Term
Units of analysis
What are we studying? |
|
Definition
Individuals
Groups
Organizations
Social Artifacts |
|
|
Term
|
Definition
| police, gang members, students, etc. |
|
|
Term
|
Definition
| multiple persons with the same characteristics |
|
|
Term
|
Definition
| formal groups with established leaders and rules |
|
|
Term
|
Definition
| Products of social beings and their behavior. |
|
|
Term
|
Definition
Drawing conclusions about groups from individual level data
- cause of stereotypes |
|
|
Term
|
Definition
Drawing conclusions about individuals from group level data
- effect of stereotypes |
|
|
Term
| Examples of hasty generalization |
|
Definition
joe a canadian is against gun control
- therefore all canadians must be against gun control |
|
|
Term
| examples of ecological fallacy |
|
Definition
presidential elections of 2000, 2004, and 2008
- wealthier states tended to vote democratic and poorer states tended to vote republican
-- so from this data, people might think that wealthier voters tend to vote democratic and poorer voters tend to vote republican
BUT |
|
|
Term
| what causes juvenile delinquency? |
|
Definition
deterministic characteristics
- lack of parental supervision
- peer group association
- early childhood experiences
- amount/kind of education
free will aspects
- why didn't you decide to slack off in school? |
|
|
Term
| Probabilistic Causal Model |
|
Definition
Necessary cause
- a condition that MUST be present for the effect to occur
-- being charged before being convicted
Sufficient cause
- a condition that, if it is present, will pretty much guarantee that the effect will occur
-- pleading guilty before being convicted |
|
|
Term
|
Definition
In order for it to be true that "john is a bachelor" it is necessary that it be also true that he is
- male
- adult
- unmarried
to state "john is a bachelor" implies john has each of those three additional predicates |
|
|
Term
|
Definition
- stating that "john is a bachelor" implies that john is male
- knowing that john is a bachelor is sufficient to know that he is male |
|
|
Term
|
Definition
being at least 30 years old is necessary of serving in the U.S. Senate. If you are under 30 years old then it is impossible for you to be a senator.
-- therefore, if you are a senator, you are at least 30 years old |
|
|
Term
|
Definition
| A U.S. president's signing a bill that congress passed is sufficient to make the bill law. note that the case whereby the president did not sign the bill e.g. - through exercising a presidential veto, does not mean that the bill has not become law (it could still have become law through a congressional override). |
|
|
Term
Studying Human Behavior
- Criteria for causality |
|
Definition
- empirical relationship between variables
- temporal order
- no alternative explanations |
|
|
Term
| Statistical Conclusion Validity |
|
Definition
hypothesis testing
is there a statistical relationship between the change in the suspected cause and the change in the suspected effect?
-- sample size is important! |
|
|
Term
|
Definition
Cross-Sectional Design
- studying sample of subjects at one point in time
- collecting data once
Longitudinal Design
- studying subjects over a long period of time
- repeated measures |
|
|
Term
|
Definition
Trend (ex - UCR)
- study changes within some general population over time
Cohort (ex - wolfgang)
- examine more specific populations as they change over time
Panel (ex - NCVS)
- similar to trend, but the same set of people is interviewed on two or more occasions |
|
|
Term
| Approximating Longitudinal Studies |
|
Definition
Retrospective research
- asks people to recall their past for the purpose of approximating observations over time
Prospective Research
- Follows subjects forward in time |
|
|
Term
Longitudinal Designs
Advantages & Disadvantages |
|
Definition
Advantages
- more representative of true phenomenon
- convergence of methodology
Disadvantages
- time
- resource demands
- multiple threats to internal validity |
|
|
Term
| No Alternative Explanations |
|
Definition
Internal Validity
the extent to which your study actually measures what it says it is measuring
- does x actually cause y?
- is there any reason why you may be able to conclude that something else (Z)? caused Y? |
|
|
Term
| Three Classes that Scientists Measure |
|
Definition
Direct Observables
- those things or qualities we can observe directly
Indirect Observables
- require relatively more subtle, complex, or indirect observations for things that cannot be observed directly
Constructs
- theoretical creations, cannot be observed directly or indirectly |
|
|
Term
Progression of Measurement
(CCOM) |
|
Definition
Conceptualization
Conceptual Definition
Operational Definition
Measurements in the real world |
|
|
Term
| Exhaustive and Exclusive Measurement |
|
Definition
Exhaustive
you should be able to classify every observation in terms of one of the attributes composing the variable
Mutually Exclusive
you must be able to classify every observation in terms of one and only one attribute |
|
|
Term
|
Definition
what is your current age
20-24
25-29
30-34 |
|
|
Term
| Not Mutually Exclusive example |
|
Definition
What is your current age?
