PSYC 316: FINAL EXAM
146 Cards in this Set
| Front | Back |
|---|---|
|
Theory and What they Seek to Prove
|
Systematic explanation for the observation that relate to a particular aspect of life.
Prove:
- Theories prevent our being taken in by flukes
- Theories makes sense of observed patterns
- Theories shape and direct research efforts
|
|
Variables
|
Logical Groupings of attributes.
|
|
Attributes
|
Characteristics of people or things
(Ex: age --> Young / old )
|
|
IV / DV
|
IV: Variable with values that are not problematical in an analysis, but are taken as simply given.
DV: Variable assumed to depend on or caused by the independent variables.
|
|
Purpose of Social Research
|
Explanatory / Descriptive
|
|
Dialects of Social Research: Idiographic
|
An approach to explanation in which we seek to exhaust the idiosyncratic causes of a particular condition of event.
|
|
Dialects of Social Research: Nomothetic
|
An approach to explanation in which we seek to identify a few causal factors that generally impact a class of conditions of event.
|
|
Qualitative Data
|
Non-numerical data
(ex: descriptions)
|
|
Quantitative Data
|
Numerical data
|
|
Pure Research vs. Applied Research
|
Gaining knowledge for the purpose of knowledge. Vs. Putting research into practice.
|
|
Paradigms
|
A model or frame of reference through which to observe and understand (A broad point of view / patterns happen).
|
|
Macro-Theory
|
Theory aimed at understanding the big picture of institutions, societies and their interaction (Class struggles).
|
|
Micro-Theory
|
Understanding at the intimate level of individuals and their actions (Jury deliberation).
|
|
Meso-Theory
|
Referencing and intermediate level between macro and micro (Social categories).
|
|
Rational Objectivity Reconsidered
|
With the growth of science and technology and the decline of superstition, rationality has more become the center of social life. However studies have shown that people aren't always rational in how they act (Asch experiment of lines / Muzafer Sherif experiment with the spot of light).
…
|
|
Elements of Social Theory
|
Observation (Seeing, hearing, touching) / Fact (A phenomenon that has been observed) / Laws (Universal generalization about classes of fact) / theory (A systematic explanation for observations that relate to a particular aspect of life) / Concepts (Abstract elements representing classes o…
|
|
Axioms and Postulates
|
Fundamental assertions on which a theory is grounded.
|
|
Hypothesis
|
A specified testable expectation about the empirical reality that follows from a general proposition.
|
|
Null (None) Hypothesis
|
Hypothesis testing and tests of statistical significance, the hypothesis that suggests there is no relationship among the variables under the study.
|
|
Deductive Theory Construction
|
Research is used to test a theory:
- Specify the topic
- Specify the range of phenomena your theory addresses
- Identify and specify your major concepts and variables
- Find out what is known about the relationships among those variables
- Reason logically from those propositions to…
|
|
Inductive Theory Construction
|
Theories are developed from analysis of data:
- Observe aspects of social life and seek to discover patterns that may point to relatively universal principles: Ground theory / field research.
|
|
Purpose of Research: Explanation
|
- Idiographic: The goal is to find an exhaustive understanding of the causes producing events and situations in a single or limited number of cases.
- Nomothetic: To find a few factors that can account for many of the variations in a given phenomenon.
|
|
Purpose of Research: Criteria for Nomothetic Causality
|
- The variables must be correlated (Correlation: empirical relationship between 2 variables such that changes in one are associated with changes in the other)
- The cause takes place before the effect
- The variables are non-spurious (Spurious relationship: A coincidental statistical co…
|
|
Purpose of Research: Nomothetic Causality and Hypothesis Testing
|
- To test a hypothesis: specify variables / measurement of variables / hypothesize correlation, strength of relationship, statistical significance / specify tests for spuriousness
- False criteria for nomothetic causality: complete causation / exceptional cases / majority cases
|
|
Spurious Relationship
|
A coincidental statistical correlation between 2 variables shown to be caused by some third variable. Ex: Shoe size and math skills
|
|
Necessary Cause
|
Represents a condition that must be present for the effect to follow.
