Response bias

Response bias (English response set, response bias), response bias or response error is a systematic distortion of, inter alia, in interviews, surveys, opinion polls social science surveys, psychological tests and questionnaire responses obtained. The data does not reflect the true ( actual and "true" ) settings and situations. The reasons for such variations are seen among respondents in the questions and in the design of the questionnaire, to look into the specific situation or the interviewer ( interviewer effect).

Response

Each interview conducted by an interviewer or a questionnaire represents a social situation that can have an impact on the answers. The examiner may inadvertently have an influence on the responses and other behaviors by their presence and by their expectations. Above all, the individual respondents' attitudes and personality characteristics are, however, influence, because every self-assessment is also a self-representation. Therefore pronounced response tendencies can be simultaneously viewed as important facets of personality traits.

The distortion caused by response tendencies is regarded as a typical problem of method of questionnaires. However, such effects will undoubtedly occur in interviews and in most other methods of psychological assessment. The effects of methodological reactivity, that is, influencing the results of the applied research methodology, are a fundamental problem for almost any psychological methodology, even many medical examination methods. Psychological and medical tests may also lead to the phenomenon of reactance, that is, defensive reactions and refusal to participate.

Response trends were examined from different disciplinary perspectives, including test methodology, differential - psychology, social psychology, cognitive psychology. On this theme, including schematic judgment tendencies characteristic error of judgment and systematic illusions of memory. A related issue form from the medical diagnosis, especially in psychiatry, known phenomena simulation, aggravation, dissimulation or trivialization of complaints and findings; the frequent discrepancies between the experienced physical symptoms and the objective findings form an important topic in the psychology of illness behavior.

Most of these response bias and Beurteilerfehler can be found already in the 1950s with Lee J. Cronbach and Joy P. Guilford, including the Yes - saying tendency ( Acquiescence ) and the tendency to undecided middle or at the ends of a scale ( extremes ).

Typical response bias

Several effects are considered to be the cause for the unity of method variance (English common -method variance ).

Method effects

• Effect of social decontextualization

Respondent characteristics

• Acquiescence: The Yes Sage tendency ( content -independent Acquiescence, acquiescence tendency ) is the tendency of people with more questions is "yes ", " true " or answer "right" regardless of the content of the questions. The Yes Sage tendency is found more often in authoritarian personalities and timid and conservative people, as adapted behavior.
• Consistency effect ( engl. consistency bias) arises from the tendency similar sounding statements to answer consistent, so that they match each other in content, though not uniformly so true.
• Retrospection effect ( engl. recall bias, retrospection effect) means that experiences and events in retrospect, the next day or after a few weeks, positive or negative will be evaluated, for example, the pain experienced in the memory -intensive than in the currently levied assessments (negative retrospection effect).
• Rezenzeffekt (English recency effect ) states that later in-depth information a greater impact on the memory capacity of a person to exercise than they used to in-depth information. Short-term memory is the last perceived information more weight (see Primacy - recency effect).
• Hindsight Bias (English hindsight bias) describes inaccurate memories when people after they have experienced the actual outcome of an event, systematically wrong remember their own previous predictions.
• Silence distortion (English non-response bias): respondents can answer another behavior than it would not answer the end when they took part in the survey. The lack of responses thus distort the overall picture.
• Social desirability ( Social Desirability Response Set ) results from the tendency not to answer items according to the personally relevant setting, but according to social norms that are desired by the subject's view. The tendency for social desirability is usually considered to be a widespread, more or less unintentional tendency to a positive self-presentation. Methods, the number of honest answers to increase by anonymization, the randomized response technique and the Unmatched -Count Technique are ( sa sensitive question ).
• Central tendency: The tendency to center (german error of central tendency ) is the tendency of respondents, rather select the middle scale points in multi-stage scales ( eg, Likert scales ).
• Tendency to clemency / Hardness: The tendency to leniency or harshness ( engl. error of extreme tendency ) is the tendency of respondents to tilt at multistage answers to extremes. This is especially true in test or test situations.

Influences the formulation and design

The answers can be influenced by the wording of the questions (English wording) and (with questionnaire and the possible answers ), by the clarity of the questions and the instructions and by the conduct of the investigation. These include unnecessary foreign words and ambiguous formulations, or those who already have a certain response suggest ( framing effect ). Also, the order of the questions may cause distortion order effect ( also: question number effect or position effect ) when a previously asked question affects how the following question will mirror construed and evaluated.

Who developed a questionnaire should be based on the appropriate textbooks and study the draft carefully before use (see questionnaire, interview technique ).

Influences the examiner and the examination situation

Many of the aforementioned effects on the side of the respondents ( examinees ) correspondences on the side of the investigator ( interviewer ), eg Observation and assessment errors as halo effect or certain characteristics of the interviewer ( interviewer bias). If the examiner can even recognize certain expectations, creating a experimenter effect (also: Rosenthal effect, Pygmalion effect, Rosenhan experiment or experimenter - artifact). The Hawthorne effect describes that already participating in an investigation can trigger particular expectations which lead to biased results.

Criticism

Conceptually different response tendencies can be distinguished, but these are empirical and methodological hardly be distinguished from each other. They are often linked to each other and are influenced by the individual cognitive style, semantic and linguistic difficulties, the order effects, etc.. Moreover, many of these trends as the Yes Sage tendency to extreme the tendency to social desirability or the tendency or mean response categories typical characteristics of certain personality traits.

In the test methodology, the opinion was sometimes common that these effects are controlled or compensated: by pairs of oppositely poled questions (what language often complicated ), by specifying a two-stage response options (only "yes" or "no"), by control issues, or by should cover a so-called lies scale, the contradictions and probably incorrect answers. The attempt to separate conscious and unconscious shares, or the desire to test methodological isolation and statistical correction overwhelm the questionnaire methodology in principle. To clarify how a person "really" is reminiscent of an earlier and outdated conception of an unchanging personality and does not correspond to the understanding of personality traits with time-and situation-dependent variability. In particular, the study of social desirability requires (in terms of interaction), which establishes a link an interactionist perspective: between test situation, test motivation, personality characteristics, stylistic features, verbal intelligence, personal expectations and motives, willingness to adapt, benefit considerations, fear of harm or fear of discovery as well as the individual and the general value dispositions such as honesty and openness. There will also be a wide spectrum of opinion which answers are desired in a particular situation or task or not.

The self-representation is an integral part of the self-assessments expressed. Who uses self-assessment in the psychological assessment, must accept also the structural subjectivity of these self- reports. To want to delimit individual facets of self-expression within the self-assessment of a man who seems to be only interpretive possible if there is objective CVs or behavioral observations can not be included. In the social psychological research setting objectification experiments reported by methods of psycho- physiology and the bogus pipeline technique with the deception about an alleged polygraph are unsuitable for various ethical or methodological reasons.

The guidelines for quality assurance in psychological assessment require that the results of the investigation as little as possible can be distorted by the candidates themselves. So far, however, there is for the analysis of questionnaire methods and interviews no rules or conventions, as is to be reliably detected or prevented.