; REFERENCES
Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out
Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

REFERENCES

VIEWS: 50 PAGES: 7

  • pg 1
									                AS Module 2 (ICT2) TOPIC 11.2 Verification and Validation


       Understand the distinction between accuracy of information and validity of data.
       Explain possible sources and types of error in data capture, transcription, transmission
        and processing.
       Describe methods of preventing and reducing such methods.
       Describe appropriate validation techniques for various types of data process, from data
        capture to report generation.

NOTES Heathcote Chapter 16
Turn to Heathcote p.85, look at the form for customers of a mail order company and
consider the possible sources of error that might arise – given on P.86.
Data is useless to the user without data integrity.
Data Quality Problems occur due to inaccurate, outdated or incomplete data. A single data error
is not the most significant problem, rather the many bits of incomplete data in massive numbers
of records. Problems might arise for a variety of reasons
         Customer enters wrong details, misses out information, incorrect calculations
         Transcription errors can occur by a person keying in data.
         Loss or accidental erasure of documents
         Bored employees make mistakes
Examples (but gather your own through reading)
          A New York Investment Bank discovered that 40% of the data in its credit risk
             management database was incomplete and had to be double-checked by its users
          Several studies have documented that 5-12% of bar code sales at supermarkets and
             retail stores are wrong and that the ratio of overcharges to undercharges runs at 4:1.
             The problem stems mainly from human errors in keeping shelf prices accurate.

INPUT CONTROLS see Heathcote p.86 to 88

Some solutions to try and reduce the number of data errors include: -
 Minimising transcription by using automated input methods such as bar code readers.
   Using turnaround documents, these are produced by computer and later become input
   documents e.g. remittance advices to customers to be returned with payments. These already
   include customers’ details such as account numbers etc.
 Design data collection and input documents in ways which encourage accurate completion
 Use clerical checking procedures such as re-calculation of totals or visual comparison of
   document with original see p.86
 Use codes of a restricted format e.g. Customer account numbers with two alpha entries
   followed by six digits.
 Use batching which allows calculation of batch totals
 Use batch or interactive screen validation techniques
 Ensure staff are well trained
 Control access to input documents e.g. input documents for changing pay rates should be
   available only to say Personnel Manager.

VALIDATION Heathcote p.87

VALIDATION IS MAKING SURE THAT DATA IS SENSIBLE.
Complete the following table
Check                        Description    Example



                                            Page 1 of 7
                   AS Module 2 (ICT2) TOPIC 11.2 Verification and Validation


Size check                               Number of characters in a     Must consist of 8 characters only
                                         data item is checked          (Telephone numbers)
Presence check                           Certain fields must exist.
Range check                              Data lies within maximum
                                         and minimum values
Format checks (also called                                             Stock code designated as having
picture check)                                                         2 alphabetic characters followed
                                                                       by 4 numeric digits must follow
                                                                       this arrangement
Parity bits (p.88)
Consistency checks                                                     Any employee aged over 25 years
                                                                       must be contributing to
                                                                       superannuation scheme,
                                                                       employees under 25 cannot.
Check Digit                                                            See example below

Check digit - modulus 11 algorithm e.g. stock code is 573148
 Stock Cod e                             5      7       3         1    4       8
 x Wei gh t                              7      6       5         4    3       2
 Produ ct                           35         42      15         4   12      16
 Su m of produ cts                12 4
 Di vi de sum by 11 g ivin g 11 remai nde r 3
 Ch eck d ig it is 11-3=8
Stock Cod e                         5          7       3          1    4      8       8
x Wei gh t                          7          6       5          4    3      2       1
Produ ct                           35         42      15          4   12     16       8
Su m of produ cts                 13 2
Di vi de sum by 11 g ivin g 12 remai nde r 0
Th ere fo re check di gi t O.K.


VERIFICATION P88

VERIFICATION IS MAKING SURE THAT INFORMATION IS ACCURATE
  1. When information is gathered from source documents, which cannot be read directly by,
     the computer needs to be converted. Key-to-disk is typically error prone, so before data
     is processed by the computer necessary to verify by re-typing the data using the same
     source documents but with the key-to-disk device in verify mode. In this mode data is
     compared with that typed in originally and differences signalled.


   2. Print out all master file changes to check authorisation and consistent with transactions.

OUTPUT CONTROLS
 Set procedures for treatment of error reports
 Reconcile control totals in all ledgers
 Order processing and invoice verification




                                                    Page 2 of 7
                AS Module 2 (ICT2) TOPIC 11.2 Verification and Validation



EXAMINATION QUESTIONS
1996.2 (4 marks)
 A local bookshop uses an interactive computer database system to store stock details and answer
customer enquiries regarding the availability of books. An enquiry for a particular book shows
that there is a single copy remaining in stock. However on searching the shelves (and stock
room) the book cannot be found. Give possible reasons for this apparent discrepancy.

