MP SET key terms

The degree to which a test or instrument measures what it is intended to measure, supported by evidence of content, construct, and criterion relationships rather than by appearance alone.

The consistency or dependability of scores across repeated administrations, parallel forms, or raters. It is a necessary but not sufficient condition for validity, and is often estimated by correlation coefficients.

Evidence that a test measures the theoretical trait it claims to measure, established through convergent and discriminant relationships with other measures, factor analysis, and experimental interventions.

A reliability estimate obtained by administering the same test twice to the same group after an interval and correlating the two sets of scores. Suitable for stable traits, not for fluctuating states.

An internal consistency estimate computed by dividing a test into two halves, correlating scores, and applying the Spearman-Brown formula to correct for the reduced length of each half.

An index of internal consistency equal to the average of all possible split-half reliabilities. Values range from 0 to 1; higher values indicate items measuring a common underlying dimension.

A statement specifying the concrete procedures or observations used to measure a variable, allowing replication. For example, defining aggression as the number of button presses delivering a noise blast.

A statistical statement that no effect or relationship exists in the population. Researchers test whether sample data provide sufficient evidence to reject it in favour of an alternative hypothesis.

Rejecting a true null hypothesis, a false positive. Its probability is controlled by the chosen significance level, conventionally 0.05, and it is balanced against the risk of Type II error.

The pre-set probability threshold for rejecting the null hypothesis, commonly 0.01 or 0.05. It defines the critical region of a test statistic and reflects the researcher's tolerance for Type I error.

A measure of dispersion giving the average distance of scores from the mean, computed as the square root of the variance. It uses the same units as the original data.

A symmetric, bell-shaped probability distribution defined by its mean and standard deviation, in which approximately 68 percent of cases fall within one standard deviation of the mean.

A standard score expressing how many standard deviations a raw score lies above or below the mean. It permits comparison of scores drawn from different distributions.

The difference between a sample statistic and the corresponding population parameter arising from chance variation in the sample drawn. It decreases as sample size increases.

The standard deviation of a sampling distribution of a statistic, indicating how much a sample estimate is expected to vary from sample to sample. It shrinks with larger samples.

A selection procedure in which every member of the population has an equal and independent chance of inclusion, reducing systematic bias and supporting generalisation of findings.

A sampling technique in which the population is divided into homogeneous subgroups and random samples are drawn from each, ensuring representation of key characteristics and reducing sampling variability.

A cyclical, practitioner-led inquiry aimed at solving a local problem through planning, acting, observing, and reflecting. It prioritises practical improvement over generalisable theory building.

A testable, falsifiable proposition about the relationship between variables, derived from theory or observation and stated before data collection so that predictions can be checked.

Presenting another person's words, ideas, data, or images as one's own without adequate attribution. It includes self-plagiarism and applies to drafts, presentations, and code as well as published text.

The formal acknowledgement of sources used in scholarly work, giving enough detail for readers to locate the original. Style varies by discipline, but accuracy and consistency are always required.

Principles governing responsible conduct of research, including honesty in reporting, informed consent, confidentiality, minimising harm to participants, and proper acknowledgement of others' contributions.

Voluntary agreement by a participant to take part in research after being told the purpose, procedures, risks, benefits, and right to withdraw at any time without penalty.

The researcher's obligation to protect participants' identities and data from unauthorised disclosure, using anonymisation, secure storage, and restricted access throughout the study and afterwards.

Pre-publication evaluation of a manuscript by independent experts who assess originality, methodological soundness, and significance, usually anonymously, to advise the editor on acceptance.

A journal-level metric equal to the average number of citations in a given year to articles published in that journal during the two preceding years. It is often criticised as a proxy for individual quality.

The theory and practice of teaching, encompassing the methods, strategies, and principles a teacher uses to support learning, together with the reasoning behind those choices.

Malcolm Knowles's model of adult learning, assuming learners are self-directed, bring experience to the task, and prefer problem-centred, immediately applicable content over subject-centred instruction.

A learning theory holding that learners actively build knowledge by connecting new information to prior understanding, rather than passively receiving transmitted content from a teacher.

Vygotsky's term for the gap between what a learner can do unaided and what they can achieve with guidance from a more capable other, the space where instruction is most productive.

Temporary support provided by a teacher or peer that helps a learner complete a task beyond their independent ability, gradually withdrawn as competence develops.

Assessment conducted during learning to gather evidence about progress and adjust teaching, typically low-stakes and feedback-oriented rather than used for grading or certification.

Assessment carried out at the end of a unit or course to judge achievement against standards, often used for grading, certification, or accountability rather than for immediate instructional adjustment.

A hierarchical classification of cognitive objectives, revised as remember, understand, apply, analyse, evaluate, and create, used to design learning outcomes and align assessment tasks.

A precise statement of what a learner should know, value, or be able to do by the end of a unit, phrased with observable action verbs so that achievement can be assessed.

Biggs's principle that intended learning outcomes, teaching and learning activities, and assessment tasks should all point toward the same verbs and level of understanding.

An assessment interpreted by comparing a learner's performance to a defined standard or set of objectives, not to the performance of other test takers.

An assessment interpreted by comparing a learner's score with those of a reference group, reporting relative standing through percentiles or standard scores rather than absolute mastery.

Awareness and regulation of one's own thinking, including planning how to approach a task, monitoring comprehension while working, and evaluating strategies afterwards.

The process by which learners set goals, select strategies, monitor progress, and adjust effort, guided by self-efficacy beliefs and feedback rather than by external direction alone.