Paul: In defense of the USPS, many residents do not submit a change of address notice to the USPS. To help clean up mailings, the USPS licenses use of the NCOA to vendors that sell address updating services to mailers, and offers reductions in postage rates to mailers that update addresses using the NCOA. Ultimately the resident that receives someone else's mail would have to return it marked 'not at this address' to let the mailer know that it has the wrong address. Local mail carriers slap 'Nixie' labels on some misaddressed mail, but don't always have enough information to make those decisions. Sig

-----Original Message----- From: owner-sas-l@listserv.uga.edu [mailto:owner-sas-l@listserv.uga.edu] On Behalf Of Swank, Paul R Sent: Wednesday, November 02, 2005 6:12 PM To: Sigurd Hermansen; SAS-L@LISTSERV.UGA.EDU Subject: RE: Test - Control Design question - Direct marketing / Measurement

I've been living in my house for more than ten years and I still get mail for the people who owned it before the people who sold it to me! Thus, I have serious doubts about the USPS change of address database.

Paul R. Swank, Ph.D. Professor, Developmental Pediatrics Director of Research, Center for Improving the Readiness of Children for Learning and Education (C.I.R.C.L.E.) Medical School UT Health Science Center at Houston

-----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of Sigurd Hermansen Sent: Wednesday, November 02, 2005 4:52 PM To: SAS-L@LISTSERV.UGA.EDU Subject: Re: Test - Control Design question - Direct marketing / Measurement

USPS National Change of Address database searches and related searches will return invalid address codes, and will return codes indicating that a name and address string almost matches an address change on the NCOA, but does not match closely enough to return an address update. Some mailers exclude one or both from mailings. The USPS also returns 1st class mail for which it has no forwarding address or the time interval for using a forwarding address has expired. One does not have to carry out the experiment to realize some potential mailing failures, but does have to conduct the test to find all.

We are currently looking closely at results of our ten-year tracing of a NCI cohort of 500,000 subjects. Any suggestions that you direct marketing people might offer for improving deliveries of surveys to a mobile population? We are trying to determine whether survey non-respondents are not receiving questionnaires or simply not responding.

Apologies for hijacking the topic, even if for a worthy purpose. Sig

-----Original Message----- From: owner-sas-l@listserv.uga.edu [mailto:owner-sas-l@listserv.uga.edu] On Behalf Of Talbot Michael Katz Sent: Wednesday, November 02, 2005 4:51 PM To: SAS-L@LISTSERV.UGA.EDU; Nick I Cc: Talbot Michael Katz Subject: Re: Test - Control Design question - Direct marketing / Measurement

Hi, Nick.

It is perfectly legitimate to apply suppressions to your original population before undertaking your experimental design. Then, your experiment applies to your post-suppression population. And most direct marketing campaigners do have a variety of suppressions they mix and match for individual campaigns. So, if you could determine who you were able to mail beforehand, then you could indeed suppress the unmailables before building your experimental design, as you suggest. But I am under the impression that the way an unmailable is determined in this case is by sending out the mail and having it returned, marked undeliverable; you have to carry out the experiment to find out.

-- TMK -- "The Macro Klutz"

On Wed, 2 Nov 2005 16:15:23 -0500, Nick . <ni14@MAIL.COM> wrote:

>Here is my 1 cent worth: > >For tracking response rates and creating lift chart why not do it this way: > >You have an entire pouplation say 100 prospects. >You know some of these will not be abled to get mailed to. Send the >ENTIRE file (100 prospects) to the vendor so they can flag who will get mailed to. >Now you know who cannot get mailed to. >Say there are 5 such prospects. >DELETE them from the 100 original prospects. >You now have 95 prospects who can get mailed to. >Now put aside your 10% control and mail to the remaining 95 - 10%*95. >There should be no measurement issue now. Right? David, TopKatz? > >NICK > > > > >vijay.jayanti@GMAIL.COM wrote: >> I need some statistician to respond to this question. >> >> SITUATION: >> In our Direct mail Campaigns, we pull a 10% random control/ Hold out >> group. We send the test/ treatment group to a mail fulfillment house.

>> They post back a variable to the database which says "Able to mail" >> which is a Yes/ NO column, which captures the information, if that >> address was mailed or not. >> >> CURRENT MEASUREMENT METHODOLOGY: >> Measure response rate for both treated and control cells and subtract

>> (TEST - CONTROL) to get the lift. Test for significance. >> >> While calculating the response rate for TEST (= RESPONSES/ BASE), we >> include ALL THE CUSTOMERS in the base (including those who were >> flagged "NO" in "ABLE TO MAIL" field. >> >> The prevalent method in analyzing the responses is to ignore this >> variable. The reason being quoted by my senior colleague is that >> "Since the control group does not have a "ABLE TO MAIL" field, (since

>> they do not go thru the mailing process), we have to include all the >> customers irrespective of their "ABLE TO MAIL" value. So if we >> exclude those who did not get a mail, we are NOT STATISTICALLY >> CORRECT, AS IT IS NOT RANDOM ANYMORE. (MEANING CONTROL IS NOT >> ANYMORE EQUIVALENT TO TEST). >> >> Also I am told that from a Financial/ ROI point of view, including >> everyone in the base gives a more correct picture. >> >> MY OPINION: >> I think it is wrong. I have over 4+ years of experience as a >> marketing analyst (I am not a statistician). I think we should >> exclude those who did not receive a mail and report the correct >> response rate. Other wise we will be depressing the true resp. rate. >> >> what is right way to compute the response rate? INCLUDE THOSE WHO DID

>> NOT GET A MAIL IN THE BASE? OR EXCLUDE THEM? > >I see that TopKatz took a not-claiming-to-be-an-expert approach on >this. > >I agree with TopKatz and your 'senior colleague'. And I like to think that >I know >everything. :-) :-) > >The key point (well, *one* key point) is your target population for any

>future marketing campaigns. You want to be able to extrapolate back to

>the real target population. And that means including everyone in the >database. Particular types of survey sample analyses might warrant this

>exclusion (with appropriate survey design and weighting consequences), >but that's not what your company >is >expressly after here. > >HTH, >David >-- >David L. Cassell >mathematical statistician >Design Pathways >3115 NW Norwood Pl. >Corvallis OR 97330 > > > > > >-- >___________________________________________________ >Play 100s of games for FREE! http://games.mail.com/