Research Methods Flashcards
1
Q
What are the 6 ethical guidelines?
A
Consent Deception Confidentiality & privacy Debrief Withdrawal Protection from harm
2
Q
What is an independent variable?
A
The variable being manipulated or compared.
3
Q
What is a dependent variable?
A
The variable being measured and the variable that the IV could affect.
4
Q
What is an extraneous variable?
A
Any variable other than the IV that might affect the results.
5
Q
What is a confounding variable?
A
Any variable other than the IV that will affect the results; varies with the IV.
6
Q
What are the 5 steps of the scientific method?
A
1 - Ask a question 2 - State a hypothesis 3 - Conduct an experiment 4 - Analyse the results 5 - Make a conclusion
7
Q
What is a hypothesis?
A
A clear, precise testable statement stating the relationship between the variables to be investigated.
8
Q
What is a non-directional hypothesis?
A
Direction is not predicted, the groups will differ.
9
Q
What is a directional hypothesis?
A
Predicts the expected direction of the results, one group will be higher than the other.
10
Q
What are the 4 experimental methods?
A
Laboratory experiment
Field experiment
Natural experiment
Quasi experiment
11
Q
What is a laboratory experiment?
A
The IV is manipulated in a controlled environment.
12
Q
What is a field experiment?
A
The IV is still manipulated but it is carried out in the real life place.
13
Q
What is a natural experiment?
A
An experiment where the difference in IV would have happened even if the researcher had not been there.
14
Q
What is a quasi experiment?
A
The IV is not manipulated, it is something that the person just is and cannot be manipulated or changed by the experimenter.
15
Q
What are the strengths of a laboratory experiment?
A
Can show cause and effect
Creates accurate measurements due to the control
Can be replicated really easily
16
Q
What are the weaknesses of a laboratory experiment?
A
Low similarity to real-life situations
Tasks are artificial
Aware being studied due to the artificial setting (may act differently)
Total control of all variables is never possible.
17
Q
What are the strengths of a field experiment?
A
High similarity to real life
Can show cause and effect because the IV is manipulated
Participants may be unaware and won’t act differently
18
Q
What are the weaknesses of a field experiment?
A
Less control over extraneous variables
Participants don’t always know they’re taking part - unethical
Harder to replicate as the environment is not controlled
19
Q
What are the strengths of a natural experiment?
A
Ethical - IV is not manipulated
Practical
More applicable to real life (naturally occurring IV)
20
Q
What are the weaknesses of a natural experiment?
A
Can’t show cause and effect because researcher has not manipulated the IV
Less control over extraneous variables
A desired event may only happen rarely, making opportunities for research less common
21
Q
What are the strengths of a quasi experiment?
A
Ethical
Practical
More applicable to real life (naturally occurring IV)
Experimental tasks often carried out in controlled environments so some studies will share some strengths of lab experiments
22
Q
What are the weaknesses of a quasi experiment?
A
Can’t show cause and effect because researcher has not manipulated the IV
Less control over extraneous variables
It may be difficult to find participants that have the correct IV required
23
Q
What are participant variables?
A
Anything that may vary between participants which may affect the DV.
24
Q
What are situational variables?
A
Anything in the research situation that may affect the DV.
25
What is an independent groups design?
Using different participants for each condition of the experiment. Participants only take part in one condition.
26
What is repeated measures design?
Using the same participants for each condition of the experiment. Participants take part in both conditions.
27
What is matched pairs design?
Using different but matched participants for each condition of the experiment. Participants take part in one condition, matched on similar characteristics that may affect the DV.
28
What are demand characteristics?
Participants being aware of what the aim or expected results are, and what this implies for how participants are expected to behave. This may make them change their behaviour, which affects the validity of the results.
29
What are investigator effects?
Any behaviour of the researcher including interaction with participants eg. selection of participants, leading questions and bias in interpretation of results that could affect the results.
30
How can reduce investigator effects be reduced?
Randomisation
| Participants to groups, tasks A or B and order of questions.
31
How can order effects be reduced?
Counterbalancing - for repeated measures design only.
| Half of each group take part in condition 1 first and the other half do condition 2 first.
32
What are the strengths of independent measures design?
