Discussion

Raciel1990

CULTPSY3_Ch04_Methods1.pptx

Home >Psychology homework help >Discussion

Lecture PowerPoint Slides

Benjamin Cheung

Chapter 4—Methods for Studying Culture and Psychology

Cultural Psychology

Third Edition

Steven J. Heine

Chapter Objectives

In this chapter, you will:

Discuss points of consideration when deciding on what cultures to sample

Understand the role of methodological importance in cross-cultural research

Explain how generalizability is of particular importance to cultural psychology

Define statistical power

Differentiate between independent variables and dependent variables

Understand the relationship between statistical power, independent variables, and dependent variables

Differentiate between back-translation and plain translation

Discuss the pros and cons of the back-translation method

Discuss the impact of, and solutions to, different biases that respondents have to questionnaires

Explain the advantage of the experimental method over straightforward questionnaire studies

Chapter Objectives

In this chapter, you will:

Differentiate between between-groups and within-groups manipulations

Recognize what constitutes neuroscientific evidence

Explain how neuroscience has contributed to cultural psychology

Describe different research methodologies that have been particularly useful for cultural psychology

Distinguish between the different analyses that situation sampling allows researchers to do

Distinguish between culture-level and individual-level measures

Discuss the utility of “unpackaging” cultural findings

Discuss the relevance of Occam’s Razor in cultural psychology

Justify the claim that Nisbett and Cohen’s study on the Culture of Honor demonstrates several methodological concepts discussed in the textbook

Understand how, through Nisbett and Cohen’s study on the Culture of Honor, large-scale sociocultural contexts can affect psychology at the individual level

Overriding Themes in This Chapter

Studying psychological phenomena is challenging in general—much more so when studying it cross-culturally.

We have trouble accessing our own psychological states, never mind trying to study someone else’s.

Whom to Study?

Discuss: How do we decide which cultures to sample?

With all the cultures around the world, which ones do we study? Discuss.

Whom to Study?

If the goal is to see if X (e.g. individualism) shapes Y (e.g. preference for uniqueness), then find two cultures that vary on X.

CultureHighX

CultureLowX

Choosing which cultures to study depends on the research question.

If the goal is to see if X (e.g., individualism) shapes Y (e.g., preference for uniqueness), then find two cultures that vary on X.

If the goal is to see the universality of X (e.g., theory of mind), then find two cultures that are maximally different on many dimensions.

Similarities between them suggest a high level of universality.

Whom to Study?

If the goal is to see universality of X (e.g. theory of mind), then find two cultures that are maximally different on many dimensions.

Choosing which cultures to study depends on the research question.

If the goal is to see if X (e.g., individualism) shapes Y (e.g., preference for uniqueness), then find two cultures that vary on X.

If the goal is to see the universality of X (e.g., theory of mind), then find two cultures that are maximally different on many dimensions.

Similarities between them suggest a high level of universality.

Issues with Comparing Cultures

Researchers must understand cultural norms and practices of the cultures being studied, especially in relation to the psychological phenomenon in question.

If not, researchers risk drawing conclusions based on faulty information and assumptions.

Such understanding is accomplished through use of ethnographies, foreign collaborators, and immersion in culture of interest.

Combination of these methods is ideal.

Issues with Comparing Cultures

Researchers need to watch for methodological equivalence.

Ensure methods are understood in identical ways across cultures

Some cultures may not understand process of completing psychological surveys

Researchers may need to use slightly different methods with different cultures (especially drastically different ones).

Issues with Comparing Cultures

Researchers circumvent many of these problems by primarily targeting university students in industrialized societies in cross-cultural research.

What are some problems that this practice may generate?

Issues with Comparing Cultures

Problems with targeting university students in industralized societies

Generalizability—do findings apply to nonstudent populations in industralized societies? What about in subsistence cultures?

Power (ability to detect an effect  cultural difference)—since university students in industrialized societies share many similar experiences, cultural effects may become diluted.

Such comparisons are conservative and less powerful.

Issues with Cross-Cultural Surveys—Translations

Translation of materials

Translation of materials is important in cross-cultural research but can be problematic.

Many psychological terms cannot be directly translated.

Issues with Cross-Cultural Surveys—Translations

If a bilingual collaborator is available, s/he can decide whether translated materials are appropriate

Or back-translation:

If a bilingual collaborator is available, s/he can decide whether translated materials are appropriate.

Translator 1 translates materials from original language to target language

Translator 2 translates materials back to original language

Original and back-translated materials are compared, and discrepancies are resolved

Question for Consideration

Discuss: How does one translate idioms, or nuanced foreign words like schadenfreude, amae, and self-esteem?

Issues with Cross-Cultural Surveys —Response Biases

Psychological surveys are usually done using number scales.

For example:

	Strongly disagree						Strongly agree
I never laugh at dirty jokes		1	2	3	4	5	6	7
I have very few bad habits		1	2	3	4	5	6	7
I never lie		1	2	3	4	5	6	7

Different cultures have different tendencies, or biases, when responding to surveys.

