Research Methods Flashcards
1
Q
Define: Aim
A
A general statement about the purpose of an investigation which is to be tested.
2
Q
Define: hypothesis
A
A precise, testable statement about the expected outcome of an experiment.
- Experimental/Alternative hypothesis: used in the context of an experiment (H1)
- Null hypothesis: shows that there will be no effect nor relationship as stated in the hypothesis (H0)
3
Q
What is a directional hypothesis?
A
~(One tailed hypothesis)
~Predicts the nature or direction of the outcome, more precise than a non-directional hypothesis.
~Specifically states the direction of the results.
4
Q
What is a non directional hypothesis?
A
~(Two tailed hypothesis).
~Doesn’t predict the direction that the result will go in.
5
Q
When will psychologists use a directional or non-directional hypothesis?
A
Directional = when past research (theory/study) suggests that results will go a particular way.
Non-directional = when past research is unclear or contradictory, or does not exist at all.
6
Q
Define: Variable
A
A variable is the precise, technical term that psychologists use for something that can change/vary. Such as a quality, a characteristic or an action.
7
Q
Define: Operationalisation
A
A statement or form that is testable involving the IV and DV. This enables the research to be replicated.
8
Q
Define: Confounding variable
A
A variable that confuses the experiment as it varies systematically with the IV and so may also have an effect on the DV.
9
Q
List the 3 features of a true experiment
A
- Randomisation
- Control
- Manipulation of an independent variable
10
Q
What is meant by randomisation?
A
A true experiment requires that participants are randomly allocated to conditions.
11
Q
When will an independent variable be used?
A
When a researcher wants to establish a cause and effect between two variables, in this case manipulation of a variable is needed to see how it affects the dependent variable.
12
Q
What is meant by control?
A
Efforts made to control or hold constant all variables.
- independent and dependent
- extraneous variable: obstacles that get in the way of the experiment
13
Q
What is meant by manipulation of an independent variable?
A
The experimenter controls the independent variable and measures the dependent variable.
14
Q
What are the four types of experiment
A
- Laboratory
- Field
- Natural
- Quasi - experiment
15
Q
Give an advantage of experimental methods
A
Experimental methods provide the most rigorous way of testing a hypothesis because it seeks to establish cause and effect relationships (causal relationship).
16
Q
Define: Natural experiment and evaluate
A
An investigation in which the researcher cannot directly manipulate the IV to measure the DV. (Naturally occurring event).
- Increased mundane realism/ecological validity.
- Cannot establish causal relationship as IV is not manipulated. Random allocation not possible (threat to internal validity).
17
Q
Define: laboratory experiment, and evaluate
A
A laboratory experiment is an experiment conducted under highly controlled conditions.
- Causal relationship can be identified. Internal validity.
- Lacks mundane realism and low ecological validity.
18
Q
What are the similarities between laboratory and field experiments?
A
Causal relationship established.
IV is directly manipulated.
19
Q
Define: field experiment, and evaluate
A
A field experiment is an experiment that is conducted in a real world situation.
- Participants are not usually aware that that they are participating in an experiment.
- More time consuming and expensive.
20
Q
Define: Quasi experiment and evaluate
A
The independent variable is actually not something that varies at all, it is a condition.
- Allows for clear comparison.
- Can only be used where conditions vary naturally.
21
Q
What are the differences between laboratory and field experiments?
A
Controlled vs. natural (lab. Is more highly controlled).
In field participants are not planned.
22
Q
What are the 6 ethical issues (acronym)?
A
Privacy Confidentiality Deception Right to withdraw Informed consent Protection from harm
23
Q
ETHICAL ISSUE: Right to withdraw.
Describe it and explain how to deal with it and it’s limitations.
A
~Description: participants should be able to leave a study at any time if they feel uncomfortable, they should also be allowed to withdraw their data.
~How to deal with it: they should be told at the start of the study that they have the right to withdraw.
~Limitation: participants may feel they shouldn’t withdraw as this may affect the study. Many participants are paid or receive course credits, they may worry they won’t get this if they withdraw.
24
Q
ETHICAL ISSUE: Protection from harm
Describe it and explain how to deal with it and it’s limitations.
A
~Description: participants should not experience negative, physical or psychological effects of participating.
~How to deal with it: you should avoid any risks that are greater than every day life. The study should be stopped if you think they are being harmed.
~Limitation: researchers are not always accurately able to predict the risks of taking part in a study.
