What are ethics in Psychology?

• Psychology distinguishes between research ethics and professional ethics
• Research ethics focus on how participants are treated and how studies are conducted
• All human and animal research must receive ethics approval before data collection
• Ethics committees evaluate whether research risks are justified by potential benefits
• Ethical oversight exists due to historical abuse in psychological and medical research

Ethical Foundations of Research

• Ethical guidelines emerged in response to serious historical violations
• International frameworks established minimum ethical standards
• Modern research follows a risk–benefit model
• Research should benefit participants, society, or knowledge without causing undue harm

Core Ethical Principles in Research

Principle	Meaning
Research merit and integrity	Research must be worthwhile and conducted honestly
Justice	Benefits and burdens of research should be fairly distributed
Beneficence	Benefits must outweigh risks

What Is Informed Consent?

• Participants must voluntarily agree to participate
• Consent requires understanding procedures, risks, and rights
• Full disclosure of hypotheses is not always required
• Limited disclosure may be justified to reduce bias
• Deception may be used only when necessary and ethically approved
• Participants must be debriefed and re-consent after deception
• Consent may be waived only in exceptional circumstances

Ethical Responsibilities of Researchers

• Report research truthfully and accurately
• Use rigorous and transparent methods
• Disclose conflicts of interest
• Treat participants and collaborators with respect
• Protect vulnerable groups
• Respect cultural and Indigenous communities
• Be accountable for research conduct and outcomes
• Promote ethical research culture

Ethical Responsibilities of Participants

• Read consent forms carefully
• Weigh risks and benefits before agreeing
• Participate honestly
• Ask questions or raise concerns if needed
• Follow agreed responsibilities
• Request study results if interested

Sampling in Psychological Research

• Research questions always refer to populations
• Data are collected from samples
• Samples should represent the population
• Biased samples lead to biased conclusions
• External validity depends on sampling quality
• Representative samples do not need to be large

Types of Sampling

Probability Sampling

Method	Description
Simple random	Every individual has equal chance
Stratified random	Population divided into strata, then sampled
Cluster	Groups selected first, then individuals

Non-Probability Sampling

Method	Description
Convenience	Easily available participants
Snowball	Participants recruit others

Sample Size

• Sample size affects statistical power
• Very small samples limit detection of effects
• Very large samples may detect trivial effects
• Typical psychology samples vary widely in size
• Sample choice has major implications for knowledge claims

Psychological Measurement

• Psychological constructs must be operationalised into variables
• Operationalisation determines what is actually measured
• Many constructs can be measured in multiple ways
• Measurement quality determines research value
• Psychometrics studies how psychological constructs are measured

Measuring Psychological Constructs

• Concrete variables are easier to measure
• Psychological constructs are abstract and indirect
• Self-report scales are commonly used
• Multiple items are used to capture complex constructs
• Composite scores combine multiple items
• Reverse coding is required when item direction differs

Measurement Decisions

• Define the construct clearly before measuring
• Use existing validated scales when possible
• New measures require extensive validation
• Measurement error is unavoidable
• Error reflects imprecision, not researcher mistakes

Reliability of Measurement

• Reliability refers to consistency of measurement

Type	Meaning
Test–retest	Stability over time
Internal consistency	Agreement across items
Inter-rater	Agreement across observers

Validity of Measurement

• Validity refers to whether a measure captures the intended construct

Type	Meaning
Face validity	Appears to measure the construct
Content validity	Covers all aspects of the construct
Criterion validity	Relates to relevant outcomes
Discriminant validity	Does not relate to unrelated constructs

Measurement Levels

• Researchers choose how variables are measured
• Many psychological constructs are continuous by nature
• Numeric measurement is usually preferable

Advantage of Numeric Measurement
More information
Greater flexibility
Higher statistical power

Variable Types

Type	Examples
Nominal	Gender, category membership
Ordinal	Education level, rankings
Interval	Temperature, pain scales
Ratio	Height, weight, counts

Summarising One Variable

Categorical Variables

• Frequency tables
• Percentages
• Bar charts
• Pie charts (less precise)

Numerical Variables

• Mean or median (centre)
• SD or IQR (spread)
• Histograms
• Boxplots

Boxplots

• Display median, quartiles, range, and outliers
• Useful for identifying skew and unusual values
• Particularly useful for comparing groups

Relationships Between Variables

• Most studies examine relationships between variables
• One variable is usually the outcome (DV)
• One variable is used as a predictor (IV)

Summarising Two Variables

Two Categorical Variables

• Contingency tables
• Row, column, or cell percentages
• Clustered bar charts

Two Numerical Variables

• Scatterplots
• Direction, strength, and form of relationship
• Pearson correlation summarises linear relationships

One Categorical and One Numerical Variable

• Group means and medians
• Group variability statistics
• Comparative boxplots

What Is Central Limit Theorem?

• Samples are drawn from populations
• Each sample produces a statistic (e.g. mean)
• The distribution of these statistics is the sampling distribution
• Regardless of population shape:
  • The sampling distribution of the mean is approximately normal
  • This holds when sample size is sufficiently large

Importance of the Central Limit Theorem

• Many statistical tests rely on means
• Normality assumptions apply to sampling distributions, not raw data
• CLT allows inference even when data are not normally distributed
• Enables probability-based conclusions about populations