Research Design

Photo of a hand holding a calculator and two graphs

Time to spend on this section: 2 hours

This section of the theme will discuss the processes involved in the selection of the most appropriate research design and methods for your project

Selecting your research design

This section is adapted from Bryman, A. (2004). Social Research Methods (2 nd Ed.) Oxford University Press, Oxford/ New York.

When selecting a framework for the collection and analysis of data, it is essential that you consider three important criteria: reliability, replication and validity.

Reliability

In general, reliability is concerned with the question of whether the results of a study are repeatable. It is an indication of the ability of a system to perform and maintain its functions consistently in routine circumstances as well as hostile or unexpected circumstances. Reliability is particularly important in quantitative research and may refer to:
  • The statistical reliability of a set of data
  • The experimental reliability of an experiment
  • Data reliability, a property of some disk arrays in computer storage
  • Reliability engineering ensures a system will be reliable when operated in a specified manner
  • Reliability theory, as a theoretical concept, to explain biological aging and species longevity
  • Reliability (computer networking) is a category used to describe protocols.

Replication

It is sometimes necessary for researchers to replicate (i.e. reproduce or duplicate) the findings of others; in order for this to happen, a study must be replicable. A study must be replicable in order that the reliability of a measure or a concept can be determined. Replications should not be confused with repeated measurements which refer literally to taking several measurements of a single occurrence of a specific phenomenon.

Validity

Validity is concerned with the integrity of the conclusions that are generated from a piece of research. A valid measure is one which is measuring what it is supposed to measure. A valid measure must be reliable, but a reliable measure need not be valid. Validity refers to obtaining results that accurately reflect the concept being measured and it implies reliability (consistency).

The main types of validity that are typically distinguished include:

  • Measurement (or construct) validity e.g. does an IQ test really measure variations in intelligence?
  • Internal validity e.g. if we suggest that x causes y, can we be sure that it is x that is responsible for the variation in y and not something else?
  • External validity e.g. can the results of a study be generalised beyond the specific research content?
  • Ecological validity e.g. are social scientific findings applicable in people’s everyday, natural social settings?

According to classical test theory, validity cannot exceed reliability.

There are five main types of research design:

Experimental

A true experiment is often used as a yardstick against which non-experimental research is assessed. A laboratory experiment takes place in a laboratory or in a contrived situation, whereas field experiments take place in real-life settings, such as in rivers, organisations or classrooms.

Cross-sectional

A cross-sectional design entails the collection of data on more than one case and at a single point in time in order to collect a body of quantifiable or quantitative data in connection with two or more variables, which are then examined to detect patterns of association e.g. survey research in which data are collected predominantly by questionnaire or by structured interview.

Longitudinal

Examine the change of a system (e.g. a study of changing relationships in a troop of chimps in Africa) over time. Examples include the British Household Panel Survey (BHPS), the National Child Development Survey (NCDS).

Case study

A basic case study entails the detailed and intensive analysis of a single case and is concerned with complexity and particular nature of the case in question e.g. studies on a single community, school, organisation, event or individual.

Comparative

This design simply entails a study using more or less the same identical methods of two contrasting cases.

IDevice Icon Activities 18 and 19: Selection of Research Design and Related Issues

Activity 18 requires you to briefly describe 5 different research designs from your discipline, or school and consider the most appropriate one to adopt for your own research. . Activity 19 encourages you to select from a list issues that may impact on your research design. Now open your log book and complete this Activity 18 and 19.