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;
100 Cards in this Set
- Front
- Back
inaccurate observation
|
self-explanatory; can be guarded against by using simple and complex measurement devices
|
|
overgeneralization
|
assuming that a few similar events are evidence of a general pattern; can be guarded against by using large sample sizes and replication of studies
|
|
selective observation
|
paying attention to only future events that correspond with the pattern you have generalized (e.g. only taking note of lazy Mexicans after hearing the generalization that all Mexicans are lazy)
|
|
ex post facto hypothesizing
|
making hypothesis after observing; perfectly acceptable as long as it is tested
|
|
ego involvement in understanding
|
being biased in the belief that your own work and techniques are effective; can be guarded against by consulting supervisors or other practitioners for an objective point of view
|
|
Bottom-up searches
|
search literature looking for any and all source that provide evidence pertaining to the practice question you formulated
|
|
Top-down searches
|
rely on the results of evidence-based searches that others have done
|
|
Evidence Based Practice Process
|
applying the scientific method to practice decisions
|
|
Evidence based practices
|
specific practices that have shown to be effective in working with a given client population
|
|
Qualitative
|
– procedures evolve as more observations are gathered; typically permit the use of subjectivity to demonstrate deeper understandings of meanings of human experiences
|
|
Quantitative
|
seek to produce precise and quantifiable findings; all or most procedures are formulated in advance; strict adherence to procedures with maximum objectivity
|
|
Exploration
|
provides a beginning familiarity on a topic; seldom provide conclusive answers
|
|
Description
|
to describe situations and events
|
|
Explanation
|
to explain a phenomenon
|
|
Evaluation
|
to evaluate the effectiveness of a policy
|
|
Constructing Measurement Instruments
|
to make measurement instruments for others to use
|
|
Cross sectional studies
|
examines a phenomenon by taking a “snapshot” of it at one time and carefully analyzing it (e.g. US Census)
|
|
Longitudinal studies
|
intended to describe processes occurring over time; observations are conducted over an extended period; important part is that observations occur at different points in time
|
|
Aspects of good research questions
|
1. narrow and specific; 2. answerable by observable evidence; 3. relevant; 4. feasible
|
|
Concept
|
mental image that symbolizes an idea, an object, an event, a behavior, a person, etc.; words that people agree upon to symbolize something
|
|
Variables
|
concepts investigated by researchers; must be expected to vary
|
|
Attributes
|
different variations of a concept
|
|
concept, attributes
|
A variable is a _________ being investigated that is characterized by different _________.
|
|
hypothesis
|
a tentative and testable statement about a presumed relationship between variables
|
|
independent variable
|
the variable being explained or caused; this is the variable we seek to measure
|
|
control variable
|
a moderating variable that we seek to control by holding it constant in our research design
|
|
positive relationship
|
both variables move in the same direction (e.g. time spent studying and test scores)
|
|
negative/inverse relationship
|
variables move in opposite directions; as one increases, the other decreases and vice versa (e.g. hours spent partying and test scores)
|
|
curvilinear relationship
|
a relationship in which the nature of the relationship changes at certain levels of the variables (e.g. anxiety and performance)
|
|
nominal level
|
attributes of variables are categorical and can be described in terms of how many but not degree of (gender, ethnicity, birthplace)
|
|
ordinal level
|
attributes may be rank-ordered according to degree (level of education in terms of: high school diploma/GED, some college, bachelor's degree, etc.)
|
|
interval level
|
differences between different levels have the same meanings (differences between IQ scores of 100 and 105 compared to 95 and 100)
|
|
ratio level
|
same attributes of interval measures but with a true zero point (number of arrests)
|
|
measurement error
|
data do not accurately portray the concept we attempt to measure
|
|
systematic error
|
type of measurement error; the info we collect consistently reflects a false picture (e.g. acquiescent response set; social desirability bias)
|
|
random error
|
type of measurement error; no consistent pattern of effects (e.g. cumbersome, complex, boring measurement procedures or measures use professional jargon which respondents are not familiar with)
|
|
triangulation
|
using several different research methods to collect the same information; useful in avoiding measurement error
|
|
Interobserver and Interrater reliability
|
degree of agreement or consistency between/among observers
|
|
Test-retest reliability
|
assessing a measure’s stability over time; acceptable reliability: above .70 or .80
|
|
Internal consistency reliability
|
assess whether the items of a measure are internally consistent (using the split halves method or parallel forms reliability)
|
|
Validity
|
whether a particular measurement measures what we seek to measure; (synonymous with “accuracy”)
|
|
Face validity
|
whether a measure appears to measure what is supposed to measure; subjective assessment
|
|
Content validity
|
the degree to which a measure covers the range of meanings included within the concept; subjective assessment
|
|
Criterion-related validity
|
based on some external criterion
|
|
Predictive validity
|
measure can predict a criterion that will occur in the future
|
|
Concurrent validity
|
measure corresponds to a criterion that is known concurrently
|
|
Construct validity
|
assess whether a measure fits theoretical expectations
|
|
Convergent validity
|
IDK
|
|
Discriminant validity
|
IDK
|
|
Factorial validity
|
whether the number of constructs and the items that make up those constructs measure what the researcher intends to measure
|
|
Guidelines for asking questions:
|
(1) Be specific and consistent;
(2) Use plain, simple language; (3) Don't ask double-barreled questions; (4) Don't lead the interviewee (no loaded questions) |