No order effects
| Less chance of demand characteristics
33
What are the weaknesses of independent measures design?
Participant variables - differ between participants
| Number of participants - need more
34
What are the strengths of repeated measures design?
Participant variables - fully controlled
| Number of participants - fewer needed
35
What are the weaknesses of repeated measures design?
Possible order effects (boredom, practice)
| More chance of demand characteristics
36
What are the strengths of matched pairs design?
No order effects
Less chance of demand characteristics
Participant variables - some are controlled
37
What are the weaknesses of matched pairs design?
Participant variables - some won't be controlled
| Number of participants - may be difficult to match them
38
What is opportunity sampling?
The researcher takes whoever is easily available at the time.
39
What is volunteer sampling?
Asking people to volunteer - individuals determine their own involvement in a study.
40
What is random sampling?
Every member of the target population has an equal chance of being selected.
41
What is systematic sampling?
Every nth person is selected from your target population.
42
What is stratified sampling?
Identify subgroups in the target population, then take a selection of participants from each subgroup but in proportion.
43
What is generalisability?
If a sample is representative of a target population then we can generalise these results and conclusions to the whole target population.
44
What are the 3 distinctions of observations?
Naturalistic/controlled
Overt/covert
Participant/non-participant
45
What is a mean?
Mathematical average (add up all the scores and divide by the number of scores)
46
What is a median?
Middle value
47
What is a mode?
Most frequent or common value in the scores
48
What is a range?
The difference between the highest and lowest score.
49
What is standard deviation?
Measures how concentrated the data are around the mean.
50
What is reliability?
Being consistent
51
What is test-retest?
The same participants do the same measure on different occasions. There should be a high correlation between the scores.
52
What is inter-observer reliability?
Consistency between different observers working on the same study by using the same behavioural category checklist by there being a positive correlation between the scores when compared.
53
What are ways to improve reliability?
Use objective measures
| Standardised method and task
54
What is internal validity?
Reducing extraneous and confounding variables such as demand characteristics, bias, lying, participant variables and situational variables.
55
What is face validity?
The measure appears to measure what it is supposed to measure.
56
What is concurrent validity?
The measure gets the same results when compared to the results of another established measures.
57
What is ecological validity?
The findings of the study can be generalised to everyday life.
58
What is temporal validity?
The findings of the study can be generalised from one time period to another.
59
What should behaviour category checklists be?
```
Observable
Measurable
Self-evident
Comprehensive
Not overlapping
```
60
Why do we use self-reports?
To capture feelings, thoughts and beliefs that are not observable.
61
What are the strengths of a self-report?
- We can capture thoughts and feelings which we can't observe.
- Standardised questionnaires and structured interviews allow easy replication. (High reliability)
- Practical questionnaires in particular allow large quantities of data to be collected quickly.
62
What are the limitations of a self-report?
- Subjective, people may interpret the questions or answer choices differently (low reliability)
- Truthfulness is an issue. Interviewer bias and social desirability bias makes it hard to gauge truthfulness (low validity)
63
What are the strengths of interviews?
- Unstructured or semi-structured interviews allow flexibility and ask further questions
64
What are the weakness of interviews?
- Strong social desirability bias because it is face to face with researcher and the interviewee wants to appeal to the researcher
- Not time efficient: only one person can be interviewed by one interviewer at a time
- Investigator bias: may contain leading questions
65
What are the strengths of questionnaires?
- Less social desirability bias than interviews
| - Time efficient: can have thousands answered at the same time, lots of data collected
66
What are the weaknesses of questionnaires?
- No flexibility, questionnaire is already created so no chance for further questions
- Investigator bias: may contain leading questions
- Doesn't provide detailed data
- Subjective questions or answer choices: people interpret them differently
67
What is an open question?
A question that allows respondents to give their own answers without limitations. Generates qualitative data.
68
What are the strengths of open questions?
- Detailed, full information is gathered
| - Responses aren't limited so participants can give true answers (more valid)
69
What are the limitations of open questions?
- Hard to compare and analyse data
| - Can do a content analysis but if there are rare responses this won't be very useful
70
What is a closed question?
A question that allows respondents to select a response that is limited.
71
What are the strengths of closed questions?
- Efficient way to collect specific data
| - Responses can be compared and analysed easily
72
What are the limitations of closed questions?
