MASTER RESEARCH METHODS

Key in case of forgetting: Delete, double, check, flip

What is a correlation?

A statistical measure that expresses the extent to which two co-variables are related

Positive Correlation

Co-variables change in the same direction

Negative correlation

Co-variables change in opposite directions

What is the correlation co-efficient?

Shows the strength and direction of the correlation. Must be above +.8 for the study to be considered reliable/valid

Evaluation of correlations

Strengths:

• May indicate connection

• Don’t require manipulation of variables

• High ecological validity

Weaknesses

• Don’t prove causal relationships

• Don’t reflect curvilinear relationships

Case studies

• In depth investigation, description and analysis of individuals, groups, or institutions

• Usually about something unusual

• Mostly qualitative data

• Happens over an extended period of time

Case Studies AO3

Strengths

• Rich, detailed, qualitative insights into behaviour

• Can e used to generate future hypothesises

Weaknesses

• Generalisability

• Subjective

• Researcher bias

• Low validity

Content analysis

• Indirect Study of behaviour by examining communications

• Observational

• Coding- Analysis of the communication, identifying instances of the chosen categories

• Thematic analysis- Identifying themes, implicit or explicit ideas within the data that are researched

Content analysis AO3

Strengths

• Informed consent- easy as data is pre-existing

• High external validity

Weaknesses

• Indirect study analysed outside of the context it occurred

• Bias risk- subjective

Reliability

A measure of consistency, if the experiment is repeated, you will get the same result

What is thematic analysis?

A qualitative research method in psychology used to identify, analyse, and report patterns (themes) within data, such as interview transcripts or text

What is validity?

How legitimate the results are

What is Internal Validity?

Does it measure what it’s supposed to

What is external validity?

Overall generalisability

What is Ecological Validity?

Form of external validity, generalisability to other settings/situations

What is temporal validity?

Form of external validity, generalisability over time

Ways of assessing validity

• Face value- does it measure what it’s supposed to

• Concurrent validity- assesses a new test's accuracy by comparing it to an established, validated measure (the "gold standard") administered at the same time

Ways of improving validity (each method)

• Questionnaires: anonymous submission. lie scale

• Interviews: Interpretive validity, triangulation

• Observations: Minimal intervention by researcher, covert observations

• Experiments: control groups, standardised procedures, single/double blind procedures

What is the mnemonic for stats tests?

Carrots Should Come Mashed With Swede Under Roast Potatoes

Chi- Squared (Carrots)

• Test Of Difference

• Nominal Data

• Unrelated Design

Sign test (Should)

• Test Of Difference

• Nominal Data

• Related Design

Chi- Squared (Come)

• Test Of Association/Correlation

• Nominal Data

Mann-Whitney (Mashed)

• Test of difference

• Unrelated design

• Ordinal data

Wilcoxon (With)

• Test of difference

• Related design

• Ordinal data

Spearman’s Rho (Swede)

• Test of Association/Correlation

• Ordinal Data

Unrelated T-Test (Under)

• Test Of Difference

• Unrelated Design

• Interval data

Related T-Test (Roast)

• Test Of Difference

• Related Design

• Interval Data

Pearson’s R (Potatoes)

• Test Of Correlation/Association

• Interval data

How to remember the table (other than the tests)

• We Urc

• Nominal (1st on 1st row)

• Ordinal (2nd on 1st row)

• Interval (3rd on 1st row)

• Unrelated (1st column)

• Related (2nd column)

• Correlational (Top 3rd column)

All are test of difference EXCEPT for third column

How to chose difference or correlation

• Should be obvious from hypothesis

• Correlation looks for an association

Experimental designs Stats tests

• Independent Groups: Unrelated

• Matched pairs/ Repeated measures: Related

What are the different levels of measurement used in stats tests?

• Nominal: Distinct individual categories

• Ordinal: Data ordered by a scale, doesn’t have to be equal intervals and is subjective

• Interval: Numerical scales of equal, precisely defined intervals

What is probability?

Likelihood of an event occurring`

What is significance?

How sure we are that a difference/correlation exists

Do stats tests work on probability or certainty?

Probability

What is the significance level

0.05% (0.01% for drugs)

What is a Type I error?

• Null hypothesis rejected and alternative hypothesis accepted when it should be reversed

• Optimistic error/ false positive

What is a Type II error?

• Null hypothesis accepted and alternate rejected when it should be reversed

• Pessimistic error/ false negative

What is a paradigm?

A standard, model, or typical example of something, acting as a framework or set of assumptions. Psych lacks a universally accepted paradigm, too much internal disagreement, conflicting approaches

What is a paradigm shift?

Scientific revolution when there’s a significant change in dominant unifying theory within a scientific discipline 

What is a theory?

