Fixed designs: some fundamental features

Fixed designs are theory driven. The only way in which we can, as a fixed design requires, specify in advance the variables to be included in our study and the exact procedures to be followed, is by having a reasonably well articulated theory of the phenomenon we are researching.

Causal models (like the one here) help to make clear the multiple causality of most things studied in social research. It emphasizes that while some variables may have a direct effect on others (e.g. that of assessment beliefs on assessment methods), others mhave an indirect effect (that of epistemological beliefs on assessment methods through assessment beliefs). Hard thinking to establish this kind of model before data collection is invaluable.

Fixed designs should always be piloted. Fixed designs are usually concerned with aggregates, with group properties and with general tendencies. Most non-experimental research also deals with averages and proportions. The relative weakness of fixed designs is that they cannot capture the subtleties and complexities of individual human behaviour. Even single case designs are limited to quantative measures of a single, or a small number, of simple behaviours. The advantage of fixed designs lies in their ability to transcend individual differences and identify patterns and processes which can be linked to social structrures and group organizational features.

Generally, researchers using fixed design methods remain at a greater physical and emotional distance from the study than those using flexible, qualitative methods.

The concept of 'variable' is part of the quantative worldview central to fixed design research. In this worldview, reality is conceptualized as consisting of properties of things, properties that can vary and thus be measured and compared. These properties may be seen as categorical (i.e. gender, nationality, teaching style) or as continuous (i.e. age, knowledge of some topic). It is very important to identify what variables are being studied, and to distinguish different types of variables (such as independent and depentdent variables).
independent variables: those which the experimenter manipulates.
dependent variables: those were we look for change.


issues of Validity, generalisability objectivity, credibility (p.100-109).

References:
Real World Research: A Resource for Social Scientists and Practitioner-Researchers, 2nd Edition. Colin Robson. (p.96-109)