|
types of survey
|
mail, online, interview (in-person), telephone
|
|
strengths of mail survey
|
cheaper and quicker; large samples; anonymity facilitates responses regarding sensitive areas
|
|
weaknesses of mail survey
|
expensive and time consuming; lower response rates than face-to-face
|
|
strengths of online survey
|
less expensive and time consuming; largest samples; automatic data entry; anonymity facilitates responses regarding sensitive areas
|
|
weaknesses of online survey
|
representativeness (especially among poor and elderly); lower response rates than face-to-face
|
|
strengths of interview (in-person)survey
|
higher response rate; minimizes "don't know" or "no" answers; allows interviewers to observe respondent while asking questions
|
|
weaknesses of interview (in-person) survey
|
more expensive and time-consuming; lack of anonymity can impede responses regarding sensitive areas; interviewer safety
|
|
strengths of telephone survey
|
inexpensive; more honest answers; interviewers have more support; opportunity to probe/clarify; personal safety; opportunity for supervision; can be computer-assisted
|
|
weaknesses of telephone survey
|
bogus surveys; survey discontinuation; ease of hanging up; answering machines; caller id; cell phones
|
|
Sample
|
subset of a population that is observed for purposes of making inferences about the nature of the total population
|
|
Probability Sampling
|
random selection; each participant has an equal chance of selection
|
|
Overgeneralization
|
sampling frames are not consistent with what we seek to generalize
|
|
Cultural bias
|
unwarranted generalization of research findings to the population as a whole when one culture or ethnic group is not adequately represented in the sample
|
|
Gender bias
|
unwarranted generalization of research findings to the population as a whole when one gender is not adequately represented in the sample
|
|
Nonresponse bias
|
substantial number of people chosen choose not to participate
|
|
Nonprobability Sampling
|
used when probability or random sampling is not possible or appropriate (e.g. homeless individuals)
|
|
Availability or convenience sampling
|
sampling from subjects who are available (e.g. college students)
|
|
Purposive or judgmental sampling
|
researcher used his or her own judgment in selecting sample members (e.g. handpicking community leaders or experts for expertise on target population)
|
|
Quota sampling
|
relative proportion of the total population is assigned for the target population’s characteristics, grouped into strata, and then sampled
|
|
Snowball sampling
|
process of accumulation as each located subject suggests other participants
|
|
Sampling frame
|
list or quasi-list of members of a population (e.g. student roster, telephone directory)
|
|
Simple random sampling
|
each element in sampling frame is assigned a number and a table of random numbers is used to select elements for the sample
|
|
Systematic Sampling
|
selection of every kth element or member of the sampling frame
|
|
Stratified sampling
|
grouping members of a population into homogeneous strata before sampling (e.g. ethnic group or gender); improve the representativeness of a sample by reducing the degree of sampling error
|
|
criteria for establishing causality
|
(1) The IV must precede the DV chronologically; (2) There must be a relationship or correlation between the IV and DV; (3) The co-variation between the two variables cannot be explained by a third variable
|
|
establish causation
|
The goal of experimental designs is to _____________.
|
|
threats to internal validity
|
history, maturation, testing, instrumentation, statistical regression, selection bias, ambiguity about the direction of causal influence [SHAMITS]
|
|
History
|
external events; changes occur outside the person
|
|
Maturation
|
passage of time; changes occur within a person
|
|
Testing
|
process of testing itself enhances performance on a test; also: social desirability, in which participants choose what they think researchers want
|
|
Instrumentation
|
change in instrumentation over time (e.g. switching from one measure to another between pre- and post-test); instrument decay (e.g. physical decay in a measurement apparatus)
|
|
Statistical regression
|
extreme scores will regress to the mean
|
|
Selection bias
|
differences between two groups prior to intervention
|
|
Ambiguity about the direction of causal influence
|
related to time order of IV and DV; DV could actually cause IV
|
|
Experimental
|
only design that can claim causality; requires random assignment to experimental and control group and comparison of results between the two
|
|
random assignment
|
r indicates:
|
|
measurement
|
o indicates
|
|
intervention
|
x indicates:
|
|
Pre-experimental
|
no randomization; cannot claim causality; commonly used in pilot studies; use of comparison group as opposed to control group
|
|
Quasi-experimental
|
similar to experimental design, except that participants cannot be randomly assigned (e.g. smokers vs. non-smokers; would be unethical to cause someone to become addicted to nicotine)
|
|
Advantages to secondary analysis
|
data is already there, less costly and time consuming
|
|
Disadvantages to secondary analysis
|
no control over sampling; must trust that data is reliable
|
|
Continuous variable
|
can theoretically have an infinite number of values between adjacent units on a scale
|
|
Discrete variable
|
one in which there are no possible values between adjacent units on a scale
|
|
Mean
|
average; add up the values to find the sum and then divide by the number of values
|
|
Median
|
the middle; remove the highest and lowest value until you arrive at the middle; if there are an odd number of values, average the two in the middle to find the median
|
|
Mode
|
most frequent value; the value that appears more often than any other value; there can be more than one
|
|
Standard deviation
|
typical distance from the mean; do not need to know how to calculate; 0 = no variation among values
|
|
p-value
|
the probability of an observed relationship being caused by chance; this value is compared to the “significance level” or “alpha (α) level” (customary “significance level” is 0.05)
|