- Data lacks detail
- The limited response options may not be able to capture the range of responses participants want to give (low validity)
73
What are closed questions - rating scales?
```
Limit responses turn them in a quantitative format giving quantitative data.
EG:
0 - never
1 - Rarely
2 - Sometimes
3 - Often
```
74
What are the strengths of closed questions rating scales?
It gives a wider range of responses yet is still easy to process and analyse.
75
What are the limitations of closed questions rating scales?
The scale points may be open to subjective interpretation, this lowers reliability.
76
What are closed questions - Likert rating scales?
```
Limit responses and turn opinions into a quantitative format giving quantitative data.
EG:
1 - Strongly agree
2- Agree
3 - Not sure
4 - Disagree
5 - Strongly disagree
```
77
What are the strengths of Likert rating scales?
- A range of responses is available
- Trying each response to a number value makes it easy to analyse
- Not forced choice option of 'not sure' means data may be more valid
78
What are the limitations of Likert rating scales?
- 'Not sure' is not informative, some participants may use this response a lot
- Descriptors are open to subjective interpretation
79
What is an experimental hypothesis?
Prediction of a difference in the DV between two conditions of the IV.
80
What is a correlational hypothesis?
Prediction of a relationship between two co-variables.
81
What is an alternate hypothesis?
The actual hypothesis you are testing.
82
What is a null hypothesis?
Prediction your hypothesis is wrong: 'there will be no difference/no relationship'
83
What is a directional (one-tailed) hypothesis?
Prediction of direction of outcome.
84
What is a non-directional (two tailed) hypothesis?
Prediction of a general difference or relationship.
| Null hypotheses are always non-directional.
85
What are the 3 levels of measurement?
- Nominal
- Ordinal
- Interval
86
What is nominal data?
- Data which is in categories
- Each participant does not have a score of their own, they are counted in categories
- All behaviour category checklists are nominal data
- All coding in a content analysis is nominal data
87
What is ordinal data?
- This is continuous numerical data (scale)
- All participants have an individual score
- The data is ranked numerically
- The numbers have a relationship to each other
88
What is interval data?
- Continuous numerical data
- This is where the difference between the numbers is the same
- There is a true mathematical meaning to the numbers
89
How do you see if our result is significant?
- Work out the probability that what we have found is due to chance.
- This is expressed as a value, p.
- The smaller the value of p is, the more significant your results are.
90
What are the steps to find out if your data is significant?
- Collect your data
- Carry out a statistical test
- This gives you the calculated value
- Compare calculated value with the critical value in a statistics table
- Read the table to see if your data is significant, and if so, what level of significance (p)
91
In psychology, what percentage do we usually aim to be sure that our results are due to a real effect?
- 95% sure
- 5% probability that the null is true
- This is expressed as p=<0.05. This is known as the significance level.
- So p=<0.05 means there is an equal to or less than 5% probability that our results are due to chance.
92
What does it mean by 'the results were significant at p<0.05'?
- This means that the result is significant at 0.05 level
| - This means there is a less than 5% probability that the difference/relationship found was due to chance
93
What is a type 1 error?
This is a false positive.
| It is accepting the alternate when we should have accepted the null.
94
What is a type 2 error?
This is a false negative.
| It is accepting the null when we should have rejected it.
95
Why are statistical tests used in psychological research?
Statistical tests are used to determine the likelihood that the effect they have found is due to chance by allowing a significance level to be determined.
96
What do statistical tests for correlations give us?
A correlation coefficient
| calculated value
97
What is a one-tailed hypothesis?
A directional hypothesis
98
What is a two-tailed hypothesis?
A non-directional hypothesis
99
What is the sign test?
- A test of difference
- Use it for nominal data
- Used to analyse repeated measures or matched pairs design (related)
100
How do you carry out the sign test?
- For each participant, note whether the sign is +, - or = depending on whether the score in each condition goes down, up or stays the same.
- Count N, which is the number of participants, excluding those with equal (=) scores.
- Count the number of the least frequent sign (- or +).
- This is s.
- Use the sign test table
- See whether the results are significant by seeing the number of N, whether the hypothesis is 1 or 2 tailed and at what level of significance it is at.