|
|
Sufficient Cause
|
Represents a condition that, if present, guarantees the effect in question.
|
|
Units of Analysis
|
Individuals (Most common in social science) / groups (Families) / organizations ( colleges / corporations) / social interactions (Chat rooms) / social artifacts (Books / songs)
|
|
Ecological Fallacy
|
Erroneously drawing conclusions about individuals solely from the observation of groups. Ex: Car accidents and income
|
|
Reductionism
|
A strict limitation of the kind of concepts to be considered relevant to the phenomenon under study. (Ex: Sociobiology)
|
|
Cross Sectional Study
|
- Study based on observations representing a single point in time, a cross section of population.
|
|
Longitudinal Study: Trend / Cohort / Panel
|
A study design involving the collection of data at different point in time:
- Trend studies: A study in which a given characteristic of some population is monitored over time.
- Cohort study: A study in which some specific subpopulation is studied over time.
- Panel study: A study in w…
|
|
Conceptualization: Real / Nominal / Operational
|
The process through which we specify what we mean when we use particular terms in research. We can't meaningfully answer questions without a working agreement about the meaning of the outcome. Conceptualization produces specific, agreed upon meaning for a concept for the purposes of resea…
|
|
Conceptualization
|
The mental process whereby fuzzy and imprecise notions are made specific and precise.
|
|
Concepts
|
Constructs derived by mutual agreement from mental images. Conceptions summarize collections of seemingly related observations and experiences.
|
|
Levels of Measurements
|
- Nominal: Variables whose attributes are merely different; they have only the characteristics of exhaustiveness and mutual exclusiveness (Gender / Religions / Majors)
- Ordinal: Variables with attributes we can logically rank in order (Socioeconomic status / Conservativeness)
- Interv…
|
|
Criteria of Measurment Quality
|
- Precision and Accuracy: Precision in the level of detail.
- Reliability : Quality of measurement method that suggest the same data would have been collected each time in repeated observations of the same phenomenon.
- Validity: Term describing a measure that accurately reflects the co…
|
|
Types or reliability: Test Retest
|
- Problems with test retest reliability procedures: differences in performance on the 2nd test may be due to the first test.
- Many constructs of interest change over time independent of the stability of the measure.
- The interval between the administration of the tests may be too long…
|
|
Types or reliability: Parallel Form
|
Used to assess the consistency of the results of two test constructed in the same way from the same content domain.
|
|
Types or reliability: Inter Rater
|
Multiple sets of randomly assigned variables should produce the same classifications when viewed by different observers.
|
|
Types or reliability: Split Half Method
|
In split-half reliability we randomly divide all items that purport to measure the same construct into two sets. We administer the entire instrument to a sample of people and calculate the total score for each randomly divided half.
|
|
Internal Consistency
|
- Average Inter item correlation
- Split Half reliability
- Cronbach's Alpha
|
|
Validity in Measurements: Face Validity
|
Quality of an indicator that makes it a reasonable measure of some variable.
|
|
Validity in Measurements: Criterion Related
|
The degree to which a measure relates to some external criterion:
- Concurrent: requires that the scale be judged against some other method that is acknowledged as the gold standard.
- Predictive: ability of a scale to predict future events and behaviors.
|
|
Validity in Measurements: Construct
|
The degree to which a measure relates to other variables as expected within a system of theoretical relationships:
- Indicates how well the scale measures the construct it was designated to measure
|
|
Validity in Measurements: Content
|
The degree to which a measure covers the range of meanings included within a concept.
|
|
Validity and Reliability
|
- Not Valid and not Reliable
- Reliable but not Valid
- Valid but not Reliable
- Valid and Reliable
|
|
Probability Sampling Theory
|
- The general term for samples selected in accord to probability theory.
