Comm Research Methods 3100 (Exam 2) Brais

Exploring Two Variables of Communication Example:

- Do children who watch a lot of cartoons behave more aggressively on the playground? - Are couples who touch less in public as happy as couples who touch more in public? - How has persuasive techniques in infomercials changed over time?

Nature of Communication Example:

- How do couples use touch to signal relationship status in public? - How much violence is depicted in Saturday morning cartoons? - What persuasive techniques are used on late night infomercials?

Types of Hypothesis Relationships:

1. Associational Relationship 2. Causal Relationship

Variables Must Be Defined

1. Conceptual definitions -how you define the variable 2. Operational definitions- how the scholars plan to measure and/or observe the variable of interest 3. "Conceptual fit" - how closely your operational definition matches your conceptual definition --> Example: " Intimacy" - Conceptually: Being physically "close" to or with another person in a romantic fashion - Operationally: Types of physical "closeness"; proximity, touch, affection, etc.

Quantitative Sampling Methods (Non-random Sampling)

1. Convenience 2. Volunteer 3. Snowball 4. Network

Example of Independent Variable and Dependent Variable

1. H: There will be a positive increase in graduation rates of at-risk high-school seniors who participate in an intensive study program. - IV: Participation in intensive study program. - DV: Graduation rates 2. H: Counseling interventions will enable clients to achieve low anxiety levels more rapidly than clients who receive normal nursing and medical care. - IV: Counseling interventions - DV: Time taken to reach a low anxiety level 3. H: Clients who maintain low levels of anxiety will be more responsive to pain medication. - IV: Anxiety level - DV: Effectiveness of pain medication

Types of variables

1. Independent Variable 2. Dependent Variable 3. Extraneous Variable

Two Types of Variables that Have an Effect on The Dependent Variable

1. Mediating Variables 2. Moderating Variables

Types of Hypothesis

1. Null 2. Non-Directional 3. Directional

Suggestions for getting more response rates

1. Offer an incentive of some kind to get more people to agree to participate 2. Follow up with people who refuse to participate, and ask them again (hoping they will change their mind the second time) 3.Make your study easy to participate in so that they will be less likely to refuse. (ex: short survey)

Types of Causal Relationships

1. Positive Linear 2. Negative (Inverse) Linear 3. Curvilinear 4. Inverted Curvilinear

Types of Research Questions about Communication

1. Question of Definition 2. Question of Fact

Types of Variable Relationships

1. Reversible relationship 2. Irreversible relationship 3. Deterministic relationship 4. Stochastic relationship 5. Sequential relationship 6. Co-extensive relationship 7. Sufficient relationship 8. Contingent relationship 9. Necessary relationship 10. Substitutable relationship

3 primary procedures for measuring variables

1. Self-report (Social desirability bias) 2. Other-report (Limited exposure) 3. Observation (Hawthorne Effect bias)

Types of Triangulation

1. Sources -multiple interviewees, multiple field sites, multiple cases, multiple observations 2. Methods -qualitative method plus quantitative methods, observation plus self-report, plus other reports 3. Researchers -multiple interviewees or observers

Variables Have Dimensions

1. Unidimensional (Variables containing only one dimension) --> Example: "Education" - Highest level completed - 1 dimension: High School Diploma; BA/BS; MA; or PhD 2. Multidimensional (Variables containing more than one dimension) --> Example: "SAT Score" - Uses writing, reading, and mathematics by combining the 3 separate scores and creating a composite score.

Quantitative Sampling Methods (Random Sampling)

1.. Simple random 2. Stratified 3. Proportional stratified 4. Cluster

How To Sample

1st - Define the population 2nd - Determine a sampling frame 3rd - Identify the Unit of Analysis 4th - Pick a sampling method (Qualitative or Quantitative)

1. Mediating Variables

A ---> M ---> B - A causes M which causes B ---> Example: High level of confidence (A) might cause a student to speak out more in class (M), which may give them a better grade (B).

Fact Pattern

A factual relationship occurring repeatedly

1. Convenience (Non-random Sampling)

A group of people that are convenient to access ---> Example: Talking to people in your class for your communication research study

1. Simple random (Random Sampling)

A group of subjects (sample) are selected for study from a larger group (population), and each member of the population has an equal chance of being chosen at any point during the sampling process. ---> Example: "Drawing a name out of a hat" OR creating a computer generated system to pick participants

2. Other Report Procedures

A method where you ask others to rate others behaviors.

Sampling

A sample of people that are chosen to be included in the study as participants. - This sample is supposed to be representative of the entire population under study.

