Acquiescence bias

Causes

Agreeableness

A prominent psychological explanation attributes acquiescence bias to social norms that encourage agreeable behaviour. Evidence indicates that respondents approach surveys as though they are common conversations. A consequence of this is that the conventions that govern conversations influence the interpretation of survey questions and responses to them. Accordingly, pressure to conform to such norms and conventions prompts people to agree with stimulus statements. Based on research into the "Big Five" personality traits, individuals are predisposed to agreeable behaviour to differing degrees.

Perceived authority of the interviewer

An explanation favoured by sociologists is that acquiescence bias is a product of the combination of the inclination to yield to the opinions of high authority individuals, and respondents' perception of the researcher/ interviewer as having higher authority. According to this explanation, when selecting answers, a tendency to agree with statements made by the interviewer arises from the respondent's intention of being polite or respectful. In support of this, there is evidence that indicates that respondents of lower social status acquiesce more frequently than respondents of higher social status. However, several studies have failed to replicate this finding.

Satisficing

Acquiescence bias is proposed to be a product of 'satisficing' behaviour. 'Satisficing' sees respondents select responses that are satisfactory or good enough, rather than engage in 'optimizing,' which produces best possible selection. This is done to conserve cognitive energy.

Four-stage model of the response process

Roger Tourangeau, Lance J. Rips, Kenneth Rasinski have developed a cognitive model that proposes four stages to the process of survey response selection. Each stage entails several specific cognitive processes. For example, the Comprehension stage entails paying attention to a question or instruction set. They propose that some cognitive processes are required in order to select an answer whereas others serve as optional aids. A respondent's decision of which processes to employ from the model is determined by a number of influences, notably speed and precision.

The four stages:

• Comprehension – understanding the question and the information required.
• Retrieval (for factual questions) – remembering or calling to mind the appropriate information.
• Judgement (for factual questions) – the processing of recalled information to form judgements.
• Selection – choosing and communicating an answer. Two sets of processes have been proposed: translating a judgement into the scales provided, and revising the response based on factors such as 'consistency'.

Survey response effects can arise at any stage of the response process.

Strong and weak satisficing

Two forms of 'satisficing' have been proposed:

• Weak satisficing: Respondents still execute all four stages of the 'Four stage model of the response process.' However, the stages are carried out less rigorously, which results in the output being satisfactory rather than highly accurate. An example of weak satisficing is a respondent who doesn't search their memory as deeply as possible. This would lead to acquiescence if the respondent only scanned their memory for information supporting the positive response option.
• Strong satisficing: Respondents implement a surface-level approach to answering the question by omitting the 'Retrieval' and 'Judgement' stages and only engaging in 'Comprehension' and 'Selection.' The respondent does not access any internal cognitive resources concerning the construct of interest for the question. A mechanism of selecting an appropriate answer may instead involve interpreting external cues such as question-wording. Following the social convention of agreeing is one alternative mechanism that would lead to Acquiescence bias.