Research Methods Flashcards
(130 cards)
1
Q
Aim
A
A general statement of what the researcher intends to investigate.
2
Q
Hypothesis
A
Statement of what researcher believes to be true relating to a study. Hypothesis should be operationalised.
3
Q
Operationalised
A
Clearly defined and measurable.
4
Q
Directional Hypothesis
A
States whether changes are greater or lesser, positive or negative.
5
Q
Non-directional Hypothesis
A
Doesn’t state the direction, just that there is a difference, correlation or association.
6
Q
Extraneous Variables
A
‘Nuisance’ variables that do not vary systematically with the independent variable. May have an effect on the dependent variable if it’s not controlled.
7
Q
Independent Variable
A
Some aspect of the experimental situation that is manipulated by the researcher - or changes naturally - so the effect on the DV can be measured.
8
Q
Dependent Variable
A
The variable that is measured by the researcher. Any effect on the DV should be caused by the change in the IV.
9
Q
Demand characteristics
A
Any cue from the researcher or research situation that may reveal the aim of the study. This may lead to a participant changing their behaviour in the research situation.
10
Q
Investigator Effects
A
Any effect of the investigator’s behaviour on the outcome of the research (the DV). May include everything from the design of the study to selection of, and interaction with, participants during research process.
11
Q
Randomisation
A
The use of chance when designing investigations to control for the effects of bias.
12
Q
Standardisation
A
Using exactly the same formalised procedures for all participants in a research study.
13
Q
Control groups
A
Used for the purpose of setting a comparison. They act as a baseline and help establish causation.
14
Q
Single blind
A
A participant doesn’t know the aims of the study so that demand characteristics are reduced.
15
Q
Double blind
A
Both participant and researcher are unaware of the aims of the study to reduce demand characteristics and investigator effects.
16
Q
Independent groups
A
One group do condition A, another group do condition B. Participants should be randomly allocated to experimental groups.
17
Q
Outline the strengths of independent groups
A
- No order effects - participants are only tested once so can’t practise or become bored/ tired as easily. This controls CVs.
- Less likely to guess the aim - only tested once, behaviour may be more ‘natural’ as a result.
18
Q
Outline the weaknesses of independent groups
A
- Participant variables - the participants in the two groups are different. May reduce the validity of the study.
- More participants - need twice as many participants as repeated measures for same data. More time spent recruiting which is expensive.
19
Q
Participant Variables
A
May act as confounding Variables in an independent groups design because people in each condition are different. This may be the cause of the change in the DV rather than the manipulation of the IV.
20
Q
Order Effects
A
Occur when participants are tested more than once (repeated measures). This might lead to a better performance through practice or worse performance due to boredom or fatigue.
21
Q
Repeated Measures
A
Same participants take part in all conditions of an experiment.
The order of conditions should be counterbalanced to avoid order effects.
22
Q
Outline the strengths of repeated measures
A
Avoids participant variables - people have to take part in both conditions so they have the same characteristics. Controls the CVs.
Fewer participants - less time spend recruiting participants, less expensive.
23
Q
Outline the limitations of repeated measures
A
Order effects - participants may do better or worse when doing a similar task twice. May reduce the validity of results.
Participants may guess the aims and change their behaviour.
24
Q
Matched Pairs
A
Two groups of participants are used but they are also related to each other by being paired on participant variables that matter for the experiment.
25
Outline the strengths of matched pairs
Participant variables - participants are matched on a variable that is relevant to the experiment. This enhances the validity of the results.
No order effects - participants are only tested once so no practice or fatigue effects can affect validity of results.
26
Outline the limitations of matched pairs
Matching isn't perfect - time consuming, can't control all relevant variables, may not fully address participant variables.
More participants - need twice as many participants as repeated measures for same data. More time is spent recruiting which is expensive.
27
Laboratory Experiment
A controlled environment where extraneous and confounding variables can be regulated. The researcher manipulates the IV and records the effect on the DV.
