MYP Design Evaluating the Relevance of Research Notes

Research Only Counts When It Changes Something

1. By this point you have interview notes, a questionnaire, two tally charts, three product analyses and a folder of saved web pages.
2. Most of it will not survive into the final folder, and that is the intended outcome.
3. A finding earns its place only if it changes a decision, sets a constraint or proves the need is real.
4. Criterion A asks you to prioritize research, and prioritizing means ranking it rather than listing it.
5. A folder containing everything you gathered reads as collecting, while a folder containing the useful six reads as inquiry.

The Relevance Test: Does It Touch the Problem Sentence?

1. Write your one line problem statement at the top of a blank page and hold every finding up against it.
2. Ask three things of each finding: does it describe my client, does it change what I will make, and can I act on it this term.
3. A finding that passes all three goes into the folder with a sentence saying what it changes.
4. A finding that fails all three goes into an appendix or out of the project altogether.
5. Findings that fail only the third test, such as a vacuum forming process your school cannot run, still deserve one line explaining why you ruled them out.
6. Rank whatever survives, because the three strongest findings should be visible on the first page of your analysis.

Bias Hides in Who You Asked and Who Wrote It

1. Bias is anything that tilts an answer before the question has even been asked.
2. Sampling bias comes from asking the wrong people, since six friends in your own form are not a sample of the school.
3. Question bias comes from wording, as in "how much easier would a proper rack make your lesson?".
4. Source bias comes from who published it, and a company page about its own material will never list that material's weaknesses.
5. Confirmation bias is your own, and it shows up as quietly skipping the two responses that contradicted your favourite idea.
6. You cannot delete bias, so name it instead: write "12 responses, all Year 8, all from one class" and let the reader weigh it themselves.

Data Ages at Very Different Speeds

1. Out of date information is still information, and the skill is knowing how fast each kind goes stale.
2. The density of birch plywood has not moved in decades, while the price of a sheet and the current version of a safety standard both move quickly.
3. Check the publication date on every figure and write that date beside the figure in your folder.
4. Where a source is old but nothing better exists, keep it and record the age as a stated limitation.
5. Repeat your own primary research if the situation shifted, since a timetable change can make last term's observation meaningless.

Cutting Research Is Part of Doing Research

1. Every page you keep is a page somebody has to read, so a shorter folder of sharper findings reads far better.
2. Cut anything you copied in but never used, including the properties page for a material you decided against.
3. Cut general background about the industry, because it tells the reader about the world rather than about your client.
4. Cut duplicated findings and keep whichever one has the stronger evidence behind it.
5. Move raw questionnaires and tally sheets into an appendix so the main body holds only analysis.
6. Cutting should be reversible, so move things to a separate file rather than deleting them, since a finding can become relevant again in week five.

Analysis Is the Sentence Between Finding and Decision

1. Analysis is the writing that connects a piece of evidence to a choice you made.
2. Use a three part move every time: the finding, what it means, and what it changes.
3. "18 of 24 students said they carry wet work with both hands. That means the solution cannot need a spare hand to open it. So the specification will require a single motion or hands free opening."
4. Do this properly for your top five or six findings instead of thinly for all thirty.
5. Where two findings conflict, say so openly and explain which one you followed and on what grounds.
6. Group findings by theme rather than by method, so everything about size sits together even though it came from an interview, a tape measure and a data sheet.
7. Finish with a short paragraph drawing the themes into the shape of the design brief you are about to write.

What a Strong Research Analysis Looks Like

1. It opens with the problem sentence, so a reader knows what everything after it is being measured against.
2. It presents five or six findings, each with its evidence, its meaning and its consequence.
3. It states the limitations honestly, naming sample sizes and the age of any source.
4. It finishes pointing directly at the design brief, which is the next thing in the folder.
5. It is considerably shorter than the pile of research behind it, and every line in it is doing a job.