10 or less
10 to 20
20 to 30
30 to 40
40 to 50
50 or greater |
|
|
Term
| Both Exhaustive and Mutually Exclusive example |
|
Definition
what is your current age?
less than 18
18-29
30-39
40-49
50 or older |
|
|
Term
|
Definition
Reliability
- assures that the scale can consistently measure something
Validity
- assures that the scale can measure what it is intended to measure |
|
|
Term
|
Definition
observed score
- true score
-- perfect measure with no internal or external influences
- measurement error
-- slop
ex - child getting 85% on a spelling test |
|
|
Term
| Catergories of Measurement Error |
|
Definition
mistakes
stable attributes
situational factors
transient states
test characteristics |
|
|
Term
|
Definition
X = T + e
X - observed score
T - true ability
e - error |
|
|
Term
| Methods of accessing reliability |
|
Definition
Test-retest method
Inter-rater method
Split-half method
- the method you use depends on the type of measure you are using |
|
|
Term
|
Definition
compare measurements from different observers who are rating the same persons, events, or places.
# of agreements over total # of ratings |
|
|
Term
| increasing inter-rater reliability |
|
Definition
standardize procedures
train observers
use good operational definitions |
|
|
Term
|
Definition
when researchers measure a phenomenon that does not change between two points separated by an interval of time.
- how similar are the scores? |
|
|
Term
| increasing test-retest reliability |
|
Definition
use good operational definitions
clarify items
pre-test potential test items
ensure that your items are valid
- important to have content validity |
|
|
Term
|
Definition
make more than one measure of any concept
see if each measures the concept differently |
|
|
Term
| increasing split-half reliability |
|
Definition
use good operational definitions
clarify items
pre-test potential test items
preset questions in a random order and test for split-half reliability
- if you have the capability to do so
ensure that your items are valid
- important to have content validity |
|
|
Term
| increasing inter-rater reliability |
|
Definition
standardize procedures
clarify items
train observers
use good operational definitions |
|
|
Term
| relationship between reliability and validity |
|
Definition
reliability does not mean validity
validity infers reliability |
|
|
Term
|
Definition
does it appear to measure what it is supposed to measure?
sometimes we don't want our measures to have face validity - why?
because sometimes people will answer questions differently trying to conceal/express something |
|
|
Term
|
Definition
does the measure cover the range of meanings included in the concept?
measures the extent to which a test represents all possible items |
|
|
Term
| criterion-related validity |
|
Definition
compares a measure to some external criterion
- concurrent
correlates with current outcomes
- predictive
correlates with future outcomes |
|
|
Term
|
Definition
whether your variables are related to ch other in the logically expected direction
- convergent validity
our measure should correlate with measures of similar constructs
- discriminant validity
our measure shouldn't correlate with measures of unrelated constructs |
|
|
Term
|
Definition
offender
victim
offense
- an individual act of burglary, auto theft, bank robber, etc.
