Overall Differences in Language and Thinking Processes:

• the blueprint for action or the specific procedures used to obtain empirical evidence about a phenomenon.
• is structured in such a way that the relationship among variables can be examined.
• restricts or controls extraneous variables that might have an impact on the study.  This is done so that the researcher can say that the outcomes are a consequence of the manipulation of the independent variable (the intervention in experimental research) or the consequence of what was observed and analyzed (non-experimental research).
• distinguishes research from everyday observations.
• requires the investigator to be objective and unbiased in his/her involvement with the phenomena of study.

Variables:

• a concept or construct to which numerical values have been assigned.
• objects or events which can take on different values. (a variable must have more than one value)

DePoy describes three types of variables:
a)  independent variable: the variable you manipulate; or your treatment.
b)  intervening variable: the confounding variables that could have an effect on the study but are not the variables of interest.
c)  the dependent variable: the variable that the investigator seeks to describe, understand, or predict.  The dependent variable is thought to respond to the independent variable.

Plan of Design:  this requires a set of thinking processes considering bias, manipulation, control, validity, and reliability.

a)  Bias:  potential, unintended, or unavoidable effects on study outcomes; the difference between what is true, and what is "observed" and recorded.
                Sources of bias:
• selection of inappropriate instrumentation.  Some sources of instrumentation bias:

1)  inappropriate data collection
using a data collector who introduces bias to the response (supervisor surveying the employees)
poor training of the data collectors
using an instrument with reliability or validity problems
2)  inadequate questions (questions worded in such a way that a subject is swayed to answer a certain way when his/her true response would have been something else)
• poor sampling technique
• deviation from the plan or structure of the design
b)  Manipulation:  the action process of maneuvering the independent variable so that the effect of its presence, absence, or degree can be observed on the dependent variable.  This allows you to measure a cause and effect relationship.  If there are extraneous variables that might have an impact on your study's outcome, you may include them in the research design for statistical analysis.  The data were analyzed by looking at the influence of all of these variables.  If this is not taken into account during the analysis, the study's outcomes will be less powerful.
c)  Control:  The extent to which the researcher can manage extraneous sources that might affect a study and lead to
incorrect scientific conclusions.  By controlling the assignment of subjects to groups and the research environment, you can better study the relationship among variables.  This is done by random assignment to groups and using control groups.  If the researcher did not control extraneous variables, he/she could not rule out those variables as a reason for the study's outcome.  The important thing to remember is that extraneous variables are evenly distributed between or among groups being studied if there is random selection of subjects and random assignment to groups.  If you control the variables too tightly, your research group is so specific that the generalizability is limited.
d)  Validity:  it is the extent to which your findings are accurate.  DePoy discusses four types:  1) internal validity, 2) external validity, 3) statistical conclusion validity, 4) construct validity.

1)  Internal validity refers to the ability of the research design to accurately answer the research question.  In other words, did the independent variable really make the difference?  Threats to internal validity are defined by DePoy in  Box 8-3:
History
Testing
Instrumentation
Maturation
Regression
Mortality
Interactive effects
2)  External validity refers to the capacity to generalize findings and develop inferences from the sample to the larger population.  Threats to external validity include:
reactivity: the extent to which subjects are responding to just being in the study and
realism: the extent to which the condition simulates real life.
3)  Statistical validity refers to the power of your study to draw statistical conclusions. The use of appropriate tests is important.
4) Construct Validity:  the fit between the actual constructs that are the focus of the study and the way the constructs are operationalized in the study.  Statistics and measurement texts go into a lot more detail on construct and other related forms of validity, but these will not be discussed here.
 
e) "Reliability":  DePoy states that this refers to the stability of the research design, but this is not the accepted use of the term reliability.  Replication is the word used to describe the process that one follows when one is attempting to repeat someone's work, under the same conditions. If the study is repeated under the same conditions, the design should yield the same results.  Reliability is a term that most texts will refer to as the consistency or reproducability of measurement.  Reliability will be discussed in the section on data collection.
 

Qualitative design is:

Context Specific:   in qualitative research, the investigator must go to the setting where the phenomena occur or seek information from the individuals who experience the phenomena of interest.

Complexity and Pluralistic Perspective of Reality:  qualitative research assumes a pluralistic perspective of reality; the end result of inductive reasoning is the development of a complex relationship among smaller pieces of information.

Transferability of Findings:  the purpose of qualitative research is not to generalize from a small sample to a larger group of persons with similar characteristics.  It is a theory-generating tool.  It may reveal unique meanings of human experience in human environments.  A researcher may need to follow up a qualitative study with a quantitative study in order to generalize findings related to the study.

Flexibility: the design in qualitative research is more fluid and flexible.  It is not the blueprint for action; the design evolves.

Language:  a shared concern is understanding the language and its meaning for the people being studied.  People use and understand language differently.  The researcher engages in a rigorous analytical process to translate the meaning and structure of the context of those being studied into the meaning in language structures represented in the world of the researcher.

Emic and Etic Perspectives:

a) emic:  this is the insider's perspective or way of understanding and interpreting experience.  This perspective is often what the qualitative inquiry is looking for.

b) etic:  this is the outsider's orientation and refers to those that are external to the group being studies.  Although external, they select an analytical "lens" through which to examine information.  Those who do not experience a phenomenon can come to know it.

Many investigators integrate both perspectives but qualitative researchers often favor only an emic perspective.

Summary:  Quantitative researchers approach their studies with intact theory and a set of procedures.  Procedures are developed and followed to eliminate the potential for factors other than those being studied to be responsible for the study's findings.  Qualitative researchers believe that knowledge of the whole can be learned by a study of the parts.  Qualitative researchers allow theory and procedures to evolve through the study.  Qualitative researchers assume that multiple realities are created by people.  Value is placed on perceptions and experiences of people.  Subjective experience is recognized by qualitative researchers as being as legitimate as the objective experience.  It considers the wholeness of humans, and qualitative researchers believe that knowledge of the whole is learned by studying the whole experience.
 

© S. Bevins (Spring, 1999), edited by T. Bevins (Summer, 1999)
Taken from Depoy, E. & Gitlin, L. (1998).  Introduction to Research.  St. Louis:  Mosby.
            and LoBiondo-Wood, G. & Haber, J. (1998).  Nursing Research. St. Louis:  Mosby.