Table 11. Commonly used quantitative data analysis methods for IR

Purpose Description Example of data analysis techniques
To describe The aim of analysis is to use summary statistics to describe study variables within a study population.

Summary statistics (also known as descriptive statistics) are used to describe the distribution of quantitative variables, in terms of: (i) location (central tendency) such as a mean, median or mode; and (ii) spread (variation) such as standard deviation, percentiles or range.

Frequency distributions are useful for categorical data (for example age groups, or responses to Likert-scale questions). They are easy to explain and interpret for audiences without specialist knowledge and can be presented graphically and in different formats to aid interpretation (e.g. tables, bar/pie charts etc.).

To compare The analysis aims to determine if observed differences between two or more groups are due to chance and to produce an effect estimate.

T-tests are used for continuous outcome data to determine if the difference between the means of two groups may be due to chance. The difference in means between the two groups can be reported alongside the associated 95% confidence interval.

Chi square tests (2) are used for categorical data to find out if observed differences between proportions of events in groups is due to chance.

Finding causality: group comparison Group comparison analysis is used to explore the if differences in the study outcomes between groups is due to chance. The groups can be categorized by exposures under study. When there is a meaningful difference between groups, we assume that the difference is due to the exposures.

Chi square is used to assess differences in nominal or ordinal outcomes between independent groups.

Independent t-test is used to assess differences in continuous or ratio outcomes between independent groups.

Paired t-tests are used to assess differences in nominal or ordinal outcomes between paired groups (such as measurements taken from the same group pre- and post-intervention)

Finding causality Regression analysis is the type of analysis used to predict study outcome from a number of independent variables.

Linear regression is used for continuous or ratio scale outcome variables

Logistic regression is used for dichotomous outcome variables, e.g. a variable with only two possible values (e.g. yes or no).