Research Topics

Always think in terms of narrowing down to very specific questions addressing very specific variables . --> Example: "I am interested in studying how physicians give terminal diagnoses to people with terminal illness.

Triangulation

The comparison of two or more forms of evidence with respect to an object of research interest.

3. Extraneous Variable

A variable that is not apart of the research design, and usually unpredictable and uncontrollable by the researcher.

Hypothesis (H)

An educated guess about what will happen in a relationship between variables based on what is known from existing theory.

Hawthorne Effect (Limitation of Observation)

An effect where people alter their behaviors because they know that they are being watched.

"Third variable problem"

An unseen or unmeasured variable that is accounting for changes seen in both the independent and dependent variable.

Variables

Any concept that varies -It can be measured and manipulated -It can take on more than one value -It can be different for each research study -->Examples: Gender, Satisfaction, Violence

2. Stratified (Random Sampling)

Breaking up your frame into multiple categories (strata) and filling them with the participants that fit best with each category until no participants are left.

CHAPTER 5

CHAPTER 5

CHAPTER 7

CHAPTER 7

CHAPTER 8

CHAPTER 8

Conceptual Fit

Conceptual and operational definitions must match

2. Volunteer (Non-random Sampling)

Consists of people who are willing to volunteer for a study. ---> Example: Advertising your study through flyers and emails. Then people reaching out to volunteer and be apart of your study.

Have you ever notice that when you are in a conversation with someone you like, or someone you need to impress, you adjust your style of speaking to match his? This is known as _______________

Convergence

4. Stochastic relationship

Dependent variable MIGHT BE PROBABLE TO result from Independent variable, but it isn't 100% gurantee that it will happen.

3. Deterministic relationship

Dependent variables MUST ALWAYS result from Independent variables

Operational Definition

How the concept (variable) that is being studies is measured and/or observed ---> Example: You will be studying "violence" so you must specify how you plan to determine whether behavior in the television programs that you watch are coded as violent. - What do you count as violent?

When you are in a conversation with someone you feel dissimilar to or dislike in some ways and you maintain your own style of communication. This is known as ___________________

Divergence

2. Negative (Inverse) Linear Relationship

EX: As "A" decreases, "B" increases

3. Curvilinear Relationship

EX: As "A" increases, "B" decreases for a while, then "B" increases

1. Positive Linear Relationship

EX: As "A" increases, "B" increases

4. Inverted Curvilinear Relationship

EX: As "A" increases, "B" increases for a while, then "B" decreases

Generalizability

Ensuring that a researcher's findings will apply to other people and situations within the population that the sample is supposed to represent ---> Example: College students enrolled in a public speaking class OR elderly individuals in assisted care faclities

Census

Find out information about the population without having to interview entire population.

2. Moderating Variables

Has a (predicted) effect on both A and B - M affects A and B ---> Example: A person's overall health (M) affects an individual's communication with healthcare provider (A), and their satisfaction with provider (B).

Sample Error

How far off your sample is from your population -Researcher try to avoid this

Conceptual Definition

How the concept (variable) that is being studied is defined ---> Example: How do we define the word "violence"? - the act of causing physical harm on others

2. Non-Directional (Two-Tailed Hypothesis)

Hypothesis does not specify a direction of relationship concepts (variables). Also known as: Two-Tailed Hypothesis --> Example H: People from the North and South speak at different rates. H: Discussing sex with a child will have an affect on their sexual habits.

3. Directional (One-Tailed Hypothesis)

Hypothesis implies specific direction of relationship between concepts (variables). Also known as: One-Tailed Hypothesis --> Examples: H: People from the South speak slower than people from the North. H: Discussing sex with a child will increase their sexual habits.

3. Proportional (Random Sampling)

Identify your category through proportion ---> Example: Looking at the UNCC communication department and eciding that we need 100 sampled participants, so we get the participants and identify what track their on. - 45 out of 100 (Mass Media) - 20 out of 100 (Organizational Comm) - 10 out of 100 (Public Relation) -25 out of 100 (Health Comm)

4. Cluster (Random Sampling)

Identifying groups (clusters) that you think represents the entire population and sample randomly withing each group, letting each group represent the population. ---> Example: Researcher evaluated the effects of a lifestyle intervention on diet, physical activity, and smoking in communities in Iran. SO he used cluster sampling by targeting three cities, and randomly sampled within each city. (The cities each served as a cluster)

