Error of Measurement

Error of Measurement – Error(s) made in a study because the measuring instrument is no entirely accurate. See also: Measuring Instrument/Standard error of measurements

Experimental Error – The variation that is probable and that may be anticipated in every instance in which an experiment is duplicated. An adequate experimental design must include adequate safeguards against experimental errors.

Fatigue Effect – If, according to the respondent, a study takes too long, his interest and motivation are lost. He respondent becomes fatigued, and his answers and reactions will be unreliable. Alternately, in order to beat time, the task is rushed. The precise moment at which the effect appears varies from one person to another, from one study to another, from one location to another, It is to be recommended that an interview not be overlong. (As a rule, 45 to 60 minutes should be a limit.)

Fear of Censure – The fear a respondent has of being thought stupid, inexperienced, poor, “different,” etc. by the interviewer. See also: Response tendency

Interviewer Bias – A distortion in the response or registered information caused by the behavior of the interviewer. The causes may be, among others, interviewing the wrong person, poor contact with the person interviewed resulting in the acquisition of incorrect information, or systematic errors due to inadequate presentation of questions or registration of answers. See also: Bias/Interviewer

Memory Error – Every error made in a research project owing to a failure of the memory. For example, mistaking one number for another.

Random Variance – Collective term for all kinds of interfering factors that, frequently imperceptibly, enter into a research project. These interfering factors may occur in the case of persons (headache, lack of interest in the study) as well as in the case of the material (illegible due to stains) and also from the research situation (power failure). See also: Source of error

Research Manager Expectation Effect – The results of a research assignment may be affected because the research manager or researcher lets his expectations (unconsciously) be known to his sample. This effect can be easily marked in small-scale studies, giving an important color or slant to the results.

Response Error – The variability in the reactions of a respondent – variability in the perception of reality by the respondent or variation in response time.

Response Tendency – Syn: Response set/ Response style The tendency of persons to give answers that have been influenced by the manner in which the questions have been put or the sequence of the questions in a questionnaire. See also: Sequential Effect

Sequential Effect – An effect that appears in questionnaires test, evaluations. The sequence of a specific number of stimuli (for example,questions) “compels” a person to judge something or someone differently from what would be the case with some other sequence. For example, a subject questioned with “no.” The subject questioned does not expect to have answer with “yes” so many times in sequence. The effect is subjective (person-related). See also: Rotation

Side Effect – Effects that are, generally, undesirable and not anticipated. For example, irritation and/or dishonest response to interview goes on for a long time.

Socially Desirable Answer – An important effect in tests and attitude-scales. Frequently, respondents give answers that are socially desirable, instead of honest, sincere answers. For example, generally, people do not like to admit favoring the reintroduction of the death penalty. This sort of effect needs to be taken into consideration in a study. See also: Editing questionnaire

Source of Error – The cause of errors in a study that may be demonstrated. A number of errors (sources) may be avoided nearly entirely, while for other errors no more can be done than recognizing them. In the latter instance, their presence may be taken into consideration. The sources of error may be schematically expressed. Examples:
1. Poor hearing; poor eye-sight; illiteracy; low level of development.
2. The interviewer influences the respondent (doesn’t get along with people of different race’ color’ social class; too enthusiastic in eliciting an answer).
3. Respondents who always give answers that are socially acceptable: “good manners freaks;” people who strive for harmonious patterns in their answers.
4. The interpreter of the research results is entirely objective.
5. Health condition of respondents (headache, etc.); familiarity with the research project (learning-effects).
6. Power failure; poor ventilation in research area, etc.
7. Respondent suddenly become ill-upset stomach, headache.
8. Questions that are not legible (due to stains); interviewer doesn’t understand answers; interviewer interprets answers subjectively. See also: Random Variance

Sympathy Effect – A response to a question that is intended to please the interviewer but is entirely honest and reliable. Usually, the respondent is not aware of this affect, and lying does not occur intentionally; neither is the interviewer aware of this effect in every instance. The effect leads to distortion in a study. It is often difficult to determine the magnitude of such an effect or to counter it. See also: Bias

Vanity Effect – A disturbed reaction from an informant: the person interviewed does not give reliable, honest answers, but, rather, answers in ways that enhance his vanity. For example: pertaining to his income, education, possessions, etc.

White Noise – Sound of different frequencies. Also: disturbances, disturbing noises in an experiment.

Leave a Comment

Web Analytics