28
Outline the strengths of a lab experiment
EVs and CVs can be controlled - the effect of EVs and CVs on the DV can be minimised. Cause and effect between the IV and DV can be demonstrated (high internal validity).
Can be easily replicated - due to the standardised procedure, the experiment can be repeated. If the results are the same this confirms the validity.
29
Outline the limitations of a lab experiment
May lack generalisability - controlled lab environment may be artificial and participants may be aware they are being studied. Behaviour may not occur naturally and therefore has low external validity and cannot be generalised to everyday life.
Demand characteristics - may affect results
30
Field Experiment
An experiment that takes place in a natural setting. The researcher manipulates the IV and records the effect on the DV.
31
Outline the strengths of a field experiment
Natural environment - participants may be more comfortable in their own environment, results are more generalisable to everyday.
Participants are unaware they are being studied - more likely to behave normally, greater external validity.
32
Outline the limitations of a field experiment
More difficult to control CVs - observed changes in DV may not be due to the IVs. More difficult to establish cause and effect.
Ethical issues - participants may not have given informed consent. Invasion of privacy.
33
Natural Experiment
An experiment where the change in the IV is not brought about by the researcher but would have happened even if the researcher had not been there. Researcher records effect on DV.
34
Outline the strengths of a natural experiment
May be the only ethical option - it may be unethical to manipulate the IV, e.g. studying the effects of institutionalisation on children, so NE may be the only causal research that can be done for such topics.
Greater external validity - involve real-life issues so findings are more relevant.
35
Outline the limitations of a natural experiment
The natural event may only occur rarely - reduces opportunity for research and limit the scope for generalising findings to other situations.
Participants aren't randomly allocated - experimenter has no control over which participants are placed in which condition as the IV is pre-existing. May result in CVs that aren't controlled.
36
Quasi-experiment
IV is based on a pre-existing difference between people, e.g. age or gender. No one has manipulated this variable, it simply exists. DV may be naturally occurring (e.g. exam results) or may be measured by the experimenter.
37
Outline the strengths of a quasi experiment
Often high control - shares some strengths of lab experiments.
Comparisons can be made between people - the IV is a difference between people e.g. people with autism and people without. This means comparisons can be made.
38
Outline the limitations of a quasi experiment
Not randomly allocated - participant variables may have caused the change in the DV.
Causal relationships not demonstrated - researcher doesn't manipulate/ control the IV. We cannot say for certain that any change in the DV was due to the IV.
39
Population
The large group of people that a researcher is interested in studying.
40
Sample
A group of people who take part in a research investigation. Drawn from a target population and it is presumed to be representative of that population.
41
Generalisation
The extent to which findings and conclusions from an investigation can be broadly applied to the population.
42
Bias
Over or under-represented sample. Limits the extent to which generalisations can be made to target population.
43
Opportunity Sample
People who are most available i.e. ones who are nearest/ easiest to obtain.
- Asking people nearby.
44
Evaluate opportunity sampling
✓ Quick method - convenient, one of the most popular sampling methods.
✗ Inevitably biased - sample is unrepresentative, drawn from a very specific area. Findings cannot always be generalised.
45
Volunteer Sample
Participants select themselves. Researchers advertise, for example, by placing an ad in the newspaper.
46
Evaluate volunteer sampling
✓ Participants are willing - know how much time and effort is involved and so are likely to engage more.
✗ Likely to be a biased sample - participants may share the same characteristics, e.g. keen and curious. Generalisation is limited.
47
Random Sample
Every person in the target population has an equal chance of being selected.
- Lottery method - all members are given a number and placed into a hat.
48
Evaluate random sampling
✓ Free from researcher bias - researcher has no influence over who is selected.
✗ Representation not guaranteed - still could produce a biased sample which limits ability to generalise.
49
Systematic Sample
Participants are selected using a set 'pattern' (sampling frame)
- Every nth person is selected from a list of the target population.