A set of general laws/principles that have the ability to explain certain behaviours. Occurs through gathering evidence- direct observation

What is deduction?

Process of deriving new hypotheses from existing theory 

Scientific testing

Theories should suggest many hypotheses- it can then be tested systematically and objectively 

What is falsifiability- Popper?

A key criterion of a scientific theory and scientific knowledge is provisional – the best we can do at the moment.

Theory of Falsifiability

A way of separating good science from pseudoscience. It suggests that for a theory to be considered scientific, it must be able to be tested and conceivably proven false. Science should attempt to disprove a theory rather than attempt to continually support theoretical hypotheses.

Replicability- Popper

If a theory is to be trusted, it must be shown to be repeatable across different contexts and circumstances

What is validity?

If repeated across numerous contexts, can generalise 

Validity, replicability, and reliability

To achieve this, psychologists need to report their studies with as much precision and detail as possible

Empirical Method

• Data on direct sensory experiences 

• Objective- no bias, emotions or personal opinions 

• Lab experiments are the most objective- most controlled

• Locke- knowledge determined by experience and sensory perception. A theory cannot be scientific unless it has been empirically tested and verified 

What is an abstract?

• First section, but written last

• Short summary of text

• 150-300 Words

• Includes major elements such as the aim, methods, results, and conclusions

Introduction components

• Literature review

• Talk about existing studies/linked to what you are studying

• Should follow logical progression: start broad, narrow down, until aim and hypothesis are presented.

Methods

• Split into the following sections:

  • Design- groups/observations and justification

  • Sample- Who, method, target population

  • Materials- questionnaires, etc

  • Procedure- Standardised instructions and debrief

  • Ethics- How were they addressed?

Results

• Summarise key findings

• Descriptive stats- graphs, tables, etc

• Inferential stats- stats tests

• If qualitative, include themes or catagories

Discussion

• Verbal/written summary rather than statistical summary of findings

• Relate back to theories and studies from the introduction

• Any limitations of the study are discussed and any ideas of how they could be addressed in future are presented

• Wider applications of results

What is a hypothesis?

A hypothesis is a testable statement about the relationship between two or more Variables. It is an educated guess or a proposed explanation for a phenomenon that can be tested through experimentation or observation

What is an aim?

An aim is a statement that describes what a study intends to achieve. It outlines the overall purpose or goal of the research, often in a broad sense.

What is the Experimental Method?

Manipulating an Independent Variable, to measure the effect on the Dependant Variable

What is H₀ ?

Null hypothesis.

What is H₁ ?

Alternate Hypothesis.

What are the 4 types of experiment?

Field, Laboratory, Natural, Quasi.

Field Experiment

A research method conducted in a natural, real-world setting instead of a lab, where researchers manipulate an independent variable to measure its effect on a dependent variable

What are the strengths of Field Experiments?

• High ecological validity: The findings are more likely to be generalisable

• Natural behaviour: Participants' behaviour is more natural and less affected by the artificiality of a laboratory.

• Causal inference: Random assignment helps establish a cause-and-effect relationship between variables.

What are the weaknesses of Field Experiments?

• Less control: Researchers have less control over extraneous variables.

• Ethical concerns: Participants might not be able to give informed consent if they are unaware they are in a study.

• Difficult to replicate: The unpredictable nature of real-world settings can make it hard to replicate the experiment.

What is a Laboratory Experiment? 

A research method conducted in a highly controlled environment where researchers manipulate one or more independent variables to measure their effect on a dependent variable.

What are the strengths of Laboratory experiments?

• High internal validity: The tight control over variables means that a clear cause-and-effect relationship between the IV and DV can be established.

• Replicability: The controlled, standardized procedures make it easier for other researchers to replicate the study and verify the findings. 

What are the weaknesses of Laboratory experiments?

• Low external or ecological validity: The artificial environment may not reflect real-world settings, making it difficult to generalize the results to everyday situations.

• Potential for artificiality: Participants might change their behaviour because they are aware they are being studied, a phenomenon known as demand characteristics

What is a Natural experiment?

an observational study where researchers leverage a naturally occurring event or pre-existing situation to investigate the effect of an independent variable (IV) on a dependent variable

Strengths of Natural experiments

• High External Validity: Real-world, natural settings, findings have high ecological validity and can be generalised.

• Reduced Bias: Participants unaware they are part of a study, which minimizes the demand characteristics

• Cost-Effectiveness: pragmatic and cost-effective research design

• Real-World Application: useful tool for evaluating effectiveness of actual policies and interventions in complex social and political conditions.

What are the weaknesses of Natural experiments?

• Difficulty Inferring Causation: It is difficult to establish a clear cause-and-effect relationship because the researcher has little or no control over extraneous or confounding variables.

• Lack of Control: The researcher cannot manipulate the IV or randomly assign participants to conditions, which lowers internal validity.

