Communication Research Final

Front 1

experiment

Back 1

study of EFFECTS OF VARIABLES manipulated by researcher where all other variables are HELD CONSTANT.

Front 2

confounding

Back 2

mixing variation from one source with variation from another souce. so that its impossible to know whether the effects are because of the impact of either variable separately or some combination of them.

Front 3

types of questions that lead to experimental research

Back 3

1) Questions asking if a cause-and-effect relationship exists:

Front 4

Requirements for a “true” experiment:

Back 4

I. Control: “methods researchers use to remove or hold constant the effects of nuisance variables”
- Pages 399-401 Major approaches to control:

Front 5

Notation for experimental designs

Back 5

O= Observation of ____________________ the dependent variable.
X= Experimental variable or _____________ the independent variable.
R= _______________________
Randomization (either r. assignment or r. sampling)

Front 6

So, a notation for a true experiment:

Back 6

*Starts with R
*Must make observations (O) of at least 2 groups *One group with X, one group withoutOne example with all 3 elements:

Front 7

One example with all 3 elements:

Back 7

R O X O (experimental group)
R O O (__________ group)

Front 8

How would you verbally describe what the experimenter did? (reveals the name)
Pretest Posttest Control Group Study

Back 8

Pretest-Posttest Control Group? Posttest Only Control Group? Solomon Four Groups Design?

Front 9

1. My question is whether gender or age has an effect on the ability to comprehend accented English. Which research method should I use?

Back 9

-

Front 10

2. If I’d like to determine if the intervention called the “Picture Exchange Communication System” is effective with individuals with severe autism, which research method should I use?

Back 10

-

Front 11

3. To determine if the intervention called the “Picture Exchange Communication System” is effective with individuals with severe autism, I find a large group of these individuals and RANDOMLY assign them to two groups: experimental and control groups, then I give them pretests and post tests. Which notation shows this experiment?

Back 11

A.

Front 12

Refer to the following article title to answer the next 2 questions:

Back 12

-

Front 13

Pries, J. (2006). The effect of picture communication symbols on the verbal comprehension commands
by young children with autism. Focus on Autism & Other Developmental Disabilities, 21, 194-210.

5) What is the dependent variable?
A. Picture communication symbols B. Verbal comprehension commands C. Young children with autism

Back 13

-

Front 14

Internal validity/invalidity

Back 14

Internal validity: Degree to which we can be sure that no confounding variables have obscured the true relationship between the variables in the experiment. The confidence we can have in the assertion that the variable(s) we have called “cause(s)” or IVs actually produced the observed effects.
Internal invalidity: the presence of contamination that prevents the experimenter from concluding that the study’s experimental variable is responsible for the observed effects.

Front 15

Exernal validity/invalidity

Back 15

External validity: describes our ability to generalize from results of the study to the real world.

Front 16

FACTORIAL DESIGNS

Back 16

“When more than one input or independent variable is studied, the input variables are called “factors” and the study is a “factorial design.”

Front 17

Factorial designs...

Back 17

To examine “interaction effects” = to see if special
combinations of variables produce unique effects.

Front 18

EXPERIMENT 1:

Back 18

BUT, what if a particular combination of gender and animation is especially strong in inducing positive (or negative) perceptions of credibility? Is it better to be a “high animation male” or a “high animation female?”




Front 19

TYPES OF QUESTIONS that lead to empirical descriptive research

Back 19

Questions focusing on interpretations of meanings of social actors.
§ Questions about naturally arising behavior that can’t be produced in laboratories.
§ Desire to observe firsthand as messages are produced.

Front 20

3. To determine if the intervention called the “Picture Exchange Communication System” is effective with individuals with severe autism, I find a large group of these individuals and RANDOMLY assign them to two groups: experimental and control groups, then I give them pretests and post tests. Which notation shows this experiment?

Back 20

A.

Front 21

Refer to the following article title to answer the next 2 questions:

Back 21

-dlfkg

Front 22

Pries, J. (2006). The effect of picture communication symbols on the verbal comprehension commands
by young children with autism. Focus on Autism & Other Developmental Disabilities, 21, 194-210.

