Section A Question 1

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/9

encourage image

There's no tags or description

Looks like no tags are added yet.

10 Terms

1
New cards

1. EXPERIMENTAL DESIGN — Randomisation

Define: Randomly allocate treatments to units.
Explain: Reduces systematic allocation bias.
Apply: Randomly assign tomato plants to fertilisers.
Connect: Works with replication to make treatment comparisons fairer.

2
New cards

1. EXPERIMENTAL DESIGN — Replication

Define: Repeat treatments across independent units.
Explain: Accounts for natural variation; improves reliability.
Apply: Multiple plants per fertiliser type.
Connect: Works with randomisation; independent units prevent pseudoreplication.

3
New cards

1. EXPERIMENTAL DESIGN — Blocking

Define: Group similar units; test all treatments per block.
Explain: Accounts for group differences.
Apply: Each greenhouse receives all fertilisers.
Connect: Works with randomisation to reduce confounding from greenhouse differences.

4
New cards

1. EXPERIMENTAL DESIGN — Control

Define: Baseline group or consistent conditions.
Explain: Helps isolate treatment effects.
Apply: Compare fertilised plants with no-fertiliser plants.
Connect: Helps control confounding when comparing the independent variable's effect on the dependent variable.

5
New cards

1. EXPERIMENTAL DESIGN — Pseudoreplication

Define: Treating non-independent measurements as replicates.
Explain: Inflates apparent sample size.
Apply: 20 plants in one treated pot ≠ 20 replicates.
Connect: Violates independent replication and can undermine statistical assumptions.

6
New cards

2. VARIABLES & STATISTICAL MODELLING — Independent and Dependent Variables

Define: IV (X) = predictor; DV (Y) = measured response.
Explain: Establishes the relationship investigated.
Apply: Fertiliser type (IV); plant growth (DV).
Connect: Their data types help determine the statistical model used.

7
New cards

2. VARIABLES & STATISTICAL MODELLING — Confounding Variables

Define: Factors linked to predictor and response.
Explain: Can distort the apparent treatment effect.
Apply: More sunlight for fertilised plants may increase growth.
Connect: Randomisation, blocking and control help reduce confounding.

8
New cards

2. VARIABLES & STATISTICAL MODELLING — Fixed and Random Effects

Define: Fixed = specific groups compared; random = sampled group variation.
Explain: Accounts for different sources of variation.
Apply: Fertiliser = fixed; greenhouse = random.
Connect: Blocking identifies group differences that can be included as random effects.

9
New cards

2. VARIABLES & STATISTICAL MODELLING — Types of Data

Define: Continuous = numerical; categorical = groups.
Explain: Data type influences statistical analysis.
Apply: Growth (cm) = continuous; fertiliser type = categorical.
Connect: IV and DV data types help determine the appropriate statistical model.

10
New cards

3. STATISTICAL ASSUMPTIONS — L.I.N.E.

Define: Assumptions checked after fitting linear regression. (L.I.N.E)
- Linearity: Does the model describe the shape of the relationship?
- Independence: Are the observations independent of each other?
- Normality: Are the model's prediction errors approximately bell-shaped?
- Equal variance: Is the spread of prediction errors consistent?
Explain: Assess model suitability and reliability.
Apply: Check plant growth against fertiliser levels.
Connect: Independence links to replication; pseudoreplication can violate it.