Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
131 Cards in this Set
- Front
- Back
modified replication
|
a replication with some major modifications such as examining a new population or using an improved measurement technique
|
|
strict replication
|
mimics the original study in all important respects to see if the same types of results as in the original study will be obtained
|
|
literature review
|
usually present in a separate chapter immediately following the chapter that contains the introduction, this is relevant information used by a researcher
|
|
3 major databases in the social and behavioral sciences
|
1. Sociological Abstracts, Planning/Policy and Development Abstracts
2. PsycINFO containing Psychological Abstracts 3. ERIC |
|
thesaurus
|
the first step in searching a database. Used to find a topic and refine a research topic
|
|
major headings
|
guides readers through a long literature review
|
|
Sampling
|
first area of critical assessment.
More often than not, samples are less than ideal |
|
Instrumentation
|
second area of critical assessment.
It is safe to presume that all instruments are flawed to some extent. Intruments have limitations. |
|
Experiments
|
Third area of critical assessment.
Experiments are often flawed by having inappropriate control conditions |
|
synthesis
|
provides a "whole" picture of what is known and what is not known as well as an attempt to show how diverse peices of information fit together and make sense
|
|
APA citing for a jorunal article
|
The title of the article is in lowercase except for the first word (or the first word following a colon), and the title of the journal is in upper-and lowercase and is italicized. It is followed by the volume number, which is also italicized
|
|
APA citing for a book
|
The title of the book is in lowercase except for the first word (or first word following a colon) and is italicized
|
|
APA citing for material found on the Web
|
Provide the date on which the material was accessed because the content of a Web site might vary from day to day
|
|
APA citing for all sources
|
Follow the punctuation used in the examples carefully. For instance, when writing in English, a commoa is not necessary between two elements in a list. However, in APA style, a comma is used between the names of two authors
|
|
population
|
the group in which researchers are ultimately interested. A population may be large or small
|
|
sample
|
a small portion of a population
|
|
unbiased sample
|
every member of a population has an equal chance of being included in the sample
|
|
simple random sample
|
an unbiased sample where names are placed in a hat and drawn at random
|
|
samples of convenience
|
a biased sample also known as accidental samples.
Ex: a psych professor wants to study a principle of learning theory as it applies to all college sophomores but only uses those students who happen to be enrolled in his course |
|
volunteerism
|
2 forms
1. researchers issue a call for volunteers 2. volunteeris might bias a sample from an entire population |
|
simple random sample
|
every member of a population is given an equal chance of being included in a sampe
|
|
sampling error
|
error created by random sampling (for example: a random sample might contain a disproportionately large number of males, high achievers, and so on)
|
|
systematic sampling
|
every nth individual is selected
(every 10th person or every 6th person, etc) |
|
precision
|
discussing the magnitude of sampling errors
Results are more precise when researchers reduce sampling errors |
|
Two major ways to reduce sampling errors
|
1. increase sample size
2. use stratification in conjunction with random sampling (stratified random sampling) |
|
stratified random sampling
|
divide the population into strata (ex: men and women) and draw at random
|
|
Number vs. percentage
|
Researchers usually draw the same percentage of participants, not the same number
|
|
multiple strata
|
increases precision
Ex: gender AND age |
|
cluster sampling
|
researchers draw groups or clusters of participants instead of drawing individuals
|
|
Major drawback of cluster sampling
|
each cluster tends to be more homogeneous in a variety of ways than the population as a whole
|
|
purposive sampling
|
researchers purposively select individuals whom they believe will be good sources of information
|
|
snowball sampling
|
can be useful when attempting to locate participants who are hard to find...
Convince one person to contact others and bring them in Thus these are biased |
|
demographics
|
background characteristics of the participants in research such as gener, age, and income
|
|
mortality
|
if some participants drop out of the experiment at mid-course, mortality is said to have occurred
|
|
sample size
|
secondary to bias
|
|
highly precise results
|
results will vary by only a small amount from sample to sample
|
|
pilot studies
|
studies designed to obtain preliminary information on how new treatments and instruments work
|
|
instrument
|
generic term for any type of measurement dvice (e.g. test, questionnaire, interview schedule, or personality scale)
|
|
instrumentation
|
term used as the heading for the section of the report where the measurement devices used in the research are described
|
|
validity
|
instrument is valid to the extent that it measures what it is designed to measure and accurately performs the functions it is purported to perform
|
|
content validity
|
researchers make judgments on the appropriateness of an instrument's contents
For achievement tests, this type of validity is essential |
|
face validity
|
judgments are made on whether an instrument appears to be valid on the face of it
In other words, on superficial inspection, does the instrument appear to measure what is purports to measure? |
|
criterion
|
"In the empirical approach to validity, researchers make planned comparisons to see if an instrument yields scores that relate to a criterion"
|
|
predictive validity
|
poses the questions: To what extent does the test predict the outcome it is supposed to predict?