5) What is the dependent variable?
A. Picture communication symbols B. Verbal comprehension commands C. Young children with autism

Back 22

-adfksdf

Front 23

Internal validity/invalidity

Back 23

Internal validity: Degree to which we can be sure that no confounding variables have obscured the true relationship between the variables in the experiment. The confidence we can have in the assertion that the variable(s) we have called “cause(s)” or IVs actually produced the observed effects.
Internal invalidity: the presence of contamination that prevents the experimenter from concluding that the study’s experimental variable is responsible for the observed effects.

Front 24

Exernal validity/invalidity

Back 24

External validity: describes our ability to generalize from results of the study to the real world.

Front 25

FACTORIAL DESIGNS

Back 25

“When more than one input or independent variable is studied, the input variables are called “factors” and the study is a “factorial design.”

Front 26

Factorial designs...

Back 26

To examine “interaction effects” = to see if special
combinations of variables produce unique effects.

Front 27

EXPERIMENT 1:

Back 27

BUT, what if a particular combination of gender and animation is especially strong in inducing positive (or negative) perceptions of credibility? Is it better to be a “high animation male” or a “high animation female?”




Front 28

TYPES OF QUESTIONS that lead to empirical descriptive research

Back 28

Questions focusing on interpretations of meanings of social actors.
§ Questions about naturally arising behavior that can’t be produced in laboratories.
§ Desire to observe firsthand as messages are produced.

Front 29

3. To determine if the intervention called the “Picture Exchange Communication System” is effective with individuals with severe autism, I find a large group of these individuals and RANDOMLY assign them to two groups: experimental and control groups, then I give them pretests and post tests. Which notation shows this experiment?

Back 29

A.

Front 30

Refer to the following article title to answer the next 2 questions:

Back 30

-dlfkg alfjsdflk

Front 31

Pries, J. (2006). The effect of picture communication symbols on the verbal comprehension commands
by young children with autism. Focus on Autism & Other Developmental Disabilities, 21, 194-210.

5) What is the dependent variable?
A. Picture communication symbols B. Verbal comprehension commands C. Young children with autism

Back 31

-adfksdfsdf sdfs df

Front 32

Internal validity/invalidity

Back 32

Internal validity: Degree to which we can be sure that no confounding variables have obscured the true relationship between the variables in the experiment. The confidence we can have in the assertion that the variable(s) we have called “cause(s)” or IVs actually produced the observed effects.
Internal invalidity: the presence of contamination that prevents the experimenter from concluding that the study’s experimental variable is responsible for the observed effects.

Front 33

Exernal validity/invalidity

Back 33

External validity: describes our ability to generalize from results of the study to the real world.

Front 34

FACTORIAL DESIGNS

Back 34

“When more than one input or independent variable is studied, the input variables are called “factors” and the study is a “factorial design.”

Front 35

Factorial designs...

Back 35

To examine “interaction effects” = to see if special
combinations of variables produce unique effects.

Front 36

EXPERIMENT 1:

Back 36

BUT, what if a particular combination of gender and animation is especially strong in inducing positive (or negative) perceptions of credibility? Is it better to be a “high animation male” or a “high animation female?”




Front 37

TYPES OF QUESTIONS that lead to empirical descriptive research

Back 37

Questions focusing on interpretations of meanings of social actors.
§ Questions about naturally arising behavior that can’t be produced in laboratories.
§ Desire to observe firsthand as messages are produced.

Front 38

3. To determine if the intervention called the “Picture Exchange Communication System” is effective with individuals with severe autism, I find a large group of these individuals and RANDOMLY assign them to two groups: experimental and control groups, then I give them pretests and post tests. Which notation shows this experiment?

Back 38

A.

Front 39

Refer to the following article title to answer the next 2 questions:

Back 39

-dlfkg alfjsdflk

Front 40

Pries, J. (2006). The effect of picture communication symbols on the verbal comprehension commands
by young children with autism. Focus on Autism & Other Developmental Disabilities, 21, 194-210.