- Often used for large scale surveys/
- If all members of a population were identical in all respects, there would be no need for careful sampling procedures. However, that is rarely the case.
- A sample of indivi…
|
|
Random Selection
|
Each element has an equal chance of selection independent of any other event in the selection process.
|
|
Advantages of Probability Sampling
|
- Probability samples are typically more representative than other types of samples because bias is avoided
- Probability theory permits researchers to estimate the accuracy or representatives of the sample
|
|
Simple Random Sampling
|
Type of probability sampling in which the units composing the population are assigned numbers. A set of random numbers is generated and the units having those numbers are included in the sample.
|
|
Systematic Sampling
|
- Type of probability sampling in which every K th Unit in a list is selected for in inclusion in the sample (Periodicity in a list to be sampled can introduce bias in the selection)
- Sampling interval: Standard distance between elements selected from a population in the sample.
- Sam…
|
|
Non Probability Sampling
|
Technique in which samples are selected in some way not suggested by probability theory.
|
|
Non Probability Sampling: Snowballing
|
When each person interviewed may be asked to suggest additional people for interviewing.
|
|
Non Probability Sampling: Quota Sampling
|
Units are selected into a sample on the basis of pre-specified characteristics, so that the total sample will have the same distribution of characteristic assumed to exist in the population being studied.
|
|
Stratified Sampling
|
- Stratification: grouping of units composing a population into homogenous groups (Strata) before sampling
- Slightly more accurate than simple random sampling
- Stratification is a modification to simple random and systematic sample methods
- It helps reduce error
- Implicit stratifi…
|
|
Weighting
|
Assigning different weights to cases that were selected into a sample with different probabilities of selection (Ex: of 30%male and 70%females).
|
|
Correlations
|
Correlations are the most common way to describe association, yet they are not fully understood.
|
|
Correlations: Issues
|
It can be misleading (Example of high IQ and Shoes).
|
|
Invariant to Scale Change
|
- Any linear scale transformation, including multiplying and dividing by a constant will not affect the correlation
- Z-scoring eliminates differences caused by nonlinear transformations
- Z-Score = (Score - Mean) / Standard Deviation
- Computing the correlation coefficient from z-scor…
|
|
Strength of Correlation
|
Small (0.1) Moderate (0.3) Large (0.5)
Proportion of variance:
- When all of the variability in Y is accounted for by X then r=1
- When only some of the variability is accounted for by X then r<1
- When none of the variability in Y is accounted for by X then r=0
|
|
Correlations: Restricted Range
|
Restricting the true range of a variable will have the effect of reducing the correlation coefficient.
|
|
Correlations: Impact of Outliers
|
Outliers have a huge impact on correlation.
|
|
Internal Validity ( 9 threats)
|
It is based on whether or not the association between X and Y is a causal one in the form in which they were measured or manipulated.
|
|
Threat to Internal Validity: Ambiguous Temporal Precedence
|
Lack of clarity about which variable occurred first may yield confusion about which variable is the cause and which is the effect (True experiments control for this).
|
|
Threat to Internal Validity: Selection
|
Systematic differences over conditions in respondent characteristics that could also cause the observed effect. People in different groups are different at the start of an experiment, therefore treatment effects can be attributed to the initial group differences (If kids are in treatment …
|
|
Threat to Internal Validity: History
|
Events occurring concurrently with treatment could cause the observed effect:
- This is controlled in laboratory experiments and can be reduced by picking groups from similar areas so history does not affect just one group
|
|
Threat to Internal Validity: Maturation
|
Naturally occurring changes over time could be confused with a treatment effect.
|
|
Threat to Internal Validity: Regression
|
When units are selected for their extreme scores, they will often have less extreme scores on other variables, an occurrence that can be confused with a treatment effect:
- If researcher requires, select a large group with extreme scores, then randomize into treatments
- Study people ov…
|
|
Threat to Internal Validity: Attrition
|
Loss of respondent to treatment can produce artificial effects if that loss is systematically correlated with conditions.