25
```
ETHICAL ISSUE (dealing with it): Debriefing
Describe it and explain how to deal with it and it's limitations.
```
~Description: If deception and a lack of informed consent have occurred, then the debrief provides a way to resolve these issues. Issues will arise if the debrief does not effectively address these issues and if it fails to relieve distress.
~How to deal with it: at the end of the study the researcher should provide detailed information about the research and answer any questions the participants may have.
~Limitation: it is not always easy to monitor the unforeseen negative effects so active intervention can be difficult.
26
ETHICAL ISSUE: Privacy
| Describe it and explain how to deal with it and it's limitations.
~Description: freedom from the observation, intrusion or attention of others.
~How to deal with it: do not observe any one without informed consent unless in a public place.
~Limitation: no universal agreement as to what constitutes a public place.
27
ETHICAL ISSUE: Informed consent
| Describe it and explain how to deal with it and it's limitations.
~Description: participants should be given comprehensive information regarding the nature and purpose of the research and their role in it. They can then make a decision about whether to participate or not.
~How to deal with it: an alternative is to gain presumptive consent. This is when others are asked how they would feel, if a planned study is acceptable or not.
~Limitation: if you gain informed consent then this involves telling the participants the exact nature of the study this means that they might change their behaviour. Presumptive consent is a problem as what people expect they will not mind may be different from actually experiencing it.
28
ETHICAL ISSUE: Deception
| Describe it and explain how to deal with it and it's limitations.
~Description: this is when a participant is not told about the true aims of the study.
~How to deal with it: The need for deception should be approved by an ethical committee and costs/benefits weighted up. Participants should be debriefed after the study, this involves telling them about the true nature of the study.
~Limitation: cost-benefit decisions are flawed as they are subjective decisions. Debriefing can't turn the clock back, people may still feel upset or embarrassed.
29
ETHICAL ISSUE: Confidentiality
| Describe it and explain how to deal with it and it's limitations.
~Description: A participants right to have personal information protected.
~How to deal with it: researchers should not record the names of any participants.
~Limitation: it is sometimes possible to work out the identity of a participant from information provided, therefore confidentiality is not always possible.
30
What is meta-analysis?
When a researcher looks at the findings from a number of different studies and produces a statistic to represent the overall effect.
31
What are the three types of extraneous variables?
* Participants variables
* Situational variables
* Investigator variables
All are controlled for a valid experiment.
32
When can ethical guidelines be ignored?
If the cost to the participant is less than the benefit to the research. Sticking to guidelines could jeopardise results.
33
How can participant variables be controlled?
Individual differences found among participants could change the results of a study but can be limited by:
•Large sample size - so that extreme cases have less effect on the overall result.
•Random allocation - so that outliers are spread over groups.
•Repeats - using the same participants in all conditions to identify any anomalies.
34
What are the 4 principles (code of conduct) of the BPS?
1. Respect
2. Competence
3. Responsibility
4. Integrity
35
How can situational variables be controlled?
Differences in the setup of the experiment such as the environment can affect results, this can be limited by:
•Standardise procedure - everyone follows the same order/timings/equipment, etc.
•Standardise instructions - simple same instructions that everyone can follow.
•Counterbalancing - order of tasks are switched in different conditions so that's differences average out.
36
How can investigator variables be controlled?
Differences due to the person running the experiment such as body language alters the response. So does confirmation bias (investigator has an opinion which they influence on participants) and demand characteristics (cues in environment which help participants to work out what the research hypothesis is). This can be limited by:
•Placebo condition - some people are left out to see if behaviour changes due to researcher.
•Single blind - participants don't to know what the study is about.
•Double blind - participants and investigator don't know what the study is about.
37
Define: Reliability
A measure of consistency in which two or more measurements/observation of the same psychological event, will be consistent with each other.
Methods: test-retest reliability and inter-observer reliability
38
Describe test-retest as a way of assessing reliability
A method of assessing the reliability of a questionnaire/psychological test/interview, by testing the same person on two separate occasions.
* Shows to what extent the test (or other measure) produces the same answers.
* There must be sufficient time between the test and retest (but not too long so that attitudes/abilities may change).
* Two sets of scores should then be correlated to see if it is significant and positive.
39
Describe inter-observer as a way of assessing reliability
The extent to which there is agreement between two or more observers involved in observations of behaviour.
