Section 3. General Research Approaches

3.1 Type of Research.
Initially, a researcher must decide what type of research is to be conducted: confirmatory or exploratory research. Hair et al. (1995) suggest that confirmatory studies are those seeking to test (confirm) a prespecified relationship, whereas exploratory studies are those which define possible relationships in only the most general form and then allow multivariate techniques to estimate a relationship(s). In the latter case, the researcher is not looking to "confirm" any relationships specified prior to the analysis, but instead allows the method and the data to define the nature of the relationships.


3.2 General Research Approaches.
A perspective on QPR is based on Jenkins (1985). Using Stone's categorization scheme (1978) as a basis, he classifies strengths and weaknesses of a wide range of research methods (also referred to as "research approaches" or "methodologies"). Evaluating these characteristics, it is possible to sub-divide these methods into QPR and non-QPR methods.

QPR Methods	Non-QPR Methods
Field experiment	Math Modeling (analytical modeling)
Lab experiment	Group feedback
Free simulation experiment	Participative research
Experimental simulation	Case study
Adaptive experiment	Philosophical research
Field study
Opinion research
Archival research
Table 1. QPR versus Non-QPR Methods (Click on the method for its definition)

It should be noted that the choice of a type of research and a general research approach does not "force" a particular data collection technique (Section 4) or data analysis technique (Section 5). It may, however, influence it, as a general research approach may be better served by a subset of all possible data collection and analysis techniques available. Figure 3 distinguishes the four basic choices quantitative researchers will need to make in the conduct of their research.

Figure 3. Type of Research, General Research Approaches, Data Collection Techniques, & Data Analysis Techniques

Let us illustrate the point. A researcher may decide to conduct a survey for data collection, but this is not a sufficient description of the research approach in that the general research approach could be either a case study or a field study, for example. Some people even refer to the "experimental instrument" that is used to gather data in an experiment as a survey, so the researcher does not clarify her/his resarch design by stating that a survey was conducted.

Moreover, there is frequent confusion in IS studies about the difference between case studies and field studies. One frequently reads claims that a researcher has conducted "multiple case studies on nine (or so) firms." Case studies are generally so intensive that the researcher cannot venture to more than one or two sites. Nine sites suggests that the data collection effort was restricted in each site and that the general research approach is not a case study, but, in fact, a field study.

Continuing confusion in what IS researchers term the general research approaches to be impedes further work in the field since meta-analyses will often take a resarcher at her/his word and misclassify the general resarch approach. More agreement with and adherence to standard definitions will benefit all concerned.