Research methods

Question 1

Define reliability

Answer 1

The extent to which the results of research are consistent. I.e. when a piece of research is repeated using the same method, the sme results are obatined.

Question 2

Define validity

Answer 2

The extent to which a test or piece of research measures what it claims to be measuring.

Question 3

Define ecological validity

Answer 3

Where a measure of behaviour accurately reflects the way in which the behaviour would occur in normal circumstances. The extent to which research findings can be generalised to situations other than the research setting.

Question 4

Define population validity

Answer 4

The extent to which results from research can be generlised to the target populaion. Can only happen if the ptps are typical of the population from whcih they are drawn.

Question 5

Define Historical/temporal validity

Answer 5

Genralising the findings to another historical period. Temporal validity relates to whether findings from a study, or concepts within a theory hold true over time.

Question 6

What is a hypothesis?

Answer 6

a general prediction about what the researcher expects to happen or find out in an investigation.

Question 7

What is a directional hypothesis?

Answer 7

Predict the way one variable will affect another in the study. Used when previous studies show similar effect and the researcher can be confident that the same results will be found. Often start with ‘participants who…’

Question 8

What is a non-directional hypothesis?

Answer 8

not specific in what they predict. Usually used when there is no previous research and therefore the researcher cannot be confident about the outcome of their own study. Often start with ‘There will be…’

Question 9

What is a null hypothesis?

Answer 9

Predict that there will be no difference between the two conditions in the IV in relation to the DV or no correlation between the two variables. Used when previous research has not shown a similar affect or there is no previous research. Often start with ‘There wil be..’

Question 10

What is the IV

Answer 10

the different conditions. It is the variable that the researcher manipulates and is assumed to have a direct affect on the DV

Question 11

What is the DV

Answer 11

The variable that is affected by the IV. It is the data collected/ what is measured

Question 12

What is a control group?

Answer 12

A group that closely resembles the people under investigation but is not subjected to the factor under study (e.g.does not recieve treatment). They are used as a benchmark that those being tested are compared to.

Question 13

Define extraneous variables

Answer 13

Any variable other that the IV that could affect the DV if not controlled. Could confound results and affect the internal validity if not controlled. E.g. situational and participant variables

Question 14

define confounding variables

Answer 14

uncontrolled EV’s that may affect the DV and therefore reduce the internal validity of the results.

Question 15

Types of experiements

Answer 15

1) Laboratory: controlled, manipulates IV
2)Field : real-world setting, deliberately manipulates the IV
3)Quasi: IV is pre-existing/ naturally occuring (e.g. gender) but researcher manipulates the task. Controlled conditions and aim is to establish if the naturally occuring IV causes difference in performance on the task (DV).
4)Natural: IV and DV are naturally occuring and researcher simply measures the DV

Question 16

Strengths (2) and limitation (1) of Quasi and Lab experiements

Answer 16

(+)Control over EVs / establish cause and effect - cntrolled conditions. Internal validity.
(+)Replicability
(-)Low ecological validity stimulus material often lacks mundane realism

Question 17

2 strengths and 2 limitations of field and natural experiments

Answer 17

(+)High ecological
(+)Reduction of demand characteristics
(-)Difficult to replicate
(-)Low control over extraneous variables.

Question 18

What are the different types of experimental designs?

Answer 18

1)Independent groups design: Different participants for each condition
2)Repeated mesures: The same participants in all conditions of the experiment
3)Matched pairs design: Participants are paired as closely as possible on important or relevant variables and then one from each pair is allocated to each condition

Question 19

2 Strengths and 1 limitation of independent groups design

Answer 19

(+)Order effects are avoided
(+)Same stimulus material can be used for all participants: equally as challenging/easy
(-)more time consuming than repeated measures to gather the sample: twice as many participants are needed

Question 20

2 strengths and limitations of repeated measures

Answer 20

(+)Participant variables (difference between ptps) do not confound the results
(+)less time consuming to get sample than independent groups design
(-)Cannot use same stimulus material
(-)Order effects may confound results (however, counterbalancing can overcome this)

Question 21

What is counter balancing?

Answer 21

half the ptps randomally allocated to condition A then B and half randomly allocated tocodnition B then A (know as AB-BA). It allows order effects to be evenly distributed across both conditions.

Question 22

How are ptps matched in matched pairs designs?