Socially desirable responding—tendency to respond in ways that elicit positive impressions from others

Moderacy bias—tendency to choose numbers toward midpoint of scale

Extremity bias—tendency to choose numbers toward the ends of the scale

Issues with Cross-Cultural Surveys —Response Biases

	Strongly disagree						Strongly agree
George W. Bush was a great president		1	2	3	4	5	6	7
George W. Bush was the greatest president		1	2	3	4	5	6	7
Stephen Colbert for president		1	2	3	4	5	6	7

Moderacy bias

Extremity bias

Another example:

Discuss: How do we counteract this?

Different cultures have different tendencies, or biases, when responding to surveys.

Socially desirable responding—tendency to respond in ways that elicit positive impressions from others

Moderacy bias—tendency to choose numbers toward midpoint of scale

Extremity bias—tendency to choose numbers toward the ends of the scale

Solution 1:These biases can be controlled for by using forced-choice questions (e.g. Yes/No)

Issues with Cross-Cultural Surveys—Response Biases

George W. Bush was a great president	No	Yes
George W. Bush was the greatest president	No	Yes
Stephen Colbert for president	No	Yes

What are some drawbacks of doing it this way?

It is harder to detect nuances and smaller differences when using forced-choice questions.

Issues with Cross-Cultural Surveys—Response Biases

Solution 2: Standardization

Turn one’s response into a standardized score (z-score)

z-score

–

Larger than the average

Smaller than the average

Frequency

Extraversion

Conscientious

Solution 2: Standardization, especially examining how one scores across a variety of different domains

This involves turning one’s responses into standardized scores (z-scores).

This creates a standard distribution curve for each participant, with a z-score of the average at 0.

Scores larger than the average have a positive z-score.

Scores smaller than the average have a negative z-score.

This process forces everyone to have the same average (i.e., 0) and creates the same distribution for everyone’s scores, thus getting rid of moderacy and extremity biases.

This allows us to look at people’s patterns of responses (e.g., people are more extraverted than they are conscientious).

Issues with Cross-Cultural Surveys—Response Biases

This does not allow conclusions based on differences in averages between groups/cultures (e.g. Chinese more introverted than Americans).

This does allow conclusions based on patterns of groups’ responses (e.g. Chinese more introverted than extraverted, opposite for Americans).

Issues with Cross-Cultural Surveys—Response Biases

	Strongly disagree						Strongly agree
George W. Bush was a great president		1	2	3	4	5	6	7
George W. Bush was the greatest president		1	2	3	4	5	6	7
Stephen Colbert for president		1	2	3	4	5	6	7

Acquiescence bias

Solution:

	Strongly disagree						Strongly agree
George W. Bush was a great president		1	2	3	4	5	6	7
George W. Bush was the worst president		1	2	3	4	5	6	7
Stephen Colbert for president		1	2	3	4	5	6	7

Another response bias is the acquiescence—referring to tendency to agree with most items on a measure

This bias can be neutralized by reverse-scoring half of the items

Half of the items is numbered normally, while the other half is numbered such that large numbers actually indicate strong disagreement.

Issues with Cross-Cultural Surveys—Response Biases

Reference group effect—one’s response to questions may depend on the group that one is using for reference.

For example, how does one respond to the item “I am tall”?

Issues with Cross-Cultural Surveys—Response Biases

The author of the textbook, Dr. Steven Heine, at 5’8”, is considered tall in Obama, Japan (reference group 1).

But in the Netherlands, his neck would hurt from craning his head all day (reference group 2).

So how will his reference group affect how tall he perceives himself to be?

Issues with Cross-Cultural Surveys—Response Biases

To control for this, use objective and concrete measures, which can be achieved by:

Providing specific scenarios as questions

Soliciting quantitative responses (e.g. frequencies of specific behavior)

Using behavioral and physiological measures

Issues with Cross-Cultural Surveys—Response Biases

Deprivation effects—tendency for people (or cultures) to value what they would like, not what they have

E.g. Americans value “humility” more than Chinese, Chinese value “choosing one’s own goals” more than Americans

No clear solution for this bias, except to interpret results with caution

Issues with Cross-Cultural Experiments

Culture is not an independent variable that can be manipulated.

Culture ≠ Independent

But other variables have been manipulated in cross-cultural studies.

Two types of possible manipulations:

Between-groups manipulations

Within-groups manipulations

Issues with Cross-Cultural Experiments

Between-groups manipulation

Participants are randomly assigned to one of the conditions of the independent variable.

Ensures that participants are statistically equivalent at the beginning of study

We can attribute any differences in dependent variable between groups to differences in independent variable.

Issues with Cross-Cultural Experiments

Within-group manipulation:

Participants go through all conditions of independent variable.

No random assignment is necessary for conditions (although researchers may randomly assign the order in which one goes through the conditions).

