Chapter
11. Can the News Influence Our Implicit Prejudice?
Introduction
Chapter 4
true
true
You must read each slide, and complete any questions on the slide, in sequence.
Nonexperimental Design
A design in which there is no control or manipulation of the variables. This design does not seek to establish cause and effect and instead focuses on describing or summarizing what takes place.
Experimental Design
A design in which the experimenter controls and manipulates the independent variable and makes comparisons between the different levels, allowing the establishment of cause-and-effect relationships between the independent and dependent variables.
Independent Variable (IV)
The variable that influences the dependent variable. In experiments the researcher manipulates or controls this variable.
Dependent Variable (DV)
The variable measured in association with changes in the independent variable; the outcome or effect.
Factorial Design
An experimental design that has more than one independent variable.
Experimental Realism
The degree to which a study participant becomes engrossed in the manipulation and truly influenced by it.
Mundane Realism
The degree to which a study parallels everyday situations in the real world.
Reliability
The stability or consistency of a measure.
Validity
The degree to which a tool measures what it claims to measure.
Sensitivity
The range of data a researcher can gather from a particular instrument.
Main Effect Hypothesis
A prediction that focuses on one independent variable at a time, ignoring all other independent variables.
Interaction Effect Hypothesis
A prediction about how the levels of one independent variable will combine with another independent variable to impact the dependent variable in a way that extends beyond the sum of the two separate main effects.
IRB
A board that reviews the ethical merit of all the human research conducted within an institution.
Descriptive
Describes what is happening.
Inferential
Tests a specific prediction about why something occurs.
Factorial Design
This activity will allow you to create a design to test the combined effect of 2 variables—the type of crimes and the race of the suspects—on racial prejudices.
Dr. Melanie Maggard
Dr. Natalie J. Ciarocco, Monmouth University
Dr. David B. Strohmetz, Monmouth University
Dr. Gary W. Lewandowski, Jr., Monmouth University
Something to Think About...
Scenario: Although progress has been made since the 1960s Civil Rights Movement, racism still exists in today’s society, often in ways that are not easily recognizable. Racial prejudice affects all of us and is influenced by a combination of learned experiences, environment, parenting, and more. No single event is responsible for creating our racial prejudice; rather, prejudice is the result of multiple factors.
Robert W. Kelley/The LIFE Picture Collection/Getty Images
Something to Think About…
Racial prejudices are evident in a wide variety of settings, including in the context of how we view those who have committed crimes. Is it possible that we judge African Americans differently than European Americans in these situations? Do our views make us more likely to believe that punishment for certain types of people should be harsher? Does the type of crime committed—such as a “white-collar” crime like embezzlement vs. a “blue-collar” crime like robbery—also contribute to our beliefs?
Dimitri Otis/The Image Bank/Getty Images
Our Research Question
Jesse Kunerth/iStock/Getty Images
With these questions in mind, we will develop a research study that examines types of crimes, races of suspects, and racial prejudice. But first, you will need a framework to help you explore this topic. Research studies all start with a question, so here is your chance to ask one of your own.
Now that you have a research question (“How do the type of crime and race of the suspect influence racial prejudice?”), you must decide which type of research design will best answer your research question. To narrow things down, consider the following:
Having decided that your research question requires a comparison between groups, you must determine the best comparison to make. However, there are multiple comparisons to be made: between types of crime and races of suspects.
Now that you have selected experimental design that compares exposure to embezzlement vs. robbery and a European-American vs. African-American suspect on racial prejudices, you can identify your independent and dependent variables.
You have developed an experiment with 2 independent variables (type of crime and race of suspect) and 1 dependent variable (racial prejudice). This means we have a factorial design.
Next, we need to operationally define the first independent variable (IV), type of crime, by determining exactly how we will manipulate it. As we do, we’ll want to be sure our study has a high level of experimental and mundane realism.
Now we need to operationally define the second independent variable (IV), race of suspect, by determining exactly how we will manipulate it. As we do, we’ll want to be sure our study has a high level of experimental and mundane realism.