Answer 22

paired on relevant variable following an assesment. Ptps with 2 highest scores are paired, then next to highest etc until they are all paired. One from each pair would be randomally allocated to a different condition of the experiment (so there was an equal number in each condition)

Question 23

2 strengths and a limitation of matched pairs design.

Answer 23

(+)Order affects are avoided - take part in 1 condition
(+)same stimulus material can be used for all ptps
(-)Difficult and time consuming to match pairs: need to conduct preliminary studies

Question 24

Define correlation analysis

Answer 24

will establish the strength and direction (or type e.g. positive or negative) of association and also the strength of the lin between the two co-variables.

Question 25

Negative vs positive correlation

Answer 25

negative: as one variable decreases the other inceases perfect is -1 Positive: as one variable increases so does the other perfect is 1

Question 26

Types of observations

Answer 26

1)Controlled: behaviour is observed and measured in a controlled environment 2)Naturalistic: behaviour is watched and measured in a natural environment. observes and records natural behaviour in a natural environment 3)Covert: observer is not clearly visible and ptps do not know that theyre being watched 4)Overt: observer is clearly visible and ptps know they are being watched 5)participant: Researcher becomes a member of the group whose behviour she/he is observing 6)Non-participant: researcher remains outside of the group and does not become a member of the group whose behviour he/she is observing

Question 27

Define event sampling and give one strength and limitation

Answer 27

Decides on a specific event relevant to the investigation and records i every time it is observed (+)aims to record all behavior so the results arelikely to be valid (-)may be difficult to record all events if they happen at once so results may not be representative of ptps behaviour and therefore would not be valid

Question 28

Define time sampling and give one strength and limitation

Answer 28

Observing and recording only behaviour that occurs at given points in time such as every five minutes (+)observer has time to record what they have seen so the results are likely to be valid (-)Some bhaviours will happen outside of the intervals so the results may not be representative of ptps behaviour and therefore would not be valid

Question 29

Two strengths and limitations of using a questionnaire

Answer 29

(+)Speed of data collection (+)Researcher bias is unlikely to confound to results: unlike an interview ptps can answer questionnaire without researcher being present (-)Self-report data may not be valid (-)Leading questions or misinterpretation

Question 30

Two types of interviews

Answer 30

1) structured 2) unstructured

Question 31

Two strengths and limitations of interviews

Answer 31

(+)easy to deal with misunderstandings: can be clarified and there could be follow-up questions (+)Replicable (structured) (+)flexible (unstructured) (-)Researcher bias: researcher could misinterpret answers to support their hypothesis (-)Interviewer effects: could inadvertly influence the ptps answers through how they ask the question and from body language etc. Interviews are prone to DC's and socially desirable answers

Question 32

Two strengths and limitations of case studies

Answer 32

(+)High ecological validity (+)Investigates situations which could not be set up for ethical reasons: they often investigate naturally occuring events that would be unethical to test during an experiment. Can provide insight to areas of psychology that we wouldn't be able to investigate in any other way. (-)Difficult toreplicate: they are unique so difficult/impossible to replicate. Not possible to test the external reliability of findings (-)Findings cannot be generalised very easily to other individuals: each individual & their experiences are unique so may only represent that person. May be difficult to generalise findings to others

Question 33

What are the 5 types of sampling techniques?

Answer 33

1)Random: everyone in the target pop has an equal chance of selection. Can use manual or computer selection 2)Opportunity: convenient members of target pop who are willing and available to take part. Researcher asks. 3)Stratified 4)Systematic: researcher gets names of everyone in the target pop and selects every nth member in the sample 5)Volunteer:uses members of the target pop who have put themselves forward to take part. Advert.

Question 34

How is stratified sampling used?

Answer 34

1) identifies different sub-groups (stratas) in the target pop 2)calculate number of each strata required for the sample (based on the proportion in the target pop) 3)Uses random sampling puts name of every member of target pop in different containers according to the sub-group they belong to 4)draws out required number of names from each sub-group

Question 35

1 benefit and 1 limitation of volunteer and opportunity sampling?

