HSC Science Extension cheat sheet

Deductive reasoning moves from general premises to a specific conclusion; if the premises are true and the form valid, the conclusion must be true.

Inductive reasoning generalises from observed instances to a broader claim; it is ampliative and never guarantees truth, only probability.

Abductive reasoning selects the hypothesis that best explains the evidence; it is inference to the best explanation and remains provisional.

Falsifiability (Popper): a scientific claim must specify observations that would refute it; unfalsifiable claims sit outside empirical science.

Paradigm shift (Kuhn): normal science accumulates anomalies until a crisis produces a new paradigm; paradigms are not discarded on single anomalies alone.

Research paradigms: positivism seeks objective, measurable law-like generalisations; interpretivism seeks meaning in context; critical theory examines power and emancipation.

Confirmation bias is seeking or weighting evidence that supports a prior belief; the remedy is pre-specified predictions and disconfirming tests.

Correlation does not entail causation; causal claims require temporal precedence, mechanism, and control of confounders.

Ethical principles: merit and integrity, justice, beneficence, and respect for persons; human research requires informed consent and ethics approval.

A research question must be answerable, specific, and feasible; operationalise variables into measurable indicators before designing the method.

A hypothesis is a testable, directional or non-directional prediction linking independent and dependent variables; a null hypothesis states no effect.

Independent variable is manipulated or selected; dependent variable is measured; controlled variables are held constant; extraneous variables are nuisance influences.

Validity types: internal (causal inference), external (generalisation), construct (measures the intended concept), and reliability (consistency of repeated measurement).

Sampling: random sampling supports generalisation; stratified sampling preserves subgroup proportions; convenience sampling risks selection bias.

A literature review establishes the gap, defines key terms, and justifies the method; it must cite sources and avoid unsupported claims.

Risk assessment identifies hazards, likelihood, severity, and mitigation; human or animal participants require ethics approval before data collection.

A feasible timeline and budget specify milestones, resources, and contingencies; scope creep is controlled by fixing the research question early.

Data types: categorical (nominal, ordinal) versus numerical (discrete, continuous); the type determines permissible statistics and graphs.

Descriptive statistics summarise centre (mean, median, mode) and spread (range, interquartile range, standard deviation); the median resists outliers.

A p-value is the probability of data at least as extreme as observed, assuming the null hypothesis is true; it is not the probability the null is true.

Statistical significance (typically p < 0.05) is not practical significance; report effect size and confidence intervals alongside p-values.

A 95% confidence interval means that 95% of intervals from repeated sampling would contain the true parameter; it is not a 95% chance the parameter lies within one interval.

Type I error rejects a true null (false positive); Type II error fails to reject a false null (false negative); power is 1 minus the Type II error rate.

Measurement error reduces reliability and attenuates observed effects; calibrate instruments and report uncertainty with appropriate significant figures.

Conclusions must be limited to the evidence: distinguish supported inference from speculation, acknowledge limitations, and state what further data would test the claim.

Standard structure: title, abstract, introduction, methods, results, discussion, conclusion, references, and appendices for raw data.

The abstract states aim, method, key results, and conclusion in a single self-contained paragraph, typically 150-250 words.

Methods must be reproducible: report participants or materials, apparatus, procedure, variables, and analysis in sufficient detail for replication.

Results report findings without interpretation; figures and tables carry numbered captions, labelled axes, units, and error bars where relevant.

Discussion interprets results against the hypothesis and literature, explains limitations, and states implications without overclaiming.

Referencing: in-text citations must match the reference list exactly; use one consistent style and acknowledge all borrowed ideas.

Scientific writing uses precise, concise, objective language, past tense for methods and results, and avoids unsupported causal verbs.

Data integrity requires honest reporting of all results, including non-significant findings, and retention of raw data for verification.

Deductive reasoning moves from general premises to a specific conclusion; if the premises are true and the form valid, the conclusion must be true.

Inductive reasoning generalises from observed instances to a broader claim; it is ampliative and never guarantees truth, only probability.

Abductive reasoning selects the hypothesis that best explains the evidence; it is inference to the best explanation and remains provisional.

Falsifiability (Popper): a scientific claim must specify observations that would refute it; unfalsifiable claims sit outside empirical science.

Paradigm shift (Kuhn): normal science accumulates anomalies until a crisis produces a new paradigm; paradigms are not discarded on single anomalies alone.

Research paradigms: positivism seeks objective, measurable law-like generalisations; interpretivism seeks meaning in context; critical theory examines power and emancipation.

Confirmation bias is seeking or weighting evidence that supports a prior belief; the remedy is pre-specified predictions and disconfirming tests.

Correlation does not entail causation; causal claims require temporal precedence, mechanism, and control of confounders.

Ethical principles: merit and integrity, justice, beneficence, and respect for persons; human research requires informed consent and ethics approval.

A research question must be answerable, specific, and feasible; operationalise variables into measurable indicators before designing the method.

A hypothesis is a testable, directional or non-directional prediction linking independent and dependent variables; a null hypothesis states no effect.

Independent variable is manipulated or selected; dependent variable is measured; controlled variables are held constant; extraneous variables are nuisance influences.

Validity types: internal (causal inference), external (generalisation), construct (measures the intended concept), and reliability (consistency of repeated measurement).

Sampling: random sampling supports generalisation; stratified sampling preserves subgroup proportions; convenience sampling risks selection bias.

A literature review establishes the gap, defines key terms, and justifies the method; it must cite sources and avoid unsupported claims.

Risk assessment identifies hazards, likelihood, severity, and mitigation; human or animal participants require ethics approval before data collection.

