Research Methods

Question 1

L1 What is independent variable

Answer 1

Variable researcher manipulates in order to determine its effects on the dependent variable, and may be divided into levels, sometimes referred to as experimental conditions

eg. investigating how chocolate effects mood, experimental conditions may be 10g of chocolate, 20g chocolate and no chocolate

Question 2

L1 What is control condition and dependent variable + extraneous variable

Answer 2

Control condition = provides standard against which experimental conditions can be compared, condition where IV is not manipulated at all eg. no choc and effect on mood

dependent variable = variable being measured (eg. Mood)

extraneous variables = Variables other than IV that could affect DV, eg. sleep, weather and personal life events could affect mood other than choc

Question 3

L1 what is confounding variable

Answer 3

confounding = any variables other than IV that have effected the DV eg. Light and noise

Question 4

L1 what and why do we use operationalisation

Answer 4

• used for an experiment to be successful, as dependent and independnet variables are defined and stated how it will be measured
  eg. Measuring aggression in children it must be defined and measured “aggression means how angry or irritated… and can be measured by displays of verbal or physical acts of aggression in a ten minute period

Question 5

L1 Laboratory Experiments

Answer 5

This is when an experiment is carried out in a controlled experiment, such as lab, allowing high level of control over the IV and eliminate EV.

Researcher can then measure change in DV caused by manipulation of IV.

Pps are randomly allocated to a condition, neither experimenter or pp decides on which condition pp is placed in, unpredictable method (eg. Flipping a coin) decides.

Other variables experimenter wants to hold constant are extraneous variables, aim of controlling EVs is to minimise possible impact on results of the investigation. Eg. extraneous variable is the pps personality

Lab experiments are conducted in an artificial setting

Question 6

L1 + and - of lab experiment

Answer 6

+
- high level of control over IV and conditions, easy to control EVs and prevent them from becoming confounding variables that may affect results collected
- researcher can monitor IV and establish cause and effect between IV and DV

-
- demand characteristics might occur within study, change behaviour as they guess the aim of the study, may behave in a more positive light than normal = social desirability bias
- establishing a high level of control over the IV and EVs means that experimental condition can often lack mundane realism, lab study does not have ecological validity

Question 7

L1 What are Field Experiments

Answer 7

• exp carried out in the real world or in a natural setting, rather than artificial setting of a lab
• IV still manipulated or controlled by researcher to see effect on dependent variable (eg. Observing ppl on the street)

Question 8

L1 + and - of field experiments

Answer 8

+
- strength is that there is more mundane realism and ecological valid, than lab, reflecting more real life than a lab experiment

• less chance of demand characteristics, as opposed to lab, as in field exp, pps might not be even aware of taking part in research, therefore behave more naturally and more valid data is collected

-
- Less control over EVs when compared to lab experiment, therefore effect on DV may not be caused by Iv but by EVs, meaning research is not valid (not measuring what intends to measure)

• difficult to replicate, meaning may be unreliable, when compared to lab exp especially

Question 9

L1 What is a natural experiment

Answer 9

Researcher takes advantage of a naturally ocurring IV to see effect on the DV
Natural experiment is a study measuring variables that arent directly manipulated by the experimenter, eg. comparing behaviour in a single sex and mixed school, meaning IV is naturally occurring
Experimenter is finding pps who already meet conditions of experiment, rather than allocating pps to condition themselves

Question 10

L1 + and - of natural experiments

Answer 10

+
- high level of mundane realism and ecological validity compared to lab experiments

• very useful when its impossible or unethical to manipulate the IV in a lab / field experiment. Eg. psychologists wanting to study a naturally occurring event which would otherwise be unethical
  to impose on pps ( group of naturally stressed men aged 60-65 who have high stress levels and cholesterol )

-
- low control over EVs compared to lab experiments

• difficult to determine cause and effect due to low levels of control

Question 11

L1 what are quasi experiments

Answer 11

contain naturally occurring variable, but the naturally occurring IV is a difference between people that already exists (eg. Gender, age etc..)

researcher examines effect of this variable on the DV

an example might be that males have higher levels of testosterone than females in a sample of 50 men and 50 women aged 40-70. Psychologists might want to study this natural diff between the genders and see what effect it might have on DV (agression levels)

