Scribd
Upload
89 views

Building A Conceptual Model

Building a Conceptual Model

Uploaded by

JeremiahOmwoyo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
89 views

Building A Conceptual Model

Building a Conceptual Model

Uploaded by

JeremiahOmwoyo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 18

Variables, Relationships, Hypotheses

How to Build a Good Conceptual Model

Jim Goes, Ph.D.


School of Advanced Studies
University of Phoenix
Induction and Deduction

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 1


1
Where Theory Fits In

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 2


2
What is a theory?

 Theory verses Ideology


 Theories
 prevent “exceptions to the rule”
 make sense of patterns
 shape research
 Theories are maps of reality. The truth they
depict may be objective facts “out there” or
subjective meanings inside our heads.
 A theory is a statement about how reality works
(Einstein)
 A theory is a systematic explanation for the
observed facts and laws that relate to a particular
aspect of life (Babbie)

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 3


3
Good theory provides…

 An explanation of the data


 Prediction of future events
 Relative simplicity (parsimony)
 Hypotheses that can be tested
 Practical utility

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 4


4
What makes a good research question?

• Is the question or problem stated in a solvable way?


• Is the question open and non-biasing?
• How important/relevant is the question?
• Does the research answer the question?
• Are definitions misused as explanations,
circular, or inadequately specific?
• Are conceptual definitions operationalized?

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 5


5
Putting Theory into a Conceptual Model

 Concept: Abstract elements representing classes


of phenomena with the field of study.
 Variables: Concepts stated in measurable terms
(more specific than concepts).
 Statements: Principles, laws, axioms,
propositions and hypotheses in theory building.
 Postulates lead to propositions
 Propositions lead to hypotheses

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 6


6
Examples

 Concept: The role of gender in communication.


 Variable: verbal arguments made by women and
men in public speaking.
 Statement: Women have more hedges and
qualifiers in their verbal arguments in public
speaking than do men (a hypothesis).

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 7


7
From Theory to Hypothesis

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 8


8
Causes and Common Knowledge

 Going to college causes people to be smarter


 Old age leads to senility
 More frequent testing of students makes schools
more accountable for results and leads to greater
student learning
 Tax breaks for corporations stimulate job
creation and lower unemployment
 Left-handed people lead shorter lives

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 9


9
Causal Relationships and Hypotheses

 Criteria for causality


 Association
 Temporal priority
 No spurious variables
 All three criteria must be met

 Hypotheses
 At least two variables
 Expresses a causal relationship
 Expressed as a prediction
 Logically linked to theory and a research question
 Is testable, i.e., with empirical evidence

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 10


10
Types of Causal Relationships

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 11


11
Stating Causal Relationships

 Family breakups cause delinquency


 Family breakups lead to delinquency
 Family breakups are related to delinquency
 Family breakups influence delinquency
 Family breakups are associated with
delinquency
 Family breakups produce delinquency
 Family breakups result in delinquency
 The higher the level of family breakups the
higher the level of delinquency
 Family breakups increase the likelihood of
delinquency

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 12


12
The Priority of Variables

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 13


13
Types of Hypotheses

 The Null Hypothesis predicts no relationship


 The Research or Alternative Hypothesis predicts a
relationship
 We say that we can “reject the null” (reject) or “fail
to reject the null” (accept)
 We DON’T say “we can adopt the research
hypothesis”
 Hypotheses are either accepted or rejected – they
are not “partially” accepted or rejected

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 14


14
Evaluating Relationships

Issues in Relationships
• “third” variables or spurious relationships
• crossing levels of analysis/measurement
• freedom to vary
• causality
Issues to Consider
• Is the analysis based on a single variable?
• Are important “third” variables missing?
• Is the level of relationship measurement the
same as level of hypothesis/conclusion?
• Are variables indeed free to vary?
• Are causal relationships stated or implied?
If so, are they defensible?
January 2002 Copyright © 2002 Jim Goes / Cybernos Page 15
15
Logical Errors in Causal Explanation

 Tautology
 Teleology
 Spuriousness (Questionable Cause)
 Provincialism
 Hasty Conclusion

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 16


16
A Spurious Relationship

January 2002 Copyright © 2002 Jim Goes / Cybernos Page 17


17

You might also like