Adult Category Norms

In 2018 and 2019, the Adult Cognition collected norming data for common categories using responses collected via Amazon’s Mechanical Turk (MTurk) program. Participants ages 18+ were recruited to participate in the study and compensated for their time.’

Categories were randomly presented to each subject. Categories lists were pseudo-randomly organized to 1) avoid similar or related categories within a list (e.g., football related categories are not within the same list) and 2) to ensure that each list contains approximately the same number of exemplars to be rated across all categories (i.e., same amount of time expected to complete each list).

Below are the analyses resulting from the norming data. They are all available for download in a comma-delimited (CSV) format.

TYPICALITY ESTIMATES

CATEGORY POTENCY

  • Potency is the proportion of participants that provided each response. We only list those responses that were given by at least 5% of the corresponding sample (overall and by specific age group).
  • Potency data (comma-delimited format)

RESPONSE VARIABILITY

  • We report both Total Types/Total Tokens and Idiosyncratic Types/Total Types metrics.
    • Total Types/Total Tokens: The total number of separate exemplars (“types”) divided by the total number of responses (“tokens”) given for a category.
    • Idiosyncratic Types/Total Types: The number of exemplars given by only one participant (“idiosyncratic type”) divided by the total number of separate exemplars (“types”) given for a category.
  • Type/token data (comma-delimited format)

FREQUENCY CORRELATIONS

RANK-ORDER CORRELATIONS

 


For more questions or more information about the data, you can email us at hertzoglab@gmail.com.