methods Flashcards
(85 cards)
1
Q
How are observations different from experiments?
A
In an experiment, the researcher manipulates the IV and measures the DV. In an observation, there is no manipulation of an IV, so you cannot draw conclusions about the causes of behaviour
2
Q
What are the 6 types of observation?
A
-naturalistic
-controlled
-overt
-covert
-participant
-non-participant
3
Q
what is a naturalistic observation? what is a strength?
A
-naturalistic observations take place where behaviour normally occurs.
-they are more ecologically valid
4
Q
what is a controlled observation? what is a strength?
A
- controlled observations take place in a set-up situtation, like a lab setting.
- they have more control over variables
5
Q
what is an overt observation? what is a limitation?
A
- Overt observations are where the particpants are aware they are being observed.
- There are more demand characteristics which could lead to participant reactivity.
6
Q
what is a covert observation? what is a strength?
A
- covert observations are when the participants are unaware they are being observed, and the researcher is hidden from view.
-there are fewer demand characteristics
7
Q
what is a participant observation? what is a strength?
A
-In a participant observation the researcher becomes part of the group being observed.
-It is useful for research on ‘closed’ groups like cults
8
Q
what is a non- participant observation? what is a limitation?
A
-In a non-particpant variation the researcher is not part of the group being observed.
-The limitation is that they are usually more objective
9
Q
how do observations compare to other methods in terms of validity?
A
-there is no manipulation of IV, so we can only observe what people do without knowing why.
-so observations have weaker internal validity than experiments
-observations measure what they are supposed to measure (what people do, not what they say they would)
-so observations have better internal validity than self reports and questionnaires
10
Q
what is a directional hypothesis? when is it used? give examples of language in a directional hypothesis.
A
- a directional hypothesis predicts the direction of an effect.
- used when earlier research suggests the difference will be in a certain direction
e.g ‘more than’ or ‘less than’
11
Q
What is a non-directional hypotheses? when is it used? give examples of language used in a directional hypotheses.
A
- predicts an effect but no direction
- they are used when research shows mixed predictions/ there is no research to suggest the direction of effect.
e.g ‘there is a difference between’
12
Q
When writing a hypothesis what must you always do?
A
-operationalise the variables
(how IV is manipulated- how DV is measured)
13
Q
what are extraneous and confounding variables?
A
- Extraneous variables are any variables over than the IV that may affect the DV
- Confounding variables are a kind of extraneous variable that varies sytematically with the IV. (so we cant tell if the change in DV is caused by the IV or the variable)
14
Q
what are demand characteristics?
A
- any cues from the researcher or the research situation that may be interpreted by participants as revealing the purpose of the investigation.
- this could lead to the participant changing their behaviour.
15
Q
what are investigator effects? how can they be controlled?
A
- any effects from the investigators behaviour (either concious or unconcious) that affects the research outcome.
- can be controlled via standardisation
16
Q
What is randomisation?
A
The use of chance methods to control the effects of bias when designing materials, and the order of conditions.
17
Q
what is standardisation? why is it used?
A
- Standardisation is using the exact same procedure and instructions for all pps in a research study.
- used to avoid investigator effects caused by different intructions/ procedures
18
Q
what are the three types of extraneous variables?
A
-
participant variables
> individual differences between pps
> may cause participant reactivity
-situational variables
>something in the research situation which affects the participants performance (e.g time of day)
-investigator effects
>effects from investigator behaviour that affects the outcome
19
Q
What is participant reactivity? what causes it? what does it lead to?
A
- participants react differently to the investigation because they know/ have clues about the experiment.
- caused by demand characteristics
- Leads to the ‘please-u effect’ which is a change of the behaviour to please the research aim… OR
> Leads to the ‘screw-u effect’ which is a change of behaviour to go against how they think the experiment wants them to behave
20
Q
How can we control extraneous variables
A
- Keep participants naive
-randomise stimulus - standardisation (reduces situational variables)
- deception (give false aim)
- single/ double blind procedures (reduces investigator effects)
21
Q
What is internal validity? what can it be affected by?
A
- Internal validity is how well the study measures what is is supposed to measure
- it can be reduced by demand characteristics (not a true measure of the variables as pps may change their behaviour) and extraneous variables (not a true show of causal relationship)
22
Q
-what is external validity? what are the types of it?
A
- external validity is if the results generalise outside of the original study.
- includes:
> ecological validity (generalised to other situations) , considers if the task is realistic compared to irl scenarios and studies
> population validity (generalised to other people), considers if the results would be the same with other types of people.
> temporal validity (generalised to other time) considers if the results would be valid over time
23
Q
what are the four types of experiment?
A
- Laboratory
- Field
- Natural
- Quasi
24
Q
What is a laboratory experiment? what are the strengths and limitations? example?
