Chapter 4 Flashcards
Researchers cannot observe
- All of a person’s behavior
* All people’s behavior
Researchers can observe
- Samples of individuals
- Samples of behavior at particular times
- Samples of different settings and conditions
Goal of sampling
- Represent larger population of
- Behaviors
- People
- Settings and conditions
Use data from a sample to represent the population
- “Generalize” the findings from sample to population
* Sample must be similar to population
External Validity
*Extent to which a study’s findings may be used to describe people, settings, conditions beyond those used in the study.
Sampling Behavior
- Extent to which observations may be generalized (external validity)
- Depends on how behavior is sampled
Two Methods of Sampling Behavior
- Time sampling
* Situation sampling
Goal of Sampling Behavior
Obtain representative sample of behavior
Time Sampling
- Choose time intervals for making observations
- Systematic
- Random
- Don’t use time sampling for observing behavior during rare events (e.g., hurricane)
- Event sampling
Situation Sampling
- Choose different settings, circumstances, conditions for observations
- Enhances external validity
- Use subject sampling to observe only some individuals within a situation
Naturalistic Observation
- Direct Observation without Intervention
- Observation in natural (real-world) setting
- No attempt to intervene or change situation
- An observer using this method of observation acts as a passive recorder of events as they occur naturally.
Goals of Naturalistic Observation
- Describe behavior as it normally occurs
- Examine relationships among naturally occurring variables
- Establish external validity of lab findings
- Use when ethical considerations prevent experimental manipulation
Direct Observation with Intervention
*Characterizes most psychological research
Gain control over observations
Three Methods in Natural Settings
- Participant observation
- disguised, undisguised
- Structured observation
- Field experiment
Problem of Reactivity
- People change their usual behavior when they know they’re being observed.
- Goal: observe people’s usual behavior
- Avoid reactivity
Indirect (Unobstrusive) Observational Methods
- Examine evidence of past behavior
- Nonreactive
Indirect (Unobtrusive) Observational Methods: Two types of methods
- Physical traces
- Use (natural or controlled)
- Products
- Archival records
- Running records
- Episodic records
Possible problems in archival records
- Selective deposit
- Selective survival
- Spurious relationships
Unobtrusive Measures
Seek converging evidence using multimethod approach.
Recording Behavior
- Comprehensive record
- Video, audio recordings; written field notes
- Select specific behaviors
- Checklists, ratings
- Method for recording behavior determines how results are
- Measured, summarized, analyzed, reported
Nominal Measurement Scale
Categorize behaviors, events
*Sort stimuli into discrete categories
Ordinal Measurement Scale
Rank-order
*Rank-order stimuli on a single dimension
Interval Measurement Scale
- Specify distance on a dimension
- Rating scales are treated as interval scales
- Specify the distance between stimuli on a given dimension
- No true zero
Ratio Measurement Scale
Specify distance plus meaningful zero
*Specify the distance between stimuli on a given dimension and express ratios of scale values
Method for analysis depends on
- Goal of the study
- How data are recorded
- Measurement scale
Analysis of Observational Data: Two types of analysis
- Qualitative
* Quantitative
Qualitative Analysis
- Data reduction to summarize comprehensive records
- Coding: identify units of behavior using specific criteria
- Emphasis on verbal summary
- Content analysis
- Identify relevant archival source
- Obtain representative sample from source
- Code content using descriptive categories
Quantitative Analysis
- Statistical summary of observations
- Descriptive statistics depend on measurement scale
- Nominal: relative frequency
- Ordinal: rank percentages
- Interval and ratio: mean, standard deviation
- Provide a numerical, or quantitative summary of observations in a study
- Calculate descriptive statistics that summarize the observational data
- Assessing the reliability of the observations
Interobserver Reliability
- Measure of agreement between observers
- Nominal: percent agreement
- Ordinal: Spearman rank-order correlation
- Interval and Ratio: Pearson correlation
Factors that affect interobserver reliability
- Characteristics of the observers
- Bored, tired, amount of experience
- Train observers and provide feedback
- Clearly define events and behaviors to be observed
- Provide examples
- Clear operational definitions
Problems in observational research
- Influence of the observer on behavior
* Observer bias
Influence of the Observer
- Reactivity: people change their usual behavior when they know they’re being observed
- Researchers want to observe people’s usual behavior.
- Demand characteristics: people pay attention to cues and information in the situation to guide their behavior.
Reactivity
- People change their usual behavior when they know they’re being observed
- Reactivity is a potential problem in most psychological research.
Controlling Reactivity
- Conceal observer
- Disguised participant observation
- Use indirect (unobtrusive) observation
- Adapt participants to observer
- Habituation
- Desensitization
- Limit information about study
Ethical issues when controlling reactivity
- Privacy and informed consent
- Observe people without their knowledge
- Potential risks
- Structured observation and field experiments
- Risk/benefit ratio
- Balance ethical obligation to benefit individuals and society with potential risks of observing people’s behavior
Observer Bias
- Observers often have expectations about behavior.
- Example: expectations based on research hypotheses
- Expectations can lead observers to look at only particular behaviors
- Observer bias: systematic errors in observation that result from expectations
- Also called experimenter expectancy effects
- Potential problem in all research
- Hard to eliminate
- &Observers must always be aware that they may be biased
- Reduce bias by keeping observers “blind” to aspects of the study:
- Reasons for observations
- Goals of the study
- Hypotheses
Situation Sampling
Involves observing behavior in as many different locations and under as many different circumstances and conditions as possible.
- By sampling various situations, researchers reduce the chance that their results will be unique to specific circumstances or conditions.
- Enhances the external validity of findings
Subject sampling
Determine which students to observe
- The research could either select students systematically or select students randomly
- Goal: obtain a representative sample
- Within situations, subject sampling may be used to observe only some individuals in the setting
Participant Observation
- Observers play a dual role:
- They observe people’s behavior and they participate actively in the situation they are observing.
- Allows an observer to gain access to a situation that is not usually open to scientific observation.
- Observers may lose their scientific objectivity if they identify too closely with the people and situation they are observing.
- Participant observers must be aware of the threat to objective reporting due to their involvement in the situation, particularly as their involvement increases.
- The observer can influence the behavior of people being studied.
Undisguised participant observation
Individuals who are being observed know that the observer is present for the purpose of collecting information about their behavior.
*Is used frequently by anthropologists who seek to understand the culture and behavior of groups by living and working with members of the group
Disguised participant observation
Those who are being observed do not know they are being observed.
Online participant observation
*Provides the opportunity to observe individuals generally not receptive to being observed, and to generate hypotheses that may
Structured Observation
- Researchers intervene to exert some control over the events they are observing.
- Often the observer intervenes in order to cause an event to occur or to “set up” a situation so that events can be more easily recorded.
Field Experiment
- When a researcher manipulates one or more independent variables in a natural setting in order to determine the effect on behavior
- Represents the most extreme form of intervention in observational methods
- Are frequently used in social psychology
Difference between field experiments and other observational methods
Researchers exert more control in field experiments when they manipulate an independent variable
Physical traces
- Use traces and products
- Remnants, fragments, and products of past behavior
- Examining physical evidence of past behavior can provide important clues about the characteristics of individuals and events
- The examination of products allows researchers to test important hypotheses about behavior.
- Offer researchers valuable and sometimes innovative means to study behavior, and the measures available are limited only by the ingenuity and resourcefulness of the investigator
- The validity of physical-trace measures must be examined carefully and verified through independent sources of evidence
- Researchers need to obtain supplementary evidence for the validity of physical traces
- Alternative hypotheses for observations of physical traces must be considered and care must also be taken when comparing results across studies to make sure that measures are defined similarly.
Use traces
What the label implies- the physical evidence that results from the use (or nonuse) of an item.
- Remains of cigarettes in ashtrays, aluminum cans in a recycling bin, and litter on a campus walkway are examples of use traces
- Classify use traces according to whether the researcher intervenes while collecting data regarding the use of particular items.
Natural Use Traces
*Are observed without any intervention by a researcher and reflect naturally occurring events
Controlled use traces
Result from some intervention by a researcher
Products
The creations, constructions, or other artifacts of behavior.
*The examination of produces allows researchers to test important hypotheses about behavior.
Archival records
- Public and private documents describing the activities of individuals, groups, institutions, and governments.
- Are continuously kept and updated are referred to as running records.
- Share some of the same advantage as physical traces
- Unobtrusive measures that are used to complement hypothesis testing based on other methods
- Plentiful
- Can avoid an extensive data collection stage- data are simply waiting for researchers
- Is often part of the public record and usually does not identify individuals, ethical concerns are less worrisome.