• Replication Difficulties: Due to the unique nature of naturally occurring events, it can be extremely difficult to replicate the study conditions.

• Potential for Bias: Even without demand characteristics, other forms of bias can occur, such as observer bias or sample bias.

• Data Limitations: Researchers often rely on retrospective data collected for other purposes, which may be inaccurate, incomplete, or difficult to access.

What is a Quasi experiment?

A type of research method that aims to establish a cause-and-effect relationship between variables but lacks random assignment of participants to groups.

What are the strengths of Quasi experiments?

• High External Validity: Because they are often conducted in real-world, natural settings rather than artificial laboratory environments, the findings tend to be more generalizable to the broader population and other contexts.

• Study of Naturally Occurring Events: Researchers can study the effects of naturally occurring events or pre-existing conditions that cannot be manipulated in a true experiment.

• Cost-Effectiveness and Efficiency: These designs are often less expensive and time-consuming than true experiments, requiring fewer resources and potentially smaller sample sizes.

• Higher Internal Validity than Non-Experimental Research: While lower than true experiments, quasi-experiments allow for more control over confounding variables. 

What are the weaknesses of Quasi experiments?

• Lower Internal Validity: The primary weakness is the inability to definitively establish a causal relationship due to the lack of random assignment.

• Risk of Confounding Variables and Bias: There is a higher likelihood of confounding environmental and extraneous variables influencing the results.

• Difficulty Ruling Out Alternative Explanations: The lack of control makes it challenging to rule out alternative explanations for the observed results, as any differences between groups could be due to pre-existing disparities rather than the independent variable.

• Reliance on Retrospective Data: Retrospective data, often used in some quasi-experimental designs, can be inaccurate, incomplete, or difficult to access, potentially affecting the reliability of the findings.

• Statistical Complexity: To account for potential biases and confounding variables, quasi-experimental studies may require the use of advanced statistical techniques such as multiple regression or propensity score matching.

Which types of experiments have high levels of Complexity?

Field, Natural, and sometimes Quasi

Which types of experiments have lower levels of complexity?

Laboratory, sometimes Quasi

Which types of experiments have high levels of Variable Control?

Laboratory Experiments

Which types of experiments have low levels of Variable Control?

Field, Natural, sometimes Quasi

Which type of experiments have high levels of Internal Validity?

Laboratory

Which types of experiments have low levels of Internal Validity?

Field, Natural, Quasi

Which types of experiments have high levels of External Validity?

Field, Natural, sometimes Quasi

Which types of experiments have low levels of External Validity?

Laboratory, sometimes Quasi

Which types of experiments have high levels of Demand Characteristics?

Laboratory, Natural, sometimes Quasi

Which type of experiments have low levels of Demand Characteristics?

Field

Which experiments have high levels of Realism?

Field, Natural, Quasi

Which type of experiments have low levels of Realism?

Laboratory

What is Non-Directional?

Increase/Decrease of the outcome is not specified.

What is Directional?

Increase/Decrease of the outcome is specified.

Are Extraneous Variables wanted?

No.

What is an extraneous variable?

Any Variable, other than the Independent Variable, that may affect the Dependant Variable. They are nuisance variables that don’t vary systematically with the Independent Variable.

What are confounding variables?

Any variable, other than the Independent Variable, that may effect the Dependent Variable. We cannot be sure of the true source of the changes. They vary systematically with the Independent Variable.

What are Demand Characteristics?

Any cue from the researcher or environment that may be interpreted as the purpose of the investigation.

What can Demand Characteristics lead to?

Participants changing their behaviour, making the findings less valid.

Investigator Effects

Any effect of the Investigator’s behaviour on the research outcome. This varies from the design of the study, to the selection of and interaction with participants during the study.

What is the Please-You Screw-You Effect?

Participant(s) conscious bias caused by known Demand Characteristics.

What is Randomisation?

The use of chance in order to control the effects of bias when choosing participants. 

What is Standardisation?

Using exactly the same formalised procedures and instructions for all participants in a research study.

What is Social Desirability Bias?

A type of response bias that is the tendency of survey respondents to answer questions in a manner that will be viewed favourably by others. It can lead participants to over-report 'good' behaviours or under-report 'bad' behaviours.

What is the Experimental design?

A systematic approach to planning and conducting scientific experiments to test hypotheses and establish cause-and-effect relationships between variables.

What is the Repeated Measures design?

An experimental method in which the same participants are used in all conditions of an experiment

What are the advantages to the Repeated Measures design?

Controls Individual Differences: Eliminates participant variables as a potential source of error.

Greater Statistical Power: By reducing variability caused by individual differences, the design increases the sensitivity to detect the true effects of the independent variable.

Efficiency and Cost-Effectiveness: Requires fewer participants and resources.