2. Casual relationship (Quantitative Studies)

Implies that one variable causes a change in the direction of the other variable. MUST have at least two variables - Use a hypothesis to guide the research - Express a hypothesis as a prediction or an expected future outcome "A" causes "B" An increase in "A" causes an increase in "B" A decrease in "A" causes a decrease in "B" --> Example: H: Watching soap operas causes people to have extra-marital affairs

7. Sufficient relationship

Independent variable is enough to cause Dependent variable

8. Contingent relationship

Independent variable is only enough to cause Dependent variable when a third variable is needed. ---> A is a third variable

Theory

It explains, predicts, and/or controls -It helps to explain or make sense of reality, answering "Why? -It allows us to forecast what will happen -And once we understand a process, we can then control it

4. Ratio Level Measurement

It has characteristics of an interval scale, but it also has an absolute zero point that cannot have negative numbers. -Used in quantitative research ---> Rules of Ratio scales: - Absolute zero - Equal intervals - Can be mathematically measured to a decimal point and even a fraction ---> Example: Researcher asked married people to report the number of months they knew their spouse prior to marriage. OR a record amount of money donated to a charity.

Limited Exposure (Limitation of Other Report Procedures)

Limited prior exposure to the other individuals behavior

9. Necessary relationship

ONLY the Independent variable can cause the Dependent variable to occur (one variable must be present for the other variable to present)

3. Observation

Observing individuals behavior in a certain setting rather it is at a bar, mall, restaurant, park or in a laboratory setting like a room set up like a living room.

Research Hypothesis

Once the researcher knows enough about the topic to make a prediction. The researcher then makes a statement about the relationship between a dependent and an independent variable. -The researcher hypothisis is symbolized by a capital letter H, followed with a numeric subscript (H1). You may have several hypothesis so numbering helps.

5. Sequential relationship

Ordering is important, and variables must appear sequentially/chronologically (first this happens then this happens)

Measurement

Process of determining characteristics and/or quantity of a variable through systematic recording/organization of observations

Research Questions (RQ)

Questions scholars ask about the way things work, and a question researchers attempts to answer. - Who, What, When, Where, and How - It looks at the nature of communication or explores the relationship between two variables.

Sampling frame

Realistic version of your population ---> Example: Communication Studies students currently enrolled as majors at UNC Charlotte

Unit of Analysis

Sampling unit(s) ---> Example: Communication Studies students OR the groups OR group meetings

Four Levels of Measurement

Scales used to obtain precise and consistent measurement of variables 1. Nominal 2. Ordinal 3. Interval 4. Ratio

Hypothesis Relationships

Specify how the value of one concept (variable) changes in relation to another. -May be either positive, negative, or the two variables may not have any relationship to one another

Likert Scale

Technique used to measure participants feelings or attitudes toward another person, issue, and event. (Seen using a 5 or 7 point scale to measure)

Semantic Differential Scale

Technique used to measure the meanings that participants assign to stimulus (groups, types of music, a person, an idea). -Scale can be from -7 or +7

10. Substitutable relationship

The Independent variable is one of many things that could cause the Dependent variable to occur.

Population

The body of people you are claiming to generalize towards based on your sample. ---> Example: Conducting a study on group communication among college students. - Population: college students

Social Desirability Bias (Limitation of Self-Report procedures )

The ideas that if participants are asked to answer questions that are sensitive in nature, people will undoubtedly feel swayed to present themselves in a particular light, regardless of whether it is indeed true.

Response rate

The number of people who agreed to participate in your study. -The higher the response rate the better

Refusal Rate

The number of people who refused to participate in your study.

Study Objective

The questions we ask

Representation

The sample accurately represents the characteristics of the people, objects, observations, etc. of the population

Inquiry

The search for understanding

Research Objectives

The statements of what you ultimately want to accomplish through your research. (So what?) --> Example: "I am studying ________ because I want to find out (who/what/when/where/whether/how) ______ is, in order to understand __________"

1. Independent Variable (IV)

The variable that causes or determines the value of another variable. (The outcome depends on this variable) -Its the manipulated and predictor variable ---> Example: Consider two variables: the number of cigarettes a person smokes and the probability of getting lung cancer. - Both of these are variables because they can take on different roles. -Cigarettes is a variable because one might have smoked zero cigarettes or one may have smoked 3,000 cigarettes in their lifetime. -Lung cancer is a variable because it can vary, in which you can have lung cancer, or not. (The Independent variable is the number of cigarettes smokes)

2. Dependent Variable (DV)

The variable that depends on what happens with the independent variable. (Its the outcome) ---> Example:Consider two variables: the number of cigarettes a person smokes and the probability of getting lung cancer. - The Dependent variable is that lung cancer is predicted to be caused by the number of cigarettes smoked

Metatheory

Theory about a theory that allows people to understand the philosophy driving their decisions about research methods, designs, and analysis.