50
Evaluate systematic sampling
✓ Unbiased - first item is usually selected at random. Objective method.
✗ Time and effort.
51
Stratified Sample
Participants are selected according to their frequency in the target population.
- Subgroups (or strata) are identified, e.g. gender, age. Relative percentages of strata in population are reflected in the sample.
52
Evaluate stratified sampling
✓ Representative method - generalisability more likely.
✗ Stratification is not perfect - strata can't reflect all the ways in which people are different. Complete representation isn't possible.
53
Ethical issues
When a conflict exists between the rights of participants and the aims of the research.
54
BPS code of conduct
A quasi-legal document which protects participants based on four principles: respect, competence, responsibility, integrity.
55
Ethics committees
Weigh up costs (e.g. potential harm) and benefits (value of research) before deciding whether a study should go ahead.
56
Informed Consent
Participants should be able to make an informed judgement about whether to take part. Alternative forms of consent:
• Presumptive - ask a similar group.
• Prior general - agree to be deceived.
• Retrospective - get consent after the study.
57
Deception
Deliberately misleading or withholding information so consent isn't informed. Participants should be debriefed.
58
Debriefing
Should include an explanation for the following:
• True aims of the investigation.
• Details that were not given during the study e.g. existence of other condition.
• What their data will be used for.
• Their right to withhold data.
59
Protection from harm
Participants should be at no more risk than they would be in everyday life. Should be given the right to withdraw. Researcher should provide counselling if participants have been distressed.
60
Privacy/ confidentiality
We have the right to control information about ourselves. If this is invaded, confidentiality should be respected. Data cannot be shared with other researchers, personal details should be protected, researcher should refer to participants using initials, numbers or pseudonyms.
61
Positive Correlation
Co-variables rise or fall together.
62
Negative Correlation
One co-variable rises and the other falls.
63
Zero Correlation
No relationship between the two variables.
64
Differences between correlations and experiments
In a correlation, there is no manipulation of variables so cause and effect cannot be established. The influence of EVs is not controlled so it may be a third 'untested' variable causing the relationship between the two co-variables (intervening variable).
65
Evaluate correlations
✓ Useful starting point for research - provides a precise measure for how two variables are related. May suggest hypotheses for future research.
✓ Economical - no need for controlled environment, no manipulation of variables. Less time-consuming.
✗ No cause and effect - there may be intervening variables that explain the relationship.
✗ Method used to measure variables may be flawed - reduces validity.
66
Observational techniques
A way of seeing or listening to what people do without having to ask them. Used within an experiment as a way of assessing the DV.
67
Evaluate observational techniques
✓ Can capture unexpected behaviour - give insight into spontaneous behaviour.
✗ Risk of observer bias - interpretation of the situation may be affected by variables. Can be reduced using more than one observer.
68
Naturalistic observation
Takes place where target behaviour would normally occur.
69
Evaluate naturalistic observations
✓ High external validity
| ✗ Low control - uncontrolled EVs make it more difficult to detect patterns.
70
Controlled observations
Some control/ manipulation of variables including control of EVs.
71
Evaluate controlled observations
✓ Can be replicated due to standardised procedures.
| ✗ Low external validity - behaviour may be contrived.
72
Covert observations
Participants are unaware they are being studied.
73
Evaluate covert observations
✓ Free from demand characteristics which increased validity.
| ✗ Ethics - right to privacy may be affected.
74
Overt observations
Participants are aware of being studied.
75
Evaluate overt observations
✓ Ethically acceptable
| ✗ Demand characteristics - reduces validity of findings.
76
Participant observations
When the researcher becomes part of the group they are studying.
77
Evaluate participant observations
✓ Can lead to greater insight - enhances validity of findings.
✗ Possible loss of objectivity - researcher may identify too strongly with those in the study (going native).
78
Non-participant observations
✓ More objective- less chance of bias, increased validity.
| ✗ Loss of insight - may be too far removed from situation. May reduce validity of findings.
