Exam #2 (AUDIENCE ANALYSIS)

What is longitudinal data?

Data where variables are measured over time, often at multiple time points, including possible retrospective data

What is longitudinal analysis?

Analysis of longitudinal data to study stability, change, and prediction over time

What is an advantage of longitudinal vs cross-sectional designs?

Separates age and cohort effects and allows study of stability and change

What is an advantage of longitudinal vs experimental designs?

Can study variables that cannot be ethically or practically manipulated

How should time span and measurement intervals be determined?

Based on theory, not convenience

What is measurement equivalence?

Ensuring the same construct is measured across time

What is heterotypic continuity?

Same construct manifests differently over time

What is a trend study?

Samples different people from the same population at different times

What is an example of a trend study?

General Social Survey

What is the first step in SEM trend analysis?

Recode Year into Decade

What is important when recoding decades?

Code irrelevant decades as missing values

What is the second step in SEM trend analysis?

Run crosstabs with Xmovie in rows and Decade in columns using column percentages

How do you analyze newspaper readership over time?

Crosstabs with News in rows and Decade in columns

What problem occurs with TV data?

Too much variability

How do you fix variability in TV data?

Run frequency analysis and recode into groups

What are the two types of recoding?

Recode into same variable or different variable

Which recoding method is recommended?

Recode into a new variable

What is the first step in recoding variables?

Create a new variable

What is the next step after creating a new variable?

Recode old variable into new variable

What should you check after recoding?

Match missing values and run frequency analysis

What percentage watched 2 hours or less of TV?

51.5 percent

What is a cohort study?

A trend study focusing on a specific subpopulation, usually age groups

What is a period effect?

Change over time affecting all age groups

What is an age effect?

Differences due to age

What is a cohort effect?

Differences due to generational group

What was the SEM viewing period effect result?

No overall change

What was the SEM viewing age effect result?

Younger people viewed more

What was the SEM viewing cohort effect result?

No clear cohort effect

What is net change?

Overall population-level change

What is gross change?

Individual-level change

What is an example of net change?

57 percent to 55 percent equals 2 percent change

What is an example of gross change?

88 percent of individuals changed opinions

What does Panel Model 1 test?

Whether X predicts later Y

What is a problem with Panel Model 1?

Reverse causality and third variable problem

What does Panel Model 2 do?

Controls for prior Y to predict change in Y

What does Panel Model 3 test?

X to Y and Y to X simultaneously

What does Panel Model 4 add?

Third variables as controls

What does Panel Model 5 use?

Structural equation modeling

What is an advantage of Panel Model 5?

Accounts for measurement error and compares paths

What are the main topics of experimental research?

Pre-experimental designs, internal validity, true experiments, factorial designs, field experiments

What is a one-shot case study?

Treatment followed by a test only

What is the problem with a one-shot case study?

No baseline for comparison

What is a one-group pretest-posttest design?

Test, treatment, test

What is the problem with one-group pretest-posttest?

Demand characteristics

What is a static group comparison?

Treatment group compared to control group

What is the problem with static group comparison?

Preexisting differences

What is a testing threat?

Pretest influences behavior

What is a history threat?

External events influence results

What is maturation?

Natural changes over time

What is an instrumentation threat?

Measurement changes across time

What is a chronology effect?

Meaning of variables changes over time

What is a human effect?

Bias from human coders

What is statistical regression?

Extreme scores move toward the mean

What is experimental mortality?

Participants drop out of the study

What is selection bias?

Groups differ before the experiment

What are the components of a true experiment?

Control group, random assignment, environmental control, IV manipulation

What is the benefit of random assignment?

Equal distribution of confounds

What is matching?

Assigning participants based on shared traits

What is a limitation of matching?

Does not control all confounds

What is the classic experimental design?

Pretest-posttest control group design

What analysis is used in the classic design?

Gain scores or ANCOVA

Why is a control group important?

Helps rule out alternative explanations

Why use a posttest-only design?

Avoid testing by treatment interaction

What analysis is used in posttest-only design?

Between-group comparison such as t-test

What is the Solomon four-group design?

Design that tests for testing effects and confirms treatment effects

What are factorial designs?

Experiments with multiple independent variables

What is an example of a factorial design?

3 by 3 design such as preview length by movie length

What is a natural field experiment?

Nature determines the independent variable without random assignment

What is a manipulation field experiment?

Researcher manipulates IV but does not randomly assign participants

What is a strength of field experiments?

High ecological validity

What is a weakness of field experiments?

Low internal validity

What are the main advantages of experiments?

Control over time order, environment, and individual differences

What are the main disadvantages of experiments?

Artificial setting and difficulty studying long-term behavior