|
|
validity coefficient
|
a correlation coefficient used to express validity
range from 0.00-1.00 1.00 is perfect validity |
|
concurrent coefficient
|
administering the test and collection the criterion data at about the same time
|
|
criterion-related validity
|
general term for both concurrent and predictive validity
|
|
construct validity
|
type of validity that relies on objective jidgments AND empirical data (i.e. data based on observations)
Provides indirect evidence |
|
construct
|
stands for a collection of related behaviors that are associated in a meaningful way
Ex: depression is a construct that stands for a personality trait manifested by behaviors such as lethargy, flat affect when speaking, loss of appetite, etc. |
|
reliable
|
test is said to be reliable if it yields consistent results
|
|
Validity vs. reliability
|
When evaluating instruments, validity is more important that reliability.
|
|
Three important principles
|
1. A test with high reliability may have low validity
2. Validity is more important than reliability 3. To be useful, an instrument must be both reasonably valid and reasonably reliable |
|
interobserver reliability
|
a researcher observes unobtrusively
|
|
reliability coefficients
|
correlation coefficients that decribe reliability
|
|
interobserver reliability coefficients
|
When researchers use reliability coeffiecients to describe the agreement between observers
|
|
test-retest reliability
|
researchers measure at two different points in time
|
|
parallel-forms reliability
|
administering one form onf the test to examinees and about a week or two later administering the other form to the same examinees, thus yielding two scores per examinee.
When the sets of scores are correlated, the result indicates the parallel-forms reliability |
|
How high should a reliability coefficient be?
|
.80 or higher especially for individuals
For group averages of 25 or more, .50 is serviceable |
|
split-half reliability
|
researcher administers a test but scores the items in the test as though they consisted of two separate tests
*Better example in Topic 33* |
|
internal consistency
|
reliability estimates
|
|
Cronbach's alpha
|
method for estimating internal consistency
-based on single admin. of a test -mathematical procedures are used to obtain the equivalent of the average of all possible split-half reliability coefficients |
|
norm-referenced tests (NRTs)
|
tests designed to facilitate a comparison of individual's performance with that of a norm group
Ex: examinee earns percentile rank of 64...this means she scored higer than 64% of individuals in the norm group |
|
criterion-referenced tests (CRTs)
|
tests designed to measure the extent to which individual examinees have met performance standards
|
|
acheievement test
|
measures knowledge and skills individuals have acquired
|
|
aptitude test
|
designed to predict some specific type of achievement
Example: SAT predicts success in college |
|
intelligence test
|
designed to predict acheievement in general, not any one specific type
|
|
3 approaches to reducing social disrability in participants' responses
|
1. administering personality measures anonymously
2. observe behavior unobtrusively 3. use projective techniques (these provide loosely structured or ambiguous stimuli such as ink blots) |
|
Likert-type scales
|
scales that have choices from "Strongly agree" to "strongly disagree"
|
|
reverse scoring
|
Ex: If 5 points are rewarded for answering "Strongly agree," to a positive statement toward school, then 5 points would be awarded for answering "strongly disagree" with a negative statement toward school
|
|
pretest-posttest randomized control group design
|
Don't know a definition for this...it's at the beginning of Topic 37
|
|
pretest sensitization
|
aka reactive effect of testing
changes observed in the experimental group may be the result of a combination of the pretest and the treatment |
|
posttest-only randomized control group design
|
There is no pretest...
more info in Topic 37 |
|
Solomon randomized four-group desidgn
|
combination os pretest-posttest and posttest-only randomized control group design
-4 rows of symbols (4 groups) -two experiments conducted at the same time |
|
true experimental designs
|
random assignment to treatments
|
|
threats to internal validity
|
Depending on the design of the experiment, there may be explanations for changes other than the treatment....that's these.
|
|
history threat
|
other environmental influences on the participants between the pretest and the posttest
|
|
maturation threat
|
participants mature during the period between the pretest and the posttest which causes a threat
|
|
instrumentation threat
|
this refers to possibly changes in the instrument from the time it was used as a pretest to the time it was used as a posttest
|
|
testing threat
|
the effects of the pretest on the performance exhibited on the posttest
|
|
statistical regression threat
|
this occurs only if participants are selected on the basis of their extreme scores
|
|
intact groups
|
previously existing groups
|
|
selection
|
researchers do not assign participants to the two groups at random, there is a very stron possibility that the two groups are not initially the same in all important respects
this is a threat |
|
selection-history interaction
|
selection of participant for two groups was not at random so they may be syustematically subjected to different life experiences
|
|
selection-maturation interaction
|
the two groups, on the average, were at somewhat different developmental stages at the time of the pretest, which would have led to different rates of maturation in the two groups, which could affect self-concept
|
|
mortality
|
differential loss of participants from the groups to be compared
|
|
generalize
|
assume that the treatment adminstered to the experimental group will work as well in the population as it did in the sample
|
|
threats to external validity
|
the extent to which the experiment is subject to threats from outside things (like the population)
|
|
selection bias threat
|
researcher's ability to generalize to a population is greatly limited....no generalizations should be made when this is the case
-you won't know whether the effects of the treatment can be expected if the treatment is administered to the entire population |
|
reactive effects of experimental arrangements threat
|
this reminds researchers that if the experimental setting is different from the natural setting in which the population usually operates, the effects that are observed in the experimental settings may not generalize to the natural setting
|
|
reactive effect of testing (pretest sensitization) threat
|
the possibility that the pretest might influence how the participants respond to the experimental treatment
|
|
multiple-treatment interference threat
|
occurs when a group of participants is given more than one treatment
|
|
internal validity vs. external validity
|
external is concerned with "to whom and under what circumstances can the results be generalized?"