Answer 35

(+)Not time consuming (-)sample may not be representative of the target pop/ may be bias: researcher selects the sample themselves (opportunity) or they volunteer

Question 36

2 strengths and limitations of random and systematic sampling

Answer 36

(+)No researcher bias: all ptps have an equal chance of being selected (random)/ the researcher has no influence over who is the every nth number (+)Sample is likely to be representative of target pop: all members of the target pop have an equal chance of selection/ sample is made from the target pop (-)sample may not be representative: all have an equal chance of selection so by chance people with similar characteristics are selected

Question 37

strengths and limitations of stratified sampling

Answer 37

(+)avoids researcher bias (+)very likely to be reprasentative of target pop (-)very time consuming (-)impossible to account for the many number of ways that people are different - might not be representative

Question 38

What is the mean (descriptive stat) incl advantage and disadvantage ?

Answer 38

average, calculated by adding all the values together and dividing by the total number of scores (+)most sensitive measure of cenbtral tendancy as it uses all the values in a set of data (-) can be unrepresentative of the data set if there are extreme values

Question 39

What is the mode(descriptive stat) ? incl advantage and disadvantage

Answer 39

The value in a set of scores that occurs most frequently (+)not affected by outlying values (anomalies) (-)Does not use all the available data and there could be more than one mode

Question 40

What is the median (descriptive statistic)? incl advantage and disadvantage

Answer 40

middle value of a set of numbers placed in numberical order (+)Not affected by anomalous results and onlt takes the middle values (-)Doesn't use all of available dat and is therefore not as sensitive as the mean

Question 41

What is the range?

Answer 41

Difference between the highest and lowest calculated scores in a given set of data.

Question 42

3 ways of improving reliability: experiments

Answer 42

1) Standardisation 2)Take more than one measure: for each ptp as this reduces impact of any anomalous scores 3) Pilot studies: discover any problems with research design such as -ptps misunderstanding of instructions -whether timings etc are adequate

Question 43

Assessing reliabilty: experiement

Answer 43

1)Test-retest: test or procedure is used several times (at least twice), correlate scores from later occasion, short interval between tests (e.g. 1 week) so answers aren't remembered, assess strength of correlation using appropriate stats test. Degree of reliabilty is determined by comparing calculated correlation co-efficient with critical value in stats table - should be a (strong) positive correlation between two sets of scores if they are reliable.

Question 44

Improving reliabilty: observations 3 ways

Answer 44

1)Behavioural categories: operationalised so that it is clear for all observers 2)Pilot studies: identify any problems with design such as poorly defines BC's or inadequate training 3)standardisation: when more than 1 investigator the way they collect and record data should be standardised. This could involve filming the behaviour so that the observers can practise the categorisation.

Question 45

Assessing the reliability: observation

Answer 45

Inter observer reliabilty (extent to which the observers agree on the observations): -2 or more devise set of categories that they intend to use to code the behaviour -carry out observation independently and tally each BC when its observed -at end of study compare and/or correlate their totals with one another -strength of correlation should be assessed using appropriate stats test -determine degree of reliability should be 0.80 or more

Question 46

How is the degree of reliabilty determined when using a stats table (e.g.for inter observer reliability or test retest)

Answer 46

by comparing the calculated correlation co-efficient with the critical value in the statistical table which indicates the extent of correlation - there should be a (strong) positive correlation between the two sets of score if the results are reliable. A correlation coefficient of 0.80 or more suggests good inter-observer relaibilty

Question 47

Improving reliabilty: self report 3 ways

Answer 47

1)Reduce ambiguity: questions may need to be re-examined and rewritten. Could replace some open questions with closed questions. 2)pilot studies: discover problems with design such as leading or ambiguous questionnaires or interviews 3)Standardisation: all ptps should be subject to the same environment, info and questions. using the same researcher can improve reliabilty (esp interviews) exposed to same approach of asking questions, body lanuage, tones etc.

Question 48

Assessing reliabilty: self-report method

Answer 48

Test-retest

Question 49

assessing reliabilty: split-half method

Answer 49

- ptps scores on one half os a test should be correlated with their scores on other half of test -strength of correlation should be assessed using appropriate stats test -reliability assessed by comparing ... should be a (strong) positive correlation between two sets of scores if results are reliable

Question 50

Assessing validity 3 ways

Answer 50

1)Face validity 2)Concurrent validity 3)Content validity

Question 51

Assessing validity - concurrent validity (CV)

Answer 51

-has CV if it produces similar findings to an existing measure. Can be assessed by: -same ptp would complete 2 tests/Qairs (your own and an existing one) that assess the same aim -correlate scores from one test with score from other test -strength of correlation by using stats test -assess degree of validity by... should be +0.8 (strong positive)

Question 52

Assessing validity: Content validity (CV)

Answer 52

asking experts in the field to check methodology of a study to see how accurately it measures the desired behaviour. If they feel it does, then it is said to have content validity.

