Data accuracy, integrity, redundancy & validation (1 Viewer)

[Huy]

I hope it's okay if I post my PM that I received this morning, incase others will (in the future) ask the same question.

aNg3L_iReNe wrote on October 16th, 2003 09:30 AM:
hey! i have a question for you (since you know your stuff pretty tite!!) and i know that you're willing to answer keke..

hmm.. i have some difficulty understanding the difference between data validation and data integrity!! please help..

oh and also.. in the communication topic.. there is something about the OSI reference model... do we need to know anything about that?

thanX!!!

We'll start by defining the abovementioned terms (in the subject).

Data accuracy - the extent to which data is free from errors.

That is, accurate data does not contain mistakes, typo's, and so on.

Data integrity - the reliability of data.

If data is accurate, then it is 'integrated' (using your words, Irene, "tite" or tight; together)

If you have reliable data, it has integrity (think about English and textual integrity). Reliable data is accurate data

Data redundancy - undesirable duplication of data, within a database.

I think everybody knows this already, but I wanted to make sure they know it by now

If you have data that is replicated time and time again, used all over the place, with multiple sources, then it is redundant data.

Data validation - checking the accuracy of the data, before it is processed into information.

If you are validating something, you're making sure it is valid. (I know, it is obvious, but say it to yourself...).

If data is valid, it is then accurate, accuracy leads to reliability, hopefully reducing redundancy and leading to a coherent source of data, which is integral and 'together'

One more thing, don't bother trying to figure out the difference between data validation and data verification, because there are conflicting definitions between the Year 9-10 Computing Studies course, and the Preliminary (Year 11) and HSC courses (HSC).

With Communication Systems (Core topic 3) and the OSI reference model, you don't have to know this for the HSC.

If you want to know something about it, or have something to write on it, you can just say:

The OSI reference model "divides data into many layers" and that should score 1 mark.

I highly doubt that the OSI reference model will be examined in the HSC, as it has 'not a lot' to do with communication systems, rather they will focus on protocols and handshaking more than the OSI itself.

 

[SamD]

Originally posted by Huy
One more thing, don't bother trying to figure out the difference between data validation and data verification, because there are conflicting definitions between the Year 9-10 Computing Studies course, and the Preliminary (Year 11) and HSC courses (HSC).

Yes many texts disagree, however the generally accepted differences are:

- Data validation ensures the data entered is reasonable and of the correct type at the time of collection. For example your IPT HSC mark should be an Integer in the range 0 to 100. Validation does not ensure that the mark entered is actually your mark, just that it is a possible HSC mark.

- Data verification takes place as the data is collected and also at various other times to ensure the data is actually correct. For example when a data entry operator enters your HSC mark they then do a double check of the mark entered against the paper form. Or if you read your credit card number to someone over the phone, they commonly read it back to you. Or you recieve a form from some organsiation and on the form it requests you to make changes to your personal details if any of them have changed or are inaccurate. These are all data verfication tasks that improve the integrity of the data.

HTH
Sam

 

[redslert]

Yes in the HSC exam you must always give a simple example just like SamD has above in his answers
this way you make sure you get full marks
no example would mean half marks

 

[Huy]

Originally posted by redslert
this way you make sure you get full marks
no example would mean half marks

You're never awarded half-marks in the HSC.

ie for a 1 mark question, you cannot receive 1/2.

It's either full marks (1/1) or no marks being awarded (0/1).

But if it's out of 2, then naturally you'll receive 1/2

 

[redslert]

Originally posted by Huy
But if it's out of 2, then naturally you'll receive 1/2

yeah i know...
i was refering more to the last part of your reply

 

[skip to my lou]

data redundancy can be good tho...ie backups

 

[Huy]

Originally posted by redslert
yeah i know...
i was refering more to the last part of your reply

Oops! my mistake

(This is my first post for a long time... since I've now finished my HSC; aced the ext eng paper yesterday )

Originally posted by skip to my lou
data redundancy can be good tho...ie backups

That's true.

I always leave my junk in /incoming
Things that are recycled/deleted remain in the bin, since I usually end up recovering them and making duplicate files (I've got the storage space, so it doesn't hurt me).

It's just a security measure, the more you have, the less headaches you'll have (unless the filenames are identical, then you're going to go through a lot of files ).

 

[rx72c]

If you backup yourself up well then the HSC markers will give you marks, by this what i mean is if your example makes sense and provides necessary evidence in makeing your explanation of data integrity, accury and validity make sense.