Of the options you explored, the task that is highest in experimental and mundane realism involves reading a local news event about a specific crime (embezzlement or robbery) involving $5,000, with the race of the suspect (European-American or African-American) identified. Because participants may figure out the intention of our study if they are exposed to multiple news stories, each participant will only be exposed to 1 condition, or combination, of our independent variables.
The following table illustrates the 4 combinations in our 2 x 2 factorial design, where each person will be assigned to 1 of these conditions:
A: Embezzlement + European-American
B: Embezzlement + African-American
C: Robbery + European-American
D: Robbery + African-American
Summary of Our Factorial Study
Race of Suspect
European-American
African-American
Type of Crime
Embezzlement
A
B
Robbery
C
D
Operationally Defining the Dependent Variable
You have now established the key comparisons between the 4 groups as variations of the type of crime and race of the suspect. Next, we need to specify the exact nature of our dependent variable, racial prejudice. First, consider the following:
We know we want to use a behavioral measure to measure racial prejudice. Now it is time to determine which type of behavioral measure to use. Keep in mind how many and what types of questions, reliability, validity, and sensitivity would be appropriate for our study.
Now that you have determined what you will manipulate and measure, you must formulate an experimental hypothesis. Because we have multiple independent variables, we will also need multiple hypotheses: main effect hypotheses and interaction effect hypotheses.
Now that you have determined how you will collect your data and your intended sample, you must submit your research procedure to the Institutional Review Board (IRB) for ethical approval. The IRB or ethics board will determine whether or not your study meets all ethical guidelines.
IRB
Each IRB has its own protocol that conforms to the national standard when a researcher submits an application for proposed research to be reviewed. In addition to the appropriate paperwork and other information submitted to the IRB, the board would consider the following description during their evaluation of your proposed experiment:
The purpose of this research is to determine whether there is an interaction effect of type of crime (embezzlement or robbery) and race of suspect (European-American or African-American) on racial prejudice. To study this topic, invitations to participate will be mailed to a random sampling of adults in the city. Participants will be told that the study is about current news events and offered $10 for their participation. Those who volunteer will provide informed consent before being randomly assigned to 1 of 4 conditions: Embezzlement + European-American, Embezzlement + African-American, Robbery + European-American, or Robbery + African-American. So that we may measure their racial prejudice, participants will read a news event based on the condition to which they are assigned and then determine the number of years in prison the suspect should receive.
Responding to the IRB
wk1003mike/Shutterstock
The IRB reviewed your submission and has 1 concern. The study involves deception, which carries the potential risk of temporary psychological distress based on whether a participant’s racial prejudice via the prison sentence they provide matches what they believe about themselves.
You must now determine how to respond to the IRB, keeping in mind the ethics of deception.
Now that we have secured the IRB’s approval, we should determine what the entire study will look like. Below are the steps of the study; can you place them in the proper order?
A.
B.
C.
D.
E.
F.
Debrief the participants.
Participants record the number of years in prison they believe the suspect should receive for the crime.
Randomly assign participants to 1 of the 4 conditions.
Obtain informed consent.
Participants read the local news events based on the condition to which they are assigned.
Give participants instructions for the article and recording of the prison sentence.
Collecting Data
Now that you have a sense of how to conduct this study, it is time to see what data from this study might look like.
If you were to run a full version of this study, you would want to have at least 30 participants in each of your 4 groups, for a total of 120 participants. Because you are using a between-subjects design, each participant will only be in 1 group.
Image Source/Photodisc/Getty Images
Example Data Set
This is an example of what your data set would look like. The top row shows the variable names; the other rows display the data for the first 5 participants in each crime condition.
In the “Crime” column, a 1 = Embezzlement, and a 2 = Robbery. In the “Race” column, a 1 = European-American, and a 2 = African-American. The number of years in prison was recorded under Years, with possible scores ranging from 0 to 15 years.