Research questions, Hypothesis, and Theory

These heuristic functions allow for us to generate new knowledge, learning, and understanding.

4. Network (Non-random Sampling)

Using social networks to locate or recruit participants.

1. Self-Report procedures

This procedure is good at measuring individual's beliefs, attitudes, and values, or in finding out about behaviors that we might not be able to observe directly. --> Example: The evaluations that you make of professors and instructors at the end of the semester . Or perhaps, you find yourself consulting ratemyprofessor.com prior to enrolling in a class with a particular professor.

3. Snowball (Non-random Sampling)

This sampling method asks study participants to make a referrals to other potential participants, who in turn make referrals to other participants and so on.

1. Nominal Level Measurement

This type of measurement makes use of classifying variables into categories. ---> Rules for categorical scales: - They must have at least 2 categories - The categories listed must be exhaustive (the category must represent the variable fully) - The categories listed must be exclusive (variable fits into one and only one category, thers no overlap) - The categories must be equivalent ---> Example:Political affiliation, religion affiliation, biological sex, and race are all variables measured at the nominal level. ---> Example: Preferred news source (self report): newspaper, TV, Magazine, none, other

2. Ordinal Level Measurement

This type of measurement uses rank ordered and ordered classifications ---> Rules for Ordered scales: -Categories are arranged in an order -There is NOT equivalent distances between the items ---> Example: Asking participants how successful the strategy was: not at all, somewhat, very. ---> Example: Top Grossing Films in the USA (2018) 1 Black Panther - $585,896,528 2 Fifty Shades Freed - $99,229,300 3 Peter Rabbit Sony - $98,660,268 4 Insidious: The Last Key - $67,347,895 5 Maze Runner: The Death Cure - $57,535,995

Quantitative

Uses the deductive method -Deductive reasoning moves us from the general (the theory) to the specific (the research study). -Researchers are supposed to be objective

Qualitative

Uses the inductive method - Inductive reasoning moved us from the specific (The research study) to the general (the theory) -Incorporates values and perspectives of both researcher and participants -Can specify one or more variables -Use a research question to guide the research -May simply wish to describe who the participants in a study are and how they act, believe, perceive the world, or look

3. Interval Level Measurement

Variables are ranked in EQUAL distances from each other. Rules for Interval scales: -Equal distances -Include a zero (neutral) point (can have positive/negative values) ---> Example: Doing teacher evaluations and using a 5 point scale. (Likert Scale) "My teacher comes prepared for every class?" - Strongly disagree = 1 - Disagree = 2 - Neutral = 3 - Agree = 4 - Strongly agree = 5

1. Reversible relationship

Variables can go either way

2. Irreversible relationship

Variables can only go in one direction

6. Co-extensive relationship

Variables occur together or simultaneously ---> Example: If we are happy then will will be smiling

Confounding Variable

When the effects of two variables cannot be separated from each other. ---> Example: Let's say a public speaking instructor not only asked students to practice their speeches an hour each day, but also suggested they visualized giving the speech before doing so. - If the students were then rated as better public speakers, there would be no way of knowing which of the two variables (practice or visualization) was responsible for the effect (better public speaking).

1. Associational relationship

Where one variable is found, the other also will be found. If A, then B (A and B are two different variables) -We are hypothesizing, so it is possible that a third variable may occur. --> Example: H: People who watch soap operas have extramarital affairs.

1. Null Hypothesis

a statement that the research hypothesis is wrong. In other words, there is no (null) relationship between the variables that the research predicted. -When written it is symbolized by a capital letter H, followed with a numeric subscript of zero (H0). -simply substitute in the phrase "no relationship" or "no association" to turn a research hypothesis into a null hypothesis. --> Example 1: H1: There will be a negative association between depression and relational quality dating relationships. H0: There will be no association between depression and relational quality dating --> Example 2: H1: The poor relational quality associated with depression will be associated with increased loneliness. H0: The poor relational quality associated with depression will not be associated with increased loneliness. --> Example 3: H1: The more people watch soap operas, the more extramarital affairs they will have H0: There is no relationship between exposure to soap viewing and extramarital affairs. --> Example 4: H1: Adolescents males report greater enjoyment of slasher films than do adolescent females. H0: Males and females do not report different enjoyment of slasher films.