incident
- one or more offenses committed by the same offender or group of offenders acting in concert, at the same time and place |
|
|
Term
|
Definition
monitoring
agency accountability
research |
|
|
Term
| theoretical relationship between crimes committed and official statistics |
|
Definition
crimes undiscovered
crimes discovered
crimes reported
crimes recorded |
|
|
Term
|
Definition
type 1 offenses - index crimes
murder and non negligent manslaughter
forcible rape
burglary
larceny-theft
robbery
aggravated assault
motor vehicle theft
arson
type 2 offenses - non index crimes
22 other crimes |
|
|
Term
uniform crime reports
positives |
|
Definition
can compare agencies
quick, easy, efficient
index offenses are valid indicators of public's crime concerns |
|
|
Term
uniform crime reports
negatives |
|
Definition
doesn't count ALL crimes
variety of completeness of data both across location and across time
variety of definition of crime across time
creative work with crime statistics
hierarchy rule
crime rate is unweighted
victim characteristics are deemphasized
demographic shifts can skew data |
|
|
Term
| national incident-based reporting system |
|
Definition
incident-based instead of summary based reports
expanded offense reporting
new offense definitions
elimination of hierarchy rule
greater specificity of data
crimes against society
attempted versus completed crimes
designation of computer crime
quality control |
|
|
Term
national crime victimization survery
positives |
|
Definition
Model for spurring international imitation
Opportunity to obtain a picture of victims and their characteristics
More accurate estimate of certain crimes like rape and assault
Assesses issues such as fear of crime, satisfaction with police services, attitude toward the police, and reasons for not reporting crimes to the police. |
|
|
Term
National crime victimization survery
negatives |
|
Definition
cost of large samples
false reporting
misteken interpretation of incidents
poor memory
telescoping
over reporting and underreporting
sampling bias
human mistakes in coding and mechanical errors |
|
|
Term
|
Definition
expanded list of questions
computer-assisted telephone interviewing technology
altered the scope of crimes measured
revised screening questions
rephrasing of many questions
increase the threshold for series victimizations |
|
|
Term
| national survey on drug use and health |
|
Definition
based on a nationwide sample of households
- 12 year olds and above
questions regarding use of illegal drugs, alcohol, and tobacco
distinguishes between lifetime, current and heavy use
measurement error?
- do people tell the truth
- non-sampled population |
|
|
Term
|
Definition
differences from others
- targets a specific population
- asks a broader variety of questions
- subset receives a follow up questionnaire (when college students) |
|
|
Term
| drug serveillance systems |
|
Definition
Arrestee drug abuse monitoring (ADAM)
- provides ongoing assessment of drug use among arrestees
Drug abuse warning network (DAWN)
collects emergency medical treatment reports for "drug episodes" from a sample of hospitals |
|
|
Term
| measuring crime for specific purposes |
|
Definition
local crime and self-report surveys
incident-based crime reports
observing crime |
|
|
Term
|
Definition
establishing existence of cause-effect relationship
- systematically manipulating one or more variables
exerting control
rule out alternative explanations |
|
|
Term
three important factors in experimental research
o o
x
o o |
|
Definition
independent or dependent variables
pretesting and post testing
experimental and control groups
O stands for observation
X stands for manipulation/variable of interest |
|
|
Term
experiemental research
advantages/disadvantages |
|
Definition
advantages
can establish causal relationships
- if extraneous variable are controlled
disadvantages
cant manipulate some variables
reduced generalizability |
|
|
Term
|
Definition
between-subjects designs
- different participants are exposed to different conditions
within-subjects designs
- same participants are exposed to different conditions
mixed designs
at least one IV is manipulated between-subjects and at least one IV is manipulated within-subjects |
|
|
Term
XO
researcher wants to study effect of a reading program on reading achievement |
|
Definition
one-shot case study
(x is the manipulated/variable of interest)
(o stands for observation)
she implements the program at beginning of year and measures achievement at end of year |
|
|
Term
do not know whether the students reading skills actually changed from the start to end of the school year
(improve by giving a pretest at the start of the study)
O X O |
|
Definition
also known as: one-group pretest-posttest
scores could be influenced by other instruction in school, the students' maturation, or the treatment |
|
|
Term
our researcher may wish to have a comparison group
OXO
O O |
|
Definition
also known as: static-group pretest-protest
end of year reading scores still could be influenced by other instruction in school, the students' maturation, or the treatment |
|
|
Term
|
Definition
static-group comparison
- if our researcher believes that the pretest has an impact on the results of the study, she might not include it. |
|
|
Term
random assignment to groups should spread the variety of extraneous characteristics that subjects possess equally across both groups
R XO
R O |
|
Definition
randomized posttest-only, control group
because our researcher did not pretest, she might wish to randomly assign subjects to treatment and control group |
|
|
Term
|
Definition
randomized pretest-posttest control group
of course, our researcher could include a pretest with her random assignment |
|
|
Term
|
Definition
randomized solomon four-group
occasionally researchers combine randomized pretest-posttest control group design with the randomized posttest-only control group design |
|
|
Term
one of the pretest groups is assigned to treatment and one of the non-pretest groups is assigned to treatment
R OXO
R O O
R XO
R O |
|
Definition
with the randomized solomon four-group design, all groups are randomly assigned and given the posttest.