* Measures reliability by correlating the observations: total no. of agreements / total no. of observations > +.80
* Content analysis = inter-rater reliability
* Interviews = inter-interviewer reliability
40
How can reliability be improved in
A) Questionnaires
B) Interviews
C) Experiments
D) Observations
A) Questionnaires - replace some of the open questions with closed questions, so that questions are less ambiguous and there is less room for misinterpretation.
B) Interviews - same interviewer each time or properly train interviewers to not ask misleading questions. Structured interviews with fixed questions are best.
C) Experiments - favour Lab as more variables are controlled which leads to preclude replication.
D) Observations - behavioural categories should be properly operationalised, measurable and self-evident. Also shouldn't overlap and all possible behaviours should be covered.
41
Define: Validity
The extent to which an observable effect is genuine and can be generalised, or represents the real world as the observed effect can go beyond its research setting.
* Internal validity - what happens inside the experiment, no confounding/extraneous variables (face, concurrent).
* External validity - can be generalised outside the experiment (ecological, population and temporal).
42
How can validity be improved in
A) Experimental research
B) Questionnaires
C) Observations
D) Qualitative methods
A) Experimental research - use of control group to see if changes in DV are due to effect of IV. Standardised procedures limit investigator effects. Use of single/double blind reduces demand characteristics.
B) Questionnaires - incorporate a lie scale (2 Qs that ask the same thing differently) to assess the consistency of the respondent and control effects of social desirability bias. Assure respondents that their data will remain anonymous.
C) Observations - Covert is best as behaviour is more likely to be natural and authentic. Behavioural categories shouldn't be too broad, overlapping or ambiguous.
D) Qualitative methods - favoured over quantitate as associated depth and detail is better at reflecting participants reality. Use of triangulation- number of different sources as evidence.
43
Define: Mundane realism
How a study mirrors the real world. The research environment is realistic to the degree to which experiences encountered in the research environment will occur in the real world.
44
Define: ecological validity
To what extent results can be generalised to different (outside) settings.
45
Define: historical/temporal validity
How well an experiment/study/research can be generalised to different times.
46
Define: Population validity
Refers to the extent to which the results can be generalised to groups of people other than the sample of participants used.
47
Define: face validity
The extent to which the measurement actually measures what it claims to be measuring, assessed at face value.
48
Define: concurrent validity
The extent to which a psychological measure relates to an existing similar measure.
50
Define: Opportunity sampling
| Give a pro and con
Consists of taking the sample from people who are available at the time the study is carried out and fit the criteria you are looking for.
- Relatively easy to create
- Can be unrepresentative as sample is taken from small section of the population.
50
Define: Random sampling
| Give a pro and con
Sample in which every member of the population has an equal chance of being chosen.
- unlikely sample will be bias, no control
- Can take a long time
51
Define: stratified sampling
| Give a pro and con
A proportional representation of the target population. The target population is broken down into smaller groups such as gender and age.
- very representative
- time consuming, difficult
52
Define: Volunteer sampling
| Give a pro and con
Consists of participants becoming part of a study because they volunteer when asked or in response to an advert.
- participants are motivated, easier access
- similar volunteers, demand characteristics
53
Define: systematic sampling
| Give a pro and con
Taking the nth person from a list to create a sample.
- no bias
- may include a frequency trait, unrepresentative sample
54
Define: experimental design
Refers to how participants are used in an experiment.
| Either as independent groups, repeated measures or matched pairs.
55
Describe independent groups and evaluate it
-Each participant is only used in one condition.
~Example: to test the effect of caffeine on mood, one group of participants are given coffee, the other group aren't. Mood is then assessed before and after.
* Lower risk of demand characteristics. No order effects such as boredom, fatigue or practice effects.
* Higher risk of participant variables. More participants are needed.
56
How can the limitations of independent groups design be dealt with?
Random allocation
57
Describe repeated measures and evaluate it
-Each participant takes part in every condition.
~To test the effect of caffeine on mood, participants have their mood assessed, drink a cup of coffee, then are assessed again.
* No participant variables as they are in 'both' groups. Fewer participant needed.
* Risk of demand characteristics. Risk of order effects which gives rise to extraneous variables.
58
How can the limitations of a repeated measures design be dealt with?
Counterbalancing - The order of each condition is switched.
| Eg. half of the participants do task A first whilst the other half do task B first.