Participant Number
Crime
Race
Years
101
1
1
5
102
1
1
2
103
1
1
2
104
1
1
5
105
1
1
5
161
2
1
6
162
2
1
4
163
2
1
5
164
2
1
5
165
2
1
5
Selecting the Proper Tool
Ines Koleva/E+/Getty Images
Now that you have collected your data, you must decide the best way to summarize your findings. The decisions you made about how to collect your data dictate the statistics you can use with your data now. First, you need to consider if your study is descriptive or inferential.
The following is an example of output for another two-way ANOVA where participants experienced only 1 condition in the study. This study was about how number of hours of television per night (1 hour or 4 hours) for a week and gender (male or female) influence self-reported happiness.
Click on the table below to learn more about each element of the output.
Tests of Between-Subjects Effects
Dependent Variable: Happiness
Source
Type III Sum of Squares
df
Mean Square
F
Sig.
Partial Eta Squared
Corrected Model
92.833a
3
30.944
120.996
.000
.758
Intercept
1267.500
1
1267.500
4956.067
.000
.977
TV
76.800
1
76.800
300.297
.000
.721
Gender
8.533
1
8.533
33.366
.000
.223
TV * Gender
7.500
1
7.500
29.326
.000
.202
Error
29.667
116
.256
Total
1390.000
120
Corrected Total
122.500
119
a. R Squared = .758 (Adjusted R Squared = .752)
This is the df or degrees of freedom. This ANOVA has 4 dfs, one for each between-subjects main effect, and one for the interaction main effect.
These are the F statistics, one for each main effect and one for the interaction. It represents the size of the difference between condition means compared to the size of the residual error.
This is the p level or the significance level. It represents the probability or likelihood that the results happened by chance. The lower the p level, the less likely the result happened by chance.
The F score and p level will only tell you whether there is a significant difference. To determine which means are different, and the nature or direction of those differences, you need to look at the means via a post-hoc test when you are comparing more than 2 means.
Since there are only 2 levels for TV and gender, we do not need to conduct a post-hoc test for these effects.
The eta squared (eta2) is the effect size. It tells us the proportion of change in the dependent variable that is associated with being in the different groups of the independent variable or the interaction of the independent variables.
Tutorial: Evaluating Output
chapter_11_table_activity_2
Alex Brylov/Shutterstock
To report these numbers in a results section, put the numbers in as follows:
F (#,#) = #.##, p = .##, eta2 = .##.
This should be done for each main effect and the interaction.
Click on the table below to learn more about each element of the output.
Descriptive Statistics
Dependent Variable: Happiness
TV
Gender
Mean
Std. Deviation
N
1 Hour
Male
3.53
.507
30
Female
4.57
.504
30
Total
4.05
.723
60
4 Hours
Male
2.43
.504
30
Female
2.47
.507
30
Total
2.45
.502
60
Total
Male
2.98
.748
60
Female
3.52
1.172
60
Total
3.25
1.015
120
This is the average or mean (M) happiness rating for 1 hour of TV.
This is the standard deviation (SD) of happiness rating for 1 hour of TV.
This is the average or mean (M) happiness rating for 4 hour of TV.
This is the standard deviation (SD) of happiness rating for 4 hour of TV.
This is the average or mean (M) happiness rating for males.
This is the standard deviation (SD) of happiness rating for males.
This is the standard deviation (M) of happiness rating for females.
This is the standard deviation (SD) of happiness rating for females.
In this case, the means tell us that happiness ratings were higher for females than males. The results from the ANOVA support this finding that females had higher happiness ratings overall.
In this case, the means tell us that happiness ratings were higher after 1 hour of TV daily for a week than for 4 hours of TV daily. The results from the ANOVA support this finding that those exposed to only 1 hour of TV had higher happiness ratings overall.
In this case, the means tell us that happiness ratings were higher after 1 hour of TV daily for a week than for 4 hours of TV daily. The results from the ANOVA support this finding that those exposed to only 1 hour of TV had higher happiness ratings overall.
