Week Four

Question 1

internal validity

Answer 1

how sound is the design, how strongly can we assert that changes in our DV are down to our IV and not other things we haven’t controlled for.

Question 2

external validity

Answer 2

how generalizable are our findings, how representative of the real world is the study.

Question 3

steps to internal validity

Answer 3

• Sound operationalisation of the DV
  ○ Measures should be reliable and valid.
  
  • Strong experimental design logic
  • Sound operationalisation of our Ivs
  • Consideration and use of appropriate remedies to control for extraneous variables.

Question 4

factorial

Answer 4

more than one IV
○ May have all repeated measures or all between groups or mixed.
○ Allows examination of interplay between two or ore Ivs and the splitting p of these effects into interactions and main effects.

Question 5

strengths of factorial designs

Answer 5

○ More than one independent variable allows for more precise hypotheses.
○ Control of extraneous variable by including as an independent variable.
Ability to determine the interactive effect of two or more independent variables

Question 6

main effect

Answer 6

○ The influence of one independent variable on the dependent variable
One main effect for each IV in a study

Question 7

interaction effect

Answer 7

○ The joint, combines or interactive effect of two or more independent variables on the dependent variable.

Question 8

factorial design notation

Answer 8

2 x 3

= 2 IVs, one 2 levels and one 3 levels

Question 9

weakness of factorial design

Answer 9

• Using more than 2 IVs may be logistically cumbersome.

- Higher order interactions are difficult to interpret.

Question 10

variability

Answer 10

In any study, what we are trying to do is establish whether the variation in the DV between groups I different from the normal variation in the DV within groups.

Question 11

how to ensure a strong experimental design

Answer 11

• In order to maximise our chances of getting a true picture of how the IV affects the DV we need to:
  ○ Maximise the impact on our dependent variable that is related to the independent variable.
  § Increase between group/condition/level variation.
  ○ Minimise variation in our dependent variable that is not related to our independent variable
  § Compress within group/condition variation.
  Done by adopting good research design and controlling variables.

Question 12

separation

Answer 12

In any experiment we are trying to establish whether the variation in the DV between groups (caused by the IV) is different from normal variation in the DV within groups.

Separation of the IV and DV: achieved by IV operationalisation

Question 13

compressions

Answer 13

achieved by controlling extraneous variables.

Question 14

types of extraneous variables

Answer 14

○ Noise creating
§ Randomly impact the DV, not related to the IV, but potentially create extra variation in the DV not due to the IV, want to minimise this.
§ Has an effect randomly on the DV
○ Confounding
§ Systematically impact the DV, related to the IV, potentially explaining changes in the DV that you would be expecting the IV to make, want to control for this by eliminating, keeping constant or building into study so can measure impact.
§ Varies systematically on the DV.

Question 15

EVs and between groups

Answer 15

• Two separate groups of people who could differ on a whole range of things. Both relevant and irrelevant to the study at hand.

Question 16

self assignment

Answer 16

Self assignment: subject selects which treatment groups

Question 17

experimenter assignment

Answer 17

experimenter selects which treatment group.

Question 18

arbitrary assignment

Answer 18

selection based on seemingly non-relevant criteria.

Question 19

random assignment

Answer 19

• Random Assignment/Allocation: best to have no criteria for selection.
  ○ Provides max. insurance that the groups are equal.
  ○ Eliminates systematic differences between groups.
  ○ Doesn’t eliminate Evs but does help to decrease.

Question 20

matching

Answer 20

○ Use of a variety of techniques to equate participants in the treatment groups on specific variables.
§ Should be done with variables thought to be related to the IV or may confound the IV
□ E.g. intelligence, age etc.
○ Advantages
§ Controls for the variables on which participants are matched
§ Increases the sensitivity of the experiment.

Question 21

individual matching

Answer 21

○ Advantage
§ Groups equated on potential EV.
○ Disadvantages
§ Identifying the variables on which to match.
§ Difficulty matching participants increases as the number of variables on which to match increases.
§ Decrease in generalisability of results.

Question 22

distribution matching

Answer 22

Run a algorithm to ensure each group is equal.

Question 23

to eliminate EVs

Answer 23

• To eliminate EVs, built it into the design (blocking).
  ○ Make the EV another IV in the design
  ○ Should only be used when you are interested in the effect of the extraneous variable.

