Purchase a pre- and post-assessment system: basic checklist

Measuring results is what turns a service from an impression into data: Did the user improve? Does the program work? But a system that memorizes the answers to a model is not an evaluation system — the assessment requires six items for each model, and the absence of any of them spoils the comparison.

Why is this requirement included in the purchasing decision?

Because the contracting authorities ask for impact reports, the administration asks about the feasibility of the programs, and the specialist needs a reference for his professional decision. All of these require properly structured evaluation from day one — data collected without proper structure cannot be fixed retrospectively.

What should the scope cover?

For each model, the system must save:

• Tool Purpose: Initial screening, outcome measurement, or periodic follow-up—the purpose determines the timing and use.
• Version: Which version of the model was used, changing a question means a new version, not a silent modification.
• Respondent: Client, guardian, specialist, or teacher — the answers are attributed to each of them, not combined.
• Timing: Before, at a stage of, or after the programme—linked to the episode of care rather than to an abstract history.
• Review: Did the specialist review the result? And when? Did he decide to share it with the client?
• Comparability: Show results over time for each respondent separately, indicating what is missing.

Checklist in demo

• Submit a form to two different respondents: Are the two answers saved separately and attributed?
• Leave unanswered questions: Did the deficiency appear or was the result calculated as complete?
• Submit the same form some time later: Are the two results displayed opposite each other?
• Review a result as a specialist: Was the review recorded? Is participation a separate decision?
• Open result with receiving account: Expected to be blocked.
• Request a filling ratio report: How many forms were sent and how many were filled?

Pay special attention to the missing answers: A form that was half filled out and then its grade was calculated as if it was complete gives a misleading number that enters your reports and remains there. Request that the system display the percentage of completion with each result, and prevent comparison between a complete response and an incomplete one without warning.

Limits that need to be adopted

• The system does not explain: Displays the score and its course but does not classify it, suggest a diagnosis, or recommend a plan. Professional reading for the specialist alone.
• What is shown to the client: Adjusts for each model — some results are not displayed until checked by a specialist.
• Right to use standards: Protected published tools require a license from their publisher, and this is an obligation for your facility that the system does not provide.
• Failure to enforce tools: Choosing metrics is a professional decision for your team, and the system accommodates what you choose.

The next step is to document the requirements

Collect the forms you already use, define each of the six elements, and submit them with your requirements. See Evaluation page and follow-up of results.