AtlasSession 2L6: Hypothesis Testing and Statistical Testsp.3–8
What is a hypothesis
Fig. 1
Adults with partial-thickness burns: skin substitute + standard care vs standard care
Fig. 1, Who, what differs, what is measured (First look): 5 cards dealt; 0 sorted right, 5 still to sort. Bins: Who is studied 0; What differs 0; What is measured 0; Leave it out 0. Hypothesis: Who: … · What differs: … · What is measured: …
- Right so far
- 0 of 5
- Still to sort
- 5
- Crossed out
- nothing yet (no part has been tried where it fits in form only)
- Hypothesis
- Who: … · What differs: … · What is measured: …
Variable Sorter · First look
Level 1 · First look · Lecture 6 · p.3–8
Turn a hunch into a statement that an experiment could show to be wrong.
Turn a research question into a testable hypothesis that names the sample, predictor and outcome variable, and decide whether a hypothesis is needed at all.
Hunch vs testable claim (the step you are on)
'this fertiliser is good' vs 'tomato plants given it grow taller in 8 weeks'.
‘This fertiliser is good’ is a hunch; ‘tomato plants given it grow taller in 8 weeks’ could be shown wrong. ‘This dressing is good’ is a hunch too: leave it out. 5 chips to sort.