A feasible timeline and budget specify milestones, resources, and contingencies; scope creep is controlled by fixing the research question early.

Data types: categorical (nominal, ordinal) versus numerical (discrete, continuous); the type determines permissible statistics and graphs.

Descriptive statistics summarise centre (mean, median, mode) and spread (range, interquartile range, standard deviation); the median resists outliers.

A p-value is the probability of data at least as extreme as observed, assuming the null hypothesis is true; it is not the probability the null is true.

Statistical significance (typically p < 0.05) is not practical significance; report effect size and confidence intervals alongside p-values.

A 95% confidence interval means that 95% of intervals from repeated sampling would contain the true parameter; it is not a 95% chance the parameter lies within one interval.

Type I error rejects a true null (false positive); Type II error fails to reject a false null (false negative); power is 1 minus the Type II error rate.

Measurement error reduces reliability and attenuates observed effects; calibrate instruments and report uncertainty with appropriate significant figures.

Conclusions must be limited to the evidence: distinguish supported inference from speculation, acknowledge limitations, and state what further data would test the claim.

Standard structure: title, abstract, introduction, methods, results, discussion, conclusion, references, and appendices for raw data.

The abstract states aim, method, key results, and conclusion in a single self-contained paragraph, typically 150-250 words.

Methods must be reproducible: report participants or materials, apparatus, procedure, variables, and analysis in sufficient detail for replication.

Results report findings without interpretation; figures and tables carry numbered captions, labelled axes, units, and error bars where relevant.

Discussion interprets results against the hypothesis and literature, explains limitations, and states implications without overclaiming.

Referencing: in-text citations must match the reference list exactly; use one consistent style and acknowledge all borrowed ideas.

Scientific writing uses precise, concise, objective language, past tense for methods and results, and avoids unsupported causal verbs.

Data integrity requires honest reporting of all results, including non-significant findings, and retention of raw data for verification.

Deductive reasoning moves from general premises to a specific conclusion; if the premises are true and the form valid, the conclusion must be true.

Inductive reasoning generalises from observed instances to a broader claim; it is ampliative and never guarantees truth, only probability.

Abductive reasoning selects the hypothesis that best explains the evidence; it is inference to the best explanation and remains provisional.

Falsifiability (Popper): a scientific claim must specify observations that would refute it; unfalsifiable claims sit outside empirical science.

Paradigm shift (Kuhn): normal science accumulates anomalies until a crisis produces a new paradigm; paradigms are not discarded on single anomalies alone.

Research paradigms: positivism seeks objective, measurable law-like generalisations; interpretivism seeks meaning in context; critical theory examines power and emancipation.

Confirmation bias is seeking or weighting evidence that supports a prior belief; the remedy is pre-specified predictions and disconfirming tests.

Correlation does not entail causation; causal claims require temporal precedence, mechanism, and control of confounders.

Ethical principles: merit and integrity, justice, beneficence, and respect for persons; human research requires informed consent and ethics approval.

A research question must be answerable, specific, and feasible; operationalise variables into measurable indicators before designing the method.

A hypothesis is a testable, directional or non-directional prediction linking independent and dependent variables; a null hypothesis states no effect.

Independent variable is manipulated or selected; dependent variable is measured; controlled variables are held constant; extraneous variables are nuisance influences.

Validity types: internal (causal inference), external (generalisation), construct (measures the intended concept), and reliability (consistency of repeated measurement).

Sampling: random sampling supports generalisation; stratified sampling preserves subgroup proportions; convenience sampling risks selection bias.

A literature review establishes the gap, defines key terms, and justifies the method; it must cite sources and avoid unsupported claims.

Risk assessment identifies hazards, likelihood, severity, and mitigation; human or animal participants require ethics approval before data collection.

A feasible timeline and budget specify milestones, resources, and contingencies; scope creep is controlled by fixing the research question early.

Data types: categorical (nominal, ordinal) versus numerical (discrete, continuous); the type determines permissible statistics and graphs.

Descriptive statistics summarise centre (mean, median, mode) and spread (range, interquartile range, standard deviation); the median resists outliers.

A p-value is the probability of data at least as extreme as observed, assuming the null hypothesis is true; it is not the probability the null is true.

Statistical significance (typically p < 0.05) is not practical significance; report effect size and confidence intervals alongside p-values.

A 95% confidence interval means that 95% of intervals from repeated sampling would contain the true parameter; it is not a 95% chance the parameter lies within one interval.

Type I error rejects a true null (false positive); Type II error fails to reject a false null (false negative); power is 1 minus the Type II error rate.

Measurement error reduces reliability and attenuates observed effects; calibrate instruments and report uncertainty with appropriate significant figures.

Conclusions must be limited to the evidence: distinguish supported inference from speculation, acknowledge limitations, and state what further data would test the claim.

Standard structure: title, abstract, introduction, methods, results, discussion, conclusion, references, and appendices for raw data.

The abstract states aim, method, key results, and conclusion in a single self-contained paragraph, typically 150-250 words.

Methods must be reproducible: report participants or materials, apparatus, procedure, variables, and analysis in sufficient detail for replication.

Results report findings without interpretation; figures and tables carry numbered captions, labelled axes, units, and error bars where relevant.

Discussion interprets results against the hypothesis and literature, explains limitations, and states implications without overclaiming.

Referencing: in-text citations must match the reference list exactly; use one consistent style and acknowledge all borrowed ideas.

Scientific writing uses precise, concise, objective language, past tense for methods and results, and avoids unsupported causal verbs.

Data integrity requires honest reporting of all results, including non-significant findings, and retention of raw data for verification.