An Imbalance in the Scales

A psychology survey had unusual scales for conspiratorial beliefs.

A recent paper involved a survey of English adults about conspiracy theories. The paper in Psychological Medicine also studied self-reported adherence to government guidelines. As an example, the paper claimed:

around one-fifth endorsed to some degree that ‘Jews have created the virus to collapse the economy for financial gain’.

This article looks at the unusual scale used to measure conspiratorial beliefs.

Four for one

For this paper, Lucid collected 2,501 responses from English adults between 4th and 11th May 2020. The survey was online. It was a non-probability sample of English adults.

Questions asked about agreement with conspiratorial statements, such as:

Bill Gates created the virus in order to reduce the world population.

Respondents had five possible response options:

  • Do not agree;
  • Agree a little;
  • Agree moderately;
  • Agree a lot;
  • Agree completely.

This is an unusual set of options. There are four options to agree, and only one to disagree. There is no means to express having a neutral or no opinion.

Image for post
Image for post
There is spurious accuracy too. (Image: Oxford University)

Likert scales and balance

A Likert scale is a scale to which respondents can rate their agreement or satisfaction. For agreement, the standard five-point Likert scale ranges from:

  1. Strongly disagree;
  2. Disagree;
  3. Neither agree nor disagree;
  4. Agree;
  5. Strongly agree.

American psychologist Rensis Likert created this type of closed responses. The purpose was to measure attitudes. The possible responses should balance between agreement and disagreement.

Image for post
Image for post
‘How satisfied are you with…?’ (Image: WP Forms)

Survey researchers disagree on whether such scales should include a middle option. People choose middle options for various reasons.

People may be getting through the survey. They could be showing ambivalence or social desirability. That option may not mean a person has a genuine neutral opinion on an issue.

Acquiescence bias

If the purpose was to measure one belief, agree-disagree questions may not be the best format. There is acquiescence bias.

This is where people agree to a statement, no matter what it says. Some people are agreeable, or want to complete the survey. Such questions are easy to write, but may not give a true reflection of public attitudes.

Courtney Kennedy gives a short guide to question wording. (Video: Pew Research Center)

Imbalanced options and acquiescence bias means likely overestimation of agreement. This survey may overestimate how many English adults hold conspiratorial beliefs.

Question wording and response options matter.

Researchers should give balanced response options. Press reports about single questions should refer to the survey’s imbalanced options.

This blog looks at the use of statistics in Britain and beyond. It is written by RSS Statistical Ambassador and Chartered Statistician @anthonybmasters.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store