- Bias in how archives are produced.
- Selective deposit
- Selective survival
Episodic records
Personal documents are more likely to describe specific events or episodes
Selective deposit
When some information is selected to be deposited in archives, but other information is not.
- Only certain information is “for the record”
- Problem of reactivity when deciding what is “for the record”, individuals are reacting to the fact that their remarks are being recorded.
Selective survival
Arises when records are missing or incomplete
*Researchers must consider whether some records “survived,” whereas other did not
Documents that are particularly damaging to certain individuals or groups may vanish, during the change from one presidential administration to another.
Spurious relationship
Exists when evidence falsely indicates that two or more variables are associated
- False evidence can arise because of inadequate or improper statistical analyses, or more often, when variables are accidentally or coincidentally related.
- An association, or correlation, between two variables can occur when another, usually unidentified, third variable accounts for the relationship
Narrative records
- When researchers seek a comprehensive record of behavior, they often use narrative records
- Provide a more or less faithful reproduction of behavior as it originally occurred
- An observer can write descriptions of behavior, or use audio or video recordings.
- Once narrative records are created, researchers can study, classify, and organize the records to test their hypotheses about the behaviors under investigation.
- Narrative records differ from other forms of recording and measuring behavior because the classification of behaviors is done after the observations are made.
- Researchers must make sure that the narrative records capture the information that will be needed to evaluate the hypotheses of the study.
- Don not always prevent inferences and impressions by the observer, nor are narrative records always meant to be comprehensive descriptions of behavior.
Field Notes
- Include only the observer’s running descriptions of the participants, events, settings, and behaviors that are of particular interest to the observer, and may not contain an exact record of everything that occurred.
- Are used by journalists, social workers, anthropologists, psychologists, and others, and are probably used more frequently than any other kind of narrative record
- Events and behaviors are likely to be interpreted in terms of the observer’s specialized knowledge and field notes tend to be highly personalized.
- The usefulness of field notes as scientific records depends on the accuracy and precision of their content which, in turn, depend critically on the training of the observer and the extent to which the recorded observations can be verified by independent observers and through other means of investigation.
Checklist
Is often used to record nominal scale measures
- “Checks” whether or not a specific behavior occurred
- Often include space to record observations regarding characteristics of participants, as well as characteristics of the setting and whether other people are present.
- Researchers typically are interested in observing behavior as a function of these participant and context variables.
- Measure the frequency of particular behaviors in an individual or group over a period of time
- The presence or absence of specific behaviors is noted at the time of each observation
- -After all the observations are made, researchers add up the number of times a particular behavior occurred.
- -In these situations, frequency of responding can be assumed to present a ratio level of measurement
- -If “units” of some behavior are counted, then zero represents the absence of that specific behavior.
Data reduction
The process of abstracting and summarizing behavioral data
- In qualitative data analysis, data reduction occurs when researchers verbally summarize information, identify themes, categorize and group pieces of information, and incorporate their own observations about the narrative records.
- Researchers use both inductive and deductive reasoning to gain understanding, always aware of the ways in which their personal background informs the interpretation of the record.
Steps when analyzing narrative records
Identifying a relevant source, sampling selections from the source, and coding units of analysis
Coding
the identification of units of behavior including categories or themes that are related to the goals of the study
Content analysis
Any objective coding technique that allows researchers to make inferences based on specific characteristics of the records.
Relative frequency
- The most common descriptive statistic for the nominal scale
- The number of times a behavior or event occurs is tallied and then divided by the total number of observations
- Measures are expressed as either a proportion or a percentage
Arithmetic mean/Average
The most common measure of central tendency
*Describes the “typical” score in a group of scores and provides a useful measure to summarize the performance of an individual or group
Descriptive statistics
*Are reported when behavior is recorded on at least an interval scale of measurement
Standard deviation
Approximates the average distance of a score from the mean
Correlation Coefficient
Quantitative index of the degree of this covariation
- Tells us how weel the ratings of two observes agree
- Indicates the direction and strength of the relationship
- Direction: positive or negative
- Strength: degree of covariation present
Demand characteristics
Research participants often try to guess what behaviors are expected, and they may use cues and other information to guide their behavior.
Expectancy effects
Lead to systematic errors in observation
*The observer has some expectations about what behavior should be like in a particular situation or following a specific psychological treatment