101
What test is used if:
- It is a test of difference
- It is unrelated
- It uses nominal data?
Chi-squared
102
What test is used if:
- It is a test of difference
- It is related
- It uses nominal data?
Sign test
103
What test is used if:
- It is a test of difference
- It is unrelated
- It uses ordinal data?
Mann-Whitney
104
What test is used if:
- It is a test of difference
- It is related
- It uses ordinal data?
Wilcoxon
105
What test is used if:
- It is a test of difference
- It is unrelated
- It uses interval data?
Unrelated t-test
106
What test is used if:
- It is a test of difference
- It is related
- It uses interval data?
Related t-test
107
What test is used if:
- It is a test of correlation
- It uses nominal data?
Chi-squared
108
What test is used if:
- It is a test of correlation
- It uses ordinal data?
Spearman's rho
109
What test is used if:
- It is a test of correlation
- It uses interval data?
Pearson's r
110
What are the 5 key features of science?
- There must be a paradigm
- There must be theories from which hypotheses are derived and tested
- Concepts must be falsifiable
- There must be investigation using empirical methods which are replicable
- There must be general laws that govern human behaviour (generalisability)
111
What is a paradigm?
An agreed central theory about a subject.
112
What is pre-science?
No paradigm exists, and there is much debate about what the subject is and the best theoretical approach.
113
What is normal science?
A generally accepted paradigm that can account for all the phenomena related to the subject, and can explain and interpret all findings.
114
What is scientific revolution?
Evidence against the old paradigm reaches a certain point, and there is a paradigm shift. The old paradigm is replaced by a new one.
115
What is deduction?
Deriving and testing new hypotheses from a theory.
116
What is the process of deduction?
```
Theory
|
Hypothesis
|
Observation
(empirical method)
|
Confirmation (or not)
```
117
What is falsifiability?
Can it be shown to be false?
118
What is the importance of falsifiability?
Scientific theories must always be stated in such a way that the predictions derived from them could potentially be shown to be false.
119
What is the role of theory in scientific research?
Theories are a set of ideas from which hypotheses can be developed and tested. Observations will then inform further development of the theory.
120
What is the role of hypothesis testing in psychology?
Hypothesis testing allows us to gather empirical evidence through observation to support or refute our theory and to allow further theory development.
121
What must a psychologist do in order to ensure replicability of their research?
Be specific and clear - operationalise variables and use standardised methods and tasks.
122
Why is replicability important?
Our research needs to be open to scrutiny by others or our findings cannot be considered reliable.
123
Why can a researcher not generalised their conclusions if the sample is small?
A small sample is unlikely to be representative of the target population, as they may be a unique or unusual group, and so the results cannot be generalised.
124
Why can't a researcher generalise their conclusions if the experiment was set in a laboratory?
Cannot be generalised to everyday behaviour because behaviour may be different in real life - it has low ecological validity.
125
Why can't a researcher generalise their conclusions to different cultures?
Cultures are different so people may behave differently. If we generalise when we shouldn't this is known as an imposed etic.
126
What are the characteristics of a normal distribution?
- Bell shaped
- Symmetrical (around the mean)
- Mean, median and mode all at the same mid point
127
What is a skewed distribution?
- "Skewed" = distorted
- What distorts the distribution is more people getting low or high values - so the data is not symmetrical - this results in the bulk of scores being on one side and a long tail on the other side
128
What does the distribution of a negative skew look like?
A long tail on the left.
129
What is a negative skew?
- A negative skew would result from a very easy test where a lot of people did well and the rarer few did badly
- But their very low scores would drag down the mean
130
What are the characteristics of negative skews?
- Bulk of distribution is to the right of the graph
- Long tail to the left
- Mode is at top point of curve
- Mean is pulled down to the left
- Median is in the middle of these
131
What does the distribution of a positive skew look like?
A long tail on the right.
132
What is a positive skew?
- A positive skew would result from a very difficult test where a lot of people did relatively badly and the rarer few did very well
- But their very high scores would drag up the mean
133
What are the characteristics of a positive skew?
- Bulk of distribution is to the left of the graph
- Long tail to the right
- Mode is at top point of curve
- Mean is pulled down to the right
- Median is in the middle of these
