Most Important Flashcards
- methods used in CC psychology?
- Ethnography
- Interviews/Focus Groups
- Cultural Product & Content
- Psychometric Measures (bias and equivalence are applicable).
- Experiments
- Priming
Psychometric measures:
Most objective method to obtain data using standardized paper-pen questionnaires or online surveys to collect data on psychological constructs.
Standardized in the sense that they have fixed formats (i.e., Likert-scale asks people to rate how much they agree or disagree to each statement on a fixed scale).
The setting in which the scale is administered will impact on the objectivity of the measurements.
The independent variable their nation or ethnicity is hypothesized to indirectly measure the effects of culture (IV).
Unless mediator variable is included (i.e., culture is unpackaged) causal inferences can NOT be made.
Experiments in CC are ___ so are always vulnerable to
quasi-experimental, vulnerable to alternative explanations
2 problems in cc with ethnography
Researcher may impose their beliefs and positionality onto their interpretations of their findings (i.e., impose feminist beliefs).
The presence of the observer changes their behaviour.
A national level dimension of culture raises two important questions: 1) will the instruments and the constructs at the nation level still have the same structure at the individual level (___) 2) does the relationship between psychological constructs and a third variable in the same nation level as it is in the individual level (__).
isomorphism, homology
Can cultural groups such as nations be compared using individual-level scale means?
and vise versa?
No & No.
unpackaging culture is used when and invloves?
when insufficient samples to do multi level modelling
he unpackaging of culture involves 1) the identification of theoretical factors or processes which may produce cultural differences in psychological outcomes of interest 2) the explicit empirical test of the proposed process leading to these outcomes.
(5) Guidelines of cross-cultural research:
- Researchers should recognize the cultural contingencies of their own beliefs and values when conducting research and should resist evaluating other cultures against our own cultural beliefs.
- Researchers should study psychological processes in their native language of participants, using English is likely to lead to cultural accommodation of responses to the perceived cultural stereotypes of the language being used (i.e., people responses are responsive to the language the items are phrased and can underestimating cross-cultural differences).
- Samples should be matched closely as possible in order to rule out alternative explanations for observed differences in outcomes being studied (i.e., confounds).
- Instruments need to be valid and reliable in all cultural groups in order to accurately detect cross-cultural similarities or differences. Researchers should try to ensure that instruments, measures, and manipulations are understood in comparable ways in each location.
- Experiments are more persuasive in their evidence if a) they include a manipulation check on the effectiveness if the crucial manipulation in all cultural samples, and b) the experimenter specifies the relevant psychological processes and relates then to the dependent variable of interest, to rule out alternative theoretical explanations for the observed cultural difference.
how do we determine 1) if an interpretation is valid? 2) If the instrument[s] producing the observed differences are valid?
Interpretation should check for (3) biases:
• Constructs Examined (construct bias)
• Administration procedures (method bias)
• Operationalisation (item bias)
How do we determine if a scale is fair across all groups? Look for equivalence- 1. functional/construct 2. structural 3. metric 4. full score or scalar
how do we test for equivalence?
measurement models using latent variable and CFA to test for measurement and stuructural invariance
measurement invariance:
configureal (same meaning and number of dimensions)
metric (same factor loadings (slope), units)
scalar (no uniform bias, intercepts)
error (error, scale percision)
Structural invariance
not nested additional steps not required for CC comparision factor covariance invaraince factor variance invariance factor mean invariance
example of construct bias and why
Marshmallow test
How does this demonstrate construct bias?
Brain scans show that the waiters had less reward activation than the non-waiters which indicates that the meaning of the test may be fundamentally different between the two groups.
It may not be measuring willpower at all. Brain scans show that for children who waited it wasn’t the brain regions associated with self-control that were activated.
example of method bias and why
Yanny or Laurel:
How does this demonstrate method bias?
If the test was to identify which word was being said. The overlay of higher pitched sounds bias this test against hearers who are more likely to detect the higher pitches, making it unfair due to the method testing recognition.
Showing the two written word for us to read primes us to hear one or the other. Otherwise we may have heard something completely different.
example of item bias and why
“In bed I get cold feet”
How does this item demonstrate item bias?
If you are not familiar with the idiomatic meanings of the saying “getting cold feet” your answers will be very different from other cultural groups and not actually reflect a meaningful difference.
Presupposes that people who are taking the survey are familiar with these aspects of figurative speech.
