LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (July 1999)Back to main SPSSX-L pageJoin or leave SPSSX-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Wed, 28 Jul 1999 14:55:35 -0400
Reply-To:     Donna Barre <dbarre@DOE.STATE.VT.US>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Donna Barre <dbarre@DOE.STATE.VT.US>
Subject:      Apparent Rounding Problem
Comments: To: "SPSSX-L@UGA.CC.UGA.EDU" <SPSSX-L@UGA.CC.UGA.EDU>
Content-Type: text/plain; charset=us-ascii

My colleagues and I have been using the Tables module and the crosstab procedures. When the figures in the tables are examined there are many instances (10% or more) where the totals do not round to 100%. Relatedly, when an individual cell in the output is viewed, it initially contains one value, say, 62. When the actual number in the cell is selected, it contains a number such as 62.5001. When the cursor is moved out of the cell, the number changes to 63.

My question: Is there a way to change the way rounding occurs in SPSS,or is this a bug? And, if it is a bug, does it affect only "simple" procedures such as crosstabs and similar calculations in Tables, or does it affect all calculations? Does SPSS use double precision in performing calculations, and if not, is this the origin of the rounding problem?

Reporting numbers to the public that can readily be seen as inexact by anyone who does the arithmetic harms the credibility of a state agency. Thank you for any help and insights you can give me on this issue.

Donna M. Barre, PhD Statistician, VT Dept of Education


Back to: Top of message | Previous page | Main SPSSX-L page