Quasi exp usually take place in lab setting

Question 12

L1 + and - of quasi

Answer 12

+
- high level of control, effects of EVs are minimised so exp can be more confident that the IV affected the DV

• very likely to replicate, due to strict controls, meaning it is easier to replicate the study to test reliability of findings

-
- lack ecological validity, setting is artificial exp that may not be a reflection of real-life behaviour

• demand characteristics may occur, pps may accurately or inaccurately guess experiments aim, and respond according to what they think is being investigated. This can affect data collected, and results of the study (invalid)

Question 13

L2 What is an observation? + Designs of Observations

Answer 13

When a researcher watches or listens to pps engaging in whatever behaviour is being studied

Designs = ways of organising observations and collecting required observational data
- recording data= making written notes, or vid / audio recordings mean you have a more accurate permanent record
- rating behaviour= use rating scale (eg. 1 to 10) could put each pps behaviour into several categories. Could use a coding system where every pp is given a number. Behaviour rated in this way, provides quantitive data in form of numbers

Question 14

L2 what is non pp observation

Answer 14

when the researcher does not get directly involved with the interactions of the pps and does not take part in their activities
eg. Observing pps in the gym and the psychologist stands around and does not actually engage in any exercise

Question 15

L2: + and - of non pp observation

Answer 15

+
-Researcher can study a situation in its natural setting, without altering the conditions, thus allows for a more objective view of what is occurring.
-has high mundane realism and ecological validity, observed in the pps real situation, not in artificial situation
-also has less chance of showing demand characteristics

-
-solely relies on pp behaviour, only observing behaviour, since researcher cannot interact in social behavioural processes, most data collected will be qualitative and interpretive; at some extent limited
-no deeper understanding of observed behaviour, lacking validity

Question 16

L2: participant observation

Answer 16

Researcher directly involved w interactions of the pps and will engage in activities that pp are doing, eg. observing pps in the gym and psychologist actually engages in exercise and uses gym equipment

Question 17

L2 + and - of participant observation

Answer 17

+
-researchers can spend more time w pps, allowing for greater validity of observation, and of pp patterns and behaviours. Researchers can see for themselves the picture of how the pp really lives

-
-risk of researcher getting too involved, giving bias data therefore, may begin to sympathise with the group or get too involved, thus give bias data, less objectivity
-small sample sizes need to be used, they are time consuming and hard to generalise from smaller data sample.

Question 18

L2 Covert Observation

Answer 18

Psychologist goes undercover and does not reveal their true identity, may even give themselvee a new identity.
Group does not know they are being observed
Eg. Psychologist introduces themselves as “jim” and does not tell people at the gym that he is really a psychologist and that he is observing their behaviour

Question 19

L2: + and - of covert observation

Answer 19

+
-less chance of pps showing demand characteristics, the pps dont know about the study taking place

-
-ethical concerns, as the pps are being deceived as they do not know that they are being observed

Question 20

L2: overt observation

Answer 20

Psychologist reveals true identity and may also state that they are observing the group.
Can mean observer effects can occur, as pps may change behaviour when they know they are being observed, leading to invalid results
Eg. Psychologist visits a gym, tells member of the public that they will be observed in the gym by the psychologist. Members of the public then become very aware that their behaviour is being observed

Question 21

L2 + and - of overt observations

Answer 21

+
-very ethical, no dishonesty or deception, completely honest the whole time, throughout the experiment

-
-less valid results, pps may know they are being observed, therefore guess the aim of the study, and show demand characteristics

Question 22

L2: naturalistic observation

Answer 22

Researcher observes pps in own natural environment, no deliberate manipulation of independent variable.
Eg. Psychologist watches people shopping at westfields, to see whether they are shopping alone or in a group

Question 23

L2: + and - of naturalistic observations

Answer 23

+
-pps usually unaware they are being observed (covert) so there is a reduced chance of observer effect and pps likely to act naturally, giving valid results
-high mundane realism, behaviour shown is likely to reflect everyday behaviour, observations also high in ecological validity, meaning results can be generalised to other settings and contexts