A
- Set-up, controlled conditions, involving manipulation of an IV.
strength- best control of extraneous variables
limitation- low ecological and population validity, more likely to react to DC causing participant reactivity.
example:
Most memory studies
25
What is a field experiment? What are the strengths and limitations? example?
- a field experiment is where the behaviour usually occurs, but the experimentors manipulate the IV
**strengths**- better population and ecological validity than lab experiments, low participant reactivity.
**limitations**- less control over extraneous variables
example: **Hofling's** study of obedience in nurses
26
What is a natural experiment? what are the strengths and limitations? example?
- a natural experiment is where behaviour usually occurs, and the IV is an **event**, not manipulated
**strengths**- good ecological and population validity
**limitations**-
- less control over extraneous varaibles
- no random alloactions to who undegoes what condition, - could not be considered a 'true' experiment because the IV is not manipulated
- they rarely occur.
example:
case studies of brain damage
27
what is a quasi experiment? what are the strengths and limitations? example?
- a quasi experiment is where the IV is pre-existing and cannot be manipulated.
**strengths**- can be conducted under lab conditions
**limitations**-
- not 'true' experiments
- no random allocation
example:
gender differences studies
28
what is the role of the BPS?
-produced the BPS code of ethics, which are guidelines for professional psychologists
29
what are the main ethical guidelines to be aware of?
-informed consent
-deception
-protection from harm
-debriefing
-right to widthdraw
-privacy/confidentiality
30
outline the informed consent guideline
- make participants aware of anything that might affect their **willingness to participate**
-not necessary for observations/field experiments because people are expecting to be observed
-under 16s parents must give consent
31
outline the deception ethical guideline
- **active** misleading must be avoided **wherever possible**
-decpetion can be justified if :
- there is no harm to pps
- there is a strong scientific or medical justification. (a gain in knowledge)
32
outline the protection from harm guideline
- particpants should experience **no more stress than in everyday life**
- stress should be monitored in experiments
- should provide help after in cases of unanticipated stress
33
outline the debreifing guideline
- identify things abount the study pps didnt know and tell them, **complete their understanding** of the study.
- explain any reasons for deception
- assess stress and any needs for reassurance or counselling
34
outline the right to widthdraw guideline
- participants have the right to widthdraw during a study, and they should be made aware of this and reminded during the study
- they have the right to widthdraw their data after as well.
35
outline the privacy/confidentiality guideline
- dont use any names or clues that make someone, or an institution, identifiable
- use pseudonyms in case studies (e.g HM and KF, zimbardo prison numbers)
36
Why does context affect how ethical some studies were?
- the SOC studies (asch, milgram, zimbardo) were criticised for ethical reasons, even though they were in ethical guidelines at the time.
- this still damaged the reputation of psychology
-they were all USA studies, so governed by the APA not BPS
37
What are the three types of informed consent? what are the issues with each?
- **Presumptive consent**: from other people in the target population
> ask them if the research should go ahead.
**issue**-not everyone is the same, may people have different views, people might not take it seriously if they know they wont take part.
---------------------
- **Prior general consent**: people give consent to be decieved at some point
**issue**- participants are not fully informed, could also lead to a response due to demand characteristics
----------------------
- **retrospective consent**: give permission to use their data afterwards
**issue**- not proper consent, because it is done afterwards and any damage may already be done
38
what is random sampling? what are the strengths and limitations?
**Process:**
-specify the target population, then identify the sampling population
-decide on the required sample size
-used a random method- E.G- computer method (allocate names numbers, then select random numbers) or manual method (using paper and
drawing from a hat)
**strengths**
- eliminates bias as the choice is out of the experimentors control
- enhances probability of a representative sample
**limitation**
- by chance the sample could be unrepresentative
39
What is stratified sampling? what are the strengths and limitations?
PROCESS:
- specify the population, the sampling population, and decide N (sample size)
- identify the appropriate strata, aka the groups you want represented in the sample.
- divide the sampling population into the strata
- work out the proportions for the sample to be representative
- use random sampling to choose N from each strata
**strengths**
- more likely to be representative as it avoids chance error
**limitations**
- very time consuming
40
what is systematic sampling? what are the strengths and weaknesses?
PROCESS:
-specify the population, identify sampling frame and decide sample size (N)
- work out the proportion of the sampling frame for N (e.g 20%)
- Choose each N'th member of the sampling frame (every 5th person)
**strengths**
- likely to be representative
- can be used to space out observations that have to be within a time frame (more practical)
**limitations**
- small chance of systematic error
41
what is opportunity sampling? what are the strengths/ weaknesses?
PROCESS:
- used if you dont have a sampling frame
- use whoever is available, e.g college professors may use their students
**strengths**
- less time consuming
- costs less money
- no need for sampling population
**limitations**
- less likely to be representative so therefore ungeneralisable
42
what is a volunteer sample? what are the strengths and weaknesses?