59
Describe matched pairs and evaluate it
-Participants in the group are matched with each other, so that the two groups are similar as possible. Ideal match=identical twins
~Example: before the study two groups of participants are matched on age, gender, etc...one group is given coffee, the other isn't. Mood is then assessed.
* Participant variables are reduced. No order effects. Less risk of demand characteristics.
* Participants are never completely matched. Time consuming and expensive. More participants needed.
60
How can the limitations of matched pairs design be dealt with?
Restrict the number of variables to match on.
| Conduct a pilot study.
61
Define: Interviews
These are face to face conversations between the researcher and participant.
Usually contain more open questions. Can be structured (verbal questionnaire) or unstructured (basic outline).
Tend to generate qualitative data, more validity.
62
Evaluate the use of interviews
Structured: easy to analyse. Easily repeated. Data is restricted. Social desirability. Harder to compare. Interview bias.
Unstructured: more detailed. Lacks objectivity. Requires skilled interviewers, expensive.
63
Define: demand characteristic
A cue that makes participants unconsciously aware of the aims of the study.
64
How can the affects of demand characteristics/investigator effects be dealt with?
Single blind or double blind.
| By participant or experimenter not being aware of the research aims, cues are prevented.
65
What are the 4 different types of questions?
1. Open questions-allow people to elaborate on their answers and to give more detail. Generates qualitative data but are difficult to analyse.
2. Closed questions- Force people to choose an option. Easy to analyse and generate quantitative data but lack detail.
3. Likert scaling-assesses the strength of a participants opinion. They are asked to rate their feelings on a particular topic.
4. Contingent questions-depend on the answer to the preceding question.
66
What are the 3 problems associated with self report?
Lack of validity
Social desirability
Demand characteristics
67
Evaluate the use of questionnaires
Easy to make. Easy to distribute to a large amount of people. Respondents may feel more willing to reveal personal/confidential info than in an interview.
Bias questions. No administrator. Social desirability bias.
68
What are the 2 types of observations?
Participant observation: The observer acts as part of the group being observed.
Non-participant observation: The observer doesn't become part of the group being watched.
69
What is the difference between overt and covert observation?
Overt: participants are aware that they are being observed.
Covert: participants are unaware of being observed.
70
What is the difference between structured and unstructured observation?
Structured: determines the behaviour to be observed - the behavioural category. Determines the sampling to be used.
Unstructured: The observer records everything.
71
What is the difference between a naturalistic and controlled observation?
In a naturalistic observation, behaviour is studied in a natural situation where everything has been left as it is normally.
In a controlled observation, some variables in the environment are regulated by the researcher. Participants are more likely to know they are being studied.
72
What 2 sampling methods are used in observations?
Time sampling: observations may be made at regular time intervals.
Event sampling: keep a tally chart of each time a type of behaviour occurs.
73
Evaluate the use of observations
High external validity. Practical method. Few demand characteristics.
No causal relationship just correlation. Hard to replicate. Ethical issues. Practical problems. Observer bias/widely subjective (pilot studies can be run to establish specific behavioural categories).
74
What is a pilot study?
A small scale practice investigation carried out prior to research to identify problems with the design, method and sampling.
75
What are the 8 steps of pilot studies?
1. Identify a problem or topic of interest
2. Gather background information
3. Identify research hypothesis or questions
4. Choose a research method to collect data
5. Conduct a pilot study
6. Collect data to test the hypothesis
7. Analyse the data
8. Draw conclusions and report the findings
76
Define: peer review
The process of subjecting psychological research to an independent body so that they can offer constructive criticism (in terms of validity, significance and originality).
77
What is the importance, function and problems of peer reviews?
~Allocation of research funding, improves rating of university and publication of research in journals.
~Keep in touch with new ways of thinking and scientific developments.
~Reviewer less likely to provide unbiased opinion of the work, impossible to separate researcher from their values, institution and gender bias, hard to find an expert to review, anonymity.
78
Define: behavioural categories (used in observations)
Dividing a target behaviour into a subset of specific and operationalised behaviours.
* Objective, saves time, mutually exclusive
* Hard to know what categories to choose from, more effort.
79
Define: Quantitative data
Data that focuses on numbers and frequencies.
80
Define: Qualitative data
Data that describes meaning, feelings and experiences.
81
What are the factors of quantitative data?
* Objective
* Precise numerical measures used
* Lacks detail
* High in reliability
* Used for behaviour
* Collected in artificial setting
82
What are the factors of qualitative data?