5) What is the dependent variable?
A. Picture communication symbols B. Verbal comprehension commands C. Young children with autism

Back 40

-adfksdfsdf sdfs df

Front 41

Internal validity/invalidity

Back 41

Internal validity: Degree to which we can be sure that no confounding variables have obscured the true relationship between the variables in the experiment. The confidence we can have in the assertion that the variable(s) we have called “cause(s)” or IVs actually produced the observed effects.
Internal invalidity: the presence of contamination that prevents the experimenter from concluding that the study’s experimental variable is responsible for the observed effects.

Front 42

Exernal validity/invalidity

Back 42

External validity: describes our ability to generalize from results of the study to the real world.

Front 43

FACTORIAL DESIGNS

Back 43

“When more than one input or independent variable is studied, the input variables are called “factors” and the study is a “factorial design.”

Front 44

Factorial designs...

Back 44

To examine “interaction effects” = to see if special
combinations of variables produce unique effects.

Front 45

EXPERIMENT 1:

Back 45

BUT, what if a particular combination of gender and animation is especially strong in inducing positive (or negative) perceptions of credibility? Is it better to be a “high animation male” or a “high animation female?”




Front 46

TYPES OF QUESTIONS that lead to empirical descriptive research

Back 46

Questions focusing on interpretations of meanings of social actors.
§ Questions about naturally arising behavior that can’t be produced in laboratories.
§ Desire to observe firsthand as messages are produced.

Front 47

Complete Participant
(p. 248)

Back 47

Subjective Sympathetic Involved Deceptive

Front 48

Participant as Observer --

Back 48

in between both

Front 49

Complete Observer
(p. 250)

Back 49

Objective Unsympathetic
Detached
Candid

Front 50

Participant as observer:

Back 50

a role in ethnography or participant observation research characterized by the investigator’s gathering data while taking part in the activities of a group – and after making his or her research identity known to the group. (p. 251)

Front 51

We’ll focus on only the 1st step in the process of ethnography:
n n n n
n n
B) STEPS IN THE PROCESS

Back 51

1. Select a position for the researcher (p. 257)
1) Ventriloquist (transmit info, try to be invisible)
2) Positionality of voices (subjects and their voices are
the focus, be vaguely present)
3) Activism stance (be advocate for marginalized people beyond scholarly researcher role)
4) Postcritical ethnography (critique one’s own acts of studying and representing people as acts of domination)

Front 52

fieldwork

Back 52

study of people acting in the natural course of their daily lives

Front 53

naturalistic studies

Back 53

inquiries completed in the natural environment

Front 54

participant observation

Back 54

researchers study groups by gaining membership or close relationships with them.

Front 55

ethnography

Back 55

researcher studies ppls lives for an extended period of time collecting data. overtly or covertly.

Front 56

ethnomethodology

Back 56

new ethnography. not rigorous. people live within ethnic group and develop insight into the culture.

Front 57

3 major purposes for ethnog/ fieldwork:

Back 57

1. when questionnaires are Innapriopriate
2. when the setting is so new that hypothesis are undeveloped
3. when developing analytic inductions. (drawing conclusions)

Front 58

full participant observation

Back 58

gathering data while taking part in the activities, keep identity UNKNOWN

Front 59

complete observer

Back 59

no contact with the individuals observing

Front 60

participant as observer

Back 60

gather data while taking part in activites and dont reveal identity.

Front 61

research positions: defined for themselves and others

Back 61

-ventriloquizing others interests (neutral transmission of facts)
- giving voice to silenced groups
(indigenous)
- engaging in direct activism

Front 62

post-critical ethnography

Back 62

critique their own subjects according to the research postitions. how their own acts of studying are like DOMINATION

Front 63

The method selected is determined by :

Back 63

the research question asked!

Front 64

CH 8: CONTENT ANALYSIS: A) TYPES OF QUESTIONS

Back 64

Questions that ask about:
a) MAIN: characteristics of msgs that can be categorized and counted.
b) comparisons of message patterns, flow c) comparisons of msg content w/ real life
(across groups, cultures; challenging assumptions of fairness, equality)

Front 65

Content Analysis: Qualitative or Quantitative Research?

Back 65

• Considered more “qualititative” even though quantitative “counting” involved.
• Because:
*”Counts” are merely means of identifying,
describing data.
*Interpretation of results is vital to the research.
Often to help evaluate whether comm patterns “are as they should be” (p.
321).

