Statistics - Chapter 2 - Collection of data

Purpose: Reach a sound a clear conclusion to a problem which is backed by facts. Ex. Government when formulating policies.

Sources of data

Primary Data

Secondary Data

Collected at source

Communication between investigator and respondent

Original data

Data is obtained in a raw form

It is reliable (since collected first hand)

It is suitable, since questions specific to the research could be asked.

Time consuming and expensive (not economical)

No special precautions or editing needs to be done as data was obtained specifically for the project itself.

Adopting data Collected by an agency or institution which already collected the data through statistical surveys.

Sources include: published journals, magazines, census, government documents, etc.

No direct communication between Investigator and respondent

Not Original

Data is obtained in a finished form (tabulated and presented)

Less reliable (not collected firsthand)

Less Suitable (may not represent exactly what was being researched by the investigator)

Quick and easy (economical)

Special Precautions and editing needs to be done since the data was collected for a different purpose.

Definitions:

Investigator: One who conducts the investigation and seeks information. It can be an individual or organization.

Investigation: Statistical enquiry

Enumerators: They help the investigators in the collection of data. (assistants)

Informant / Respondent: They are the respondents who supply information to the investigator or enumerators. (people being surveyed)

Methods of collection of Primary Data:

Direct personal investigation

Indirect Oral Interview

Telephonic Interview

Information through correspondent

Questionnaire filled by enumerator

Mailed questionnaire method

Direct personal investigation

The investigator personally visits the respondents to collect information or data.

Merits

Demerits

Precautions

Reliability and Original

Correspondent: Local agents trained to collect information from respondents

Indirect oral investigation

Merits

Demerits

Information through correspondents

Questionnaire filled by Enumerator

Mailed Questionnaire Method

Telephonic Interview

Flexibility

Additional Information

Not suitable for wide area

Time-consuming

Expensive

Personal Bias

Social, laborious and experienced investigator

Well-versed in language, trade and customs

Limited number of questions

No personal bias

Establish a relationship (rapport)

The interviewer interviews several other people who are directly or indirectly in touch with the informants.

Wide coverage

Economical

Indirect information (less accurate)

Lack of Accuracy

Lack of uniformity

Possibility of bias to suit personal interests

Local agents or correspondents are appointed and trained to collect information from the respondents.

Merits

Demerits

Wide coverage

Time-saving

Suitable for special-purpose investigations

Continuity

Lack of uniformity

Lack of reliability

Personal Bias

Expensive

Lack of accuracy

Data is collected by an interviewer over the telephone.

Merits

Demerits

Wide area coverage

Time saving

Economical

Reliable

Limited use (Low accessibility - Only people with phones can be interviewed)

Visual feedback is not possible

An enumerator personally visits the informants with a questionnaire, asks questions and notes down responses in his own language.

A questionnaire containing a number of questions relating to the investigation is prepared. It is sent to the informants by post along with instructions on how to fill it. After filling out the questionnaire, the informant sends it back to the investigator.

Merits

Demerits

Wide coverage

Economical

Originality

Free from bias

Maintains secrecy as it filled out anonymously

Limited Accessibility

Low response (due to lack of interest)

Time consuming

Chance of misinterpretation

Merits

Demerits

Reliable and original

Better response (than mailed questionnaire method)

Useful in case of illiterate respondents

Expensive

Time consuming

Inefficiency

Personal Bias