research methods year 1

definitions and what not

experimental method

manipulation of IV to measure the effect on the DV while controlling extraneous variables.

aim

general statement of what the researcher will investigate. based on previous research

hypothesis

a testable statement that describes a relationship between the variables being investigated

directional hypothesis

states the direction of the differemce

or

states the direction of the relationship

non-directional hypothesis

does not predict the direction of the difference or relationship

only states that a difference or relationship exists

in what situation would you use a directional hypothesis?

when there has been previous research into the area you are investigating so you have an idea of what the relationship/difference will be

in what situation would you use a non-directional hypothesis?

when there has been no previous research into the area you are investigating so you don’t know what the nature of the relationship/difference will be

control condition

the level of the IV that is kept constant/has no change to compare with the experimental condition

acts as a baseline to determine the magnitude of the effect of the change in IV

experimental condition

the level of the IV that is altered/changed in an experiment to observe its effect on the DV

operalisation

the process of defining variables into measurable factors

independent variable

the variable that is manipulated/ changes naturally

dependent variable

the variable that is measured

extraneous variables

any other variables that may effect that DV that isn’t the IV. are controlled during experiments.

confounding variables

a type of EV that varies systematically with the IV, making it difficult to determine the effect of the IV on the DV

demand characteristics

features of an experiment that may inform participants of the purpose, potentially influencing their behavior

can be cues from the researcher or the experimental set up/method

investigator effects

effects of the researchers behaviour on the DV

can be conscious or unconscious

randomisation

using chance methods during the design of the study to avoid the effect of uncocious bias from the investigator ensuring all participants/ conditions are as equal as possible

may be used in the design of materials, allocation of participants to conditions and the order of conditions

standardisation

using the same formalised procedures and instructions on all participants to ensure consistency in the study (increases validity)

experimental design

the way that participants are organised in relations to the experimental groups

can be related or unrelated

independent groups

where participants are assigned to one condition or group and only take part in that one

unrelated design

repeated measures

where all participants take part in all conditions

related design

matched pairs

where participants are paired based on similar characteristics/ variables that may effect the DV one member of each pair is assigned to each group

participants only experience one condition

related design

random allocation

assigning participants to different groups in a random way so that each participant has an equal chance of being placed in any group

used for independent groups design

minimises bias (increased validity)

counterbalancing

half of the participants experience the conditions in one order and half do it the other way around

used for repeated measures designs

reduces order effects (increases validity)

lab experiments

takes place in a controlled environment

researcher manipulates IV and measures DV

strengths of lab experiments

+high internal validity as confounding variables are controlled

+can be easily replicated to test for reliability, a reliable study tends to be valid

limitations of lab studies

-low external validity as tasks and environment are artificial (lack mundane realism) so cant generalise

-subject to demand characteristics as the participants know that they’re being assessed

field experiments

takes place in a realistic environment

researcher manipulates IV and measures DV

strengths of field experiments

+high external validity as they take place in natural settings (has mundane realism) so can generalise

+less likely to be effected by demand characteristics in cases where participants don’t know that they’re being studied

limitations of field experimen

-harder to establish cause and effect relationships as there is less control over EVs

-ethical issues surrounding consent when people are unaware they’re being observed

natural experiments

researcher doesn’t manipulate IV (it is naturally occuring) and measures DV

strengths of natural experiments

+allows for research into areas that may not usually possible due to ethical or practical reasons (e.g. romanian orphan studies)

+high external validity as they make sue of naturally occurring events/reactions

limitations of natural experiments

-naturally occurring events happen rarely so follow up research may not be possible and generalisability may have a very small scope

-participants may not be randomly allocated to conditions so it may be unclear whether IV effected DV

-EVs aren’t controlled so it may be unclear whether the IV effected the DV

-research may be conducted in labs so may lack mundane realism, giving it low external validity meaning it isn’t generalisable

quasi-experiments

Iv is naturally occuring (e.g. presence of a mental illness)

DV is measured by researcher

strengths of quasi-experiments

+controlled environments means they have a high internal validity and are replicable

