tossing 84% of the data ;-(
Study this, fill in the <blanks> and ...< be well and prosper or as Dr. Who would quip RUN>!

GET FILE <yourdata> /KEEP <your6vars>.
COMPUTE ID=$CASENUM.
DATASET NAME rawdata.
*------*.

DATASET DECLARE picker.
SET MXLOOPS=200.
MATRIX.
LOOP #=1 TO 100.
SAVE GRADE(UNIFORM(1,6)) /OUTFILE picker/VARIABLES rank1 TO rank6.
END LOOP.
END MATRIX.

DATASET ACTIVATE picker.
COMPUTE ID=$CASENUM.
VARSTOCASES MAKE rank FROM rank1 TO rank6/ INDEX=index.
SELECT IF rank=1.
EXECUTE.

MATCH FILES FILE */FILE rawdata / BY ID.
VECTOR vars=<youristvar TO your6thvar>.
COMPUTE Keep=vars(rank).

tino wrote

Dear All,

I have repeated measures of the same observational units; specifically, for each observational unit, there are 6 measurements.

From these six measurements, I need to randomly select one measurement for each person.

For example, if there were 100 persons measured six times, there would be 600 measurements. From these 600 measurement, I would need 100 measurements, but each person should only be sampled once.

Say "person" is numbered from 1 to 100, and "measurements" is numbered from 1 to 6. Which sytnax could I use to make the kind of selection outlined above? Many thanks for your help!

Tino

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD