Culture and the Individual Kimberly Porter Martin.

Culture and the Individual

Kimberly Porter Martin

Overview

Scientific Paradigm

Research Design

Populations & Sampling

Data Collection

Data Analysis & Interpretation

Validity

Reliability

Inferences from Data

Scientific Paradigms

• Modern/Positivism – there is a reality and through science we can discover what that one, true reality is by PROVING what is real.

• Post positivism – there is a reality out there, but it can only be imperfectly understood through FALSIFYING hypothesis scientifically.

• Post Modern/Constructivist – there is no one reality, only perspectives that are socially constructed. So-called “scientific” results are socially constructed as well, and are therefore just another perspective among many.

Aspects of Research Design

Is the goal nomothetic or idiographic?

Does data need to be collected cross-sectionally or longitudinally?

Should quantitative or qualitative data be collected?

Should emic or etic data be collected?

Should data be collected in an experimental or a natural setting?

Nomothetic and Idiographic

• Nomothetic refers to studies that produce generalizations about a concept or population.

• Idiographic refers to studies that give detailed, descriptive information about individuals or groups.

Time Frames

• Cross-sectional research collects data during a single time.

• Longitudinal research collects data at two more different time periods to show how how something has changed over time.

Quantitative vs Qualitative Data

• Quantitative data is collected in the form of numbers or of categories that can be labeled with numbers. Quantitative data is analyzed using descriptive and inferential statistics.

• Qualitative data is collected in the form of descriptions or narratives that are reported in essay form with selected quotes or examples.

Emic vs Etic Data

• Emic data presents an insider’s view that may not be comparable to the views of outsiders. Subjective data.

• Etic data presents an outsider’s view that allows for easy comparison, but may not truly reflect the different perspectives of the groups being compared. Objective data.

Experimental vs Natural Settings

• Experimental settings remove people from their daily contexts in order to try to control variables that might confuse the results of the research.

• Natural settings allow data to be gathered in the context in which people actually live insuring that results reflect how people actually behave in the real world.

Populations

The group to which findings will be generalized. For cross cultural studies this means what

cultures will be studied.

• Selected on the basis of similarities (to control for confounding variables).

• Selected on the basis of differences (where the differences are the independent (causal) variables).

• Culturally defined level depends on the amount of diversity within the society; national, regional, local, ethnic, language group, organizational membership, etc.

Samples

The set of individuals or objects from which data will actually be collected. This means which groups within the society and which individuals within the groups should participate in the research. The best type of sample is randomized.

• Cross-cultural studies rarely use randomized samples.

• Biased samples produce biased data.• Galton’s problem says that neighboring societies

may have superficially similar characteristics because of diffusion.

Data Collection Methods

• Experiments

• Observation

• Participant Observation

• Interviewing

• Psychological Testing

• Text Analysis

Experiments• Identify the independent variable (the variable that will

make a change) and the dependent variable (the variable that will be changed).

• Select participants and randomly divide them into two groups: an experimental group that will be exposed to the independent variable, and a control group that will NOT be exposed to the independent variable.

• Place both groups in an experimental setting away from normal daily activities.

• Pretest both groups on the dependent variable.• Expose the experimental group to the independent

variable.• Post test both groups.• Compare the post test scores of the two groups. If there

is a significant difference, then it can be attributed to the independent variable as the cause.

Experiments: The Pros

• Many confounding variables are eliminated by the experimental setting.

• The randomly selected control group provides a baseline comparison against which change in the dependent variable can be measured.

• The researcher controls the exposure to the independent variable for the experimental group, making sure that it comes BEFORE the post test.

• Cause and effect can be directly inferred from a significant change in the dependent variable in the experimental groups post test scores.

Experiments: the Cons• Removing people from their normal daily context may

change their behavior/performance.• Many variables of interest cannot be controlled by the

researcher.• In many cases one cannot use randomization in the

selection of participants at all.• It is sometimes not possible to pretest individuals before

they are exposed to the independent variable.• Experiments can only provide results at a single point in

time and cannot be used to document processes of change.

• Experimental results are usually quantified and cannot be used for idiographic research purposes.

• Can only be used to collect etic data.

Observation & Participant Observation

• Spot Observation – the observer records activity as soon as he/she first enters the context.

• Pre-coded Observation – observers agree ahead of time on a set of targeted behaviors that will be counted and/or described.

• Time Sampling – activities are recorded at set time intervals (every five minutes for one hour = 12 observations)

• Event sampling – Observing on a given number of occasions for an established period of time.

Observation & Participant Observation: Pros

• Takes place during the normal daily activities of participants, and is therefore a more valid measure of what people actually do.

• Has the potential to produce either emic or etic data depending on the type of observation that is done.

• May be the only way to gather valid data on behaviors, as individuals frequently don’t or can’t answer questions about their behaviors accurately.

• Both quantitative and qualitative data can be collected.

Observation & Participant Observation: Cons

• The presence of the researcher may change the behaviors of participants, who will not behave normally in the presence of an outsider.

• Needs to be focused by a research question; can be unfocused and invalid if goals are not explicit.

• Observer bias can be a problem; the investigator may see what he/she wants to see, and ignore unwanted data.

• There may be contexts in which the observer is not welcome and so the sample of behavior will not be representative.

• Unusual behaviors will be difficult to document.

Interviewing

• Unstructured Interviews – interviews in which questions are not developed before the interview and the interviewee’s lead is followed by the interviewer.

• Structured interviews – questions are developed before the interview and are asked in the same way and in the same order for each participant.

Interviewing: Pros

• The best way to get emic data about meanings and cultural explanations.

• Both emic and etic data may be obtained this way.

• Can collect both quantitative and qualitative data.

• May get kinds of information of which the researcher was completely unaware and would not have asked about.

Interviewing: Cons

• Participants will frequently say what they think the interviewer wants them to say.

• People may not be consciously aware of their behaviors or attitudes and may not be able to report what they are.

• Structured interviews may ask questions that are not relevant in the culture of the participant.

• Unstructured interviews depend on the discretion of the interviewer, and the data collected may not be reliable.

Psychological Testing

EXAMPLES

• Optical Illusions

• Rorschach Tests

• TAT Tests

• MMPI

• State/Trait Anger Inventory

Projective Tests

Rorschach Tests

Optical Illusions

Psychological Testing: Pros

• Used to test the validity of established modern western theories about personality and about universal human traits.

Psychological Testing: Con’s

• It is ethnocentric to believe that modern western theories will fit all humans.

• The format of test items reflects modern western traditions and practices.

• The content of test items sometimes reflects modern western practices and traditions.

• Testing individuals does not work in many traditional societies, where all tasks are done in groups.

Content (Text) Analysis

• Extrapolating cultural meanings, values and personality traits from documents, art or other material cultural products.

Content (Text) Analysis: Pros

• People may unconsciously structure things that they write or design in ways that they cannot articulate.

• Widely used material products may unconsciously influence people’s values, behaviors and beliefs.

Art and Design Example

Navaho Textile Design

Egalitarian Society, dispersed population, autonomy and self-reliance valued

Art and Design Example Contrast

Indian Textile PatternsHighly Stratified Society with clear social classes/casts that cannot be escaped during a lifetime.

Content (Text) Analysis: Cons

• A single type of product may only be representative of a small portion of a population causing overgeneralization.

• Selection of a type of product requires the researcher to choose a representative type of product, from which the researcher will then extrapolate cultural patterns; this can cause a self-fulfilling prophecy.

Aspects of Cultural Bias• Differences in the way a concept is measured - eg. How big a

person is measured in one culture by weight, in another by height, and in another by spiritual criteria.

• Differences in physical testing/data collection environment.• Differences in the relevance of the concept to the groups.• Differences in familiarity with testing materials.• Differences in observer ratings for behaviors.• Problems in communication between researcher and participant

due to language, role or gender.• Differences in social consequences of participation in the study.• In the way questions are worded and/or translated from one

language to another – Back Translation = different individuals translate first from language A into B, and then from language B back into language A and the two A’s are compared.

• In the way data is interpreted eg. If a child is classified in his society as big, will the researcher interpret that as meaning the child is tall or heavy or both?

Addressing Reliability

When repeated trials of the same research procedure yield the same results.

• Reliability is more of a problem with

qualitative data and methods. • Controls in experimental and quantitative

methods make reliability easier to

achieve.

Addressing Validity

when you measure what you say you are measuring.

1. Interpretive validity

2. Ecological validity

3. Theoretical validity

Interpretive Validity

Understanding the participants and their culture and context well enough that you can design an appropriate and meaningful research project.

Ecological Validity

The data to be collected and the methods of data collection are relevant to participants and to the participants daily lives and contexts.

Theoretical Validity

Is the data that you are collecting an accurate measure of what you are studying?

Making Inferences from Data

• Low Level Inferences – can be made when the concept is well understood and measured using instruments that address all aspects of the concept.

• Medium Level Inferences – are made when certain behaviors are assumed to reflect abstract characteristics (eg. Personality traits).

• High Level Inferences – are made when concepts are not amenable to measurement (eg. acculturation)