Final answer:
You might need to use matching instead of random selection when assigning participants to groups is not feasible or ethical. Matching helps to control for lurking variables, while convenience samples are easier but less generalizable. Non-random techniques ensure representation or cater to practical constraints in the study.
Step-by-step explanation:
You might need to use matching instead of random selection and assignment in a situation where you cannot ethically or practically assign people to control or experimental groups, such as when studying the effects of exercise on health. Instead, you study people who naturally fall into these groups and try to match or control for other variables, known as lurking variables. Matching is also used when you want to ensure that specific characteristics are evenly distributed across groups, such as when interviewing married couples or collecting data twice from each individual. Alternatively, researchers may use convenience samples, which can have implications like lack of generalizability from the sample to the population.
For example, a soccer coach using stratified sampling when forming a team, or a high school educational researcher who interviews an equal number of male and female teachers, is applying non-random sampling techniques to ensure specific representation within the sample. Similarly, when a woman interviews only travelers sitting near gates, avoiding those who are busy, she is applying convenience sampling, which may lead to bias in the data collected.