Date: Wed, 27 Mar 2002 22:10:07 -0300
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Hector Maletta <firstname.lastname@example.org>
Subject: Re: Looking for a better way
Content-Type: text/plain; charset=us-ascii
SPSS can use Access data, as it reads DBF, XLS and many other formats,
and can process them without any trouble. However, spending money on a
heavy package like SPSS just to get some frequency tables and other such
simple statistics would be an overkill.
As for data entry, SPSS's Data Entry is not clearly superior to other
options. Most users seem to use other options (such as database
management software like Access) to enter their data, and then import
them into SPSS.
Your dataset seems a simple flat one, cases by variables. However, you
may happen to have several connected tables in your database, e.g. a
file whose units are persons, another whose units are households, and so
on. The latest versions of SPSS can handle complex databases directly.
In previous versions you had to define several flat tables with only one
kind of unit, read them with SPSS and thus obtain several flat SPSS
files (one for personal data, one for household-level information, and
so on) that you could merge or rearrange later in several ways.
Universidad del Salvador
Buenos Aires, Argentina
Peter Gardiner wrote:
> I currently look after a business opinion survey (around 650 respondents
> quarterly, with 50 odd questions). At present all the data entry, checking,
> storage of the unit record data and summary statistic calculations are
> performed in Access 97, then the summary statistics are exported to excel to
> compile charts and tables. There are obvious problems with this approach,
> but none more so than an inability to perform any meaningful analysis on the
> data. I also have a lack of confidence in many of the access queries that
> are used to check and manipulate the data because they are build on one
> another and are difficult to audit.
> My point is, I am considering changing to SPSS. Specifically, using Access
> to enter the data store the unit record data, then use SPSS for data
> checking, manipulation and computation of summary statistics.
> I would like to here views from people who have had experience with this
> sort of thing. What traps there are for new players. Would it be better to
> enter and check data in SPSS? Do other database packages (eg paradox)
> interact with SPSS better. Issues that you have had to resolve.
> Thanks in advance, Peter