Front 66

An example of Content Analysis...
• Brewer, P. R., & Macafee, T. (2007). Anchors away: Media framing of
broadcast television network evening news anchors. Harvard International Journal of Press/Politics, 12 (4), 3-19.

Back 66

ABSTRACT: Between 2002 and 2006, six newcomers took the anchor chairs at the evening news programs of ABC, CBS, and NBC. Collectively, they received extensive news coverage. This study uses content analysis to examine how three national newspapers framed the new anchors. A frame casting the anchors as competitors in a ratings game was especially common. At the same time, the newspapers regularly framed the anchors in terms of their reporting experience and reporting style, as well as in terms of personal characteristics such as personality, appearance, age, and sex. The newspapers were more likely to frame female anchors in terms of their sex; apart from this, no consistent differences across sex emerged. All three newspapers followed broadly similar patterns in coveri

Front 67

content analysis

Back 67

any of several research techniques used to describe and systematically analyze the content of written, spoken or pictorial communication such as books newspapers, television programs, or interview transcripts

Front 68

From Reinard, Chapter 1:
dont have to know

and from Alex:

Back 68

Rather than just accepting what others (and now, textbooks) tell you, you have learned how scholars (a term that includes you now) draw conclusions from our field. You have seen how research should look so that you can evaluate the stuff that often gets passed off to you.

You now are equipped with the skills to stay current with new research and developments in the field, remaining a “communication” expert!

Front 69

ADRIAN TOOMEY

Back 69

1) 2) 3)
4) 5)
6) 7) 8)

Area of inquiry?
Why interviews?
2 types of interviews:
a) structured
b) unstructured
Sample questions?
“Probes" - to get additional detail, elaboration, or clarification.
The process: quantitative or qualitative? Challenges?
Preliminary conclusions/further research?

Front 70

Descriptive Empirical Research”

Back 70

n “Descriptive”
– Describing relationships, not manipulating them via
experimental methods.
n “Empirical” – “Observable”

Front 71

) TYPES OF QUESTIONS that lead to empirical descriptive research:

*3 emphases:

Back 71

1) Asking about current descriptions of things, what is occurring
n 2) Exploring explanations that characterize things as they are now.

**a) Ascertaining norms (What is?)
b) Establishing goals (What is best?)
c) Developing methods (What is best to do?)

Front 72

2 main types of surveys:

Back 72

A) Questionnaires
B) Interviews (we’ll cover AFTER Midterm 2)

Front 73

INFERENTIAL STATISTICS (p. 457)

Back 73

Allows an investigator to generalize results from a particular sample to the population from which the sample was drawn.

Front 74

Questions:

Back 74

What are the chances that my results were just a fluke?
What are the chances that the difference between the two groups I tested (control and experimental) are a result of chance and not due to the experimental treatment?
What are the chances that if I conduct the same experiment with a different sample, I’ll get the same results?

Front 75

Probability distribution
-Probability=
Prob. Distribution=
Usually, the standard normal distribution (bell curve) is the

Back 75

1. tendency or likelihood with which an
event occurs in a population
2. The distribution a researcher expects to see ordinarily with regard to the variable.

-probability distribution.

Front 76

A. Probability distribution

Back 76

So, when a researcher gets a different distribution from the probability distribution, something must have caused it to differ. Could it be the independent variable(s)? What are the chances that the IV caused those scores to be so different from a standard normal distribution?

Front 77

Null hypothesis (p. 492

Back 77

Predicts: “There is no relationship between __ and __.”
The null hypothesis is actually tested statistically, not the predictive hypotheses.

Front 78

EXAMPLE: H0 “Type of speech therapy will have no relationship to level of stuttering after therapy.”

Back 78

§ 3 groups: receive different treatments (A, B, and C)
§ Measure the level of stuttering (M and SD)- this is the
DV!
§ Compare the Ms and SDs between the groups
§ TEST NULL HYP: “There will be no differences between groups’ scores” = predicting 3 bell curves with the same M and SD.

