Surveys

Surveys Definition

A survey is a systematic method for gathering information from a defined social group, often involving structured written questions administered verbally via telephone or face-to-face interviews, designed to yield data amenable to statistical analysis. These methods are employed across various domains to provide descriptive statistics at local, national, and international scales, examine complex social phenomena such as poverty, discrimination, or social stratification, and investigate underlying causal processes and hypotheses.

Methodology and Components

Survey methodology comprises several key components that must be considered for validity: sample design and selection, scope and coverage, the format and content of the questionnaire, the survey procedure, and the response rate. The use of multivariate modelling methods has significantly influenced how sociological surveys are constructed and analysed. A critical challenge remains in accurately reducing complex social issues into quantifiable numerical data.

Sampling and Generalisation

Most empirical surveys are conducted on a specified sample drawn from a larger population. To ensure results accurately reflect the entire group, researchers employ techniques such as stratification—dividing the population based on characteristics like gender, race, or socio-economic status—and weighting. Weighting is the process of adjusting sample results to infer the likely outcomes for the total population by assigning values that indicate how many population units are represented by each sample unit.

Modes of Data Collection

The method chosen for data collection significantly impacts the responses received, leading to various ‘mode effects’. Primary modes include self-administered questionnaires, telephone interviews, and face-to-face interviews, though mixed methods (e.g., postal questionnaire followed by an interview) are also possible.

Self-Administered Surveys

Self-administered formats offer advantages such as avoiding interviewer effects, being suitable for sensitive topics, reducing response-order effects, and allowing respondents ample time for considered answers. However, disadvantages include a lack of flexibility, limited control over the response process, potential disadvantages for those with lower literacy levels, and higher rates of non-response.

Interview-Based Surveys

Telephone and face-to-face interviews involve the interviewer, which introduces potential social desirability bias and interviewer effects. While these modes offer greater flexibility and allow the interviewer to clarify questions, they also present challenges regarding the presence of third parties and potential social influence. Face-to-face methods can facilitate longer or more complex surveys than other modes.

Technological Advancements and Critique

Modern research increasingly utilises Computer Assisted Telephone Interviewing (CATI) and Computer Assisted Personal Interviewing (CAPI), which allow interviewers to input responses directly using personal computers, thereby reducing data entry time. Researchers must ensure that the efficiency gains from these technologies are not compromised by errors introduced during scheduling or filtering.

Despite their utility, surveys offer only a partial description of complex social issues; Harold Garfinkel noted this tendency towards aggregate-based research as a shift away from ‘indexical’ to ‘objective’ expressions. Furthermore, surveys may be inappropriate for populations in developing countries where assumptions about literacy or personal record-keeping (such as dates of birth) cannot be made, and they risk invalidation through biases like acquiescence bias if disagreement with an interviewer is deemed impolite.