* Subjective
* Imprecise non-numerical measures used
* Rich and detailed
* Low in reliability
* Used for attitudes, opinions and beliefs
* Collected in real life settings
83
What are the characteristics of primary data?
Also known as field research, information has been obtained first hand by the researcher. Original data.
84
Evaluate quantitative data
```
Can use data to produce graphs
More likely to be objective, less bias
Easy to analyse as averages and ranges are easy to produce
Can be compared with different studies
----
Can oversimplify complex behaviour
```
85
Evaluate qualitative data
Represents the complexity of human experience.
Participants have freedom of expression.
Data is more rich - external validity.
-----
Difficult to draw conclusions and detect patterns.
Can be affected by subjective analysis - researcher bias.
86
What are the characteristics of secondary data?
Also known as desk research.
This data could be found in journal articles, books or websites. It has already been subject to statistical testing and has already been collected by previous researchers.
87
Evaluate primary data
Authentic data which is direct from participants.
-----
Requires time and effort in planning, preparing and carrying out the research.
88
Evaluate secondary data
```
Inexpensive.
Requires minimal effort.
-----
Could be outdated and irrelevant.
Information could vary in quality and accuracy.
```
89
Define: standard deviation
The measure of dispersion around the mean. Takes all scores into account.
90
What does the size of standard deviation tell you?
A large SD indicates that scores are widely spread around the mean.
A small SD tells us that the data was closely clustered around the mean. More valid/reliable as consistency is shown.
91
What does correlation imply?
The strength (correlation coefficient) and direction of a relationship between two variables.
92
How can psychology reduce welfare spending?
* Mental health, effective therapies so people can go back to work and contribute to the economy through their taxes.
* Biological care (eg drugs).
* Use of cognitive interview ensures citizens aren't wrongfully imprisoned, ensures criminals are caught.
* Positive social influence used to influence behaviour, eg. Campaigns to stop smoking, impacts the economy.
* Developmental policies introduced based on the work of bowlby's attachment theories, this allows children to become productive members of society.
93
What are the measures of central tendency?
Mean: Average, can only be used with interval data. Easily distorted by extreme values.
Median: middle value in an ordered list use with interval/ordinal data. Exact values not reflected in median, not affected by extremes.
Mode: most common, used with nominal/interval. Unaffected by extremes, useless when there is more than one mode.
94
What are the measures of dispersion?
Range: arithmetic distance between top and bottom values in the data. Affected by extremes and doesn't account for distribution of numbers.
Standard deviation: distance between each data item above and below the mean. Precise, but may hide extremes.
95
What data representations should be used for continuous and discrete data?
Continuos: Histograms (area proportional to frequency), line graph.
Discrete: Bar chart (height=frequency, categorical/nominal data).
* Tables: raw/summary data, uses measures of central tendency and dispersion.
* Scattergram: correlational analysis.
96
What are the 3 levels of measurement?
* Nominal - discrete data that can be put into tally charts/distinct categories. E.g. Colours.
* Ordinal - used when data can be put into order/ranked. E.g. Competitions.
* Interval - continuous data that has units of equal intervals/measurements. E.g. Temperature.
97
Why is interval data preferred?
Most precise/stringent.
Hence why t-tests and Pearson's rho are the most stringent statistical tests, data is assumed to be normally distributed.
98
Define and describe: case studies
~An in-depth investigation, description and analysis of a single individual, group, institution or event.
* Case studies often involve the production of qualitative data, researchers will construct a case history of individual(s) concerned - perhaps using interviews, observations, questionnaires or a combination of all.
* Longitudinal, tend to take place over a long period of time to gain developed insights.
99
Evaluate the use of case studies
* Offer rich detailed insights that may shed light on very unusual or atypical forms of behaviour (that couldn't be observed otherwise).
* May contribute to our understanding of 'normal' functions. E.g. HM case study and memory processing.
* May generate hypotheses for future study, one contradictory instance may lead to the revision of an entire theory.
- Generalisation of findings cannot occur with such a small sample size.
- The information of the case study that makes it to the final report is subjective, as it is what the researcher selects and interprets as important.
- Personal accounts from participants and their family/friends may be prone to inaccuracy and memory decay, especially for childhood memories - low validity.
100
Define and describe content analysis
~A type of observational research technique that enables the indirect study of behaviour through examining communications that people produce. This can be in the form of texts, emails, books, TV and other media.