Question 24

problems with repeated measures deigns

Answer 24

• order effects
• practice effects
• fatigue effects

Question 25

order effects

Answer 25

○ When participants are exposed to Ivs in a particular order

Question 26

practice effects

Answer 26

Differences or improvements may occur simply due to practice rather than the IV.

Question 27

fatigue effects

Answer 27

Participant is bored and pays less attention.

Question 28

control for effects

Answer 28

- counterbalancing - randomised counterbalancing - intra-subject counterbalancing - complete counterbalancing - incomplete counterbalancing

Question 29

counterbalancing

Answer 29

○ Breaking our sample into subsets and presenting the conditions in different orders. ○ By collapsing this data again it will hopefully counteract any sequencing effects.

Question 30

randomised counterbalancing

Answer 30

Participants are randomly exposed to levels of the IV in a random sequence.

Question 31

intra-subject counterbalancing

Answer 31

○ Participants take treatments in more than one order

Question 32

complete counterbalancing

Answer 32

Every participant is exposed to every possible sequence of the IV.

Question 33

incomplete counterbalancing

Answer 33

most commonly used.

Question 34

simple carry over effects

Answer 34

○ Where performance on the DV in one condition is contaminated by the effects of the previous condition

Question 35

differential carry over effects

Answer 35

○ Unique to one condition only. | ○ Is different depending on the order in which the participants are exposed.

Question 36

time as an IV

Answer 36

``` - - Occurs if you measure people at different times Can cause design effects - maturation - external effects - statistical regression - mortality ```

Question 37

maturation

Answer 37

changes due to the natural development, natural expected improvement over time. § Is it because of the IV or just because people are getting better at it over time.

Question 38

external effects

Answer 38

§ External events that affect participants during the study. § Socio-historical-economic changes relevant to outcome. □ i.e hearing about a shooting can impact your view on gun laws.

Question 39

statistical regression

Answer 39

§ Refers to the tendency to move up or down towards the mean over time People are likely to move towards the mean eventually.

Question 40

mortality

Answer 40

§ Attrition from the study (leaving the study): not all participants will remain in the study until the end. § Common in longitudinal studies Do the people who remain in the study differ systematically to those who drop out?

Question 41

experimenter effects

Answer 41

- measurement issues - attribute effects - experimenter expectancies

Question 42

measurement issues

Answer 42

problems with equipment or errors. | § To control this: training, careful measurement, attention to detail.

Question 43

attribute effects

Answer 43

participants responding differently depending on the experimenter. § Control: randomise the experimenter or use the same experimenter for the entire time.

Question 44

experimenter expectancies

Answer 44

likely response of participants and its manipulations may lead to subtle differences in the way the experimenter interacts with participants which leads to differences in outcomes. § Rosenthal effect: differential attitude or attention conveyed to participant expected to respond most favourably to the study. § Golem effect: differential attitude or attention conveyed to participant expected to respond least favourably to the study.

Question 45

controls for experimenter effects

Answer 45

§ Double blind: neither the experimenter or the participant is aware of the treatment condition administered to the participant. □ i.e. a drug and a placebo where the experimenter and the participant doesn’t know what they were given. § Partial blind □ Knowledge of each participants treatment condition is kept from the experimenter through as many stages of the experiment as possible § Automation □ No contact □ i.e. experiment done via a computer.

Question 46

participant effects

Answer 46

- demand characteristics - social desirability - hawthorne effect

Question 47

demand characteristics

Answer 47

participant gets inkling about what study is aiming for an performs in a way to conform to those expectations.

Question 48

social desirability

Answer 48

participant performs in a way they thin will be most pleasing to experimenter or will paint them in the best light.

Question 49

hawthorne effect

Answer 49

participant improves or changes performance on the outcome purely as a function of the attention received for the study and not as a function of the nature of the manipulations of the IV.

Question 50

situational effects

Answer 50

○ The impact of environmental and timing difference on participants' outcome scores. Can include time of day, weather, lighting, etc. ○ Can be controlled by keeping these constant for all participants.

Question 51

control groups

Answer 51

○ Allows you to directly examine the extent to which changes would have occurred without your intervention. ○ Make sure you endeavour to have an 'equivalent' control group. ○ May have multiple control groups § No intervention § Placebo intervention