79
Behavioural categories
Target behaviour to be observed should be broken up into a set of observable categories. This is similar to the idea of operationalisation.
80
Evaluate behavioural categories
✗ Difficult to make clear and unambiguous - shouldn't overlap which isn't possible to achieve.
✗ Dustbin categories - all forms of behaviour should be in the list. 'Dumped' behaviours go unrecorded.
81
Time Sampling
Observations made at regular intervals - e.g. once every 15 seconds.
82
Evaluate time sampling
✓ Reduces the number of observations - more structured and systematic.
✗ May be unrepresentative - researcher may miss important details outside of the time-scale so may not reflect whole behaviour.
83
Event sampling
A target behaviour/ event is recorded every time it occurs.
84
Evaluate event sampling
✓ May record infrequent behaviour - researcher will still pick up behaviours that don't occur at regular intervals. These are missed in time sampling.
✗ Complex behaviour oversimplified - if the event is too complex, important details may go unrecorded. May affect the validity of findings.
85
Unstructured observation
Everything is recorded which can be quite difficult if a lot is going on.
86
Structured observation
Includes behavioural categories and sampling methods.
87
Questionnaires
Made up of a pre-set list of written questions to which a participant responds. Can be used as part of an experiment to assess the DV.
88
Evaluate the use of questionnaires
✓ Can be distributed to lots of people - reduces effort, cost effective.
✓ Respondents may be willing to open up - less self-conscious than when in an interview. Less chance of social desirability bias.
✗ Responses may not always be truthful - social desirability bias.
✗ Respondents may favour a particular kind of response e.g. they always agree.
89
Interviews
Face-to-face interaction between an interviewer and interviewee.
90
Structured Interview
List of pre-determined questions asked in a fixed order.
91
Evaluate the use of structured interviews
✓ Easy to replicate - standardised format.
| ✗ Interviewees cannot elaborate.
92
Unstructured Interview
No set of questions. There is a general topic to be discussed. Interaction is free-flowing and interviewee is encouraged to elaborate.
93
Evaluate the use of unstructured interviews
✓ Greater flexibility - insight into interviewee's worldview.
✗ Difficult to replicate - greater risk of interviewer bias.
94
Semi-structured interviews
List of questions that have been worked out in advance but interviewers are free to ask follow-up questions when appropriate.
95
Normal Distribution
Symmetrical, bell-shaped curve. Most people are in the middle area of the curve with very few at extreme ends. Mean, median and mode all occupy the same mid-point of the curve.
96
Skewed Distributions
Distributions that lean to one side or the other because most people are either at the lower or upper end of distribution.
97
Negative Skew
Most of the distribution is concentrated towards the right of the graph, resulting in a long tail on the left.
98
Positive Skew
Most of the distribution is concentrated towards the left of the graph, resulting in a long tail on the right.
99
Correlation coefficient
Represents the strength of the correlation. Statistical tests of correlation produce a numerical value somewhere between -1 and +1 = correlation coefficient. This value tells us the strength of the relationship between the two variables.
The closer the coefficient is to 1, the stronger the relationship between co-variables.
Closer to zero = weaker.
100
Correlation Coefficient - direction of the correlation
Value of +1 represents a perfect positive correlation.
Value of -1 represents a perfect negative correlation.
Correlation coefficient of +.50 is as strong as -.50. Sign just informs us of the direction. Calculated using an inferential test, such as Pearson's or Spearman's.
101
Case Studies
* Detailed, in-depth analysis of an individual. May be longitudinal, may involved gathering data from family and friends of the individual as well as the person themselves.
* Often involve analysis of unusual events/ individuals e.g sequence of events that led up to 2011 London riots.
* Usually involve qualitative data - case history of person/ event constructed.
102
Evaluation of Case Studies
✓ Rich, detailed insight - preferred to more 'superficial' forms of data that might be collected (from an experiment assessing one aspect of behaviour). Increases validity of data.
