Page 78 - Academic and Research Book_WEB

natural sciences. This foundation was applied to the survey and quantitative approaches in social
science research. Positivism is often interpreted as empiricism, which is based on the premise of seeing
is believing. Qualitative research is an outcome of the challenge to the positivist paradigm. Thus,
qualitative research is said to be based on post-positivist stance, where knowledge could be based on
human conjectures.

While, some areas of social science research such as, psychology, social psychology, education, and
behavioural economics make use of an experimental approach, other disciplines such as, sociology
and linguistics make use of the non-experimental approach. This could be survey-based or could
utilize the designs of the qualitative approach.

Designs

Five types of research designs are described here: experimental; cross-sectional or survey; longitudinal;
case study; and comparative. The details of these designs can be found in any good reference book on
research methodology (See Bryman 2008b).

Experimental Design

As already discussed experiments are undertaken in disciplines such as psychology, social
psychology, behavioural economics, and geography, among others. Some researchers, however, also
utilize experimental design to examine the impact of policies on various sections of society. These
experiments differ from the laboratory experiments typically used in pure sciences. Laboratory
experiments are undertaken, mainly in pure sciences, in a closed environment. Field experiments, on
the other hand, are undertaken in real-life settings. Experiments in social science research are also
undertaken in real-life settings. Randomized control trials, in which the samples are organized in two
groups — controlled and experimental, is one of the popular experimental designs in social science
research today.

Cross-sectional Design

Cross-sectional design is also known as survey design. Those with a limited exposure to research
often construe survey research in the narrower context to include only questionnaires and structured
interviews. Therefore, the terminology “cross-sectional design” is preferred to provide a wider
dimension to the survey research. Cross-sectional/survey design includes, but is not limited to,
structured observation, content analysis, official statistics, and diaries, over and above questionnaire
and structured interviews.

Longitudinal Design

When a sample is surveyed on more than one occasion, at least twice, it is said to be a longitudinal
design. Longitudinal designs are broadly classified as: (1) the panel design, and (2) the cohort. A
panel design administers the same survey on the same set of samples (which implies, same number of
samples) twice or more. The British Household Panel Survey (BHPS) is an example of panel design. A
cohort, on the other hand, is a sample selected from the people sharing the same characteristics, for
example, the millennials. The sampled element and the sample size in a cohort may remain the same
or may vary in each subsequent round of the survey. But the sample continues to be drawn from the
same cohort, for example, millennials, in each round. In either design, the selection of the sample is

ACADEMIC INTEGRITY AND RESEARCH QUALITY | 67