Couples Satisfaction Index (CSI-4)
IRT-Optimized Ultra-Short Form
Happy in your relationship? 4 quick questions that capture the most telling signs of relationship satisfaction. IRT-optimized ultra-short form based on Funk & Rogge (2007), selecting the 4 most informative items from the original 32.
What four questions can tell you
More than you'd expect, and less than you'd hope. CSI-4 is the ultra-short form pulled from the 32-item CSI using item response theory, keeping the four items that carried the most information. One asks about overall happiness; three cover consistency, rewards and satisfaction in the relationship. What you get back is a rough reading, fine for a quick check-in or for tracking whether things are moving.
Don't mistake it for a replacement for the 16-item version. If you want a real read on where the relationship stands, the standard form carries far more information. The value of CSI-4 is that it's short enough to fill in repeatedly, and satisfaction is exactly the kind of thing that moves. Trends beat single readings.
How it's scored
Item 1 uses a 7-point scale; items 2-4 use 6-point scales. This assessment scores from 1 per item, so totals run 4-25. The paper's clinical cutoff was 13.5 on a 0-21 scale, which corresponds to 17.5 here. Below that suggests relationship distress.
Internal consistency came out at α = .94 in the original paper, .954 with ω = .956 in an independent replication (Quinn-Nilas, 2022, N = 640), and .89-.91 in the German version. Factor loadings run .82-.97, and the single-factor structure holds up across samples.
One detail worth knowing: Ogan and Kanter (2024) found income-related differential item functioning, with lower-income respondents averaging 1.38 points lower. When a score lands right on a boundary, that's not the moment for a verdict.
About the scale
- Funk, J. L., & Rogge, R. D. (2007). Testing the ruler with item response theory: Increasing precision of measurement for relationship satisfaction with the Couples Satisfaction Index. Journal of Family Psychology, 21(4), 572-583.
- Quinn-Nilas, C. (2022). Independent replication in midlife married adults: α = .954, ω = .956
A note
This is a self-assessment, not a diagnosis. On a four-item scale a single point carries real weight, so read the number as a direction rather than a ruling.
Result Interpretation
After completing the 4 questions, you'll receive an immediate, detailed report with:
- Your score — calculated automatically based on your responses
- Score interpretation — what your score means in practical terms
- Context — how your results compare to general population norms where available
All results are displayed on screen. No account or login needed.