✓ Enable study of unusual behaviour - some behaviours are rare (e.g. HM) and cannot be studied using other methods. Can help understanding of 'normal' functioning.
✗ Prone to researcher bias - subjective interpretations may reduce validity.
✗ Participants' accounts may be biased - may be prone to inaccuracy/ memory decay. Low in validity.
103
Content Analysis
• People are studied indirectly through their communications - type of observational research. May include spoke interaction (conversation), written forms (texts, email), examples from the media (books, TV).
• Coding may produce quantitative data - first stage of content analysis. Information needs to be categorised into meaningful units. May involve counting the number of times a particular word/ phrase appears.
E.G. newspaper reports may be analysed for the number of times they refer to mentally ill as 'mad' or 'crazy'.
• Thematic analysis produces qualitative data - a theme in content analysis refers to any idea that is recurrent. These themes are more descriptive. E.G. mentally ill may be referred to as a 'threat to our children'/ 'drain on the NHS'. Can be developed into broader categories such as 'stereotyping' of the mentally ill.
104
Evaluation of Content Analyses
✓ Many ethical issues may not apply - material to studies (TV, film) may already be in public domain, no issues with obtaining consent.
✓ A flexible method - can be adapted to produce both quantitative and qualitative data. Can be adapted to suit the aims of the research.
✗ Communication is studied out of context - researcher may attribute motivations to the speaker/ writer that were not intended. Reduces validity of conclusions.
✗ May lack objectivity especially when more descriptive forms of thematic analysis are used. Bias may threaten validity. However, reflexivity is a method of addressing lack of objectivity.
105
Reliability
Reliability is a measure of consistency. If a particular measurement is repeated and the same result is obtained then that measurement is described as reliable.
106
Assessing reliability
* Test-retest: test the same person twice. The results should be the same or very similar each time it is administered if it is reliable.
* Inter-observer: compares observations from different observers. Two or more observers compare the data by conducting a pilot study to check that observers are independently applying behavioural categories in the same way.
* Reliability is measured using a correlation - in test-retest and inter-observer reliability, the two sets of scores are correlated. The correlation coefficient should exceed +.80 for reliability.
107
Improving reliability
* Questionnaires - rewrite questions. Researcher may replace some open questions with closed, fixed choice alternatives which may be less ambiguous.
* Interviews - improved training. Use the same experimenter each time, if this isn't possible, all interviewers must be trained so they avoid leading questions.
* Experiments - standardised procedures.
* Observations - operationalisation of behavioural categories. Should be measurable (e.g. pushing is less open to interpretation than aggression) Categories shouldn't overlap and all possible behaviours should be included.
108
Validity
Whether an observed effect is genuine and represents the real world.
• Data can be reliable but not valid.
• Ecological validity - can findings be generalised to other settings in everyday life.
• Temporal validity - findings should be consistent over time. E.G. Asch's study may lack temporal validity because it was conducted in a conformist era in American history.
109
Assessing Validity
* Face Validity: whether a test looks like it measures what it should. This is achieved by passing it to an expert to check.
* Concurrent Validity: whether findings are similar to those on a well-established test.
110
Improving Validity
* Experiments - control group (researcher can be confident that changes in DV are due to IV) and standardisation (minimise the impact of participant reactivity and investigator effects).
* Questionnaires - lie scale and confidentiality control for the effects of social desirability bias, respondents are assured all data is confidential.
* Observations - good categories, well defined, not overlapping.
* Qualitative research - interpretive validity demonstrated through coherence of the reporting and inclusion of direct quotes from participants, triangulation involves using a number of different sources as evidence (interview data, personal diaries, etc.)
111
Nominal Data
Categories.
Each item can only appear in one category. There is no order.
E.G. people naming their favourite football team.
112
Ordinal Data
Placed in order, intervals are subjective.
Data is collected on a numerical, ordered scale but intervals are variable. Ordinal data lacks precision because it is based on subjective opinion rather than objective measures.
113
Interval data
Units of equal size.