Question 53

Improving validity

Answer 53

1)Revise/remove questions -poor face v then Q's should be revised so they relate more obviously to the topic -concurrent V low then researcher should remove some Q's that may seem irrelevant and assess concurrent V again 2)Improve research design to control/overcome: Investigatore effects, demand characteristics, confounding variables, social desirability bias and poorly operationalised BC's

Question 54

Improving external validity: ecological and population

Answer 54

Ecological: -give ptps a task with high mundane realism -if possible, conduct a field/ natural experiment Population: -use a large sample -use sampling technique that is likely to produce a representative sample

Question 55

Improving validity: experiment

Answer 55

1)Single blind: reduces DC's - ptps do not know what group or condition they are in and therefore cannot change their response to suit or foil the research. 2)Double blind: reduces DC's and researcher bias - Neither the ptps nor the experimenter know what each group or condition represents.

Question 56

Define Content analysis

Answer 56

a type of observational study in which behaviour is observed indirectly in visual, written or verbal material, it is technique of analysing qualitative data. It turns qualitative data into quantitative data.

Question 57

How is Content analysis carried out?

Answer 57

-Read the source -Identify potential categories which emerged from the data -Read source again and count the number of examples which fell into each category to provide quantitative data (coding) -compare the different categories to see if the results support the hypothesis

Question 58

Define thematic analysis

Answer 58

qualitative data is summarised by identifying repeated themes in the material to be analysed.

Question 59

How is thematic analysis carried out?

Answer 59

-read and re-read source -code the transcript by annotating it -recurring themes of the trancript will be identified from the annotations -re-read transcript to ensure that vthe themes are appropriate and cover all the different key features. -conclusion based onthe themes and contextualising the analysis in relation to previous research

Question 60

Evaluation of content and thematic analysis

Answer 60

(+)produce qualitative and quantitative data depending on the aims of research (flexible) (+)high ecological validity - based on observations (-)could be making artificial distinctions or could miss certain forms of behaviour, reducing validity (-)observer bias reduces the objectivity and validity of findings as different observers may interpret the meaning of behavioural categories differently. Risk researcher may attribute opinions and motivations that were not intended origionally

Question 61

Define what is meant by a case study

Answer 61

A detailed study of a single individual, group of people, institution or event. Usually carried out in the real worls. They are ideographic, very individualistic and are usually longitudinal

Question 62

Strengths and limitations of case studies

Answer 62

(+)investigate naturally occuring events that would be unethical to test during an experiment. Can provide insight to psychology that we wouldn't be able to investigate otheriwse. (+)Usually longitudinal - rich in depth info (-)difficult to generalise from individual cases (-)ethical issues such as confidentiality, consent (-)replicability (-)self report

Question 63

Features of science

Answer 63

-objectivity and the empirical method -replicability and falsifiability -theory construction and hypothesis testing -paradigms and paradigm shifts

Question 64

Nominal data

Answer 64

A 'head count' of how many people are in each category

Question 65

Ordinal data

Answer 65

Rough numerical score but no proper/recognised units have been used to measure the score.

Question 66

Interval data

Answer 66

Recognised units of measurement e.g. IQ score

Question 67

choosing a statistical test (what is the table?)

Answer 67

Nominal ordinal Interval repeated measures, Sign test, wilcoxon, related t test matched pairs, sign test, wilcoxon, related t test Independent measures, Chi2, mann-whitney, unrelated t test correlation, chi2, spearman's rho , pearson's r

Question 68

mnemonic to remember stats table

Answer 68

research super wouldn't Rodger methods sonic wear rabbitt is cheese marks& uses cool cake spencers protection

Question 69

What is a type 1 error

Answer 69

When you accept the hypothesis when it is false (also when you reject the hypothesis and its true). More likely to make a type 1 error if the level of significance is too lenient (too much room for chance) (e.g. 0.1 or 10% rather than 5%)

Question 70

What is a type 2 error

Answer 70

Would occur where a real difference in the data is overlooked as it is wrongly accepted as being not significant, accepting the null hypothesis where it is false (i.e. rejecting your hypothesis when it is true) in error (a false negative). More likely too occur when level of significance is too stringent (e.g.0.01 or 1% rather than 5%)