Re: Total Information Quality and Data Quality?
Date: Mon, 24 May 2004 22:11:11 +0200
Message-ID: <40b256e6$0$49150$e4fe514c_at_news.xs4all.nl>
Jan Moeller wrote:
> mAsterdam wrote:
>> Jan Moeller wrote: >>> ... This made me thinking about a very basic question. I know the >>> difference between "Data" and "Information",
>
>> It may help you to make your understanding of this distinction >> explicit. If you randomly pick some disparate definitions for data and >> information, chances are you are going to have some unneccessary >> loops in your definitions. For now I'll make an assumption as to what
>
> Hmm, I never thought it's possible to find disparate definitions for
> these terms. Do you have an example?
From other posts I know you are not a native English speaker.
Neither am I, so maybe I used the word 'disparate' wrongly.
What I meant was this: If you search
for some definitions you will see that data is a word used
and assumed to be known in some definitions of information,
and vice versa. So, if just just pick two out of different
contexts, you have a good chance of having a circular set
of definitions.
Old joke:
Recursive. See recursive.
>> ... information is what the receptor of communications receives in >> order to decide/act, while data would be all meaningfull stuff that is >> available.
[snip]
>> Wrong data may yet lead to correct decisions and actions.
>
> Ok, this is right. But if you seperate both notions like this, is it
> possible that correct data also lead to wrong decisions?
Sure. Some marketing research is infamous for this. Looking for 'marketshare' everywhere a lot of highly regarded (and highly paid) researchers completely overlooked 'mindshare'. Not too long ago apache and linux were considered irrelevant technologies because of it.
>> Focus on data quality would still try to get the data right, focus on >> information quality would not.
>
> This is hard to understand for me... what is the main focus on
> information quality? It's possible to ignore the data foundation and
> find other ways to enhance IQ? Which ways?
Just an example:
Many american webapplications assume the customer is an US
citizen, even if there were no such requirements.
When the stuff becomes operational, it turns out non-US people
tend to give impossible values for say the ZIP-code.
A few tricks modeling this behaviour leaves your database
partly filled with incorrect data.
Nevertheless it is possible to distill the information
needed to make correct deliveries (action).
In this example the DQ sucks, but the IQ is ok.
>> More data is just more data. More information leads, explosively, to >> complicating the making of decisions. The key difference here is >> relevancy. In other words: information overload is a IQ problem, not a >> DQ problem.
>
> That is, more information lead to complicating decision-making because
> there're more things to concern/think about? Even if all information (or
> better data, can "information" be relevant?!)
That depends on your definition. In the definiton above there would be room for irrelevant (but new) information, but I've also seen definitions of information where relevancy is part of what discriminates information from data. Your choice. So, again: please make your understanding of the distinction explicit, don't join the shallow talk.
> is relevant and usable?
> I thought it's also possible to clearify things if more information is
> available (e.g. to eleminate ambiguities).
What kind of ambiguities? How you deal with this again - getting boring, I'm affraid - depends on your understanding of the distinction.
HTH Received on Mon May 24 2004 - 22:11:11 CEST