Keywords and Phrases

Definition: In a statistical sense, bias refers to any type of error or systematic distortion that occurs within a study. 

Source: Piedmont, R. (2014). Bias, Statistical. Encyclopedia of Quality of Life and Well-Being Research. (pp. 382-383).

Definition: Within the context of an experimental study, the term control group refers to those individuals who, for the purposes of comparison, do not receive the treatment that is being studied. 

Source: Everitt, B., & Skrondal, A. (2010). Control group. The Cambridge Dictionary of Statistics. (pg. 104). Cambridge: Cambridge University Press.

Definition: An effect size is a measure of the magnitude of the difference between two groups in a study for a given attribute or outcome. In this context, this term generally refers to the difference between outcomes of the intervention group (i.e. those that received the treatment) and the control group (i.e. those that did not receive the treatment). 

Source: Everitt, B., & Skrondal, A. (2010). Effect size. The Cambridge Dictionary of Statistics. (pg. 148). Cambridge: Cambridge University Press.Centre for Evidence Based Intervention (CEBI). Effect Size. Accessed from https://www.cebi.ox.ac.uk/for-practitioners/glossary.html

Source: Centre for Evidence Based Intervention (CEBI). Effect Size. Accessed from https://www.cebi.ox.ac.uk/for-practitioners/glossary.html

Definition: Results First uses findings from impact evaluations and outcome analyses generated from rigorously implemented experimental and quasi-experimental designs.

Source: Petrosino, A. Boruch, R. (2014). Evidence-Based Policy in Crime and Justice. Encyclopedia of Criminology and Criminal Justice. (pp. 1437-1446). Springer New York.

Definition: This research design is used to evaluate how treatment variables and study participants interact under controlled conditions. Experimental research often uses carefully designed and implemented controls such as randomization to minimize statistical bias.

Source: Everitt, B., & Skrondal, A. (2010). Experimental study. The Cambridge Dictionary of Statistics. (pg. 156). Cambridge: Cambridge University Press.

Definition: This term refers to the statistical process for combining the results of similar studies to arrive at a more powerful conclusion about the overall effects of a treatment or program.

Source: Everitt, B., & Skrondal, A. (2010). Meta-analysis. The Cambridge Dictionary of Statistics. (pg. 276). Cambridge: Cambridge University Press.
Source: Centre for Evidence Based Intervention (CEBI). Meta-analysis. Accessed from https://www.cebi.ox.ac.uk/for-practitioners/glossary.html.

Definition: The probability value, or p-value, can help determine the statistical significance of a set of results.

Source:Dodge, Y. (2008). P-value. The Concise Encyclopedia of Statistics. (pp. 434-436). Springer New York.

Definition: This research design is used to evaluate how treatment variables and study participants interact under controlled conditions. Rather than use randomization, this research design uses alternative strategies to control for and minimize statistical bias.

Source: Everitt, B., & Skrondal, A. (2010). Quasi-experiment. The Cambridge Dictionary of Statistics. (pg. 348). Cambridge: Cambridge University Press.

Definition: This term refers to the random allotment of participants into a group within an experiment (i.e. the intervention group or the control group) to reduce statistical bias.

Source: Everitt, B., & Skrondal, A. (2010). Clinical trials. The Cambridge Dictionary of Statistics. (pg. 85). Cambridge: Cambridge University Press.

Source: Centre for Evidence Based Intervention (CEBI). Randomisation. Accessed from https://www.cebi.ox.ac.uk/for-practitioners/glossary.html

Definition: This term refers to the number of individuals who were involved in an experiment. 

Source: Everitt, B., & Skrondal, A. (2010). Sample size. The Cambridge Dictionary of Statistics. (pg. 376). Cambridge: Cambridge University Press.

Definition: In the context of impact evaluations, this term is often used to indicate whether or not the studied treatment is thought to have had an effect on the participants of a study. It essentially refers to the probability that the results of a study are not merely a consequence of random chance. However, it does not provide an indication of the magnitude of the effect of a treatment.

Source:Everitt, B., & Skrondal, A. (2010). Significance level. The Cambridge Dictionary of Statistics. (pg. 393). Cambridge: Cambridge University Press.

Source:Wesstein, E. Significance. MathWorld–A Wolfram Web Resource. http://mathworld.wolfram.com/Significance.html

Definition: A systematic review consists of a comprehensive review on all of the available information and research pertaining to the efficacy of a treatment or program. The results of a systematic review can be used to facilitate a meta-analysis of a particular treatment or program.

Source: Centre for Evidence Based Intervention (CEBI). Systematic review. Accessed from https://www.cebi.ox.ac.uk/for-practitioners/glossary.html.