Interval data is based on numerical scales that include units of equal, precisely defined size. Interval data is 'better' than ordinal data because more detail is preserved as the scores are not converted to ranks.
114
Probability and Significance
If the statistical test is not significant then the null hypothesis is accepted.
The null hypothesis is accepted or rejected at a particular level of probability.
115
Using Statistical Tests
• The usual level of significance is 0.05 (5%). This means the probability that the observed effect occurred by chance is equal to or less than 5%.
• The calculated value is compared with and a critical value based on probabilities.
• To find the correct critical value:
Hypothesis one-tailed (directional) or two-tailed (non-directional)
Number f participants or degrees of freedom.
Level of significance (p value).
116
Type I error
The null hypothesis is rejected and the alternative hypothesis is accepted when the null hypothesis is true.
This is an optimistic error or false positive as a significant difference or correlation is found when one doesn't exist.
More likely to be made if the significance level is too lenient.
117
Type II error
The null hypothesis is accepted, but, in reality, the alternative hypothesis is true.
This is a pessimistic error or false negative.
More likely if the significance level is too stringent, as potentially significant values may be missed.
118
Features of Science: Paradigms and Paradigm Shifts
Kuhn said that what distinguishes scientific disciplines from non-scientific disciplines is a shared set of assumptions and methods (a paradigm).
Social sciences lack a universally accepted paradigm and are best seen as a 'pre-science' unlike natural sciences such as biology.
Paradigm shifts occur when there is a scientific revolution. A handful of researchers begin to question the accepted paradigm when there is too much contradictory evidence to ignore.
119
Features of Science: Theory Construction
A theory = a set of general laws or principles that have the ability to explain particular events or behaviour.
Testing a theory depends on being able to make clear and precise predictions on the basis of the theory.
A hypothesis can then be tested using scientific methods to determine whether it will be supported or refuted.
The process of deriving a new hypothesis from an existing theory is known as deduction.
120
Features of Science: Falsifiability
Popper argued that the key criterion of scientific theory is falsifiability. Genuine scientific theories should hold themselves up for hypothesis testing and the possibility of being proved false.
Pseudosciences cannot be falsified.
121
Features of Science: Replicability
If a scientific theory is to be trusted, the findings from it must be shown to be repeatable across a number of different contexts. By repeating a study, we can see the extent to which the findings can be generalised and the validity of the research.
122
Features of Science: Objectivity
Scientific researchers must keep a 'critical distance' during research. They must not allow personal opinions or biases to 'discolour' the data or influence the behaviour of patients.
Those methods in psychology that are associated with the greatest level of control (lab) tend to be most objective.
123
Features of Science: Empirical Method
Empiricism = experience.
Empirical methods emphasise the importance of data collection based on direct, sensory experience.
The experimental method and the observational method are good examples of the empirical method in psychology.
Early empiricists (John Locke) saw knowledge as determined only by experience and sense perception. A theory cannot claim to be scientific unless it has been empirically tested.
124
Measures of central tendency
Averages which give us information about the most typical values in a set of data.
125
Mean (measures of central tendency)
✓ Most sensitive of measures - all scores/ values in data set are included within the calculation. More representative of the data as a whole.
✗ Easily distorted by extreme values
126
Median (measures of central tendency)
✓ Extreme scores do not affect it. Also easy to calculate.
✗ Less sensitive than the mean, not all scores are included in final calculation.
127
Mode (measures of central tendency)
✓ Easy to calculate. For some data (categories) the mode is the only method you can use.
✗ Crude measure - not representative of data as a whole.
128
Measures of dispersion
Based on the spread of scores.
129
Range (measure of dispersion)
✓ Easy to calculate.
✗ Only takes into account the two most extreme values - may be unrepresentative of the data as a whole.
130
Standard Deviation (measure of dispersion)
✓ More sophisticated measure of dispersion. More precise measure than the range as it includes all values within final calculation.
✗ Like the mean, it can be distorted by a single extreme value.
