Computer assisted telephone interviewing: effects on interviewers and respondents

Area of Science:

• Survey methodology
• Social sciences research methods
• Data collection techniques

Background:

• Traditional paper questionnaires are widely used in surveys.
• Computer Assisted Telephone Interviewing (CATI) offers potential efficiencies and standardization.
• Evaluating the impact of CATI on data quality and interviewer behavior is crucial for methodological advancement.

Purpose of the Study:

• To compare the effects of CATI versus paper questionnaires on survey response distributions.
• To assess CATI's influence on interviewer behavior and nonsampling errors.
• To evaluate differences in response rates between CATI and paper-based surveys.

Main Methods:

• A survey experiment was conducted, randomly assigning cases to either a paper questionnaire or a CATI system.
• Interviewer behavior, response distributions, and nonsampling errors were measured.
• Response rates for both methods were compared.

Main Results:

• Few significant differences were observed in response distributions between CATI and paper methods.
• Some evidence suggests reduced interviewer variance in the CATI group.
• Response rates showed little difference between the two data collection methods.
• Interviewers noted CATI's ease of questionnaire flow but preferred paper for editing prior answers.

Conclusions:

• CATI systems do not substantially alter response distributions or rates compared to paper questionnaires.
• CATI may offer benefits in reducing interviewer variance.
• Interviewers perceive distinct usability advantages for both CATI and paper methods, indicating potential trade-offs.