-
-impossible to have any control over EVs, meaning other variables might affect dependent variable other than IV, could make results invalid
-risk of observer bias, due to lack of control, observer might be very subjective and might be very biased when trying to interpret behaviour. Interpretations might be incorrect and could lead to unreliable results (if observation was to be repeated again, same results would not be gained again)

Question 24

L2 Controlled observation

Answer 24

Researcher observes pps in a controlled environment and this allows for manipulation of the IV
eg. Watch pps in lab setting, record their behaviour, such as inducing high levels of stress in pps by giving them a very difficult IQ test and then watch their reactions

Question 25

L2 + and - controlled observation

Answer 25

+
-cause an effect easily determined, observation highly controlled, thus psychologists can identify whether IV caused a change in the DV
-EV can be controlled for in this type of observation, therefore means results will be more valid as we can be more certain that the IV is having an effect of the DV

-
-low levels of mundane realism and ecological validity, due to high control in observation, likely to be restrictive and means results might not be accurate reflection of everyday life and may not be able to be generalised to other settings and contexts
-observer effects can occur, pps usually know they are being observed, (overt) thus pps may show social desirability bias and may behave unnaturally which means data collected may be invalid

Question 26

L2 Observer bias

Answer 26

-if observer knows purpose of study, then may observe behaviours they think meet their aims and hypothesis
-this can influence how they record their data, from the study, which might be inaccurate and subjective
-observers need to be reliable, one way to check this is to have 2 observers who each record data separately, they then correlate their observations and data together and if kappa score of +0.8 is gained, data gained from each researcher is reliable (inter-rater reliability)

Question 27

L2 Observational design concepts : behavioural categories

Answer 27

Specific types of behaviour that are being looked for in an observation.
Eg. If we wanted to observe young children playing in the school playground, make a list of some behavioural categories that we might observe, shouting, laughing, crying etc…

Question 28

L2 Sample procedures: event sampling

Answer 28

Observer decides in advance what types of behaviour they are interested in and records all occurrences
All other types of behaviour, are ignored.
Eg. If we are investigating whether people at a conference drink coffee or not. Every time a pp goes to get a cup of coffee (the event), we would log this info, eg. Make a note of the event and what happened

Question 29

L2 Sampling procedures: time interval sampling

Answer 29

Observer decides in advance, that observation will take place only during specified time periods ( 10 mins, every hour etc.. ) and records the occurrence of the specified behaviour during that period only
Eg. We might decide that we want to observe people at the conference every 30 minutes, so if we start observing at 1pm, we might decide at 1:30pm to count up how many people are drinking coffe
We would do the same at 2pm, 2:30pm etc..

Question 30

L3 self report techniques

Answer 30

Research methods in which pps give information abt themselves without researcher interference.
The pps self report info possibly w out the researcher being present for some of the techniques

Question 31

L3 Interviews

Answer 31

Involve researches asking qs in face to face situations
There are 3 types
Structured, unstructured and semi-structured

Question 32

L3 Structured Interview

Answer 32

• all pps are asked the same qqs questions in the same order
• provide quantitive data, data expressed numerically, usually consists of closed qs that have restricted answers

+
-quite quick compared unstructured interviews, require quick yes or no answers, less time consuming than longer qualitative answers in unstructured
-if pps misunderstand a q this can be clarified in an interview so that the q can be restructured/repeated to help aid understanding

-
-risk of interviewer effect (where interviewer may inadvertantly affect responders answers), can be unintentional and may even be result of the interviewers appearance, manner or gender and will affect the data collected which might be invalid
-lack quantitive quality, and in depth data, the answers are restrictive and factual, lacking meaning and depth, therefore the results might be invalid and might not be accurately measuring key variables stated in the aim

Question 33

L3 Unstructured interview + eval

Answer 33

Informal in depth conversational exchange between interviewer and interviewee
Qs are not pre planned, might consist of themes interviewer wants to discuss
Provides qualitative data, non-numerical data that uses words to give a description of what people think and feel
It usually consists of open questions requiring an in depth abt opinions eg. What are your views on recycling

+
-very useful in investigating sensitive or controsersial topics, may be a better research method than questionnaires. Unstructured interviews allow for probing of info to gain real understanding of topic being investigated