Tutorial: Evaluating Output
chapter_11_graph
Alex Brylov/Shutterstock
This graph shows us that females had higher happiness ratings when only 1 hour of TV was watched, but this difference went away when they were exposed to 4 hours of TV. Also, both males and females had lower ratings of happiness after 4 hours of TV than 1 hour of TV. There was an interaction between gender and hours of TV as the happiness ratings differed based on these variables.
Happiness
Hours of TV
Your Turn: Evaluating Output
Alex Brylov/Shutterstock
Below is the output from your study:
Tests of Between-Subjects Effects
Dependent Variable: Years
Source
Type III Sum of Squares
df
Mean Square
F
Sig.
Partial Eta Squared
Corrected Model
345.292a
3
115.097
107.933
.000
.736
Intercept
3424.008
1
3424.008
3210.873
.000
.965
Crime
200.208
1
200.208
187.746
.000
.618
Race
118.008
1
118.008
110.663
.000
.488
Crime * Race
27.075
1
27.075
25.390
.000
.180
Error
123.700
116
1.066
Total
3893.000
120
Corrected Total
468.992
119
a. R Squared = .736 (Adjusted R Squared = .729)
Your Turn: Evaluating Output
Alex Brylov/Shutterstock
Below is the output from your study:
Descriptive Statistics
Dependent Variable: Years
Crime
Race
Mean
Std. Deviation
N
Embezzlement
European-American
3.53
1.224
30
African-American
4.57
1.104
30
Total
4.05
1.268
60
Robbery
European-American
5.17
.834
30
African-American
8.10
.923
30
Total
6.63
1.717
60
Total
European-American
4.35
1.325
60
African-American
6.33
2.047
60
Total
5.34
1.985
120
Your Turn: Evaluating Output
chapter_11_multiple_choice_2
Alex Brylov/Shutterstock
Based on the results of your statistical analyses, match the correct number in the “Answer” column to the term requested under “Prompt”:
Tests of Between-Subjects Effects
Dependent Variable: Years
Source
Type III Sum of Squares
df
Mean Square
F
Sig.
Partial Eta Squared
Corrected Model
345.292a
3
115.097
107.933
.000
.736
Intercept
3424.008
1
3424.008
3210.873
.000
.965
Crime
200.208
1
200.208
187.746
.000
.618
Race
118.008
1
118.008
110.663
.000
.488
Crime * Race
27.075
1
27.075
25.390
.000
.180
Error
123.700
116
1.066
Total
3893.000
120
Corrected Total
468.992
119
a. R Squared = .736 (Adjusted R Squared = .729)
F for the ANOVA test – main effect of crime
p for the ANOVA test – main effect of crime
F for the ANOVA test – main effect of race
p for the ANOVA test – main effect of race
F for the ANOVA test – interaction effect
p for the ANOVA test – interaction effect
eta2 – main effect of crime
eta2 – main effect of race
eta2 – interaction effect
Please move the correct answer to the left.
187.746
0.000
110.663
0.000
25.390
0.000
0.618
0.488
0.180
Activity: Graphing Results
chapter_11_graph_2
By pulling the top of the bar graph, please indicate the mean for each race.
Descriptive Statistics
Dependent Variable: Years
Crime
Race
Mean
Std. Deviation
N
Embezzlement
European-American
3.53
1.224
30
African-American
4.57
1.104
30
Total
4.05
1.268
60
Robbery
European-American
5.17
.834
30
African-American
8.10
.923
30
Total
6.63
1.717
60
Total
European-American
4.35
1.325
60
African-American
6.33
2.047
60
Total
5.34
1.985
120
Type of Crime & Race
Mean Years Prison
Type of Crime
Your Turn: Results
gguy/Shutterstock
Now that you have worked with your data, you must determine the best way to express your findings in written form. You must be sure that how you describe your findings accurately represents the data.
Now that you have determined how to express your findings in a scientifically responsible way, you also need to be able to talk about what your findings mean in everyday terms so that the world can benefit from your science.