Select appropriate design

The purpose of research design is to ensure the research will meet your objective(s) accurately and impartially. There are various designs that can be applied, some of which are listed below:

Cross sectional

A cross sectional study, perhaps the most common approach to undertake surveys, involves researching to observe and understand what is happening at a specific point in time. The advantage of this type of study is that several variables can be measured at the same time, providing a “snapshot” of the specific situation. Although it cannot determine causality, it can determine “correlation” ie what is linked. For example, the Public Library Users Survey (PLUS) deprivation report 2012/13 identified a correlation between visitors living in deprived areas and library computer use. PLUS has in fact been a succession of such snapshots for over a decade and this provides evidence to show changes in the profile of users, increases or decreases in the usage of services and changes in satisfaction to cite a few examples..


Longitudinal studies involve the measurement of changes over time for an individual or group of individuals (cohort). Questions may be asked to the same cohort on a number of occasions, with the advantage of being able to explore how respondents have changed or developed their actions, characteristics or attitudes. This type of study could be used to analyse changes after alterations to a library or archive service; however, like the cross-sectional studies, it cannot prove causality but can illustrate correlation.

Case Study

Case studies are forms of research that observe individuals or groups, often within a set location such as a community or an organisation. Researchers may be required to examine the activities of a particular department, which could be used as an example of best practice. A case study can comprise of desk research and interviews from respondents who are familiar with the case, and typically describe a scenario that the individual or group has faced and /or addressed.