Date: Tue, 27 Jun 2000 18:02:23 +0200
Reply-To: Walter.Smetsers@eomdata.nl
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Walter Smetsers <Walter.Smetsers@EOMDATA.NL>
Subject: Re: SAS vs Access
In-Reply-To: <0541a2f0.04f1016a@usw-ex0106-044.remarq.com>
Content-Type: text/plain; charset="iso-8859-1"
Karyn,
Here are some answers.
>> How does SAS differ from Access, aside from the fact that one is a
>> database and one is a tool for analysis.
-> SAS can be a database too and Access can be a tool for analysis aswell.
When looking at the possibilities of both products, I think SAS has the
best cards, but one should NEVER change a winning team. I understand you
have data in Access and want this data cleaned, scubbed and made ready
for
analysis. The SAS System can read the data for you and clean, scrub and
transform the data into a structure which can be used to analyse the
data.
Conclusion: Keep the data in Access and use the SAS System to analyse the
data and purhaps return the data into Access after cleaning and
scrubbing.
>> Basically what I am trying to do is show everyone the basic
>> differences and why we need BOTH Access and SAS.
-> You don't change a winning team. You are used to Access I guess and so
are
the end users. They only need to know that Access is not perfect for
cleaning
and scrubbing of data, that's why the SAS System is used. After using the
SAS
System they can still have the possibility to use Access to analyse the
data if
that is what they want. (Flexibility for the end users)
>> While using SAS, if we use the data pulled directly from Access,
>> does SAS have the capability to filter out errors that already may
>> exist in the database.
-> Yes, the SAS System can be used for cleaning and scrubbing data and
filter out
the errors in the Access database (and return it to Access if needed).
>> Basically what are it's limitations on analyzing data. For example,
>> does the data have to come over from Access a certain way for SAS
>> to successfully compute a statistical analysis of it.
-> No, the SAS System can directly access the Access database for analysis.
The
performance aspect is one of the bottlenecks I think in the
configuration. The
SAS System has lots of possibilities to summarize, combine and
statistically analyse
data in a way the performance of analysis will increase.
>> If the data is already erroneous is Access or the main txt file,
>> will it not be wrong in SAS as well?
-> When you just read the data it will be as erronous as Access, BUT when
you have
cleaning and scrubbing decision build in the reading of the data then the
data will
be less erronous. You can define your own cleaning and scrubbing rules
for the kind
of data you are read/analysing and decisions what to do with erronous
data. For example
you could write it to a separate file and analyse this file as well.
>> Does anyone know the maximum records SAS will store? I think my
>> main database has over 200,000 records.
-> SAS has no limits of what so ever on maximum records / rows, platform
or operating system. (200.000 records is called a small amout of data)
I hope these answers may be of any assistence to you. Let me know if you
need more info.
Greetings,
Walter Smetsers
EOM Data - The Netherlands
http://www.eomdata.nl