-
-social desirability bias risk, pps lie to present themselves in a positive light particularly when discussing topics that may be socially sensitive, thus results may not be accurate or valid
-interviewers need to be well trained, to accurately conduct unstructured interview, to gain relevant data.
-Can be time consuming and expensive, make sure training carried out, more training needed for unstructured than structured

Question 34

L3 semi structured interview + eval

Answer 34

Combines mix of structured and unstructured techniques, producing both quantitive and qualitative data
Consists of both open and closed qs
some qs will be pre planned and some themes later on will be ones the interviewer wants to discuss

+
-allows to develop a pattern, by using qualitative, getting more relevant info from the pp, establish pattern in pps behaviour, interviewer gains real understanding of topic being investigated

-if pps misunderstand closed q/quantitive, it can be clarified in an interview, so q can be restructured and clarified / repeated to help aid understanding

-
-open, unstructured qs are more time consuming and expensive, as you need experienced experimenter, training for unstructured is more than training for structured

-risk of interviewer effect w the structured interview qs, when interviewer may inadvertantly affect responders answers), may be unintentional, may even be result of appearance, manner or gender, and affects the data collected which may be invalid

Question 35

L4 Correlations + strength of correlations

Answer 35

Correlation is a technique for analysing the strength of the relationship between two quantitive variables, known as co-variables

Correlational analysis will show one of the three things
- positive correlation, meaning as one variable increases, the other one increases, as well (eg. As temperature increases, the number ice cream sold increases)
-negative correlation, meaning that as one variable increases the other variable decreases, eg. More time spent playing video games the less grade As they will achieve in their exams
-no correlation = no relationship between two variables, eg. Between how big your head is and how high your IQ score is

Strength of correlations
-strength of correlation can be between -1 and 1
-strong positive = number closer to +1
-0 means no correlation, coefficient closer to 0 usually means no correlation
- -1 means strong negative correlation
strength of correlation is known as “correlation coefficient”

Curvilinear = parabola

Question 36

L4 + and - of correlations

Answer 36

+
- allows psychologists to establish strength of relationship between two variables, and measure it precisely, eg. Might be strong positive relationship of +1 between variables of being happy and being wealthy

• allows researchers to investigate things that may be not able to be manipulated experimentally, for ethical or practical reasons, eg. Natural relationship between high stress and cardiovascular disease may occur, would be unethical to intentionally cause stress or CVD upon pps, but if occur naturally, its fine to study this issue

-
- cannot demonstrate cause and effect, cannot tell which variable influences the other, eg. high stress and high cholesterol levels

-correlations only measure linear relationships, does not detect curvilinear relationships, where there is a positive relationship up to a certain point, and then after the relationship becomes negative or vice versa

Question 37

L5 aims + operationalisation + hypotheses

Answer 37

Deciding upon an aim, is the first step when conducting research, an aim is a precise statement about the purpose of the study, and what it intends to find out
The aim should include what is being studied and what the study is trying to achieve

Eg. An investigation to study the effects of time and day on mood at school

operationalisation
- 2nd step, decide upon the IV and DV and operationalise them (define variable in such a way as to make them easy to measure)
Eg. In study of effect of choc on mood, operationalised IV would be type and amount of choc, operationalised DV would be mood, on scale of 1-10 (10 v happy, 0 very unhappy)

hypotheses
- 3rd step, decide upon hypothesis and operationalise it
-Hypothesis is a precise, testable statement, about expected outcome of investigation
-Differences in DV result from manipulation of IV are known as significant differences, if it has been statistically shown that the differences are highly unlikely due to chance

2 types of hypotheses
- null = states IV has no effect on DV ( choc has no sig effect on a persons mood)
- experimental = predicts IV will have effect on DV (eg. Choc will have an effect of mood), has two types

• non directional hypothesis / two tailed = does not state direction of the predicted diff between conditions ( eg. eating choc will sig affect a persons mood )
• directional hypothesis / one tailed = states direction of predicted diff between conditions eg. Eating choc will increase a persons mood
• hypothesis should also be operationalised, eg. “Eating a 250g bar of milk choc will mean that pps are sig more likely to score between 7 - 10 on a mood scale from 1-10 than if they have not consumed choc