+allows for research into naturally occuring variables

limitations of quasi-experiments

-participants cant be randomly allocated to conditions so there may be confounding variables

-IV is not changed by experimenter so it cant claim to be the cause of change in DV

population

a large group of people that a researcher wants to stdy

this is the group that the sample is drawn form

target population

the specific group of individuals within a population that share certain characteristics that a researcher aims to study or from which a sample is drawn.

sample

a group of Popole who take part in a study

drawn from the target population

assumed to be representative of the whole population

sampling technique

a method used to select a sample form a population that is representative

random sample

all members of the target population have equal chances of being selected

1. get list of all individuale in target population

2. everyone on the list is assigned a number

3. sample is selected via a lottery method

systematic sample

people are selected from a list at regular intervals

1. sampling frame is produced (alphabetised list of target population)

2. sampling system is chosen (e.g. every 3rd person)

3. may start count from a random point in the list to avoid bias

4. work through sampling frame until sample is complete

stratified sample

composition of sample reflects the proportions of people in different subgroups in the target population

1. researcher identifies strata making up the population

2. proportions needed for the sample to be representative are calculated

3. the right amount of people are selected from each strata by random sampling

opportunity sample

selecting anyone who is available and willing

volunteer sample

“self selection”

may involve advertisement via a poster or social media

biased sample

when a certain groups are over or under represented in the sample

limits how generalisable findings are

generalisation

the extent to which findings/ conclusions from an investigation can be broadly applied to the population

this only happens when sample used was representative

ethical issues

arise from a conflict between goals of research to produce valid data and the rights of the participants

informed consent

participants should be aware of aims of the research, the procedure involved, their rights (e.g. right to withdraw) and what there data will be used for

participants can then make an informed decision about whether to participate or not without being coerced.

deception

deliberately withholding information form the participants at any stage of the study

participants who have not been fully informed cannot understand the true nature of the study and so cannot give informed consent

protection from harm

a study should not put participants at any more risk than they would encounter daily anyway.

they should be protected from physical and psychological harm.

psychological stress includes feeling embarrassed, inadequate or under pressure

participants should be reminded that they have the right to withdraw at regular points in the study

privacy and confidentiality

ppts have the right to control information about themselves- privacy

ppts have the right to have personal data protected by law- confidentiality

the place name/institution name may be withheld to protect participants' identities.

PBS code of conduct

British Psychological Society has a set of ethical guidelines to ensure research is conducted responsibly

guidelines are implemented by ethics committees in research institutes that uses a cost-benefit approach to decide whether proposals for research are ethically acceptable

consent form

details all information that may effect someone’s decision whether to participate (e.g. aim, procedure, duration, rights)

children under 16 must have parental consent

may require participants to sign

debrief

ppts should be made aware of any elements they were deceived on during the study (e.g. the existence of other experimental/control groups)

ppts should be told what their data will be used for and have the right to withhold their data (especially relevant in cases of retroactive consent)

ppts should be told their behaviour was normal and counselling should be provided where required

anonymity

for many studies researchers keep no record of ppts personal information in order to stay in line with confidentiality laws

initials/ numbers may be used to identify ppts

ppts should be reminded that their data will be protected and not shared with anyone else

pilot study

small scale trial of a study before it takes place in order to identify potential issues and modify the design

involves a small sample and is designed to test the effectiveness of the method

self report methods (interviews and questionnaires) can also have pilot versions to check that questions aren’t ambiguous or confusing

for observations the pilot study involves checking that the coding systems all work

single blind procedure

A research design where participants don’t know the aim of the study/which condition they are in, to reduce bias.

any info that could lead to demand characteristics is withheld until the end of the study

double blind procedure

A research design where the participants and the experimenters don’t know what the aim of the study is, helps eliminate bias

experimenters are usually a third party (not the people who designed the study)

often used for drug trials

since the researchers don’t know who is receiving the real treatment and who has the placebo their expectations cant influence the ppts reports

observations

non experimental method

allow researchers to study observable behaviour in natural or controlled settings without the ppts knowing they’re being observed (removes demand characteristics)

naturalistic observation

watching behaviour in a setting within which it would normally occur

controlled observation

watching behaviour in a structured environment where variables are manipulated or monitored

e.g. strange situation

covert observation

ppts behaviour is observed without their knowledge or consent

the behaviours must be public and occur naturally for the study to be ethical

overt observation

participants are aware they are being observed and have given informed consent.