Front 79

What if there ARE differences between these scores?
n Remember the questions that inferential statistics answer:

Back 79

– What are the chances that my results were just a fluke?
– What are the chances that the difference between the three groups I tested are a result of chance and not due to the different treatments?
– WhatarethechancesthatifIconductthesameexperiment with a different sample, I’ll get the same results?
– How strongly can I predict that the population at large will reflect the same results that I got from my sample?

n ANSWER: It all boils down to whether or not your results were SIGNIFICANT!

Front 80


n B. Statistically significant relationship or difference (p. 495)

Back 80

Significance: The crucial word in inferential statistics!!!
– “A difference or relationship that is beyond what might be expected to occur by chance alone.”
– The “odds” are set or determined by the researcher.

Front 81

tatistically significant relationship or difference (p. 495)
Reported in two ways:

Back 81

§ 1 CALCULATED VALUE - the number resulting from the
statistical calculation.
– Can essentially be ignored when skimming the results section; does not relate to M or SD.
§ 2 CRITICAL VALUE = p value
– A standard or level set by the researcher.
– Indicates the likelihood that the results are due to chance.
– Always reported in the literature; see results sections for examples.

Front 82

1. p<.05 often designated *
–

Back 82

Researcher has determined that the different results had less than a 5% (1/20) chance of being a fluke (and a 95% chance the results are due to manipulation of the IV and/or differences between groups. And 95% isn’t bad, so the researcher can reject the null hypothesis!).

Front 83

2. p<.01 often designated **
–

Back 83

Researcher has determined that the differences in the M and SD between the groups have a less than 1% (1/100) chance of being just random luck (and a 99% chance the results are due to manipulation of the IV/differences between groups). Researcher can reject the null hypothesis.

Front 84

3. p<.001 often designated ***
–

Back 84

Researcher has determined that the different scores on the DV have less than a .1% (1/1000) chance of being due to chance (and a 99.9% (999/1000) likelihood the different scores are due to manipulation of the IV. Yep, the research obviously rejects the null hypothesis!

Front 85

Q: Which 3 types of messages can we count on being significantly different in perceived amount of usage between genders?

Back 85

TABLE 3: Sex Differences in Perceived Instructor use of Verbally Aggressive Messages Instructor Mean (SD)
Message Male Female
Character attack .68 (1.07) Competence attack .56 ( .97) Background attack .33 ( .79) Physical appearance .23 ( .64)
Malediction
Teasing
Ridicule
Threats
Swearing
Nonverbal emblems .60 (1.06) .55 ( .92) .45
.09 ( .47) .84 (1.16) .49 ( .96)
.25 ( .61) .08 ( .43) .18
.21 ( .72) .44 ( .93)
.34 ( .83) 1.49
.07 ( .42) 4.59**
Note: Degrees of freedom are 1,302. *p<.05 **p<.01

Front 86

Inferential statistical formulas (2 common tests

Back 86

1. t test: a test of statistical significance designed to assess the difference between the means of two groups.
– Used to compare 2 or more groups. § Males, females
§ TBI, not TBI

2. Analysis of variance (ANOVA) or F value or F test: a test of statistical significance that compares the means of two or more groups
– Used to compare 3 or more groups
§ Scores on social skills in high, moderate, low autism
groups
§ Scores on self-esteem for groups high, moderate, low in communication apprehension
§ Scores on speech production for treatment groups A, B, and control group

Front 87

Sometimes p values are reported that

Back 87

confirm the null hypothesis:

Kim, E. (2005). Cooperative small group communication: Effects of communication apprehension and pre-academic achievement. International Communication Association Conference Papers, 2005 Annual Meeting, 1-39.
H1: Students with higher PAA will show higher class satisfaction within a cooperative learning context.
Q: What’ s the IV? What’ s the DV?
n n

n n
n n
H1: Students with higher PAA will show higher class satisfaction, within a cooperative learning context.
From the results section:
Class Satisfaction. To test our hypotheses, a series of
ANOVA analyses were performed....
Although the score of “class satisfaction” of the students with higher Pre Academic Achievement was higher (mean=4.26, sd=.81) than those with lower PAA (mean=3.81, sd=.83), the difference was not statistically significant (F=2.06, p=.13). Thus, hypothesis 1 (H1) was not supported.


N 18