|
|
Threat to Internal Validity: Testing
|
Exposure to a test can effect scores on subsequent exposures to that test, an occurrence that can be confused with a treatment effect.
|
|
Threat to Internal Validity: Instrumentation
|
The nature of a measure may change over time or conditions in a way that could be confused with a treatment effect.
|
|
Threat to Internal Validity: Additive and Interactive effects of threats
|
The impact of a treat can be added to that of another or may depend on the level of another threat.
|
|
Statistical Conclusion Validity (9 threats)
|
Based on 2 points:
- Whether or not the presumed cause and effect covary
- How strongly they covary
|
|
Threats to Statistical Conclusion Validity: Low Statistical Power
|
An insufficiently powered experiment may incorrectly conclude the relationship between treatment and outcome is not significant
- Statistical power: The ability of a test to detect true differences in the population. It is the probability that a test will reject the null hypothesis when…
|
|
Threats to Statistical Conclusion Validity: Violated Assumptions of Stat Tests
|
Can lead to either overestimating or underestimating the size and significance of an effect:
- All statistical test have assumptions about the underlying population
- One is that scores are distributed in a normal distribution but many tests are robust to violations of the assumption (T…
|
|
Threats to Statistical Conclusion Validity: Fishing and the Error Rate Problem
|
Repeated tests for significant relationships, if uncorrected for the number of tests, can artificially inflate statistical significance:
- We use conventions for rejecting the null hypothesis by saying that if something happened by chance only 5 in a 100 times, we'd be willing to say it …
|
|
Threats to Statistical Conclusion Validity: Unreliability of Measures
|
Measurement error weakens the relationship between 2 variables and strengthens or weakens the relationship of 3 or more variables.
|
|
Threats to Statistical Conclusion Validity: Restriction of Range
|
Reduced range on a variable usually weakens the relationship between it and another variable (Variables too similar / Floor or Ceiling effect).
|
|
Threats to Statistical Conclusion Validity: Unreliability of Treatment Implementation
|
If a treatment that is intended to be implemented in a standardized manner is implemented only partially for some respondents, effects may be underestimated compared with full implementation (happens in observational studies).
|
|
Threats to Statistical Conclusion Validity: Extraneous Variance in Experimental Settings
|
Changes in the Experimental studies.
|
|
Threats to Statistical Conclusion Validity: Heterogeneity of Units
|
Increased variability on the outcome variable within conditions increases error variance, making detection of a relationship more difficult.
|
|
Threats to Statistical Conclusion Validity: Inaccurate Effect Size Estimation
|
Some statistics systematically overestimate or underestimate the size of an effect.
|
|
Construct Validity ( 13 threats)
|
- Scientists study specific instance of units, treatments, observations, and settings
- Construct validity is making inference to the higher order concepts that the specific are used to address
|
|
Threats to Construct Validity: Inadequate Explication of Construct
|
Failure to adequately explicate a construct may lead to incorrect inferences about the relationship between operation and construct:
- General mistakes: Too general / Too specific / Wrong / Describing 2 constructs as one
|
|
Threats to Construct Validity: Construct Confounding
|
Operations usually involve more than one construct, and failure to describe all the constructs may result in incomplete construct inferences.
|
|
Threats to Construct Validity: Mono-Operation Bias
|
Any one operationalization of a construct both underrepresented the construct of interest and measures irrelevant constructs, complicating inference.
|
|
Threats to Construct Validity: Mono Method Bias
|
When all operationalization use the same method, that method is part of the construct actually studied.
|
|
Threats to Construct Validity: Confounding Constructs with Levels of Constructs
|
Inference about the constructs that best represent study operations may fail to describe the limited levels of the construct that were actually studied (If only a very low level "strength" of the treatment were used, then the experiment might conclude that the treatment was ineffective, w…
|
|
Threats to Construct Validity: Reactive Self-Report Changes
|
Changes in Self Reported Answer.
