Skip to content
phdguide

Knowledge Hub

A free, plain-English encyclopedia of research — concepts, statistical tests, terminology and tools, explained for scholars at every level.

A–Z index

Jump to a letter

A

Abstract

A concise summary of a study — its aim, method, key findings and contribution — usually 150–300 words, written last but read first.

AMOS

IBM software for covariance-based Structural Equation Modelling (CB-SEM), used to test path models and confirmatory factor analysis with graphical model building.

ANOVA

Analysis of Variance: a test comparing the means of three or more groups to see whether at least one differs significantly.

B

Bibliometrics

The quantitative study of publications and citations — for example mapping a field with tools like VOSviewer or Biblioshiny.

C

Conceptual Framework

A diagram or narrative showing the variables in a study and the hypothesised relationships between them, derived from theory and literature.

Confirmatory Factor Analysis (CFA)

A technique that tests whether measured items load onto the latent constructs a theory predicts, assessing construct validity.

Construct Validity

The extent to which a measurement instrument actually captures the abstract concept it claims to measure.

Cronbach's Alpha

A coefficient (0–1) estimating the internal-consistency reliability of a multi-item scale; values of 0.7 and above are commonly considered acceptable.

Cross-sectional Study

A design that collects data at a single point in time, giving a snapshot rather than tracking change.

D

Dependent Variable

The outcome a study seeks to explain or predict — the variable expected to change in response to the independent variable(s).

DOI

Digital Object Identifier: a permanent, unique link to a published article or dataset that does not break when a URL changes.

E

Effect Size

A standardised measure of the magnitude of a relationship or difference (e.g. Cohen's d, r), reported alongside statistical significance.

Exploratory Factor Analysis (EFA)

A data-driven technique that uncovers the underlying factor structure among a set of variables when it is not known in advance.

G

Grounded Theory

A qualitative methodology that builds theory inductively from data through iterative coding, rather than testing a pre-set hypothesis.

H

h-index

A metric of an author's productivity and impact: the largest number h such that h of their papers each have at least h citations.

Hypothesis

A testable, falsifiable statement predicting a relationship between variables, stated as a null (no effect) and an alternative (an effect).

I

Impact Factor

A journal-level metric reflecting the average number of citations to recent articles in that journal; one signal of journal standing, not article quality.

Independent Variable

A variable the researcher manipulates or observes as a presumed cause of change in the dependent variable.

Informed Consent

A research-ethics requirement that participants voluntarily agree to take part after understanding the study's purpose, risks and their rights.

L

Likert Scale

An ordered response scale (e.g. 1 = strongly disagree to 5 = strongly agree) used to measure attitudes and perceptions.

Literature Review

A structured, critical synthesis of prior research that establishes what is known, what is contested, and where the gap lies.

Longitudinal Study

A design that collects data from the same subjects over multiple time points to observe change or causal ordering.

M

Mediation

A relationship in which a third variable explains the mechanism through which an independent variable affects a dependent variable.

Meta-analysis

A statistical synthesis that pools effect sizes from multiple independent studies to estimate an overall effect.

Mixed Methods

A design that deliberately combines quantitative and qualitative data to answer a question more completely than either alone.

Moderation

A relationship in which a third variable changes the strength or direction of the link between two other variables.

N

Null Hypothesis

The default statement that there is no effect or no relationship; statistical tests seek evidence to reject it.

NVivo

Software for managing and analysing qualitative data — coding interviews, documents and media into themes and running queries.

O

Operationalisation

The process of defining an abstract construct in measurable terms — specifying exactly how it will be observed or scored.

P

p-value

The probability of obtaining results at least as extreme as observed if the null hypothesis were true; below a threshold (often 0.05) it is deemed statistically significant.

PLS-SEM

Partial Least Squares Structural Equation Modelling: a variance-based SEM approach (run in SmartPLS) suited to prediction, complex models and smaller samples.

Population

The entire group about which a study seeks to draw conclusions; the sample is drawn from it.

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses: a checklist and flow diagram for transparent, reproducible review reporting.

Q

Qualitative Research

Inquiry that explores meaning, experience and context through non-numeric data such as interviews, observation and documents.

Quantitative Research

Inquiry that measures variables numerically and tests relationships or differences using statistics.

R

Regression

A family of techniques modelling how one or more predictors relate to an outcome, estimating the size and significance of each effect.

Reliability

The consistency of a measure — whether it yields the same results under repeated or equivalent conditions.

Research Design

The overall plan linking research questions to data — deciding what will be measured, from whom, when and how it will be analysed.

Research Gap

An unanswered question, untested context or unresolved contradiction in the literature that a study can credibly address.

Research Question

The precise, answerable question a study sets out to investigate, guiding design, data and analysis.

S

Sample Size

The number of observations in a study; adequate size is needed for statistical power and credible estimates.

Sampling

The method of selecting units from a population — probability methods (e.g. random) support generalisation; non-probability methods (e.g. convenience) do not.

Scopus

A large abstract-and-citation database of peer-reviewed literature, widely used to assess journal indexing and research impact.

SmartPLS

Software for PLS-SEM, used to estimate measurement and structural models, run bootstrapping and assess predictive relevance.

SPSS

IBM Statistical Package for the Social Sciences: widely used software for descriptive and inferential statistics such as t-tests, ANOVA and regression.

Structural Equation Modelling (SEM)

A framework that simultaneously tests measurement (items to constructs) and structural (construct to construct) relationships.

Systematic Literature Review (SLR)

A review using an explicit, reproducible protocol to search, screen, appraise and synthesise all relevant studies on a question.

T

Thematic Analysis

A qualitative method for identifying, analysing and reporting patterns (themes) across a dataset.

Theoretical Framework

The established theory (or theories) a study is built upon, used to justify hypotheses and interpret findings.

Turnitin

A similarity-detection service that compares submitted text against a large corpus; a tool to support originality, not a verdict on plagiarism by itself.

V

Validity

The degree to which a study measures what it intends to and supports sound, defensible conclusions.

Variable

Any characteristic that can take different values across subjects or conditions — independent, dependent, mediating, moderating or control.

Viva Voce

The oral examination in which a candidate defends their thesis and research decisions before examiners.

W

Web of Science

A selective, multidisciplinary citation database used to identify indexed journals and track citation impact.

Not sure where to start?

Book a free 15-minute consult. We'll map your next three steps — no obligation.

Ethical, compliant guidance

We provide academic support, mentoring, analysis, editing and structuring — not authorship. Your work stays compliant with university policies.