Variables, Reliability, Validity

What is an extraneous/nuisance variable?

All other factors that may vary in an experimental setting/between participants, and may affect the DV.

E.g. noise, temperature, mood, tiredness

What is a confounding variable?

Varies systematically with the IV & confounds the results

When an extraneous variable hasn’t been properly controlled and interferes with the DV(results) = confounding variable

E.g. order effect, demand characteristics, time of day

Give 2 examples of extraneous variables and how they may be overcome

RESEARCHER BIAS: (interprets behaviour in a way to support hypothesis)

• Double-blind technique: neither the person conducting the research & participant know the study aim

NOISE DISTRACTION:

• Undertake experiment in an area where no noise distractions can occur.

Give 2 examples of confounding variables and how they may be overcome

SOCIAL DISABILITY BIAS: (participants change behaviour to prevent themselves more positively)

• Reinforce the need for honesty

PARTICIPANT VARIABLES: (participants may have certain characteristics that enhance their ability, like IQ)

• Matched pairs design or repeated measures design

Define validity

How true or legitimate something is; does it represent reality?

What is internal validity?

About control & realism- what happens inside the study

High internal validity can be gained through carefully designed research and control extraneous/confounding variables

• Did ONLY the IV change the DV?

• Did the experiment test what it intended to test?

• Mundane realism achieved?

What is external validity?

How generalisable the experiment is. External validity can be affected by internal validity.

• ECOLOGICAL VALIDITY: where

• POPULATION VALIDITY: who

• HISTORICAL VALIDITY: when

What is inter-rater reliability, and its issues?

We’d expect 2 observers watching the same behaviour to produce to same data.

Measured by correlating the observations of 2 or more researchers

^ 80%+ agreement=data has inter-rater reliability

• Observers should be trained in the use of a coding system/ behaviour checklist

What is internal reliability, and how to assess it?

A measure of the extent to which something is consistent within itself

E.g. all questions on an IQ test (questionnaire) should be measuring the same thing

May not always be relevant if you’re looking at experiences that differ between people

• SPLIT-HALF METHOD: a group of participants are given a test, of which the questions are divided in half, and the answers are compared by calculating the correlation coefficient. (Scores on both halves should be similar.)

What is external reliability, and how to assess it?

A measure of consistency over several occasions.

• TEST-RETEST METHOD: a test is given to the same participant on 2 seperate occasions, with enough time inbetween so they forget, and answers can be compared by calculating a correlation coefficient.

How do you overcome issues in internal and external reliability?

INTERNAL:

• Low internal reliability questions can be removed to see if the split-half test returns a high reliability score.

EXTERNAL:

• Poorly written questions may cause confusion & need to be rewritten

• If an interview has low reliability, the interviewer may need to be retrained so it’s more consistent.