Please enable JavaScript.
Coggle requires JavaScript to display documents.
Research Design, (Elimination relies on eliminating extra variable by…
Research Design
It is a comprehensive plan for data collection in an empirical research project. It is aimed at answering specific questions or testing specific hypotheses and must specify at least three processes-data collection process, instrument development process and the sampling process.
Data collection methods can be grouped into two categories: positivist methods and interpretive methods.
-
-
-
Elimination relies on eliminating extra variable by holding them constant across treatment such restricting the study to a single gender
Statistical control allows for extra variables to be measure and used as covariables during the testing process
Inclusion is includingextra variables in the research design and estimating the effect on dependent variable
-
Manipulation of the independent variables in one or more level ("treatments") and comparing the effects of treatments against a control group. Achieved in experimental or quasi-experimental designs
The quality of research design can be defined in terms of design attributes: internal validity, external validity, construct validity and statistical conclusion validity
External validity refers to whether the observed associations can be generalized for the sample to the population or to other people or organizations
Statistical conclusion validity : examines the conclusions derived using statistical procedure is valid
Construct validity examines how well a given measurement scale is measuring the construct that it is expected to measure
Internal validity also called causality examines whether the change in a dependent variable is indeed caused by a corresponding change in hypothesized independent variable. Requires three conditions: covariation of cause and effect, temporal precedence and no plausible alternative explanation
Positivist Methods: Methods aimed at theory (or hypotheses) testing and employ a deductive approach to research starting with a theory and testing postulates using empirical data. Examples include surveys, laboratory experiments.
Interpretive Methods: Employs an inductive approach that starts with data and tries to derive a theory from the observed data. Relies heavily on qualitative data. Examples include ethnography and action research
Experimental Studies : Intended to test cause-effect relationships (hypotheses) in a controlled setting by separating the cause from the effect and administering the cause to one group ("treatment group" and not the other group ("control group")
Field Surveys: a non-experimental design that does not control or manipulate independent variables or treatment but measures the variables and tests their effects using statistical methods. Can capture snapshots of practices, beliefs or situations from random sample of subjects
Secondary data analysis: an analysis of data that has previously been collected and tabulated by other sources such as employment statistics from Labor Services
Focus group research : involves bring a small group of subject and having them discuss a phenomenon of interest for a period about two hours. Discussion is moderated by a trained facilitator
Action research : introduces interventions or "Action" into the phenomena and observes the effect of those actions. Usually a consultant or organization member who initiates action such as new procedures or technologies in response to a problem.
Ethnography: an interpretive design inspired by anthropology that emphasizes that phenomenon must be studied with the context of culture. Research is immersed in culture for an extended period of time and engages, observes and records in great detail. Data is collected primarily via observational techniques and formal and informal interactions for participants.
Case research: an in-depth investigation of a problem in one or more real life setting over an extended period of time. Data is collected using interviews, personal observations and internal and external documents.
-