All frameworks and cheat sheets

Comparison

Jingle vs jangle: two naming fallacies in measurement

How metric names hide one construct, or merge two

Comparison of the jingle fallacy, two different things sharing a name, and the jangle fallacy, one thing with two names, with a product example of each.
Three dashboards called trust, satisfaction and loyalty may be one construct with three names
Download PNGRead the full guide

A side-by-side comparison of the two naming fallacies that break construct validity. The jingle fallacy is assuming two different measures are the same because they share a name, like two teams that both say activation. The jangle fallacy is assuming one thing is two because it has two names, as with grit and conscientiousness, which a meta-analysis found correlated at .84. For product, research and analytics teams who name metrics.

Where this appears

What's in this comparison

Jingle fallacy

  • Two different things share one name
  • People assume they are the same thing
  • Example: a growth team's activation event vs a research team's felt value

Jangle fallacy

  • One thing goes by two names
  • People assume there are two different things
  • Example: grit and conscientiousness, which correlate at .84

Use this image

Free to use in your docs, decks and posts under CC BY 4.0, as long as you credit Koji with a link to this page. To embed it, paste this into your page:

html
<a href="https://www.koji.so/visuals/jingle-vs-jangle-fallacy-comparison"><img src="https://www.koji.so/visuals/jingle-vs-jangle-fallacy-comparison/jingle-vs-jangle-fallacy-comparison.png" alt="Comparison of the jingle fallacy, two different things sharing a name, and the jangle fallacy, one thing with two names, with a product example of each." width="1200" height="630" /></a>
<p>Source: <a href="https://www.koji.so/visuals/jingle-vs-jangle-fallacy-comparison">Jingle vs jangle: two naming fallacies in measurement</a>, Koji</p>

Related

Table of six kinds of construct validity evidence, from face validity to response process, with the question each answers and a cheap way to test it.

Six kinds of validity evidence, and a cheap test for each

Table of four metascience findings on measurement, including 40 to 93% of measures lacking validity evidence and 79% of Many Labs 2 scales looking ad hoc.

How often research measures lack validity evidence

Comparison of a digital shelf scrape and a real shopper across login, location, timing, search phrasing, how much of the grid they see and intent.

What a digital shelf scrape sees vs what the shopper sees