|
|
Threats to Construct Validity: Reactivity to Experimental Situation
|
Perception to the experimental situation and those perceptions are part of the treatment construct actually tested (Participants want to obey or please the researcher).
|
|
Threats to Construct Validity: Experimental Expectancies
|
They want to please the experimenter.
|
|
Threats to Construct Validity: Novelty and Disruption Effect
|
New Situations
|
|
Threats to Construct Validity: Compensatory Equalization
|
Both Groups are compensated equally.
|
|
Threats to Construct Validity: Resentful Demoralization
|
Giving false results because you know you are in the control group.
|
|
Threats to Construct Validity: Treatment Diffusion
|
Participants may receive services from a condition to which they were not assigned, making construct descriptions of both conditions more difficult.
|
|
Threats to Construct Validity: compensatory Rivalry
|
Rivalry between groups.
|
|
External Validity (5 threats)
|
How a study would hold over variations in persons, settings, treatments, and outcomes.
|
|
Threats to External Validity: Interaction of the Causal Relationship with Units
|
An effect with certain kinds of units might not hold if other kinds of unites had been studied.
|
|
Threats to External Validity: Interaction of the Causal Relationship over Treatment Variations
|
An effect found with one treatment variation might not hold with other variations of the treatment, or when that treatment is combined with other treatments, or when only part of the treatment is used.
|
|
Threats to External Validity: Interaction of the Causal Relationship With Outcome
|
An effect found on one kind of outcome observation may not hold if the outcome observation were used.
|
|
Threats to External Validity: Interaction of the Causal Relationship with Settings
|
An effect found in one kind of setting may not hold if other kinds of settings were to be used (Urban versus rural / high income neighborhood versus low income neighborhood).
|
|
Threats to External Validity: Context Dependent Mediation
|
An explanatory mediator of a causal relationship in one context may not mediate in another context.
|
|
Classical Experiments
|
Experiments involve: Taking action / observing consequences.
In experiments, social researchers typically select a group of subjects, do something to them, and observe the effect of what was done.
|
|
PreTesting and PostTesting
|
- Pre-testing: The measurement of a DV along subjects
- Prost-testing: The measurement of a DV among subjects after they have been exposed to an IV
|
|
Double Blind Experiment
|
Experimental design in which neither the subjects nor the experimenters know which is the experimental and which is the control group.
|
|
Experimental vs. Control Groups
|
- Experimental group: subjects to whom an experimental stimulus is administered
- Control group: A group of subjects to whom no experimental stimulus is administered and who should resemble the experimental group in all other respects.
|
|
Self Report Questionnaires
|
A document containing questions and other types of items designed to solicit information appropriate for analysis.
- Open ended questions: questions for which the respondent is asked to provide his/her own answers.
- Close ended questions: survey questions in which the respondent is ask…
|
|
Self-Administered Questionnaire
|
Questionnaires in which respondents are asked to complete the questionnaire by themselves
- Mail distribution and returns: Monitor returns / Follow up mailing
- Response rates: the number of people participating in a survey divided by the number selected in the sample. (Ideal is higher …
|
|
Interview Survey
|
Interview: a data-collection encounter in which one person (interviewer) asks questions of another (respondent).
Guidelines for survey interviewing:
- Appearance and demeanor
- Familiarity with the questionnaire
- Following question wording exactly
- Recording responses exactly
- Pr…
|
|
Telephone and Online Survey
|
Telephone surveys
- Advantages: 95.5% of homes have a phone / it saves time and money
- Disadvantages: Unlisted phone numbers / cell phones / people hang up
- Random Digit Dialing: a sampling technique in which random numbers are selected from within the range of numbers assigned to ac…
|
|
Advantages of Survey Methods
|
- Useful in describing large populations
- Surveys are flexible
- Standardized questions
|
|
Disadvantages of Survey Methods
|
- Round pegs in square holes
- Seldom deal with context of social life
- Inflexible
- Artificial
- Weak on validity
|
|
Field Research: Role of the Observer
|
- Participant / researcher / observer
Degree of participation:
- Reactivity: the problem of social research subjects potentially reacting to being studied, thus altering their behavior from what it would have normally been
- Problem: How much to participate? Is it under your control?
