Skip to Main Content

Good review practice: a researcher guide to systematic review methodology in the sciences of food and health



Attrition bias: a bias introduced by missing data or study subjects which will lead to bias in analysis.

Bias: the deviation of outcomes from the truth.

Body of evidence: the totality of evidence that is used to substantiate certainty in the outcomes of independent studies.

Conflicts of interest: a set of circumstances by which a reviewer’s ability to influence the results are judged to conflict with their other interests, either professional or personal.

Critical appraisal: the process of systematically evaluating the outcomes of independent studies (evidence) using explicit and standard measures of validity.  

Detection bias: a bias introduced by an artefact that results in erroneous estimation of the outcomes.

Eligibility criteria: the acceptability measures by which the individual studies are judged for inclusion or exclusion from a systematic review.

Evidence hierarchies: a ranking approach that assigns different levels of validity or trustworthiness to different research methods or study designs based on the principles of rigor.

Evidence statements: short summaries of evidence from published literature that are gathered to help decision making in policy settings.

External validity: the degree to which the outcomes of an independent study reflect the truth beyond the subject study.

Grey literature: literature published by entities that are not commercial publishers.  Not peer-reviewed, it includes research reports, white papers, and policy documents produced by governmental agencies, industry, and research organizations; theses; conference proceedings; and preprints.  

Internal validity: the degree to which the outcomes stand true in an individual study. 

Meta analysis: a statistical technique that combines the results of individual studies to provide a more accurate estimation of their outcomes. 

Performance bias: a bias that results in systematic differences between investigated subject groups or participants due to poor compliance or poor conduct.   

Plain language: a writing that is clear and easy to read, understand and use entirely. 

Risk of bias: the likelihood of a systematic error or deviation from the truth in outcomes.

Secondary analyses: re-analysis of existing data collected from a review study for a purpose distinct from the original study.

Secondary outcomes: any important outcomes that are not pre-specified but tend to arise from the review in addition to primary outcomes.

Selection bias: a bias introduced when subjects or participants of a study are selected in a way that they are not truly representing the subjects or participants intended to be studied.

Selective outcomes bias: a bias that results in systematic differences between reported and unreported results by tendency to select and report findings based on the results.