RM2 Basics of Experimental Design

What is a fact

A statement about a direct observation of nature, consistently repeated with no doubt about its truth.

What is a theory

A collection of propositions and hypotheses attempting to explain a set of observed phenomena.

What is a hypothesis

A clear but tentative explanation for an observed phenomenon.

what 4 things do theories do

• define

• explain

• organise

• interrelate

purpose of theories for hypotheses

makes general predictions which specific hypotheses can be based

What 5 things must hypotheses be

Falsifiable, testable, precisely stated, rational, and parsimonious.

what does it mean if a hypothesis is precisely stated

all terms are clearly defined

what does it mean if a hypothesis is rational

it is consistent with known information

what does it mean if a hypothesis is parsimonious

it is the simplest explanation

What is a construct

building blocks of theories

Theoretical concepts formulated to explain phenomena, not directly measurable.

What is a variable

A characteristic that assumes multiple values, observed or measured, and operationally defined.

Difference between constructs and variables

Constructs are theoretical definitions; variables are operational definitions.

4 scales of measurement

• nominal

• ordinal

• interval

• ratio

What are nominal scales

Scales based on category membership, with no numerical relationship (e.g., gender).

What are ordinal scales

Scales where data can be ranked, but intervals between ranks are not equal.

What is interval data

Data with equal intervals between values but no true zero point (e.g., temperature).

What is ratio data

Data with equal intervals and a true zero point (e.g., height).

define experimental method as a research design

active manipulation of variables which allows us to make causal inferences

What is an independent variable (IV)

The variable manipulated to cause changes in the dependent variable.

what must an IV have

at least 2 levels

What is a dependent variable (DV)

The outcome variable measured and compared under different IV levels.

what is subjects design

the assignment of participants to experimental conditions (levels of the IV)

3 types of subjects design

• between-subjects/independent groups

• within-subjects/repeated measures

• mixed-designs

What is a between-subjects design

A design where participants are exposed to only one level of the IV.

What is a within-subjects design

participants are exposed to all levels of the IV.

What are mixed designs

Experimental designs that contain both within-subjects and between-subjects IVs.

what is are 2 considerations for between-subjects design in terms of the participants

greater number of participants are required

participant variables - individual differences

how are effect of participant variables minimised in between-subjects design

random allocation - to spread the influence of participant variables across the different levels of the IV

what is one consideration with within-subjects design

order effects

method to address order effects in within-subjects design and why?

counterbalancing - to evenly distribute order effects across IV levels.

What is a factorial design

An experimental design with two or more IVs, examining their effects independently and in combination.

2 considerations when deciding what experimental design to choose

• concerns (order effects or individual differences)

• no. of participants (availability)

What are quasi-experimental designs

Designs where participant assignment to IV levels is pre-determined, not random.

why is it an issue if random allocation cannot occur in quasi-experimental designs?

likely to be differences between groups other than the variable of interest which means inferring causality must be done with caution

2 possible solutions to X random allocation in quasi-experimetnal designs

• matching

• matched pairs

What is matching in quasi-experiments

Controlling for differences by matching participants on moderating variables.

solution if counterbalancing cannot be used in a within-subjects design

pre-test post-test control group design

What is a pre-test post-test control group design (mixed design)

A design dividing participants into treatment and control groups, where only one group experiences manipulated IV, to account for order effects.

What is Occam's Razor

The simplest explanation is preferred.

What is the hypothetico-deductive method

Observation/intuition → theory → hypotheses → Testing → Results.