week 2: advanced issues in experimental research methods

Question 1

What is a hypothesis?

Answer 1

a specific, testable claim or prediction about what you expect to observe given a set of circumstances
- statement about the assumed relationship between two or more variables

Question 2

What is the research hypothesis?

Answer 2

the statement you are testing (what you expect to find)
- can be directional (one-tailed)
- or non directional (two-tailed)

Question 3

What is a two-tailed hypothesis

Answer 3

non-directional

Question 4

what is a one-tailed hypothesis?

Answer 4

directional

Question 5

what is the null hypothesis?

Answer 5

provides a baseline against which to evaluate our alternative hypothesis (states an effect is absent)

Question 6

What does correlation mean?

Answer 6

means that two ariables vary together - as one changes, so does the other

Question 7

What is an extraneous variable?

Answer 7

anything other then the independent variable that could affect the dependent variable

Question 8

what is a confounding variable?

Answer 8

a type of extraneous variable that not only effects the dependent variable, but also varies with the independent variable in a systematic manner

Question 9

what is a between-subjects design?

Answer 9

multiple test groups, where each participant is tested in only one condition
- looks at differences in results between groups

Question 10

what is a within-subjects design?

Answer 10

each participant is tested in all conditions
- look at differences in performance in levels of test

Question 11

what are the advantages and disadvantages of between-subjects designs?

Answer 11

• useful when impossible for an individual to participate in all conditions. no carry-over effects
• noise from random individual differences between groups may cause less statistical sensitivity
• greater expense (more participants, time, money)

Question 12

What are advantages and disadvantages of within-subjects designs?

Answer 12

• economical (less participants, time, money)
• less noise (from differences between groups) causes better statistical sensitivity
• need strategies to avoid carry-over effects
• not suitable for cases where participant must be naive for each condition

Question 13

What is effect size?

Answer 13

• a statistical measure of the magnitude of an observed effect in a population
• how big the difference between two experimental groups is, and how strong a correlation is
• to detect a smaller effect size in a population, a larger sample size Is needed

Question 14

What is type II error?

Answer 14

failing to detect an effect that actually exists (we incorrectly fail to reject the null hypothesis, and reject the experimental hypothesis

Question 15

what is statistical power?

Answer 15

the probability of detecting a true effect when it actually exists in your population

Question 16

how is power calculated (equation)?

Answer 16

power = 1 - B
- B = the probability of making Type 2 error

Question 17

What is validity?

Answer 17

whether the test/questionnaire measured what it intended to measure

Question 18

What is internal validity?

Answer 18

• the extent to which we can be sure that the changes we observe have actually been caused by our manipulation, rather than other factors
• how confident we are that the cause-and-effect relationship cannot be explained by other factors

Question 19

What are maturation effects?

Answer 19

participants behaviour changes over time naturally

Question 20

what are history effects?

Answer 20

something changes about the participants circumstances that influences the variables

Question 21

What are testing effects?

Answer 21

merely having been tested before may have changed how they do on the post test

Question 22

What is regression towards the mean?

Answer 22

an extreme score is likely to become more average
- i.e. applied research attributing improvement to intervention. selection effects (e.g. participants are selected because of their extremity on the variable of interest). for example, because the anxiety is so bad that it cannot get worse, any change can only be an improvement

Question 23

Why are control groups important?

Answer 23

help minimise any influence of maturation/history/testing effects on conclusions we reach

Question 24

What is a passive control group?

Answer 24

participants do nothing/an alternate task. a no-treatment control group

Question 25

what is an **active** control group?

Answer 25

participants do something that they could reasonably assume might have an effect, but the researchers assume doesn't (e.g. sugar pill, placebo medication)

Question 26

What is a **wait-list** control group?

Answer 26

participants are waiting to take part in the intervention and believe they will be at some point

Question 27

what ways can confounding variables be eliminated?

Answer 27

- within-participants design - Random assignment - matched pairs design

Question 28

What is **random assignment**?

Answer 28

randomly assign participants to experimental groups. this 'spreads' these differences randomly across the groups to minimise any systematic influence

Question 29

What is a **matched-pairs** design?

Answer 29

when a third variable is expected to correlate with both IV and DV - matching levels of that variable across groups - e.g. age matching: every participant of a given age in group A has a matching participant of the same age in group B

Question 30

What is **differential attrition** in terms of threats to internal validity?

Answer 30

when people leave one condition or treatment more than others (drop out rates) - data becomes more biased to those who complete it - particularly relevant in **longitudinal designs**

Question 31

What is the primary disadvantage of within-subjects designs?

Answer 31

order effects

Question 32

what are order effects?

Answer 32

changes in behaviour because of 'when' a certain condition is completed

Question 33

What are **practice effects**?

Answer 33

participants perform a task better in later conditions

Question 34

What are **fatigue effects**?

Answer 34

participants perform a task worse in later conditions because they become tired or bored (especially children)

Question 35

What is **habituation**?

Answer 35

participants may become less sensitive to a stimulus through repetition

Question 36

How are order effects reduced?

Answer 36

- **counterbalancing** - testing different participants in different orders, assigned randomly - can use **latin-square design** when the experimental design has many conditions (e.g. 4 conditions would mean 24 possible orders)

Question 37

what are **demand characteristics**?

Answer 37

participants behave differently because theyre aware of being tested or because they try to work out what the experimenter expects to find

Question 38

What are **experimenter effects**?

Answer 38

experimenter's expectations can affect participant behaviour - double blinding can control for this

Question 39

What are **pygmalion effects**?

Answer 39

changes in childrens academic performance produced by teachers expectations about how well those children are likely to perform

Question 40

What is single blinding?

Answer 40

Participants do not know which experimental group they are in (controls for **demand characteristics**)

Question 41

what is double blinding?

Answer 41

neither the participants nor researchers know which group the participants are in

Question 42

What is **construct validity**?

Answer 42

the method of measurement actually measures the construct you intent to measure

Question 43

What are **ceiling effects**?

Answer 43

the task is too easy or stimulus is too extreme

Question 44

What are **floor effects**?

Answer 44

the task is too difficult, or the stimulus is too weak