Data input error must be qualified by an example appropriate to this information system e.g.
wrong ISBN (1)
Nothing for typing error or vague unspecified inputs
Initial stock was wrong (1)
Updates are batched (1) each evening whereas enquiries are interactive (1). Thus ageing of
information (1)


1993.5 (12 marks)
A stock record consists of four fields
stockcode        e.g. 765981
description      e.g. Dot-Matrix Printer
costprice        e.g. 245.00
supplier code e.g. CP
        (a) Describe FOUR appropriate but different validation techniques which could be used
        to validate the different fields of the record. In EACH case explain why the technique is
        appropriate. (8)
        (b) When completed this database is transmitted via the public telephone network
        communication link. On receipt it is found to have been corrupted during transmission.
        Describe how this may have occurred and how it may be detected. Suggest ONE method
        of reducing the occurrence of such an error. (4)

 (a) must be 4 different techniques
stockcode            check digit, maximum error trapping on possible errors (transcription,
                     transposition or it is a unique field)
Description          length check or ‘picture’ character, default for text fields, presence check
costprice            range check, item likely to be within reasonable limits
supplier code        lookup table, suppliers are known thus must be on lookup database
(b) Reason for corruption:
 noise on the line during transmission has corrupted database (1)
 nothing for hackers
 Detected by using a CRC check (2) or parity (1) detailed explanation (2)
 reduce by: sending it twice ( some protocols do this automatically) (1) or send at a slower
     rate (1)
 Booster signals (1)
 dedicated lines (private line is weak) (1)
 fibre optic lines (1)




                                             Page 3 of 7
                  AS Module 2 (ICT2) TOPIC 11.2 Verification and Validation



1997.6 (8 marks)
A well-designed information system should be able to check that input data is valid, but it can
never ensure that information is accurate.
        (a) Explain the distinction between accuracy of information and validity of data.
        Illustrate this distinction with a suitable example. (4)
        (b) Describe two ways in which data capture errors may arise, together with techniques
        for preventing or reducing these errors.

a) NOT ICT1 answers. NOT ‘if sample do not know’ NOT ‘value judgements’
 Validation ensures data is reasonable or sensible or complete (1)
 Data can be sensible but is still not correct (1)
 e.g. A person’s DOB may be entered as 13/02/55. This is valid/sensible but the person may
    have been born on 13/05/55 (2)
(b)
 transcription errors from source document (1) prevented by double keying or self validating
    fields (1)
 deliberate falsification or corruption (1) prevented by batch totals or audit procedures (1)
 transposition errors (1) prevented by double keying or self validating fields (1)
 hardware failure/malfunction (1) prevented by validation against acceptable values (1)

1998.5 (12 marks) and 2000.6 Exemplar
SupaGoods is a home sales company. Catalogues are left at people’s homes. A local agent calls
two days later to take orders and collect the catalogues. The agent sends the details of the goods
ordered to the Head Office where they are processed. The completed order is returned to the
agent who distributes the goods and collects payments.
        (a) Describe two distinct methods of data capture for the agent. State one advantage and
        one disadvantage of each method. (6)
        (b) The orders are validated at Head Office.
        (i) Explain what is meant by validation. (2)
        (ii) Describe briefly two validation checks that might be carried out on an agent’s order.
        (4)

(a)       Allow sensible descriptions of OMR, OCR, bar coding keyboard entry or paper based
          collection. Descriptions must be relevant to the technique chosen and the situation.

          Three marks per technique. Description (1), Adv. (1) and Disadv. (1) max 2x3
(b)(i)    Use of the computer/software (1) to detect data, which is incomplete or unreasonable
(1).
   (ii)   Allow, Check digits for catalogue numbers or agent’s ID code
          Reasonableness or Range check on quantities
          Presence check on agent’s code/name or quantity
          Format or character check on catalogue number/agent’s code
          Hash total for catalogue numbers
          Control total on quantities ordered or number of entries on the order
          Lookup on agent’s code or catalogue number
          (check and field).




                                             Page 4 of 7
                AS Module 2 (ICT2) TOPIC 11.2 Verification and Validation


1999.8 (6 marks)
Most banks and building societies now offer cash withdrawal facilities through the use of
Automatic Teller Machines (ATMs). The data that needs to be entered before a transaction can
take place will include the customer's account number and Personal Identification Number (PIN).
(a) (i) State three validation checks that should be made on the customer's account number. (3)
(ii) State one validation check that should be made on the PIN. (1)
          (b) These transactions will normally involve transmission of data from and to an ATM
          via a communications link. State one security precaution that should always be taken,
          giving a reason why it is needed.        (2)