Some participants go through Condition A and then B, while others go through Condition B and then A.

New Research Methodology

Neuroscience

Neuroscience has emerged recently as a new technique used in cultural psychology, by studying how the brain responds to stimuli differently based on cultural background.

Much of this work is done using fMRI – using magnets to detect activity in the brain.

Research Methods for Cultural Psychology

There are methods that are particular to cultural psychology.

Two methods will be discussed:

Situation sampling

Cultural priming

Research Methods for Cultural Psychology

Situation-sampling—a two-step method:

Participants from each culture generate situations during which they experience some psychological phenomenon.

Another group of participants assesses full compiled list of situations generated by both cultures in Step 1.

This allows for two types of analyses:

Examining cultural differences in how participants respond to the same situations

Examining cultural differences in the types of experiences/situations that people have

Cultural priming—inducing cultural ways of thinking in participants who were not enculturated by the actual cultural group

Assumes that while some ways of thinking may be different between Cultures A and B, Culture A’s way of thinking may still be present to some degree in Culture B

E.g. Priming individualism by forcing participants to use first-person singular pronouns (I, me), vs. priming collectivism by forcing the use of first-person plural pronouns (we, us)

Research Methods for Cultural Psychology

Studying “Culture,” Not “Cultural People”

Most methods study people in different cultures

Also important to study actual cultures

This is primarily done on the things found in cultural messages

Studying “Culture,” Not “Cultural People”

Testing hypotheses with cultural messages often requires using trained coders to turn messages into quantifiable data.

Coding = creating themes/categories based on theoretical expectations, then trying to fit cultural messages into those categories

E.g. do Japanese newspaper articles contain more collectivistic themes than do American newspaper articles?

Due to the subjective nature of coding, multiple coders are usually used, while being kept blind to the hypothesis.

Unpackaging Culture

Cultural differences are embedded within vast networks of cultural practices and symbols.

To understand what actually creates cultural differences amid the many cultural practices, we need to unpackage culture.

Unpackaging = identifying underlying variables that create cultural differences

Find a theoretically viable variable that can explain a cultural difference.

Confirm cultural difference in the proposed underlying variable.

Show that underlying variable is related to cultural difference in question.

Unpackaging Culture

3 steps to unpackaging

Find a theoretically viable variable that can explain a cultural difference.

For example, more concern with interdependence in Japan than in the United States may explain higher embarrassability in Japan.

Confirm cultural difference in the proposed underlying variable.

For example, Japanese more interdependent than Americans

Show that underlying variable is related to cultural difference in question.

For example, show correlation between interdependence and embarrassability, in the proper direction

More Is Better: Using Multiple Methods

No single study design is perfect, due to alternative explanations and methodological flaws.

Best way to counter such problems is to use multiple methods.

Using multiple methods to replicate findings while disproving alternative accounts = very compelling evidence

This means that, by principle of Occam’s razor, the simplest explanation that accounts for the same results as more convoluted explanations, is the preferred explanation.

Southern Honor

Nisbett and Cohen’s investigation serves as a good example of using multiple methods to unpackage a cultural phenomenon.

Many have noted that the U.S. South is more violent than the North…but why?

Nisbett and Cohen argue thus:

Herding was more prevalent in the South.

Herding made it easy for others to steal from herders.

To protect assets, herders had to build up and protect reputation by showing aggression.

Southern Honor

Nisbett and Cohen argue that:

Herding was more prevalent in the South.

Herding made it easy for others to steal from herders.

To protect assets, herders had to build up and protect their reputation by showing aggression.

To test their hypotheses, Nisbett and Cohen used a variety of methods.

The researchers’ methodologies include:

Archival data

Survey data

Physiological data

Behavioral data

Field experiment

Southern Honor

The researchers’ methodologies include:

Archival data:

More homicides, especially argument-related murders, occurred in herding regions than in farming regions in the South.

Survey data:

Southerners were more likely than Northerners to condone violence in the defense of honor.

Physiological data:

Southerners who were insulted had a spike in testosterone level, but Northerners did not.

Behavioral data:

Southerners who were insulted were likelier than Northerners to confront an oncoming stranger.

Field experiment:

Southern employers were more sympathetic than Northern employers toward fake job applicants who were convicted of a crime that was made to sound like it was committed to defend one’s honor.

Summary

Studying cultural differences requires one to be vigilant for issues and problems in designing cross-cultural studies.

Many of these issues have specific fixes, but the best fix is to mix multiple methods.

To truly understand the nature of a cultural difference/phenomenon, one must unpackage it.

Cross-cultural studies can be daunting but very rewarding and informative when done properly.

Discussion

image2.png

image3.emf

image4.jpeg

image5.jpeg

image6.jpg

image7.jpg

image8.jpg

image9.jpg

image10.jpg

image11.jpeg

image12.png

image13.png

image14.JPG

image15.jpg

image16.jpg

image1.png