• The aim is to summarise and describe this communication in a systematic way so that conclusions can be drawn. Two ways to do this:
1. Coding (quantitative data) - the initial stage of content analysis in which the communication to be studied is analysed by identifying each instance of the chosen category (e.g. a particular word). Some data sets may be too large, e.g. a lengthy interview transcript, so this is useful.
2. Thematic analysis (qualitative data) - an inductive, descriptive approach to analysis that involves identifying implicit/explicit ideas within the data. Themes will often emerge once the data has been coded.
101
Evaluate content analysis
* Can circumnavigate many of the ethical issues surrounding psychological research. Most of the material analysed already exists within the public domain so many ethical issues can be avoided, e.g. deception.
* High external validity through real life situations, e.g. Texts.
* Flexible, can produce both quantitative and qualitative data depending on the aim of the research.
- By studying people indirectly, communication that is analysed is often taken out of context. Researcher may attribute opinions/motivations to the individual which they did not originally intend.
- Despite many modern analysts being clear about their own biases and preconceptions that may influence research (found in final report), content analysis still suffers from a lack of objectivity.
102
Define: probability
A measure of the likelihood that a particular event will occur, where 0 indicates statistical impossibility and 1 statistical certainty.
The probability that the observed effect occurred by chance is less than or equal to 0.05 or 5%
Probability levels double when two-tailed tests are being used.
103
Define: significance
A statistical term that states how certain a difference or correlation exists.
A significant result means that the researcher can reject the null hypothesis.
The significance level is the point at which the researcher can claim to have found a significant result. Usually 0.05 or 5%
104
Define: critical value
When testing a hypothesis, the numerical boundary or cut-off point between acceptance and rejection of the null hypothesis.
Once the statistical test has calculate the calculated/observed value, it is compared to the critical value to check for statistical significance.
105
When is a lower level of significance (e.g. 0.01) used?
More stringent levels are used if studies where there may be a human cost - e.g. Drug trials.
In all research, if there is a large difference between the observed and critical values, in the preferred direction, the researcher will check more stringent levels, as the lower the p value the more significant the result.
106
Define: Type 1 error
A type 1 error is when the null hypothesis is rejected and the alternative hypothesis accepted even though the null hypothesis is correct.
E.g. A false positive
Likely to be made when the significance level is too lenient/high (e.g. 0.5 or 5%).
106
Define: type 2 error
A type 2 error is when the null hypothesis is accepted and the alternative hypothesis rejected, even though the alternative hypothesis is correct.
E.g. A false negative
Likely to be made when the significance level is too stringent/low (e.g. 0.01 or 1%).
107
What is the acronym for all statistical tests?
Carrots should come mashed with suede under roast potatoes
Chi-squared, sign test, chi-squared
Mann-Whitney, wilcoxon, spearmans rho
Unrelated t-test, related t-test, Pearson's r
108
What 3 things must be considered when choosing a statistical test?
1. Is the researcher looking for a difference or a correlation?
2. If looking for a difference, is the experimental design related (RM + MP) or unrelated (IG)?
3. The level of measurement/type of data: nominal, ordinal or interval?
Also consider:
Is the hypothesis directional or non-directional?
What is the number of participants (N)?
109
What are the two parametric tests?
Related t test and unrelated r test.
*Parametric = interval data, natural observation and homogeneity of variance/normal distribution.
Parametric tests are more robust!
110
What must be included when making a conclusion from a statistical test?
1. Is the result significant? (How)
2. Which hypothesis is accepted and rejected
3. Difference/correlation stated in context
111
When is the sign test used?
The sign test is a test of difference between variables.
It is used in a repeated measures design (or matched pairs) meaning that both sets of data came from one group of participants. This is to see if there is a difference between the two conditions - only applicable to nominal data.
111
What are two non-parametric tests?
Wilcoxon and Mann-Whitney
112
Describe how to do a sign test
1. First take the data and turn the difference between the two variables for each participant into either a plus or minus.
2. count up the number of times that the less frequent value occurs, this gives the value of S. Consider the N value
3. Use this information compare the calculated value of the critical value of the given table.
4. If the calculated value is equal to or less than the critical value, reject the null hypothesis and accept the hypothesis.
113
When is the chi-squared test used?
A test of difference or correlation.
Used when data is nominal, unrelated design (independent groups - difference) or in correlation.
