Research design – types

Five different types

  1. Experimental design
  2. Cross sectional or social survey design
  3. Longitudinal design
  4. Case study design
  5. Comparative design

EXPERIMENTAL DESIGN

Experiment is often used as a yardstick against which non-experimental research is assessed. It engenders considerable confidence in the robustness and trustworthiness of casual findings. Experiments tend to be very strong in terms of validity.

Business research do not make far greater use of experiment because in order to conduct a true experiment it is necessary to manipulate the independent variable in order to determine whether it does in fact have an influence on the dependent variable. Experimental subjects are likely to be allocated to one of two or more experimental groups, each of which represents different types or levels of the independent variable. Differences between the groups are responsible for variations in the level of dependent variable. Manipulation then, entails intervening in a situation to determine which of two or more things happens to subjects. However, the vast majority of independent variables with which business researchers are concerned cannot be manipulated. Example – if we are interested in the effects of gender on work experience, we cannot manipulate gender so that some people are made male and other female. If we are interested in the effects of variations in the economic environment on organizational performance, we cannot alter share prices or interest rates. As with huge majority of such variables, the levels of social engineering that would be required are beyond serious contemplation.

Distinction between the laboratory experiment and field experiment:

  1. The laboratory experiment takes place in a laboratory or in a contrived setting, whereas field experiment occurs in real-life setting, such as in workplace or retail spaces.
  2. It is experiments of the latter type that are most likely to touch on areas of interest to business and management researchers. However, in business and management research it is more common to find field experiment in which a scenario is employed as a substitute for a real life setting.

Classic experimental design

The classical experimental design, two groups are established and this forms the basis for experimental manipulation of the independent variable. The experimental group, or treatment group, receives the treatment, and it is compared against the control group, which does not. The dependent variable is measured before and after the experiment manipulation, so that a before and after analysis can be conducted. The groups are assigned randomly to their respective groups. This enables the researcher to feel confident that any difference between the two groups is attributable to manipulation of the independent variable.

T₁ T₂
OBS₁ EXP OBS₃
OBS₂ NO EXP OBS₄

 

OBS – An observation made in relation to the dependent variable; there may well be two or more observations, before (the pre-test) and after (the post-test) the experimental manipulation.

Exp – The experimental treatment (manipulation of the independent variable). NO Exp refers to the absence of an experimental treatment and represents the experience of the control group.

T – The timing of the observations made in relation to the dependent variable.

The classical experimental design comprises the following elements:

(A) Random assignment to the experimental and control groups.

(B) Pre-testing of both groups at T₁

(C) Manipulation of the experimental treatment so that the experimental groups receive it (EXP) but the control group does not (NO Exp);

(D) Post-testing of the two groups T₂.

The difference between each group’s pre and post-test scores is then computed to establish whether or not Exp has made a difference.

QUASI EXPERIMENT – studies that have certain characteristics of experimental designs but that don not fulfill all the internal validity requirements. A particularly interesting form of quasi experiment occurs in the case of natural experiments.

Cross sectional design

The cross sectional design is often called a social survey design but the idea of the social survey is so closely connected in most people’s minds with questionnaire and structured interviewing that the more generic-sounding term cross-sectional design is preferable. While the research methods associated with social surveys are certainly frequently employed within the context of cross sectional research, so too are many other research methods, including structured observation, content analysis, official statistics and diaries. A cross sectional design entails the collection of data on more than one case (usually quite a lot more than one) and at a single point in time order to collect a body of quantitative or quantifiable data in connection with two more variables(usually more than two) which are then examined to detect patterns of association. Types:

  • More than one case
  • At a single point in time
  • Quantitative or quantifiable data
  • Patterns of association

Longitudinal design

Longitudinal design represents a distinct form of research design that is typically used to map change in business and management research. Pettigrew has emphasized the importance of longitudinal study in understanding organizations as a way of providing data on the mechanics and processes through which changes are created. However partly because of the time and cost involved, longitudinal design is relatively little used in business and management research.

Case study

The basic case study entails the detailed and intensive analysis of a single case. This case study approach is a very popular and widely used research design in business research, and some of the best-known studies in business and management research are based on this design. A case can be:

  • A single organization
  • A single location-such as a factory, production site, or office building,
  • A person
  • A single event

What is a case? What distinguishes a case study from other research designs is the focus on the bounded situation or system an entity with a purpose and functioning parts. The emphasis tends to be on intensive examination of the setting. There is a tendency to associate case studies with qualitative research, but such identification is not appropriate. It is certainly true that exponents of the case study design often favor qualitative methods. However case studies are frequently sites for the employment of both quantitative and qualitative research. Types of case studies are:

  • Critical case- here the researcher has a clearly specified hypothesis, and a case is chosen on the grounds that it will allow a better understanding of the circumstances in which the hypothesis will and will not hold.
  • The unique case – the unique or extreme case is a common focus in clinical studies.
  • The revelatory case – the basis for the revelatory case exists when an investigator has an opportunity to observe and analyze a phenomenon previously inaccessible to scientific investigation.
  • The representative or typical case – this type seeks to explore a case that exemplifies an everyday situation or form of organization
  • The longitudinal case – this type of case is concerned with how a situation changes over time

Comparative design

This design entails the study using more or less identical methods of two or more contrasting cases. It embodies the logic of comparison, in that it implies that we can understand social phenomena better when they are compared in relation to two or more meaningfully contrasting cases or situations. The comparative design may be realized in the context of either quantitative or qualitative research.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s