Statistics

Unit 1 DB

Data collection is the systematic gathering of information for a specific purpose. Data can be collected in various ways.

The simplest method is from existing records. The records can be internal records like the employee records of a corporation or external records like Government statistics. Today the Internet is a popular and useful source of existing records. However, all required information may not be available from existing records and data may have to be "generated".

A simple non-interactive way of generating data is observation. An example would be counting the number of vehicles passing through a particular road. This would be useful in establishing peak traffic conditions. However, data collection usually involves interaction with people who have the required information.

The most common technique is the survey through questionnaires. A set of structured questions, usually requiring multiple-choice answers, is prepared and sent to selected respondents. This can be done by post, e-mail and even over the phone. Questionnaires are handed out at shopping malls, restaurants and airports. The advantage of this type of survey is that the responses are focused and easily amenable to statistical analysis. Sometimes more complex and unstructured responses are required, in which the next question depends on the previous answer. The appropriate data collection technique in this case is the interview. For an interview to be useful the interviewer needs considerable skill.

The final and most complex technique is experiments. These are useful when one wants to isolate the effect of a particular stimulus from several stimuli faced by respondents. The most common example of this is in testing of new drugs.

Examination of collected data is important because it enabled informed, transparent

...