114
Describe the chi-squared test
1. Find the observed x^2 value for the data
2. Find the critical value t from the tables. Df=(Rows-1)(Columns-1)
3. Compare the observed and critical values - x^2 has to be GREATER than or equal to the critical value to be significant
4. Make a conclusion
116
When is the Mann-Whitney U test used?
This is a test of difference.
Used when data is ordinal and in an unrelated (IG) design.
117
Describe the Mann-Whitney U test
1. Find the observed U value for the data
2. Find the critical value from the table. Df=N
3. Compare the observed and critical value - U has to be LESS than or equal to the critical value to be significant
4. Make a conclusion
118
When is the Wilcoxon T test used?
This is a test of difference.
Used when data is ordinal and in a related design (RM + MP).
119
Describe the Wilcoxon T test
1. Find the observed T value for the data
2. Find the critical value from the table. Df=N
3. Compare the observed and critical value - t has to be LESS than or equal to the critical value to be significant.
4. Make a conclusion
120
When is Spearman's rho test used?
This is a test of correlation/association.
This is used when data is ordinal.
122
Describe Spearman's rho test
1. Find the observed rho value for the data
2. Find the critical value from the table. Df=N
3. Compare the observed and critical value - rho has to be GREATER than or equal to the critical value to be significant
4. Make a conclusion
122
When is the unrelated t test used?
This is a test of difference.
Used when data is interval and in an unrelated design (IG).
124
Describe the unrelated t test
1. Find the observed t value from the data
2. Find the critical value from the table. Df = (Na + Nb)-2
3. Compare the observed and critical value - t has to be GREATER than or equal to the critical value to be significant
4. Make a conclusion
125
When is the related t test used?
This is a test of difference.
Used when data is interval and in a related design (RM + MP).
126
Describe the related t test
1. Find the observed t value from the data
2. Find the critical value from the table. Df = N-1
3. Compare the observed and critical value - t has to be greater than or equal to the critical value to be significant
4. Make a conclusion
127
When is Pearson's r test used?
This is a test of correlation/association.
Used when data is interval.
128
Describe Pearson's r test
1. Find the observed r value from the data
2. Find the critical value r from the table. Df = n-2
3. Compare the observed and critical value - r has to be greater than it equal to the critical value to be significant
4. Make conclusion
129
What are the 6 sections of scientific report?
1. Abstract
2. Introduction
3. Method
4. Results
5. Discussion
6. Referencing
130
What are the 2 types of statistics? (Included in the results section of a scientific report)
Descriptive statistics: tables, graphs and charts, measures of central tendency, measures of dispersion
Inferential statistics: statistical tests
131
What are the 7 features of science? (acronym)
```
Theory construction
Hypothesis testing
Empirical method
Paradigm
Replicability
Objectivity
Falsifiability
```
132
What is theory construction and hypothesis testing?
* A theory is a set of general laws or principles that can explain a particular event/behaviour.
* This information can be gathered via direct observation and should be scientifically tested.
* It should also be possible to make clear precise hypotheses.
E.g. Social learning theory
133
What is the empirical method?
* The experimental/observational methods in psychology are good examples of the empirical method.
* Data should be based on direct and sensory experiences, a theory cannot claim to be scientific unless it has been empirically tested and verified.
e.g. Milgram's shock experiment
134
What is a paradigm?
* Kuhn suggested that paradigms are a shared set of assumptions and methods, these separate science from non-sciences.
* Psychology lacks a universal accepted paradigm (evident from all the different approaches).
* Paradigm shift can occur once the accepted paradigm is critiqued and questioned, causing there to be too much contradictory evident to ignore.
E.g. Psychodynamic approach shifted to behaviourist
135
What is replicability?
* The extent to which scientific procedures and findings can be repeated by other researchers.
* Can determine validity of a finding.
E.g. Ainsworth's strange situation
136
What is objectivity?
• Scientific researchers must keep a critical distance during research. • Objectivity is only achieved when all sources of personal bias are minimised.
Objective methods such as laboratory experiments tend to be the most objective.
E.g. Asch's line experiment
137
What is falsifiability?
* Popper argued that the key criterion of a scientific theory is its falsifiability. A theory cannot be considered scientific unless it admits the possibility of being proved untrue.
* Theories should hold themselves up for hypothetical testing, alternative hypothesis must always be accompanied by a null hypothesis.
E.g. Multi store model of memory
