调查

A survey is a research method where participants answer a set of questions, administered verbally, on paper, or electronically. Surveys allow researchers to collect information quickly from large numbers of people, making them efficient for gathering data on diverse populations. This approach enables researchers to capture the diversity within a given population and generalize findings more broadly than smaller-scale methods.

Likert scales are numbered rating scales that participants use to respond to survey questions or statements. Typically ranging from 1 to 5, these scales allow respondents to indicate their level of agreement or disagreement with a statement. For example, a participant might rate agreement with 'I'm unhappy with most of my classes' on a scale where 1 means strongly disagree and 5 means strongly agree.

Convenience sampling recruits easily accessible participants, such as a professor's own students, creating a biased sample that limits generalizability. Random sampling selects participants across the entire population, giving every individual an equal chance of participation. Random sampling produces more representative results that can be generalized to the broader population, whereas convenience sampling yields biased opinions specific to the accessible group.

A primary limitation of surveys is that participants may not provide honest responses. People may lie, misremember, or answer in ways they believe make them look good rather than truthfully. For instance, survey respondents might underreport alcohol consumption or hide prejudicial attitudes. This self-reporting bias can compromise the accuracy and reliability of survey data, even when responses are anonymous.

A larger sample better reflects the actual diversity of the population, allowing researchers to generalize findings with greater certainty. When a sample is sufficiently large and diverse, the data collected is more representative of the broader population. Conversely, smaller or biased samples limit generalizability because they may not capture the full range of characteristics present in the population researchers aim to study.

Surveys can include both direct and indirect questions to uncover hidden attitudes. Direct questions ask about overt prejudice, while indirect questions assess behavioral intentions, such as willingness to interact with specific groups. Research on Arab-American prejudice found participants denied prejudicial attitudes but showed less willingness for social interaction with Arab-Americans compared to other groups, revealing subtle prejudice despite explicit denials.

Surveys collect information from larger samples, providing better generalizability and capturing population diversity. However, this breadth comes at the cost of depth—researchers cannot collect the same detailed, nuanced information about each individual that case studies provide. The choice depends on research goals: surveys prioritize breadth and representativeness, while case studies prioritize in-depth understanding of individuals or small groups.