Build an Evidence Matrix That Preserves Disagreements
An evidence matrix should make studies easier to compare without making their differences disappear. For a nursing assignment, organise it around the claim you need to support, not around how many papers you have collected.
The worked examples are teaching aids. Check them against your own source material and the requirements of your specific task.
Start with the task
Cochrane's data-collection chapter provides a methodological reference for structured extraction.
Use columns that preserve meaning
A practical coursework matrix can include study identity, design, population and setting, intervention and comparator, outcome and time point, relevant finding, limitation, and the care-plan claim it may support.
Keep the exact evidence location in a final column. A page, table, or section reference saves you from searching again when revising.
Do not compress every result into “positive” or “negative.” That removes magnitude, uncertainty, and outcome differences.
Try a fictional comparison
Suppose one paper measures immediate recall and another explores participants' experiences of written information. Their rows should retain those different contributions.
The first may inform a narrow effectiveness question. The second may help explain barriers to understanding. A shared topic does not make the results suitable for numerical pooling or vote counting.
The following is an entirely fictional extraction example, with no empirical findings claimed:
For real sources, add the verified citation and exact page or table location to each row; replace every hypothetical entry with what the paper actually reports.
If two reports concern the same study, link them under one study identity rather than count them as independent evidence.
Try a fictional comparison — comparison
Comparison row 1. Record: Teaching example A; Design and context: Trial of supported information in a clinic; Outcome and time: Recall immediately after teaching; Finding to preserve: Hypothetical report describes a between-group difference; actual extraction would retain its estimate and interval; Limitation: No later outcome supplied; Rationale role: A bounded claim about immediate recall
Comparison row 2. Record: Teaching example B; Design and context: Interviews about written discharge information; Outcome and time: Experiences discussed after discharge; Finding to preserve: Hypothetical theme concerns uncertainty about instructions; Limitation: Does not estimate prevalence or intervention effect; Rationale role: Explain an experiential issue worth considering
Write the synthesis after extraction
Read across the matrix and identify what is comparable, what conflicts, and what remains unanswered. Then draft the rationale with those distinctions intact.
A matrix is not a meta-analysis, and completing every cell does not establish that a source is trustworthy. Mark unavailable information rather than inventing it.
Use the evidence-chain guide to transfer only the relevant claim-source connection into Nursevi. Keep the fuller matrix as your working evidence record, following the assignment's requirements.