Follow Us On:

   

• Internet population demographics are not directly representive of the off-line population.

• For general population surveys, there is a need to control for the Internet's demographic skews (i.e., income, education).

• Non-response bias can affect sample representation and validity (passionate positive/negative responder bias).

• Verification of survey respondent identity is more difficult.

• As a passive methodology, forcing necessary quota groups can be difficult.

• A moderate to high relationship with the respondent is needed to achieve necessary response rates.

• The survey subject needs to be relevant and interesting to the respondent.

• The ability to probe a respondent is often not as effective compared to an interview with a live person.

• Respondents often have a lower level of trust regarding the guarantee of anonymity and confidentiality.

• There is a greater likelihood of technical "gliches," for example, due to required 24/7 availability and tendency for large volume of respondents to reply within first 48 hours.

• Response rates can get low if not carefully managed, leading to representation issues.

• Some respondents perceive telephone surveys as more invasive.

• Surveys can only be conducted during certain times of day.

• Data collection often takes longer compared to an Internet methodology and is less convenient to respondents.

• Unskilled interviewers can affect results by leading/influencing respondents.

• Unsupervised interviewer cheating can occur.

• Telephone surveys are more expensive than an Internet methodology.

• Open-ends need to be transcribed.

• The same survey often takes a little longer on the phone because it takes longer to have questionnaire read to respondent.and some respondents can be wordy.