Induction Flashcards

what are the 3 variables

what is operationalisation

independent variable: change, often manipulated by researcher
dependent variable: measured, result of IV
extraneous variables: controlled variables so you can infer cause and effect

operationalisation: making variables memorable
DV: number of....in a time frame, rating on a scale of 1-5
IV: e.g less than 3 hours sleep vs 8+ hours

lab experiments

IV changed by researcher, environment: lab or artificial, randomly allocate each condition

+can replicate to check if reliable, control extraneous variables
-risk of demand characteristics, psychological harm from IV, environment different to in real life so low ecological validity

field experiments

IV changed by researcher, in natural environment
+natural so more realistic so more valid, less risk of demand characteristic
-psychological harm from IV, harder to replicate, difficult to change extraneous variables

Quasi experiments

Iv is a pre-existing difference e.g age, gender
+more ethical as IV isn’t manipulated by researcher so no psychological harm
all other ± are the same as lab/field depending on environment

natural experiments

IV changed naturally and would still occur without researcher
+ethical bc psychological harm caused by researcher manipulating IV
+less demand characteristics→internal validity
+natural so higher ecological validity

-difficult to replicate→cant check if valid or reliable
-difficult to control extraneous variables to infer cause&effect

demand characteristics

when participants guess the aim of the experiment and change behaviors to how they think helps participants
‘screw you effect’- demand characteristics to ruin the experiment

define: ecological validity

whether results can be generalized(applied) to real life life outside of the experiment or not
lab experiments have low ecological validity

define: cause and effect

if control over extraneous variables is high then we can infer cause and effect, meaning the IV caused the DV

ethics

• issues mentioned by British Psychological Society

• for all issues, a cost-benefit analysis should assess if ethical issues are worth it for the results

• acronym ‘Drip C’

Deception & debriefing- researchers should avoid deception unless reasonably acceptable to prevent demand characteristics. After, participants should be debriefed(told the true aim of the study) and have the right to withhold their data

Right to withdraw- participants should be clearly told that they can leave whenever they want without negative consequences. This should be written on the consent form

Informed consent-participants should be told the aims of the study(when possible) and complete consent forms first

Psychological and physical harm- researcher should stop the study if any more harm than in everyday life occurs. Participants have the right to withdraw. Debriefing after should reassure embarrassed or concerned participants

Confidentiality- participants should be known by numbers not names

what is a hypothesis

• a clear statement predicting an experiment’s outcome

• written in future tense

• has operationalised variables

you should always do 2 hypothesis : one null and one alternative

-pick 2 tailed if previous research is inconclusive/mixed, or there is no previous research
-pick 1 tailed if previous research shows the direction of the effect

1 tailed/directional hypothesis

predicts the direction of the effect/difference
write a ‘DV sandwich’
IV condition 1 will have a greater/lower number of DV in a time frame than IV condition 2

2 tailed/non directional hypothesis

predicts a difference but not in a particular direction
There will be a difference in the DV between IV condition 1 and IV condition 2

null

predicts IV/difference between condition 1 and 2 has no effect on the DV
There will be no difference in the DV between IV condition 1 and IV condition 2

research methods vs research designs

research methods- lab, field, quasi, ect
(experimental)research designs- matched pairs, repeated measures, ect

repeated groups design

same participants used in both conditions, so all experience condition A and then all experience condition B

+individual differences/participant variables are constant as all participants are in both conditions so cause and effect can be inferred

-demand characteristics→exposed to both parts of IV so easier to guess the aim

-need different tests of the same difficulty→unlikely so could become an extraneous variable

-order effects e.g bored or better bc of practice on the second condition
this is reduced by counterbalancing: splitting the group into 2 and first half do condition A first, then condition B, and second half do condition B first and then condition A

independent groups design

different participants in each condition e.g half experience condition A and the other half only experience condition B, randomly allocated to present the sample being unrepresentative.

+no order effects bc each participant only experiences one condition
+demand characteristics aren’t a problem bc participants are only exposed to one part of IV

+same test e.g. same words used in each condition→not an extraneous variable
-participant variables aren’t kept constant bc different participants in each condition

matched pairs design

matched participants w important characteristics that may affect performance so different but similar participants for each condition e.g memory tests w different music but both have the same IQ

+participant variables are kept constant bc important characteristics are matched between participants
+no order effects bc each participant only experiences one condition
+no demand characteristics bc participants are only in one part of IV

-participant variables can never be fully matched in all respects bc they are different people
-matching participants is time consuming and difficult so rarely used in real life

random sampling

all members of target population have an equal chance of being picked e.g pick out names from a hat

+best chance of being unbiased
+representative sample bc everyone has an equal chance of being picked

-time consuming & expensive to compile a list of everyone in target population so rarely used

stratified sampling

divide the target population into important subgroups and select members from each category in the right proportionn

+representative of target population so results can be generalised

-impractical bc its difficult and time consuming to identify subgroups in target population

opportunity sampling

select participants who are available at the time e.g unis using psychology students

+quick, convenient and cheap→no advertising or complicated selection process

-unrepresentative and biased→usually students who take part(more educated than other groups)

volunteer/self selecting sampling

people volunteer e.g. those who respond to ads in the newspaper

+convenient→just wait for replies
+ethical→informed consent before study

-people who volunteer are usually more kind and outgoing→unrepresentative→can’t generalise results

systematic sampling

select every nth person(n=consistent number) e.g every 6th

+uses an objective system→unbiased→unrepresentative

-not truly random unless you select a random number to start as the first participant

define: target population, representative, sample, sample size

target population-who the study is aimed at
representative-unbiased, has the right proportion of each subgroup in sample
sample-people taking part in the research
sample size-usually 30 people but can vary a lot depending on the research method
too small→unrepresentative, too big→expensive and time consuming