PROCESS:
- recruiting via advertisements, sometimes aimed at specific groups
**strengths**
- less time consuming
- cheaper
- no need for sampling population
**limitations**
- less likely to be representative
- specific people might volunteer and they are more likely to be helpful, caring, extraverted (personality differences)
43
what are the 3 types of experimental design and what are they?
- **independent groups**
> participants allocation to different groups with different experimental conditions
- **repeated measures design**
> all participants take part in all conditions of the experiment
- **matched pairs design**
> pairs of pps are first matyched on some variables that could affect the DV (e.g IQ), then one pp is assigned condition A and one assigned condition B
44
what are the strengths/ limitations of independent groups design?
STRENGTHS
- easy to conduct, cheap
- reduced demand characteristics
- no order effects
LIMITATIONS
-individual differences/participant variables (confounding variable- affects results)
> this can be fixed with random allocation, but not always.
45
what are the strengths/weaknesses of repeated measures design?
STRENGTHS
- no individual/participant differences
- need fewer participants
LIMITATIONS
- time consuming
- order effects (boredom or pratice)
- demand characteristics since pps do both conditions
> can be controlled with counterbalancing or deception
46
what are the strengths/limitations of the matched pairs design?
STRENGTHS
- no order effects
- no demand characteristics
- controls participant variables
LIMITATIONS
- time consuming
- perfect matching is impossible
47
what are order effects?
A confounding variable arising from the order in which the conditions are presented
>practice effect (pp gets better/improves in 2nd condition)
>boredom effect (pp gets tired/uninterested in the study and perform worse, dont give a full reult)
48
what is counterbalancing?
- a way of controlling order effects
- changes the order of conditions so half the pps experience one condition first, and the other in an opposite order
49
what is a correlation study?
-measures the relationship between two co-variables
-does not measure cause and effect, only relationship.
50
what is the difference between a correlation and experiment?
- in an experiment, the researcher manipulates the IV and measures the DV
- but in a correlational study, they just measure the relationship between the covariables
51
what are the 3 types of correlation?
- (strong) positive correlation: when as one variable increases the other increases too.
- (strong) negative correlation: when as one variable increases, the other decreases.
- no correlation
52
what is the correlation coefficient scale?
positive values:
**1.0** = perfect correlation, **0/0.2** = very weak correlation
negative values:
**-1.0** = perfect negative correlation, **0/-0.2** = very weak negative correlation
53
when can outliers have big effects on correlations?
- if the sample size is small
- if they are extremely different from other results (if they're extreme)
54
what are the weaknesses and strengths of correlations?
WEAKNESSES
- cannot draw conclusions about causal relationships because there is no manipulation
STRENGTHS
- useul when an experiment isnt practical, e.g if the manipulation of the IV would be hard to achieve
- useful when an experiment isnt ethical, e.g the iv would be harmful like stress or illness
55
what are the three interpretations of causality in correlations?
- A causes B
- B causes A
- another factor (C) could cause A and B
56
What is quantitaive data? what are the strengths?
-measures things in numbers
- collect data through things such as experiments, correlations and **closed questions** (but there is an exception)
- analysed through descriptive statistics: e.g calculations to find means, ranges
- also analysed throuhg inferential statistics: e.g generalising the results from the sample to the populatation
STRENGTHS
-more reliable
-collected in a standardised way
57
what is qualitative data? what are strengths?
- data through words, descriptions etc.
- collect data through unstructured observations, open questions, case studies
- analysed through content analysis (convert to quantative) or thematic analysis (categories)
- can also be used as **preliminary data** to provide ideas for quantitative research
STRENGTHS
-more valid
-get data in own words of participants so more detail
58
what are the exceptions for closed questions producing quantitative data?
- "yes/no" or other responses in categories such as 'what religion are you'
>in this case the data is qualitative
59
what are the exceptions for open questions producing qualitative data?
-some open questions produce a numerical answer, e.g: 'how many times did you....'
>this is then quantitative data
60
what are 3 types of closed questions:
-likert scaes (symmetrical scale)
-rating scales (choice of numbers/words)
-fixed choice options (choice of categories)
61
when writing good questions for pps, what should you avoid?
- jargon
- leading/emotive language that would guide pps to responses
- '2 in 1' questions- ambiguous
- double negatives- hard to process
- missing categories
- overlapping categories
62
in which type of observation is there a risk that relevant behaviour will not be recorded?
- **Naturalistic**
- there is no intervention by researcher so the behaviour is not guaranteed to occur
63
In which type of observation can the researcher be sure that they will obtain data?
- **controlled observation**
- there is a deliberate intervention from the researcher which should make the relevant behaviour occur
64
what are good features of good questions?
