Re: Phone Number as Random

There are random number tables and random number generators on the internet.

Maybe you have a look?

Am Montag, 14. Januar 2019, 19:49:53 MEZ hat Peter Spangler <[hidden email]> Folgendes geschrieben:

Hi Stats Colleagues!

I’m running an experiment and hoping that someone with Census or other government experience might be helpful on. I am trying to select a random identifier to select treatment candidates in a call center experiment. I am considering using the last digit of phone number as that identifier. Is this random? Other comments?

Peter Spangler

Sent from my iPhone

=====================

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

===================== 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

[Ki Park]

The short answer is "no." Yet, it will really depend on how those phone
numbers have been generated or acquired from an existing list. If the number
is from a list, it is not random. Also, some 'random' phone numbers are
generated within an 'area' (e.g. 614-319-XXXX), so even though the number
could be random, this is not representative of the population that you are
interested in because the area and population do not match.
I would say that the call center should know how those numbers have been
generated/acquired. You may also have a problem that the last 4 digits could
be in more than one case.

Ki



--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
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

[Art Kendall]

try something like this untested syntax. Just create a new pseudo-random
number in SPSS
set seed=20190115.
compute ranorder = uniform(1).
sort cases by ranorder.
compute newgroup= mod($casenum,2)+1.

or
compute ranorder = uniform(1).
rank variables = ranorder /ntiles = 2 into newgroup



-----
Art Kendall
Social Research Consultants
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

=====================
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