two of the groups are given pretests |
|
|
Term
|
Definition
experimental participants are distributed into various groups on a random (nonsystematic) basis
all participants have an equal change of being assigned to any of the treatment groups
- drawback? cannot guarantee equality if only have small numbers |
|
|
Term
| threats to internal validity |
|
Definition
experimenter expectancy effects
subject reactivity
history
maturation
instrumentation
statistical regression
selection bias
experimental mortality
causal time order
diffusion/imitation of treatment
compensatory rivalry
demoralization |
|
|
Term
threats to internal validity
- experimenter expectancy effects |
|
Definition
experimenter influences DV
- expectancies
controls
- double blind testing
- automation |
|
|
Term
threats to internal validity
subject reactivity
controls |
|
Definition
subject reactivity:
subjects behavior a function of both the independent variable and something else
- demand characteristics: participants believe they know the hypothesis or how they are supposed to behave in the course of the study
- hawthorne effect: changes in behavior given knowledge that you are being observed
controls:
double blind testing
concealment/deception
habituation
use involving research designs |
|
|
Term
threats to internal validity
history effects
maturation effects |
|
Definition
history effects:
changes over time resulting from events that occur during the course of the experiment
control: "control" group
maturation effects:
changes over time resulting from the aging process
control: reduce time between measurements |
|
|
Term
threats to internal validity
instrumentation effects
statistical regression to the mean |
|
Definition
instrumentation effects
changes in the measurement process
control: reliability checks
statistical regression to the mean
extreme scores tend to regress toward the mean on subsequent testings
control: control group / large sample size |
|
|
Term
threats to internal validity
selection effect/group composition bias
experimental mortality |
|
Definition
selection effect/group composition bias:
the way in which subjects are chosen
control: random assignment to conditions
experimental mortality
differential loss of subjects across conditions
control: replace and report / pretest and compare dropouts to continued participants on earlier measures |
|
|
Term
threats to internal validity
causal time order
diffusion/imiation of treatment |
|
Definition
causal time order
ambiguity about order of stimulus and DV
control: longitudinal design
diffusion/imitation of treatment
when experimental and control groups communicate, experimental group may pass on elements to control group
control: separated groups with matched-design |
|
|
Term
threats to internal validity
compensatory treatment
compensatory rivalry |
|
Definition
compensatory treatment
control group is deprived of something to be of value
control: compensate control group afterwards, double blind testing
compensatory rivalry
control group is deprived of the stimulus may try to compensate by working harder
control: keep experimental and control groups physically separated |
|
|
Term
threats to internal validity
demoralization |
|
Definition
feelings of deprivation result in control group giving up
control: keep experimental and control groups physically separated |
|
|
Term
other types of validity:
generalization
construct validity |
|
Definition
construct validity:
how well the observed cause and effect relationship represents the underlying causal process in which a researcher is interested
- remember we talked about convergent and discriminant validity?
related to: experimental realism
the extent to which experimental procedures have an impact on participants
the extent to which events in the experiment are credible, involving, and taken seriously |
|
|
Term
generalization
external validity |
|
Definition
generalizability from a relationship observed in one setting to the same relationship in another setting
- replication enhances external validity
related to: mundane realism
the extent to which an experiment is similar to situations encountered in everyday life |
|
|
Term
| quasi-experimental designs |
|
Definition
when randomization isn't possible for legal, ethical, or practical reasons
- a research design that includes most (but not all) aspects of an experimental design
inherent threats to internal validity |
|
|
Term
| nonequivalent groups designs |
|
Definition
if we cannot randomize, we cannot assume equivalency
match subjects in experimental and control groups using important variables likely related to DV under study
- comparison group |
|
|
Term
| nonequivalent group designs |
|
Definition
aggregate matching
cohort designs
- necessary to ensure that two cohorts being examined against one another are actually comparable |
|
|
Term
variable-oriented research
advantages/disadvantages |
|
Definition
advantages:
rich source of ideas for developing hypotheses.
complement to nomothetic study of behavior.
disadvantages:
difficulty drawing cause and effect conclusions.
external validity.
experimenter biases |
|
|
