Descriptive Research Design: Survey, Observation and Internet

Descriptive Research Design: Survey, Observation and Internet

Survey Methods

Ü  Respondents are asked questions regarding their behaviour, intentions, attitudes, awareness, motivations, and demographics and lifestyle characteristics

Advantages of Surveys

Ü  Standardisation

Ü  Ease of administration

Ü  Ability to tap the ‘unseen’

Ü  Large sample

Ü  Low cost

Ü  Increases geographic flexibility

Ü  Suitable for tabulations and statistical analysis

Ü  Generalisability

Ü  Sensitivity to subgroup differences

Disadvantages of Survey

Ü  Difficult developing questionnaires

Ü  Need to develop constructs, scale measurements, questionnaire design

Ü  Respondents may be unable/unwilling to provide the desired information

Ü  Structured and fixed responses

Ü  Difficult to probe

Classification of Survey Methods

Ü  Nature of survey interaction

Person to person

Computer assistance


Ü  Mode of administration

Personal interviews

Telephone interviews

Mail interviews


Figure 5.2 Classification by Mode of Administration

Personal Methods

Personal face-face in-home interviews

Ü  Interview conducted at the respondent’s home


Ü  When personal contact is essential

Ü  Conducive environment to questioning process

Personal Methods cont.

Central location personal interviews

Ü  Interview conducted in shopping centres


Ü  Travel costs are eliminated

Ü  Interviewer can interact with respondents


Ü  Non-representative sample

Ü  Uncomfortable environment

Personal Methods cont.

Computer-assisted personal interviewing (CAPI)

Ü  Direct entry of research information into a computerised database

Telephone Methods

Traditional hard-copy telephone interviews

Ü    Phoning a sample of respondents and asking them a series of questions.

Ü    Interviewer records answers on hardcopy of questionnaire.


Ü    Inexpensive

Ü    Yields a very high quality sample

Ü    Quick


Ü    Inability of respondents to see questions

Ü    Inability to observe respondents

Ü    Limitations on information quality and quantity

Ü    Growing use of answering machines

Ü    People associate surveys with telemarketing

Telephone Methods cont.

Computer-assisted telephone interviewing (CATI)

Ü   Computerised questionnaire administered to respondents over the telephone


Ü   Inexpensive

Ü   Computer dials phone number

Ü   Computer skips questions

Ü   Can customise questions

Ü   No editing required

Ü   Analysis can be done at any stage

Ü   Cost savings

Ü   Quality control

Ü   Time savings

Telephone Methods cont.

Computer automated telephone systems (CATS)

Ü  Computer-synthesised voices are used to ask questions over the phone

Ü  Respondents select numbers on the telephone keypad to answer questions

Ü  Voice recognition is likely to be used in the future to record and count responses

Mail Methods

Mail interviews

Ü  Questionnaire is developed and mailed to pre-selected respondents who return the completed surveys by mail

Ü  Mail interview package consists of the outgoing envelope, cover letter, questionnaire, return envelope

Mail Methods cont.


Ü  No interviewers to recruit, train, monitor and compensate

Ü  Inexpensive to implement

Ü  Can reach many people


Ü  Low response rate

Ü  Self-selection bias

Ü  Slow form of collection

Ü  Possible misunderstanding of skipped questions

Mail Methods cont.

Mail Panels

Ü   Large representative sample of households that have agreed to participate in periodic mail questionnaires, product tests and telephone survey


Ü   Panel can be tested prior to the survey to obtain a representative sample

Ü   Produces a higher response compared with direct mail

Ü   Allows for longitudinal research


Ü   May not be a representative sample

Electronic Methods

E-mail interviews

Ü A survey using plain text which is e-mailed for the respondent to read, complete and return

Electronic Methods cont.

Internet interviews

Ü  Use HTML to write the questionnaire.

Ü  Survey can be found on the web or emailed to a potential respondent

Ü  Can contribute to higher quality data

Table 5.2 Advantages and Disadvantages of Internet surveys


n    Higher response rates

n    Greater response accuracy

n    More enjoyable

n    More aesthetically pleasing

n    Less expensive

n    Faster turnaround

n    Trend-leader population

n    Instantaneous global reach

n    Customised surveys

n    Ability to find highly specific research targets


n    Self-selection

n    Unrepresentative population

n    Anonymity bias by Internet users

n    Respect for multicultural responses

n    Anxiety in divulging information

n    Shorter attention span

n    Lack of interpersonal nuances

n    Possibility for multiple and repeated polling

n    Novelty bias will diminish

Observation Methods

Ü  Recording the behaviour of people, objects and events in a systematic manner to obtain information relevant to the problem

Ü  Can be classed as a qualitative or quantitative research technique

Ü  Observational methods include:

Structured vs unstructured

Disguised vs undisguised

Natural vs contrived


Ü  Researcher specifies what is to be observed and how the measurement will be recorded

Ü  Reduces potential for observer bias and enhances the reliability of the data

Ü  Appropriate when problem has been clearly defined and the information needed has been specified

Ü  Suitable for conclusive research


Ü  Observer monitors all aspects of the phenomenon that seems relevant to the problem at hand

Ü  Appropriate when the problem has yet to be formulated precisely and flexibility is needed in observing to identify key components of the problem and to develop hypotheses

Ü  Observer bias is high

Ü  Suitable for exploratory research


Ü  Respondents are unaware that they are being observed

Ü  Respondents behave naturally

Ü  Props include one-way mirrors, hidden cameras or inconspicuous mechanical device


Ü  Respondent are aware that they are under observation

Ü  May bias behaviour patterns


Ü  Observation takes place in the environment [supermarket]

Ü  Observed behaviour will more accurately reflect true behaviour


Ü  Respondent’s behaviour is observed in an artificial environment

Ü  Do not need to wait for behaviour to occur in a natural environment

Figure 5.4 A Classification of Observation Methods

Case: Recycling Behaviour

A municipal council in Australia examined household rubbish to identify whether households were disposing of rubbish appropriately, in particular, if they were recycling most recyclable items.  Although surveys indicated that households were recycling, trace analysis was deemed the best method  in obtaining information on actual behaviour.

Is this ethical?  Would you like it if someone went through your rubbish?

Comparison of Survey and Observation Methods

Relative Advantages of Observation

Ü  Measures actual behaviour

Ü  No interviewer bias

Ü  Useful when respondent is unaware/unable to communicate feelings

Relative Disadvantages of Observation

Ü  Little is known about the underlying motives, beliefs, attitudes, and preferences

Ü  Time consuming and expensive

Ü  Borders on being unethical