·   Deduction is one form of finding answers to problems.  Deduction is in the form of an argument that must be both true and valid to be correct.  The premises must be true and the conclusion that follows the premise must be valid.  A deduction is valid only if the conclusion cannot be proven false.  Deduction can take the form of an if/then statement, or several premises, which are true statements, followed by a conclusion.

·   Induction is different in that the conclusion is drawn from a particular fact or piece of evidence.  The conclusion will explain the facts; like investigating a crime scene and putting the pieces of evidence together to form answers, keeping in mind the conclusion is your hypothesis.

 

Research

Types of Research Designs

	Experimental (& Quasi-Experimental)
	Correlational
	Survey
	Grounded Theory
	Ethnographic
	Narrative
	Mixed Method
	Action

	To reduce coverage error, have a good sampling frame list on which to select individuals.
	To reduce sampling error, select as large a sample from the population as possible.
	To reduce measurement error, use a good instrument with clear, unambiguous questions and response options.  This encourages response & correct answers.
	To reduce non-response error, use rigorous administration procedures to achieve as large a return rate as possible.

	understand problems
	develop hypotheses
	design and implement controlled experiments
	identify independent and dependent variables
	analyze data
	draw conclusions
	think analytically
	communicate results (using the proper data tables and graphs)

 

Pre-experiment		True experiment		Quasi-experiment
·      Weak in scientific measurement ·       Fails to adequately control threat on internal validity   (Cooper, 2006).		Randomly assigns participants to different conditions. •         Experimental group will get experimental treatment •         Control group does not get treatment, it remains untouched •         Most threats to internal validity do not arise •         The researcher randomly assigns participants to different conditions of the experimental variable		Design looks like a pre-test/post-test design but it lacks the random assignment of the control groups. •  No random assignment •  Uses existing or intact group for study •  More threats to internal validity than true experiments •  No random assignment of participants or groups •  This approach introduces more threats to  internal validity: •  Maturation •  Selection •  Mortality •  Interaction

 

Choosing Statistical Test
Experimental Design

 

 

Measurement

 

 

Methods

 

Threats To Validity

Maturation   (Cooper, 2006) has to do with the passage of time and occurs when the subjects being studied have no specificity to any particular event within the experiment. History – causes confusion within the relationship being studied due to a particular event that occurs during the experiment. Testing or Pretest gives unfair advantage for post test answers in experiment. Presence of confusing variables if the researcher uses too many variables at once and cannot determine which variable affected the experiment. Hawthorne Effect  (Clark, 1999) could alter the effect of the experiment if the subject(s) know they are being watched or studied. Experimenter Effect –experiment only works with current experimenter or researcher.  With another person performing the experiment the effect is not the same. No repeatability.