Elements of Good Design



Elements of Good Design

Setting

Subjects/sample

Treatment

Measurement

Quantitative vs. Qualitative

Depending on the researcher’s view, a research problem may be addressed with two different approaches

Quantitative: positivist paradigm and directed toward the discovery of relationships and cause and effect

Qualitative: naturalistic paradigm and directed towards the discovery of meaning

Quantitative Vs Qualitative

Quantitative

Hard science

Narrow focus

Reductionistic

Logical approach

Objective

Tests Theory

Instruments

Statistical analysis/numbers

Generalization

Qualitative

Soft science

Broad focus

Holistic

Intuitive approach

Subjective

Develops Theory

Communication/ observation

Individual interpretation

Uniqueness

Quantitative Design

Concepts Important to Quantitative Design

Rigor

Control

Bias

Validity

Rigor

Discipline, accuracy, precision, and order in the research design

Control

Definition: Holding the conditions of the study constant and establishing specific sampling criteria

Established by ruling out extraneous variables that compete with the independent variables as an explanation for a study’s outcome

Means of establishing control:

Homogeneous sample

Use of consistent data collection procedures

Manipulation of the independent variable

Randomization

Bias

Any influence or action in a study that distorts the findings or slants them away from the true or expected

Validity

Internal validity: The extent to which the effects detected in the study are a true reflection of reality rather than being the result of extraneous variables

External validity: The extent to which study findings can be generalized beyond the sample used in the study

Threats to Internal Validity

History

Maturation

Testing

Instrumentation

Selection

Mortality

Threats to External Validity

Hawthorne Effect

Experimenter Effect

Novelty Effect

Reactive Effects of Pretest

Types of Quantitative Designs

Experimental Design

Pretest-posttest control group design

After only design

Solomon-four group design

Factorial design

Matching samples design

Randomized clinical trials

Quasi-experimental

Non-equivalent pretest-posttest control group

Time series design

Control group time series design

Correlational Designs

Allow the researcher to determine the strengths of relationships among two or more variables

Values range from -1.0 to +1.0

Descriptive Studies

Used to discover new meaning or new knowledge

Seeks to identify and describe possible independent variables

Exploratory Studies

Also conducted when very little is known about a problem

Can use quantitative or qualitative approach

Usually used to begin using research to study a problem

Survey Designs

Descriptive research used to obtain information regarding prevalence or distribution of variables or to describe characteristics, opinions, attitudes, or behaviors as they exist in a population – typically based on self – reports

Methods of survey research

Data are collected typically through face – to – face interviews , telephone interviews, or questionnaires distributed through the mail

Longitudinal Studies

Trend studies

Panel studies

Follow-up studies

Ex-post Facto Studies

Means “after the fact”

Looks at factors after the problem has occurred

Methodological Studies

Psychometrics

Concerned with development or refinement of data collection instruments/scales

Metaanalysis

A research method that takes the results of many studies and synthesizes the findings to draw conclusions

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download