Final answer:
The term for a type of selection bias in surveys, polls, or telephone questionnaires caused by high nonresponse rates is called response bias. This can lead to unreliable data that doesn't reflect the entire population's views. Factors like the Bradley effect and non-responses exacerbate this issue, highlighting the importance of representative sampling methods like probability sampling.
Step-by-step explanation:
The term you’re looking for, which refers to a type of selection bias concerning high rates of nonresponse in surveys, polls, or telephone questionnaires, is known as response bias. Self-selected samples are a common example of response bias, where responses are only collected from individuals who voluntarily choose to participate, such as seen in online surveys. This can lead to unreliable data as it may not accurately represent the views of the entire population.
Several factors contribute to response bias in surveys. For instance, during the 1982 California gubernatorial race, the phenomenon known as the Bradley effect occurred, where polls inaccurately reflected public opinion due to social desirability bias. This occurs when individuals provide responses they deem socially acceptable rather than their true beliefs, potentially skewing poll outcomes.
Other considerations highlighting the challenges of polling include the unavailability of certain demographics during telephone surveys due to lack of telephone access or unpublished numbers, as well as respondents' unwillingness to admit lack of knowledge on the topic. Furthermore, non-responses can make the results unrepresentative of the entire population, especially if those who respond have strong opinions that are not shared by the majority.
Effective survey methods such as probability sampling aim to overcome this bias by randomly selecting participants from the larger population to obtain more representative data. However, sampling errors can still occur if the chosen sample is not adequately representative of the population being studied.