Conceptualizing Communication Research

1
Conceptualizing Communication Research
2
Learning Objectives

• Define communication research
• Explain the differences and similarities between
  basic and applied communication research
• Define the following concepts variable,
  independent and dependent variables, interaction
  effects, ordered and nominal variables, and
  relationships.
• Identify and justify research topics the novel
  idea considered worth studying and being better
  understood.
• Construct research questions and generate
  hypotheses

3
Whats communication research?

• Disciplined inquiry that involves studying
  communication phenomena in a planned manner and
  reporting it so that other inquirers can
  potentially replicate the process if they choose.

4
What is Communication?

• The processes by which verbal and nonverbal
  messages are used to create and share meaning.
• For the ISC professional - the use of verbal and
  nonverbal messages to inform, educate, persuade
  or motivate change or maintenance of attitudes,
  opinions or behavior.

5
Basic vs. applied communication research

• Basic research research designed to test and
  refine theory. Purpose is to increase our
  knowledge about communication by testing,
  refining, and elaborating theory.
• Applied Research Research designed to solve a
  practical problem. Purpose is to solve a
  "real-world," communication problem.

6
Research Topics Where do they come from?

• For both communication scholars and
  professionals, research topics begin as tentative
  ideas. However, relevant literature on the idea
  is identified and reviewed as soon as possible to
  shape the idea into a topic worthy of research.
  Even in applied research where the research idea
  is triggered by a problem or need in a specific
  context, relevant literature must be reviewed to
  see how others have defined and tried to solve
  the problem.

7
Justifying Communication Research Topics

• Clear and convincing rationale for why the topic
  is worth studying
• For the ISC professional, the client or
  management of the organization who'll pay for the
  research is a very important audience, i.e.
• What will the findings of the research do for
  the organization or client?

8
Constructing Research Questions and Hypotheses

• Formal questions/statements posed to guide
  research
• Usually designed to accomplish one of two goals
• 1. describe communication behavior
• OR
• 2. relate communication behavior to other
  variables

9
Describing communication behavior What is the
nature of communication behavior 'X'?

• These are questions designed to produce
  descriptive data. They usually ask who, what,
  where, when, how, why.
• E.g.
• 1. What is the structure of memorable nonsmoking
  PSAs for adolescents?
• 2. How extensive is rape among college women?

10
Relating communication behavior to other
variables How is communication variable 'X'
related to other variables?

• What is a variable?
• Any concept that can have two or more values.
• A single object, (e.g. college women) is not a
  variable unless it comes in different types,
  categories or amounts and we can identify and
  understand those different states.

11
How is communication variable 'X' related to
other variables?

• Turning a communication concept into a variable
  makes it possible to examine the relationship
  between that communication behavior/variable and
  other important variables.
• E.G.
• RQ Is there a relationship between college
  womens membership in Greek systems and their
  readiness to report rape on campus?

12
Variables (cont.)

• Independent variable The variable thought to
  influence changes in another variable
• Dependent variable The variable thought to be
  changed by another variable

13
Variables (cont.)

• Ordered variable These can be assigned numerical
  value that indicates how much of the concept is
  present
• E.g.
• age, temperature, income and weight.

14
Variable (cont.)

• Nominal variable Can be differentiated only on
  the basis of type (in name only) e.g. gender,
  race, political affiliation.
• - Dichotomous or binomial variable a nominal
  variable such as gender that can only be divided
  into two categories.
• - Polytomous variable a variable that can be
  divided into more than two categories.

15
Relationships

• Causal relationship changes in the independent
  variable cause observed changes in the dependent
  variable.
• E.g. smoking cigarettes (independent variable)
  causes cancer (dependent variable).

16
Relationships (cont.)

• Recursive causal models one-way causal
  relationship. One variable influences another but
  not the other way around.
• E.g. Age influences the amount of information
  people self-disclose about themselves.

17
Relationships (cont.)

• Non-recursive causal models The causal
  relationship is reciprocal in that a variable can
  be both a cause and an effect.
• e.g. studying may lead to better grades, but
  better grades may well make it more likely that a
  person studies

18
Relationships (cont.)

• Non-causal relationship The variables are
  thought to be associated, or occur together
  without one necessarily causing changes in the
  other.
• e.g. People like people they know more about.
  i.e. self-disclosure and liking are related.

19
Relationships (cont.)

• Positive (direct) relationship Increases in the
  independent variable is associated with increases
  in the dependent variable.
• Negative (inverse) relationship Increases in the
  independent variable are associated with
  decreases in the dependent variable

20
Relationships (cont.)

• Interaction effects Effects due to the unique
  combinations of independent variables that make a
  difference on the dependent variable(s).
• Interaction effects are due to the effects of
  multiple independent variables working together,
  in contrast to the effects of each independent
  variable working alone (main effects).

21
Research Questions vs. Hypotheses A Revisit

• Research studies answer research questions or
  test hypotheses about relationships between
  variables.
• Questions are typically posed when researchers
  don't have enough evidence, on the basis of
  literature reviewed, to predict the nature of the
  relationship.

22
Research Questions vs. Hypotheses A Revisit
(cont.)

• When researchers feel confident enough to make a
  prediction, their studies test hypotheses.
• A hypotheses is a tentative statement about the
  relationship between the independent and
  dependent variable

23
Two-tailed hypothesis

• When a hypothesis simply predicts a relationship
  between variables without specifying the nature
  of the relationship it is called a two-tailed
  hypothesis.
• E.g. "Men and women self-disclose differently."

24
One-tailed hypothesis

• If the hypothesis predicts the specific nature of
  the relationship, it is called a one-tailed
  hypothesis.
• E.g. "Men self-disclose less than women."

25
Posing Research Questions and Hypotheses about
Relationships between Variables

• whether the independent variable is nominal or
  ordered and
• whether a researcher wishes to pose a research
  question or a hypothesis about the relationship
  between the independent and dependent variable

26
Posing a research question for a nominal
independent variable

• When the independent variable is nominal, divided
  into categories, the research question asks
  whether there is a difference between a (the
  first category of the nominal independent
  variable) and b (the second category of the
  independent variable) with respect to c (the
  dependent variable).

27
Posing a one-tailed hypothesis for a nominal
independent variable

• A hypothesis for a nominal independent variable
  predicts the nature of the difference between the
  two (or more) categories of the independent
  variable. It takes the form a (the first
  category of the nominal independent variable)
  will be greater (or less) on c (the dependent
  variable) than will b (the second category of the
  nominal independent variable).

28
Posing a research question for an ordered
independent variable

• When the independent variable is ordered,
  measured in sequenced numbers, the research
  question asks whether there is a relationship
  between x (the independent variable) and y (the
  dependent variable).

29
Posing a one-tailed hypothesis for an ordered
independent variable

• A hypothesis for an ordered independent variable
  specifies the nature of the relationship between
  the independent and dependent variables.