Editing Likert scale (section)

== Composition ==
{{More citations needed section|date=June 2023}}
[[File:Example Likert Scale.svg|thumb|An example questionnaire about a website design, with answers as a Likert scale]]
A ''Likert scale'' is the sum of responses on several ''Likert item''s.  Because many Likert scales pair each constituent Likert item with its own instance of a [[visual analogue scale]] (e.g., a horizontal line, on which the subject indicates a response by circling or checking tick-marks), an individual item is itself sometimes erroneously referred to as being or having a scale, with this error creating pervasive confusion in the literature and parlance of the field.

A Likert item is simply a statement that the respondent is asked to evaluate by giving it a quantitative value on any kind of subjective or objective dimension, with level of agreement/disagreement being the dimension most commonly used.  Well-designed Likert items exhibit both "symmetry" and "balance". Symmetry means that they contain equal numbers of positive and negative positions whose respective distances apart are bilaterally symmetric about the "neutral"/zero value (whether or not that value is presented as a candidate). Balance means that the distance between each candidate value is the same, allowing for quantitative comparisons such as averaging to be valid across items containing more than two candidate values.<ref>{{cite book|last1=Burns|first1=Alvin|title=Basic Marketing Research|year=2008|publisher=Pearson Education|location=New Jersey|isbn=978-0-13-205958-9|pages=[https://archive.org/details/basicmarketingre0000burn/page/250 250]|edition=Second |last2=Burns |first2=Ronald |url-access=registration|url=https://archive.org/details/basicmarketingre0000burn/page/250}}</ref>

The format of a typical five-level Likert item, for example, could be:
# Strongly disagree
# Disagree
# Neither agree nor disagree
# Agree
# Strongly agree

Likert scaling is a bipolar [[Scale (social sciences)|scaling method]], measuring either positive or negative response to a statement. Sometimes an even-point scale is used, where the middle option of "neither agree nor disagree" is not available. This is sometimes called a "forced choice" method, since the neutral option is removed.<ref>{{cite magazine |last1=Allen |first1=Elaine |last2=Seaman |first2=Christopher |title= Likert Scales and Data Analyses|magazine=Quality Progress|year=2007|pages=64–65|url=http://asq.org/quality-progress/2007/07/statistics/likert-scales-and-data-analyses.html}}</ref> The neutral option can be seen as an easy option to take when a respondent is unsure, and so whether it is a true neutral option is questionable. A 1987 study found negligible differences between the use of "undecided" and "neutral" as the middle option in a five-point Likert scale.<ref>{{cite journal |last=Armstrong |first=Robert |year=1987|title=The midpoint on a Five-Point Likert-Type Scale |journal=Perceptual and Motor Skills |volume=64 |issue=2 |pages=359–362 |doi=10.2466/pms.1987.64.2.359|s2cid=145705789 }}</ref>

{{anchor|distortion}} Likert scales may be subject to distortion from several causes. Respondents may:
* Avoid using extreme response categories (''[[central tendency]] bias''), especially out of a desire to avoid being perceived as having extremist views (an instance of [[social desirability bias]]). This effect may appear early in a test due to an expectation that questions which the subject has stronger views on may follow, such that on earlier questions one "leaves room" for stronger responses later in the test. This expectation creates bias that is especially pernicious in that its effects are not uniform throughout the test and cannot be corrected for through simple across-the-board normalization;
* Agree with statements as presented (''[[acquiescence bias]]''), for example, agreeing with both Statement A, and its opposite. This effect especially strong among children, people with developmental disabilities, elderly people, and individuals who are subjected to a culture of [[institutionalization]] that encourages and incentivizes eagerness to please;
* Disagree with sentences as presented out of a defensive desire to avoid making erroneous statements and/or avoid negative consequences that respondents may fear will result from their answers being used against them, especially if misinterpreted and/or taken out of context;
* Provide answers that they believe will be evaluated as indicating strength or lack of weakness/dysfunction ("faking good");
* Provide answers that they believe will be evaluated as indicating weakness or presence of impairment/pathology ("faking bad");
* Try to portray themselves or their organization in a light that they believe the examiner or society to consider more favorable than their true beliefs (''[[social desirability bias]]'', the intersubjective version of objective "faking good" discussed above);
* Try to portray themselves or their organization in a light that they believe the examiner or society to consider less favorable/more unfavorable than their true beliefs (''norm defiance'', the intersubjective version of objective "faking bad" discussed above).

Designing a scale with balanced keying (an equal number of positive and negative statements and, especially, an equal number of positive and negative statements regarding each position or issue in question) can obviate the problem of acquiescence bias, since acquiescence on positively keyed items will balance acquiescence on negatively keyed items, but defensive, central tendency, and social desirability biases are somewhat more problematic.