Quant #05: Hypotheses and Research Questions

Every survey research starts with hypotheses and research questions. Hypotheses refer to predictive statements about the relationship between two or more variables. Hypotheses can arise from or be informed by past research. For instance, assume that a researcher learns that past research has established that people's level of education can influence their decision to use paper bags as a means to protect the environment. If the researcher wants to study the same behavior among a company's customers, he or she may formulate the hypothesis by predictively stating that "customers' level of education influences their decision to use paper bags as a means to protect the environment".

Research questions have the same goals as hypotheses. However, they are stated in a question format. For example, using the same example as above, one can formulate the research question as follows: "Does customers' level of education influence their decision to use paper bags?" Research questions can typically be formulated as difference questions, association questions, or descriptive questions. Difference questions seek to assess the difference among at least two groups (e.g., does one group tend to engage in a given behavior more often than another group?). Comparatively, association questions focus on the relational aspect (does an increase in one variable lead to an increase or decrease in another variable?). Both difference and association questions deal with at least two variables. Descriptive questions, in contrast, solely focus on one variable (e.g., on average, how many people responded "yes" to a given question).