internal is concerned with "is the treatment, IN THIS PARTICULAR CASE, responsible for the observed changes?" |
|
3 pre-experimental designs
|
-all have poor internal validity
1. one-group pretest-posttest design 2. one-shot case study 3. static-group comparison design |
|
one-shot case study
|
gropud is given a treatment (X) followed by a test (O)
X O |
|
one-group pretest-posttest design
|
O X O
where O is test and X is treatment |
|
static-group comparison design
|
two groups, but participants are not assigned to the groups at random....the dashed line indicates that they are intact groups
X O ------- O |
|
static-group comparison design
|
two groups, but participants are not assigned to the groups at random....the dashed line indicates that they are intact groups
X O ------- O |
|
quasi-experimental designs
|
these are of intermediate value for exploring cause-and-effect
|
|
nonequivalent control group design
|
widely used quasi-experiment design with two intact groups
O X O ------------ O O |
|
equivalent time-samples design
|
has only one group (or possibly one one participant). Treatment conditions are alternated
X0O X1O X0O X1O |
|
confound
|
a source of confusion regarding the explanation for a given difference
|
|
Hawthorne effect
|
"attention effect"
There are two intertwined explanations for the differences observed because participants know they're being watched. To control, some researchers use an experimental gropu, a control group that receive no special attention and a control group that receives attention |
|
John Henry Effect
|
refers to the possibility that the control group might become aware of its "inferior" status and respond by trying to outperform the experimental group
|
|
placebo effect
|
the tendency of individuals to improve (or at least feel that they are improving) simply because they know they are being treated
|
|
placebo
|
"pill" that contains only inert ingredients
|
|
blind procedure
|
researchers do not disclose to the participants whether they are receiving an active or inactive substance
|
|
double-blind experiment
|
neither the participants nor the individual dispensing the drug know which is the active drug and which is the placebo
-this is done to prevent the possibility that the individual dispensing the drug will subtly communicate to the participants their status as being either "control" or "experimental" participants |
|
demand characteristics
|
-source of confounding
-a cue that lets participants know the expect outcome of an experiment |
|
descriptive statistics
|
summarize data so they can easily be comprehended
|
|
frequency distribution
|
shows how the scores are distributed
|
|
frequencies
|
descriptive statistics...describe how many students earned each score
|
|
percentages
|
descriptive statistics...describe how many students PER ONE HUNDRED had each score
|
|
inferential statistics
|
-helps researchers draw inferences about the effect of sampling errors on the results that are described with descriptive statistics
-help researchers make generalizations about the characteristics of populations based on data obtained by studying samples |
|
margin of error
|
amount plus/minus
|
|
significance tests
|
help researchers decide whether the differences in descriptive statistics they identify are reliable
|
|
parameters
|
statistical values derived from a census
p-opulations yield p-arameters |
|
statistics
|
statistical values based on the results obtained from a sample
s-amples yield s-tatistics |
|
null hypothesis version A
|
the observed difference was created by sampling error
|
|
null hypothesis version B
|
there is no true difference between the two groups
|
|
null hypothesis version C
|
the true difference between the two groups is zero
|
|
significance tests
|
determine the probability that the null hypothesis is true
|
|
probability (p)
|
p < 0.05
|
|
statistically significant
|
states that the null hypothesis has been rejected
|
|
nominal level
|
lowest level of measurement
-"naming" level -do not put participants in any particular mathematical category |
|
ordinal level
|
-measurements place participants in order frmo high to low
|
|
interval and ratio levels
|
-have equal distances among the scores they yield
-ratio scale is higher level than the interval scale because the ratio scale has an absolute zero point that researchers know how to measure (weight is on the ratio scale, for instance) -interval does not have an absolute zero |
|
NOIL
|
no oil in rivers
Nominal Ordinal Interval Ratio |
|
number of cases
|
"N"
also known as frequencies |
|
univariate analysis
|
researcher is analyzing how participants vary on only ONE variable
|
|
bivariate analysis
|
researchers examine a relationship between two nominal variables
|
|
proportions
|
percentage 47.8% corresponds to proportion of 0.478
|