Relationship Test: What Is It Really For, and Which Ones Are Worth Taking?

What is a relationship test for?

When you ask yourself “how are we really doing?”, you almost always answer in one lump: it’s fine, or it isn’t. The problem is that a relationship is never good or bad all at once. It is doing well in some areas and struggling in others, often without the two of you using the same words for it. One person says “we never talk anymore,” the other says “we fight too much,” and they may be describing the same thing, or something completely different.

A serious test is first of all for this: separating the dimensions. Communication, how you get through disagreements, attachment, affection, intimacy, trust, a shared project. When you see that communication is going well but arguments are costing you a lot, or that trust is solid but desire has gone quiet, you know where to look. It stops being “our relationship has a problem” and becomes “there is one specific area to work on.” That is much less frightening, and much more useful.

It also gives you a starting point. A score is not a verdict, it is a dated snapshot. It helps you choose where to begin, and later it lets you see whether anything has moved. And if your partner takes it too, on their own, you often discover that you don’t have the same picture of the same relationship. That difference in perspective is valuable information in itself, sometimes the most valuable part of the test.

What separates a serious test from a quiz?

Online, “relationship test” covers very different things: a compatibility calculator that runs on two first names, a ten-question magazine quiz, a personality test that promises to tell you whether you’re “meant for each other,” or a questionnaire built on psychology research about couples. Five criteria help you sort them out.

1. It relies on identifiable research tools. A serious test tells you where its questions come from: John Gottman’s work on what predicts a couple’s stability, the ECR-R scale by Fraley, Waller, and Brennan for attachment style, the desire models of Basson or Nagoski. If it cites nothing, be wary.
2. It measures several dimensions separately. A single “compatibility” score tells you nothing you can act on. A useful test gives you one score per dimension and explains what each one covers.
3. It is long enough, or adaptive. Ten questions can’t cover a relationship. An adaptive test, whose questions change based on your answers, can stay short while digging where it matters.
4. It is confidential. Your answers about your relationship are among the most intimate data there is. A serious test doesn’t require sign-up to start and says plainly what it does with your answers.
5. It states its limits. It makes clear that it measures how you see the relationship, not the relationship itself, that it doesn’t predict your future together, and that it doesn’t replace a professional. A test that promises to know whether “it will last” is lying.

What can a relationship test measure, and what can’t it?

Any test you take on your own relies on what you report. So it measures your perception of the relationship. You might think that is a weakness; the research suggests the opposite. In 2020, Samantha Joel and her colleagues analyzed 43 longitudinal studies covering 11,196 couples, using machine learning methods, to identify what best predicts relationship quality. The finding: what a person perceives of their partner’s commitment, appreciation, and satisfaction matters more than their own personality traits. Perception isn’t background noise, it is the signal.

What a test can’t do: predict a breakup, make a clinical diagnosis, or tell you who is right. It also can’t see what you can’t see yourself: as early as 1977, Richard Nisbett and Timothy Wilson showed that we are fairly poor at explaining our own preferences and reactions. A good test works around this bias by asking about concrete situations rather than general opinions, but it remains a snapshot taken from the inside.

Should you take the test together?

No, and it is often better to start on your own. A test filled out together, side by side, quickly turns into a negotiation: you answer what the other person can handle hearing. On your own, you answer what you actually live. Afterward, if your partner takes it separately, comparing your two results is a conversation in its own right, and a good chance to apply a simple rule: talk about what you felt when you saw your score, not about what the other person “should have” answered.

How is the MyLoveLab test built?

The MyLoveLab test is adaptive: about fifty questions, chosen as you answer, across seven dimensions. It measures your attachment style with questions inspired by the ECR-R scale, along the two axes that research uses, fear of abandonment (anxiety) and avoidance of closeness. It then combines that style with your dominant relationship pattern to give you one of 24 profiles. And it is upfront about its limits: the score hasn’t yet been statistically validated on a large sample, so that is a goal, not something already achieved. Everything is detailed on the method page.

What should you do with the results?

First, don’t read them as a verdict. A low score in one dimension doesn’t mean you don’t love each other or that the relationship is in danger: it means that is probably the first area to work on. Next, pick just one area. Trying to improve everything at once is the best way to change nothing. Finally, act small: couples who do well don’t do extraordinary things, they do specific, regular things that can be learned. That is exactly what the MyLoveLab program offers after the test, dimension by dimension, one short session a day.

And if the result shows you something hard, that is exactly where it helps. The goal isn’t to judge your relationship, it is to give you clear angles and concrete actions. You decide what to do with it, at your own pace. If your relationship is going through a serious crisis, or if you feel unsafe, a professional is still the right door to knock on, and a test changes nothing about that.

Frequently asked questions

References

1. Joel, S. et al. (2020). Machine learning uncovers the most robust self-report predictors of relationship quality across 43 longitudinal couples studies. PNAS, 117(32), 19061-19071.
2. Fraley, R. C., Waller, N. G. & Brennan, K. A. (2000). An item response theory analysis of self-report measures of adult attachment. Journal of Personality and Social Psychology, 78(2), 350-365.
3. Gottman, J. M. & Silver, N. (1999). The Seven Principles for Making Marriage Work. Crown.
4. Nisbett, R. E. & Wilson, T. D. (1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84(3), 231-259.