Glossary - Research Basics and Terminology


This glossary explains common terminology used in research articles. It helps you to understand the articles which are included in the following chapters of this course. 

If you are looking for more terminology, the glossary of 'The Cochrane Collaboration' is a useful resource which you can find here

Browse the glossary using this index

Special | A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z | ALL

Page:  1  2  3  4  5  (Next)
  ALL

A

Alpha-Level (α-level)

(Last edited: Dienstag, 14 März 2017, 8:55 )

  • Significance level
  • Probability of rejecting H0 when H0 is true
  • Usually set at α=0.05 or α=0.01
Entry link: Alpha-Level (α-level)

Alternative Hypothesis (H1 or Ha)

(Last edited: Dienstag, 14 März 2017, 8:55 )
  •  Experimental hypothesis
  • The population parameter differs from H0
  • When H0 can be rejected, it does not mean that H1 can be accepted, because of the remaining probability of α- and β-error

 

Entry link: Alternative Hypothesis (H1 or Ha)

B

Bias

(Last edited: Dienstag, 14 März 2017, 8:55 )

  • Systematic error or deviation in results
  • Example: selection bias

Systematic differences in the groups that are compared: group A has 50% more smokers than group B 

→ groups are not comparable


Entry link: Bias

Blinding

(Last edited: Dienstag, 14 März 2017, 8:55 )

 

 

  • Process of hiding which comparison group a particular participant belongs to
  • Used to minimize risk of bias
  • Common methodology in clinical trials
  • Types:
  1. Single blind: participants unaware
  2. Double blind: participants+ outcome assessors/ testers unaware
  3. Triple blind: participants+ outcome assessors/ testers+ data analysts unaware

 

Entry link: Blinding

C

Confidence Interval (CI)

(Last edited: Dienstag, 14 März 2017, 8:55 )

 

 

  • Measure of uncertainty around main finding of statistical analysis
  • Usually set at CI=95%, also CI=99% or CI=90%

 

Entry link: Confidence Interval (CI)

Continuous Data

(Last edited: Dienstag, 14 März 2017, 8:55 )

  • Information that can take any value within a certain range
  • Opposite of discrete data
  • Example: Weight of a person: The weight can be any value within the range of people´s weight (e.g. 60.3 kilogram).

Entry link: Continuous Data

D

Dependent or Outcome Variable

(Last edited: Dienstag, 14 März 2017, 8:55 )

 

 

Clinical trial

Dependent variable= Outcome (ill or healthy)

Independent variable= Treatment arm (new drug)

 

Entry link: Dependent or Outcome Variable

Discrete Data

(Last edited: Dienstag, 14 März 2017, 8:55 )

  • Information that can only take certain values
  • Opposite of continuous data
  • Example: Number of people in a room: It is impossible to have 40.3 people in a room.

Entry link: Discrete Data

F

False Negative

(Last edited: Dienstag, 14 März 2017, 8:55 )

= Negative test when individual actually has the disease


Entry link: False Negative

False Positive

(Last edited: Dienstag, 14 März 2017, 8:55 )

= Positive test when individual does not actually have the disease


Entry link: False Positive


Page:  1  2  3  4  5  (Next)
  ALL