Appropriateness of interviews, surveys and observation as methods of collecting data to determine needs and requirements

SD (2011) U3O1 KK05

The first state of the PSM is analysis where an organisation and its systems are examined to determine the nature of the problem to be solved. At the end of analysis, a logical design can be drawn up which sets out the feature set of the finished system.

To collect information, different methods may be used. Not all methods are appropriate to all information.

Empirical information (fact-based information derived from experimentation or observation rather than theory) is the strongest form in many cases. This involves measuring, timing, counting, adding up, averaging etc - all fact-based rather than opinion-based subjective information.

Examples:

• Use a stopwatch to time how long it takes a program to print 10,000 invoices.
• Scan an error log to count how many breakdowns a system had last month.
• Summarise financial records over the past year to add up the total cost of ownership of a communications system.
• Add up the total amount of time a worker spent being productive in an 8 hour shift instead of working around system faults.

Where possible, collect empirical evidence when at all possible. It leave little room for doubt or argument.

Other information, however, cannot be so scientifically collected: opinions, for example. For these cases, it would be appropriate to conduct interviews, surveys, questionnaires to gather workers' feelings.

Examples:

• Survey staff on their feelings of stress when dealing with a program.
• Ask customers how attractive they think a system's output is.
• Conduct a poll to summarise how easy users think a tool is to use.
• Observe workers' demeanour when using a system.

Do not confuse the two methods. Don't survey people whether they think the existing system is expensive to run; don't try to measure people's feelings of comfort when using some software.