Capturing, Selecting, Managing and Exchanging Data

Overview

Recall questions on methods of capturing data and automatic data capture, validation at the point of capture, selecting data, managing and archiving data, exchanging data in CSV and XML, and protecting personal data in transfer.

What you need to know

Answer each one before you reveal it.

Questions from this subtopic

Each one with the model answer underneath.

1. 1

   Describe three methods of capturing data for a database and give a situation suiting each.

   Model answer

   Optical mark recognition reads pencil marks in printed boxes and suits multiple choice tests and lottery slips, being very fast and accurate for fixed-choice answers. Optical character recognition scans printed or handwritten text and converts it into characters, which suits digitising completed paper forms or postal addresses. Barcode or QR scanning reads an encoded identifier directly and suits supermarket checkouts and stock control, where speed matters and the item already carries a code. Other valid methods include magnetic ink character recognition on cheques, sensors feeding readings automatically into a system, and online forms typed in by the user.

2. 2

   Why is automatic data capture usually preferred to manual keying?

   Model answer

   It is far faster, so far more records can be handled in the same time and staff costs fall. It is much more accurate, because it removes transcription errors made by people copying from paper. It also feeds data straight into the system without an intermediate step where forms could be lost or misread.

3. 3

   What role does validation play at the point of data capture?

   Model answer

   Validation checks that entered data is reasonable and of the right form before it is accepted, using rules such as presence checks, range checks, format checks, type checks and lookups against existing records. It cannot prove the data is true, but it stops many errors entering the database at the earliest point, which is much cheaper than correcting them later once the wrong data has spread through reports and linked tables.

4. 4

   What is meant by selecting data, and how is it done?

   Model answer

   Selecting data means extracting only the particular records and fields that are needed for a given purpose, rather than working with the whole database. It is done by writing a query that names the required fields and states conditions the records must meet, so that irrelevant data is filtered out. This reduces the volume handled, speeds up processing and reporting, and, where data is being passed on, means only the necessary information is disclosed.

5. 5

   A charity wants to send a mailing to supporters in one region who gave in the last year. In terms of selecting data, what should it extract, and why not simply use the whole database?

   Model answer

   It should extract only the name and address fields, and only the records where the region matches and the last donation date falls within the year. Using the whole database would waste processing and postage on people outside the criteria, and would expose personal data such as bank details and donation histories that the mailing task does not need, which is both a privacy risk and unnecessary.

6. 6

   What does managing data involve once it is in a database?

   Model answer

   Managing data covers keeping it accurate and current by amending and removing records as things change, controlling who may read or alter which parts of it, enforcing the validation rules and referential integrity so that it stays consistent, indexing and organising it so queries stay fast as it grows, taking regular backups and being able to restore from them, archiving data no longer in active use, and handling several users working on it at once without them interfering with each other.

Other ways to work this

The same content, four other ways.

What is inside Levo

Six ways to work the same material.