(a) (i) Account Number: allow Check digit but not check sum [1] length or format [1] and check
the account number exists on the master file [1]
(ii) PIN: allow Length or format Check [1] a check for the PIN accessed by the account number
matching with the PIN entered manually [1]
 (b) Encryption of data or secure lines, specifically the PIN by hardware before transmission, to
ensure secrecy if intercepted [2]
Also allow… checks to ensure data received is identical to that sent to preserve accuracy of the
data [2]
Authentication of the client-ATM hardware [1] + reason [1]

2000 .6 (11 marks)
An electricity supply company needs to arrange for householders' meters to be read regularly.
Meter readers visit each house and record the current meter reading for each account on a paper
data capture document. At the end of the day all the documents are read directly into the
computer system to avoid transcription errors.
        (a) State two items of data that should be printed on the data capture document before it
        is given to the meter reader.       (2)
        (b) Describe two validation checks that should be performed when the data is read into
        the computer system. (4)
        (c) How is it still possible for incorrect data to be stored in the computer system? (2)
        (d) Since the company has a large number of accounts, the billing process is run
        overnight at regular times. Explain why batch processing would be appropriate. (3)

 (a) The computer would print the
• Address / Address and Name
• Account no./Reference No
• Previous meter reading
• Expected high reading
• Round code/date for document batching
• Password for entry
 (b) 2.2 Verification and validation
ONLY HAND WRITTEN ENTRIES NEED VALIDATION I.E. METER READING
NO MARK FOR JUST NAMING A CHECK
 Range check on meter reading[1] to (may also involve looking up the previous reading),
     check that current reading >= previous reading [1]
 Format/Type check on meter reading [1] to check that each hand written entry is a digit [1]
 Presence check on meter reading [1] has meter reader completed the reading or marked as
     no access to property [1]
 Don't accept presence checks or file lookup checks for the account no as it is a turnaround
     document.



                                            Page 5 of 7
                AS Module 2 (ICT2) TOPIC 11.2 Verification and Validation


 (c) Because the meter reading may still be incorrect even if they are valid [1] e.g. Meter Reader
has written the numbers down incorrectly/OCR has misinterpreted the hand written
numbers/meter incorrectly recording units used 2
(d) Topic Identify appropriate contexts for different modes of processing data
 Because it can be run without intervention from the user and
 Batch of records will need to be processed periodically
 Because the output is not time critical
 Records of consumption of electricity do not need to be up to date at all times. (in fact it
     cannot be with current systems)
 Large volume of documents all of the same type


January 2001.7 A video and games rental shop uses a database package to record membership
details of customers who wish to borrow videos or games. The following details must be entered
for each new member; full name, address, postcode and date of birth.
(a) Name and describe a validation check that could be used when data is first entered into the
following fields for a new member. Your validation check must be different in each case.
(i) Full_Name (2 marks)
(ii) Postcode (2 marks)
(iii) Date_of_Birth (2 marks)
(b) The database package automatically generates a membership number. Explain why this is
required. (2 marks)
VALIDATION CHECKS MUST BE DIFFERENT FOR EACH PART
(a) (I)
• Format/Character check (1) check for appropriate range of characters/Relevant Example. (1).
• Presence check (1) check that data is put into the field (1)
Any 1 °— (2, 1, 0) 2
(a) (ii)
• Format check (1) check for uppercase alpha numeric/picture given (1)
• Lookup (1) on range of local postcodes (1).
• Length check (1) check that there are 7/8/9 characters (1)
• Cross field check (1) with address (1)
Any 1 °— (2, 1, 0) 2
(a) (iii)
• Format check (1) check for appropriate structure and range for a date e.g. DD/MM/YY (1).
• Range check (1) check for over 16/18 (1)
Any 1 °— (2, 1, 0) 2
(b)
• Membership number/Customer ID etc. to uniquely (1) identify the member (1)
MAX 2 marks 2




                                            Page 6 of 7
                AS Module 2 (ICT2) TOPIC 11.2 Verification and Validation


January 2002.6. A company stores its staff records in a computer file. Examples of records from
the file are shown below.
Surname              Forename             Department           Name of            Extension
                                          Number               Department         Number
Chan                 Dan                  132                  Accounts           572
Wendy                123                  Training             467
Jones                Sarah                121                  Training           468
Chan                 Dan                  132                  Accounts           572
Patel                Paul                 132                  Acounts            573
(a) The records in this file have been input with the validation checking turned off. Referring to
the examples above, describe the problems that have occurred. (4 marks)
(b) Name and describe three different validation checks that would have prevented these
problems. (6 marks)

(a)
• No entry in Surname field for second record (I)
• Training has two different department number (II)
• Dan Chan has two identical records (III)
• Accounts spelt incorrectly in last record (IV) 4
(b)
• Presence check (1) and description (1) (I)
• Cross field check/Look-up list (1) and description (1) (II) (IV)
• Uniqueness check (1) and description (1) (III)
• Cross field check/Look-up list (needs to be different response to II (1) and description (1) (IV)




                                             Page 7 of 7

								
To top
;