Put care into arrangements you can actually keep
Review capacity before agreeing, updates when plans change, and practical follow-through. Distinguish a missed arrangement from resource limits and a repeated need for chasing.
Capacity checks and timely updates keep their 1–5 values; overpromising and leaving updates to be chased use 6 minus the response. All four frequency responses give a sum of 4–20 and a mean shown to one decimal. Any skip leaves this group insufficient. The eight groups are not combined or ranked.
What these four items observe
- This group concerns what happens around ordinary agreements: checking your capacity before saying yes, giving a useful update when a plan becomes impractical, and avoiding a situation in which the other person must repeatedly chase you. It describes actions you recall in one chosen relationship during the past four weeks. It does not require perfect performance or unlimited availability. A clear, timely refusal can be dependable: it gives both people information they can use.
- Circumstances matter. Unexpected caring responsibilities, illness, changes at work or disrupted travel can affect a plan. These answers cannot identify the cause of each change or show who cares more. For a useful reading, return to one small event: what was agreed, what became known later, and how the arrangement was revised. Those details are more informative than a general impression that you are always reliable or always disappointing.
Reconstruct the mean from four responses
- In the order capacity check, update, overpromising and leaving an update to be chased, responses of 5, 4, 2 and 1 become 5, 4, 4 and 5 after reversing the last two items. Their sum is 18, so the displayed mean is 18 ÷ 4 = 4.5. Reversal is not a punishment. It aligns statements about overpromising or leaving someone to chase an update with statements about checking and following through. The resulting mean is a summary of coded responses, not a population percentile, success probability or moral grade.
- If any of the four items is marked not experienced or cannot judge, this group remains insufficient. A skipped item is not replaced by 3, and the remaining answers are not rescaled into a complete result. Other fully answered groups remain readable. A mean such as 13 ÷ 4 = 3.25 displays as 3.3. That rounding does not establish that a difference of 0.1 between two people is meaningful. Do not use the highest of the eight means to assign a personality type.
Recognize the pattern in everyday arrangements
- Responses nearer 5 might fit an occasion when you checked your diary before agreeing to shop together, then gave an early update about unexpected work and asked whether another day would work. Responses nearer 1 might fit agreeing mainly to end a conversation, then providing no update until the other person asked. A middle mean may reflect a mixture of those actions, or a split between particular items. Read the individual statements before deciding what the number describes.
- Counterexamples matter. Someone may make few commitments and handle each clearly. Another person may need frequent changes because resources are tight, while still communicating promptly and negotiating respectfully. Neither should be called unreliable merely because fewer plans were completed. Conversely, arriving on time does not demonstrate respect for boundaries in other situations. This group cannot replace separate attention to autonomy, responsiveness or safety.
Distinguish dependability from constant availability
- Declining every new arrangement to protect a score would not automatically improve a relationship. Neither would sacrificing rest, money or important responsibilities to appear dependable. Sustainable commitments require both people to know what is settled and what still needs checking. It can help to distinguish wanting to help from having made a definite agreement. Good intentions are easier to act on when they are not mistaken for guarantees.
- If every reasonable change is treated as a betrayal, trying harder to keep plans cannot address the whole problem. Where it is safe and mutually wanted, discuss how updates should be given, when to reschedule, and which emergencies are difficult to predict. This result does not ask you to prove loyalty by showing messages, movements or a private calendar. Dependability is not a reason to surrender control over personal information.
Try one low-stakes agreement
- One optional example is: “I can help for half an hour this week, but I cannot take on the whole task. I will confirm a time tomorrow evening and let you know early if it will not work.” The other person can accept, decline or suggest another arrangement. Notice whether the task, limit and next update are clear, rather than whether the sentence produces instant approval. Adapt the wording to your circumstances; it is an example, not a prescribed script.
- For a private observation, note one ordinary agreement, what you knew at the time, and whether an update was timely. Do not record the other person’s secrets or turn the note into a case against them. On a later review, ask whether the arrangement became easier to understand and which limitation could have been mentioned earlier. Keeping a record and retaking the questionnaire are optional. You may simply finish here.
What this reading cannot guarantee
- This is FermatMind original reflection content, not a translation of the IDRlabs bank or a validated psychological scale. Four items cannot cover every meaning of dependability, and recent opportunities, memory and self-presentation can affect responses. A higher mean does not certify a good partner. A lower mean does not show that a person is toxic, unlovable or headed for a breakup. The eight readings coexist without becoming an overall relationship-health verdict.
- If an interaction includes threats, coercion, stalking or violence, do not reduce it to a scheduling problem or a shared lack of communication skills. You do not need to start a potentially unsafe conversation in order to complete an exercise. Prioritize safety and trusted support. You may read the other boundary and repair guides or consult public relationship resources. Those resources offer general background; they do not validate this questionnaire’s scoring.