…
|
|
Reflexivity
|
Setting could influence knowledge gained / Difficulty interviewing certain folks and what that does to the interaction with the interviewer (Female to Male / Primitive societies).
|
|
Field Research: Grounded Theory (Example of Inductive Research)
|
An inductive approach to the study of social life that attempts to generate a theory from the constant comparing of unfolding observations.
- Think conservatively
- Obtain multiple viewpoints
- Periodically step back
- Maintain an attitude of skepticism
- Follow the research procedur…
|
|
Case Studies
|
The in-depth examination of a single instance of some social phenomenon.
|
|
Extended Case Studies
|
A technique in which case study observations are used to discover flaws in and to improve existing social theories.
|
|
Naturalism
|
An approach to field research based on the assumption that an objective social reality exists and can be observed and reported accurately.
|
|
Ethnography / Institutional Ethnography
|
- A report on social life that focuses on detailed and accurate descriptions rather than explanations (People describe world not as is, but to make sense).
- A research technique in which the personal experiences of individuals are used to reveal power relationships and other characteris…
|
|
Ethnomethodology
|
An approach to the study of social life that focuses on the discovery of implicit, usually unspoken assumptions and agreement. (Breaching experiments).
|
|
Emancipatory Research
|
Research conducted for the purpose of benefiting disadvantaged groups.
|
|
Qualitative Interview
|
Contrasted with survey interviewing, the qualitative interview is based on a set of topics to be discussed in depth rather than based on the use of standardized questions (Miner or traveler).
|
|
Focus Groups
|
A group of subjects interviewed together, prompting a discussion.
- Advantages: real-life data, flexible, high degree of face validity, fast, inexpensive
- Disadvantages: not representative, little interviewer control, difficult analysis, interviewer/moderator skills, difficult logistic…
|
|
Unobtrusive Research
|
Methods of studying social behavior without affecting it (Durkheim and suicide).
- Content Analysis: the study of recorded human communications (i.e., books, websites, paintings, laws).
- Analysis of Existing Statistics.
- Comparative and Historical Analysis.
|
|
Evaluation Research
|
Research undertaken for the purpose of determining the impact of some social intervention, such as a program aimed at solving a social problem.
- It is a form of applied research.
- Needs assessment studies - studies that aim to determine the existence and extent of problems, typically …
|
|
Quasi-Experimental Designs
|
Quasi: Having a likeness to something; resembling.
First reaction to the idea of a design without a control group or without a pretest may be that it is useless - to fraught with threats to internal validity. The point is to evaluate the threats carefully and then determine if the threat…
|
|
Designs without Controls or Pretests
|
- The one-group Posttest only design
- The one-group posttest only design with multiple substantive posttests
- The one-group Pretest posttest design
- The one group pretest / posttest design using a double pretest
- The one group pretest posttest design using a nonequivalent dependen…
|
|
The one-group Posttest only design
X O1
|
All threats to internal validity apply
|
|
The one-group posttest only design with multiple substantive posttests
X1 (O1a, O1b, ... O1n)
|
The trick here is to use Pattern Matching as a tool by collecting multiple outcomes. But note that this is the reverse of typical studies: Cause is known, not effect.