- familiar language
- unbiased wording
- questions that are easy to process
- exhaustive categories that cover all options
- exclusive categories that are easily distinct from one another
**reliable** - they are consitent with one another, measure in the same way.
**valid** they measure what is intended to.
65
difference between true random and quasi random sampling
True random is a fully random process such as chosing names from a hat or using a random generator.
Quasi random involve chosing people from an alreadt selective group, for example: systematic sampling chooses from a sampling population.
66
what is randomisation
The use of **chance methods** to control for the effects of bias when designing materials or the order of conditions
67
correlation coefficient scale
Perfect: 1.0
Very strong: 0.8-0.99
Strong: 0.6-0.8
Moderate: 0.4-0.6
Weak: 0.2-0.4
Very weak: 0-0.2
(either positive or negative correlation)
68
3 types of interviews
**structured**
- same questions used for all participants
- usually closed questions, quantitative data
**semi-structured**
- core areas for discussion but questions can be changed added after
- mix of open and closed questions
- quantiative and qualitative data
**unstructured**
- flexibility for the interview to go in whatever direction
- more open questions
- qualitative data
69
2 types of self reports
- interviews
- questionnaires - online or in person
70
advantages and disadvantages of self reports
**advantages**
- Less time consuming because data can be gathered quickly
- people are often the best source of information about their own lives.
**disadvantages**
- questionable validity- what people **say** isnt necessarily what they would do
- social desireability- pps may be unwilling to tell the truth
71
advantages of questionaires over interviews
- easier to use, researchers do not need training
- more people may disclose personal information in a questionnaire than in a face to face situation
- can be given to a large number of people so is more likely to be **representative**
- questionnaires use mainly closed questions which is easier to analgse
72
advantages of interviews over questionnaires
- researchers can check participant understanding by asking follow up questions
- researchers can adapt to the individual
- researchers may get more honest responses if they are able to build rapport with the participant
73
advantages of closed questions and open questions
**closed questions**
- yield numerical data that can be analysed statistically
- allows researchers to test hypothesis
- more reliable and comparable between pps, more precise meanings
**open questions**
- provide depth and detail, meaningful responses
- avoid participant frustration with fixed-choice responses
74
advantages of quantitative and qualitative data
**quantitative**
- numerical data that can be analysed statistically to test hypotheses
- objective, not open to interpretation or investigator bias
**qualitative**
- more detail, so more likely to measure what it is supposed to measure
(BUT- questionnaires can do this by being longer)
75
advantage of structured interview , unstructure interview, and semi-structured interview
**structured**
- use of standardised questions means responses between pps can be analysed and compared
**unstructured**
- can be personalised or adapted to capture unique experiences
**semi-structured**
- balance between structured and unstructured to include both standardised and unique elements.
76
what are descriptive statistics
- the use of graphs, tables, and summary statistics to identify trends and analyse sets of data
77
what are measures of dispersion and what are examples
general term for measures of the spread or variation in a set of scores
e.g:
**range**: subtracting lowest from highest score and adding 1
**standard deviation**: calculates how much each score deviates from the mean
78
advantages and disadvantages of mean, median, and mode
**mean**
- advantage: sensitive to all the data
- disadvantage: can be easily effected by extreme results
**median**
- advantages: not effected by extreme results
- disadvantages: less sensitive as does not use all the data
**mode**
- advantages: the only way to find 'typical' categories
- disadvantages: may not show the centre or majority of data
79
what would high and low numerical values of standard deviation indicate?
high: a wide variety of responses or results OR anomalies
low: similar responses or results
80
positive and negatively skewed distributions
examples of where you may get them
**positive skew**
- long tail is on the **positive/right** side of the peak
- most of the data is on the left
example: a difficult test
**negative skew**
- long tail is on the **negative/left** side of the peak
- most of the data is concentrated on the right
example: an easy test
81
characteristics of a negative skew
- more data on the right
- mode > median > mean
82
characteristics of positive skew
- most of data on the left
- mean > median > mode
83
what is primary and secondary data
**primary data**
- information obtained first hand by a researcher
- specific to the study
- original
**secondary data**
- information that has already been collected by someone else and pre-dates the current project
84
strengths and limitaitons of primary and secondary data
**primary**
strengths: authentic, more control of variables, internal validity
limitations: could be subjective/bias, time consuming and expensive
**secondary**
strengths: faster and cheaper
limitations: may not full match the research needs, there may be variations in quality and accuracy, could be outdated (lack temporal validity)
85
strengths and limitations of meta-analyses
**strengths**
- brings together all available evidence to produce an overall effect size
- less likely to have bias by single studies
- not impacted by outliers
- large population validity
**limitations**
- **publication bias/file drawer effect**: if research studies not supporting the hypothesis are not published they won't be included in meta-analysis. This could **distort the effect size**
- **cherry picking**: researcher only selects evidence that supports their aims
