Research methods - terms + AO3

Experimental method: aim + hypothesis (one + two tailed)

Aim = general expression of what researcher intends to investigate

Hypothesis = predictive statement of what researcher believes they will find

Directional (one tailed) hypothesis = states whether changes will be greater or lesser

Non-directional (two tailed) hypothesis = just predicts a difference/correlation

Experimental method: variables

IV = variable manipulated by researcher → to see effect on dv

DV = variable measured

Extraneous variable = other things not measured that might affect DV

Experimental method: research issues

Demand characteristics → cue from researcher or research situation may cause ppt to change behaviour

Investigator effects → investigators’ behaviour may affect outcome of research/study

Experimental method: research techniques

Randomisation

Standardisation

Control groups

Single blind

Double blind

Experimental method: pilot study

Experimental design

Types of experiment

Sampling

Ethical issues

Correlations

Correlation = statistical measure of relationship between two variables

Correlational study =

Observation

Observational design

Self report techniques

Peer review

Case study + methods used in case studies

Detailed + in-depth analysis of a single individual, group, institution or event

High in validity

Type of descriptive research where IV + DV are not manipulated → cannot conclude c+e

Usually leads to further research ~ Phineas Gage → Broca, Tan → Wernicke

Methods used in case studies (both quantitative + qualitative forms of data):

• Questionnaires

• Interviews

• Experiments

• Observations

• Case histories

• Secondary data ~ past medical records

Case study AO3

+ Case studies = valid → gather in-depth + detailed + rich data about one unique individual (from as many sources as possible)

+ Ecological validity → individuals are usually in natural setting = inc ecological validity

- Subjective→ rely on interpretation = opened to bias + researcher subjectivity ~ Freud + Little Hans

- Hard to replicate → one unique situation = lacks reliability + replicability

Content analysis

Type of observational study → behaviour is observed indirectly in visual, written or verbal material

Qualitative + quantitative data can be involved

Most commonly done in media research → analyse what people produce ~ TV

Content analysis method

Researcher needs to consider:

Sampling method ~ every fifth page of a book

Coding the data ~ process of placing qualitative or quantitative data

Method of representing the data → counting instances (quantitative analysis), describing examples in each category (qualitative analysis)

Content analysis steps

1. Coding system is created

2. Pilot study often conducted

3. Conduct the analysis

4. Turn the data in quantitative displays

5. Reliability might be checked

Thematic analysis

Technique used when analysing qualitative data → themes are identified

Psychology + the economy

Examines how findings of psychological research can benefit our economy

Attachment research into role of the father →

Features of science: connected words

Scientific:

• Quantitative data

• Objective

• Structured method

• Reliable

~ lab experiments

Non-scientific:

• Qualitative data

• Subjective

• Unstructured method

• Valid

~ case studies

Features of science: scientific methods (acronym)

Theory construction

Hypothesis

Empirical method

Paradigms

Replicability/reliability

Objectivity

Observable

Falsifiability

Features of science: definitions

Empiricism = info gained through direct observation or experiment

Objectivity = observations + experiments → should be unaffected by bias

Replicability = important that research can be repeated → similar results obtained (standardised procedures allow this) = adds to reliability of study

Falsification =

Features of science: theory construction

Deduction = reasoning from general to particular → start with theory then look for instances to support it

~ Darwin’s theory of evolution

Hypothetico-deductive model → Karl Popper

• Theories + laws should come first → used to generate expectations + hypothesis → to then be falsified

“No amount of observations of white swans can allow inference that all swans are white, but the observation of a single black swan is sufficient refute that conclusion”

Features of science: paradigms

Kuhn

• Psychology ≠ science → no single paradigm = shared set of assumptions

Features of science: can psychology claim to be a science?

• Kuhn = no → no single paradigm like other sciences

• Miller = no → ‘dressing it up’

• Heisenberg = yes → found you cannot even measure a subatomic particle without altering its behaviour (uncertainty principle)

Features of science: does psychology want to be a science

• Scientific approach = desirable but reductionist

• Determinist → looks for causal relationships (X → Y)

• Nomothetic approach → perhaps idiographic approach is more suitable

Reporting psychological investigations (scientific report)

Abstract

Intro

Method

Results

Discussion

Referencing

Validity: definition + types

Accurately measures what it is intended to measure

Internal validity = extent to which a study demonstrates a causal relationship between variables → whether DV is affected by IV (not extraneous/confounding variables, demand characteristics, investigator bias)

External validity = extent to which findings can be generalised out of the study

• Population validity = degree to which results can be generalised to other populations/wider society → dependent upon sampling technique ~ Milgram

• Temporal validity = degree to which results can be generalised across different time periods or contexts ~ Adorno + facism

Threats to validity (internal + external)

Internal validity:

• Confounding variables

• Demand characteristics

• Investigator bias

• Experiment/observation design

• Order effects (repeated measures)

• Ppt variables (independent measures)

External validity:

• Researchers are focused on experimental realism = degree to which results reflect realistic behaviour

Improving validity

Reduce demand characteristics + investigator effects + order effects

Testing validity

Face validity = the extent to which a test appears to measure what it claims to measure based on subjective interpretation

Concurrent validity = comparing results of a new test with that of an old test known to have good validity

Predictive validity = ability to predict performance on future tests

Reliability: definition

Consistency of research study or measuring test, ensuring it produces stable + similar results over repeated trials or uses

Importance of reliability

Unreliable measures introduce random error + reduce ability to detect true relationships + undermine validity of research findings

Testing reliability

Inter-observer reliability = the degree to which different observers give consistent estimates of the same phenomenon

Test-retest reliability = repeat test → same results produced each time = high external reliability

Improving reliability

Pilot studies

Standardisation → uniform procedures + data collection ~ fully operationalised behaviour categories in observations

Probability

Numerical measure of chance

0 = will not happen

1 = def will happen

Role of statistical tests

Find out how likely it is what has been found in the sample = accurately reflects whole population

Statistical significance

Chance that the results are down to error → 5% error rate = accepted as reasonable

Expressed as p=0.05

Calculating sign test

(used when experiment = test of difference, related = repeated measures + matched pairs, nominal data = category data)

Give the highest number a +

Give the lowest number a -

Ignore category of data → ‘the same’ or ‘neither’

S = lowest number

Sign test