participant observation

researcher becomes a member of the group that they are observing

non-participant observation

the researcher remains separate from the group they are observing and watches from the outsidewa

unstructured observation

researcher writes down every behaviour they see

structured observation

the researchers predefine behavioral categories and tally how many times they occur

behavioural categories

fully operationalised components of behaviour

ensures that different observerswill interpret and record behaviour consistently.

continuous recording

used in unstructured observations

event sampling

researcher records every time an event in a behavioral category occurs

time sampling

researcher records behaviour at specific intervals

self report technique

any method that asks a ppt to state or explain their feelings, oppinnions, behaviours or experiences related to a certain

questionnaires

set of written items (questions) used to asses a persons thoughts

can be used to asses the dv in an experiment

open questions

there are no fixed answers and respondents can answer in their own words

leads to rich qualitative data

closed questions

there are fixed answers set by the researcher

e.g. yes/no, a scale of 1-5

produces less detailed quantitative data

interviews

a live encounter (on the phone or F2F) where the interviewer asks questions to assess the interviewees thoughts/experiences

structured interview

have a predetermined set of questions that are asked in a set order

unstructured interview

more like a conversation

there is an aim/topic of discussion

interaction is free flowing and interviewer prompts the interviewee to expand answers

semi-structured interview

list of questions but interviewer may add new ones based on the interviewee’s pervious answers

Likert scales

rating scale ranging from strongly agree to strongly disagree

rating scales

respondent selects a value that represents how strongly they feel about a topic

fixed-choice option

item includes a list of possible options and respondents select the ones that apply

interview design

interview schedule- a standardised list of questions the interviewer intends to cover (reduce interviewer bias)

interviews are usually 1:1 although group interviews are used in some situations

in 1:1 interviews:

• quiet room away from others

• start with neutral questions to build rapport

• remind interviewee that their answers are confidential

good interview questions

• little jargon

• avoid emotive language

• avoid leading questions

• avoid double barrelled questions

• avoid double negatives

this will prevent misinterpretation maintaining the quality of information received

correlation

a mathematical technique that shows the relationship (strength and direction) between two covariables

what type of graph is used to represent correlational data?

scarttergram

co-variables

variables investigated in the correlation

NOT IV and DV because correlations show associations not cause and effect relationship

positive correlation

as one co-variable increases so does the other

negative correlation

as one co-variable increases the other decreases

zero correlation

no relationship between co-variables, changes in one do not affect the other

the difference between correlations and experiments

in an experiment the Iv is maniptulated and the Dv is meaaured by the researcher. cause and effect is established. test of difference (between conditions with different IVs)

in a correlation there is no manipulation. cause and effect cant be established established, just a relationship. test of association (between covariables)

Qualitative data

Data that was expressed in words

Non-numerical

Harder to analyse

E.g. transcript of interview, notes from unstructured interview, yes/no are also qualitative data

Quantitative data

numerical data

Data can be analysed by statistically

E.g. time, scores

Primary data

information that has been obtained first hand by the researcher for the purposes of the study.

Gathered directly from the participants via experiment, self report or observation

Secondary data

information that has been gathered by someone else before the study begins.

Usually has already undergone statistical analysis so the significance is known

E.g. from government census, other studies

Meta-analysis

combining findings from multiple studies on the same topic and each study is equally weighted

Aims to produced a statistical conclusion that summarises the overall effects or trends across studies.

descriptive statistics

the use of graphs and tables to identify trends and analyse sets of data

Measure of central tendency

averages

Mean, mode, median

Mean

Calculated by adding up all of the numbers in a data set and dividing the sum by the number of data sets there are

+The most sensitive measure of central tendency (most representative)

-Easily distorted by extreme values (outliers)