Worries about Type I errors (saying its there when it isn't) and human's tendency to "see" patterns - whether there or n…
|
|
The one-group Pretest/ Posttest design
O1 X O2
|
Threats to internal validity: History / Maturation. Alternative is that some people get O1 and others get O2.
|
|
The one group pretest / posttest design using a double pretest
O1 O2 X O3
|
- Can help with the assessment of Regression to the Mean, because you'd see this trend prior to the intervention
- Can help with assessing Maturation
|
|
The one group pretest posttest design using a nonequivalent dependent variable
(O1a, O1b) X (O2a, O2b)
|
OA and OB assess similar constructs, but only OA is supposed to change because of X
|
|
The Removed Treatment Design
O1 X O2 O3 no X O4
|
Here there is an opposite prediction depending upon the addition or subtraction of X around its respective Os: the is pattern prediction
- Only makes sense if treatment effects are short-lived (not permanent), else there could be no change after the first X
- Caution: Removing X can be …
|
|
The repeated Treatment Design
O1 X O2 O3 X O1
|
- Internal: Cyclical Maturation, as when O1 and O3 are measured on Tuesday and O2 and O4 on Saturday: this could cause the pattern observed
- Internal: Unique Historical Events corresponding with removal of treatments
- Construct: If participants notice the change in Xs, they could reac…
|
|
Designs With Control Group but no Pretest
|
- Posttest only design with Nonequivalent groups
-Posttest only design using an independent pretest sample
- Posttest only design using proxy pretest
|
|
Posttest only design with Nonequivalent groups
|
- Perhaps researcher not consulted before upper half of study was run, so adds bottom half later to improve the design
- Pretest differences in group (Selection) is a major threat, but better than nothing
- Having Pretest data for Nonequivalent groups is very important
|
|
Posttest only design using an independent pretest sample
|
Now adding a Pretest for both groups to the design, but from independent groups
- Design is used by epidemiologists, public health, marketing, and polling researchers
- Threats:
The two samples must be from same population, else Selection
If groups are run at different times, then oth…
|
|
Posttest only design using proxy pretest
|
Subscript A represents the proxy for subscript B, where A is a measure related to B, but is not the same measure
|
|
Case Control Design
|
Useful when something unusual happens to people and you are trying to figure out the cause of whatever it is
Those who are affected are "Cases"; those whom you select to match to the Cases are the "Controls"
- Useful for rare events
- Useful for Xs that can't be ethically administered
…
|
|
Designs that use Controls and Pretest
(6 types)
|
- Control groups are of minimal help unless Pretests are also available
- Pretests tell us about how the groups were prior to the treatment (X), and that knowledge is hugely important for making inferences
- Pretests also help with statistical analyses
Types of designs
Types:
- Untr…
|
|
Untreated Control group design with dependent pretest and posttest samples using switching replications
|
- Tx is removed in first group, and added to second group
- Between O1 and O2, second group is a control for first group
- Between O2 and O3, first group in a modified control for the second group
|
|
Untreated Control group design with dependent pretest and posttest samples using reversed treatment control group
|
- X+ is supposed to improve performance
- X- is supposed to decrease performance
- Ex: In an office environment, making the first groups' job tasks more challenging and complex, whereas making the second groups' task simpler
- Obtaining Pretest on performance and then a Posttest
- Cle…
|
|
Summary of Design Elements that may be Available
|
Assignment:
- Random Assignment
- Cutoff-Based Assignment
Measurement:
- Posttest observations: Single / Nonequivalent DV/ Multiple substantive posttests
-Pretest observations: Single Pretest / Proxy pretest / repeated pretests over time / pretests on independent samples
Comparison…
|
|
Interrupted Time Series Design
|
- One of the most effective quasi-experimental design, especially when combined with features from the other designs we have discussed
- Time-series: a large number of observations on the same variables consecutively over time
- Interrupted: A special time-series with an event that occu…
|
|
Simple Interrupted Time-Series Design
|
Very few rival hypotheses
- Regression to mean no plausible because of the very long series before the intervention
- Selection not plausible unless there had been a big change in population making calls in that city
- Attrition seems unlikely given the big drop
- Testing not likely b…
|
PSYC 316: EXAM 3