All frameworks and cheat sheets

Reference table

Nunnally's reliability thresholds, by how the score is used

Table of Nunnally's reliability thresholds: 0.70 for early-stage work, 0.80 for basic research, 0.90 minimum and 0.95 desirable when decisions rest on scores.
The 0.70 everyone quotes was meant for pilot work. Scores that drive decisions need 0.90. Source: Jum Nunnally, Psychometric Theory, 2nd edition, 1978
Download PNGRead the full guide

What Jum Nunnally actually recommended for reliability in Psychometric Theory (1978): 0.70 is enough while you develop an instrument, 0.80 for basic research comparing group means, and 0.90 as the minimum, with 0.95 desirable, when important decisions rest on individual scores. Use it to pick the right bar for a Cronbach's alpha or omega before a composite score goes on a dashboard.

Where this appears

What's in this reference table

How the score is usedReliability needed
Early-stage instrument development0.70 or higher will do
Basic research comparing group means0.80
Important decisions about specific scores0.90 at minimum
Important decisions, desirable standard0.95

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/nunnally-reliability-thresholds-by-use"><img src="https://www.koji.so/visuals/nunnally-reliability-thresholds-by-use/nunnally-reliability-thresholds-by-use.png" alt="Table of Nunnally's reliability thresholds: 0.70 for early-stage work, 0.80 for basic research, 0.90 minimum and 0.95 desirable when decisions rest on scores." width="1200" height="630" /></a>
<p>Source: <a href="https://www.koji.so/visuals/nunnally-reliability-thresholds-by-use">Nunnally's reliability thresholds, by how the score is used</a>, Koji</p>

Related

Checklist for auditing a composite score: name one construct, list the items, check correlations, run a factor analysis, set the bar, look for ceilings.

Audit a composite survey score in five minutes

Table matching six questions about a survey scale to the right tool, such as omega, factor analysis or invariance testing, with alpha as the wrong tool.

What Cronbach's alpha can't tell you, and what can

Table of four usability test types, moderated, unmoderated, remote and in-person, with the best scenario and main trade-off for each.

Usability test types: when to use each