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;
64 Cards in this Set
- Front
- Back
Qualitative Measurement
|
measure data DURING data collection & analysis; convert data (words, symbols); inductive reasoning (concrete-> abstract)
|
|
Quantitative Measurement
|
measure variables BEFORE data collection; focus on convert data (#'s); deductive reasoning (abstract->concrete)
|
|
Conceptualization
|
process of carefully thinking through the meaning of a construct; VERY abstract
|
|
operationalism
|
process of defining a concept so that it can be measured & repeated; observations can be made that are reliable & valid; practical; links a conceptual definition to a specific set of measurement techniques or procedures; fit your measure to your specific conceptual definition to the practical constraints within which you must operate and to the research techniques you know or can learn; links the language of theory w/ the language of empirical measures
|
|
Relability
|
dependability or consistency; consistency in results of a test or measurement
|
|
Validity
|
capacity of an instrument to measure what it was designed to measure; truthfulness
|
|
Quantitative: Improve Reliability
|
conceptualize constructs-"designed to eliminate noise"; specfic level of measurement; use multiple indicators of a variable; adminster pretest/pilot test
|
|
Quantitative: Improve Validity
|
face validity; content validity; criterion validity; concurrent validity; predictive validity
|
|
face validity
|
(quantitative)definition & method of measuremen are good fit;a judgement by the scientific community that the indicator really measures the construct ie. not many people would accept 2 +2= 4 as a sufficient way to test a college student's math abilities
|
|
content validity
|
a special type of face validity; is the fully CONTENT of a definition represented in a measure?
|
|
criterion validity
|
outside source of measure will verify your validity; uses some standard or criterion to indicate a construct accurately
|
|
concurrent validity
|
type of criterion validity; doing well on one test you should do well on another
|
|
predictive validity
|
predict some future behavior; ie. SAT results are a predicator of whether you will do well in college
|
|
Qualitative: reliability
|
Data is an interactive process; bc it is collected in a unique process; consistency through interviews, participation, photographs, document studies; evolving relationship w/ subject matter
|
|
Qualitative operationalization
|
often precedes conceptualization; describes how specific observations and thoughts about the data contributed to working ideas that are the basis of conceptual definitions and theoretical concepts; an after-the-fact description (more than a before-the-fact preplanned technique); data gathering occurs w/ or prior to full operationalization; describes how the researcher collects data, but it includes the researcher's use of preexisting techniques & concepts that were blended w/ those that emerged during the data collection process
|
|
Qualitative conceptualization
|
a process of forming coherent theoretical definitions as one struggles to "make sense" or organize the data & one's preliminary ideas; refine rudimentary "working ideas" DURING the data collection and analysis process
|
|
Quanitative conceptualization
|
refining abstract ideas into theoretical definitions early in the research process
|
|
Quantitative Operationalization
|
(after conceptualization); developing an operational definition or set of indicators for it
|
|
continuous variables
|
(quantitative measurment)infinite # of values/attributes (along a continuum); can be made discrete --> temp by saying it's "cold" and "hot"; ratio/interval levels
|
|
discrete variables
|
(quantitative measurement) fixed set of values/attributes (marital status); can be made continuous--> temperature; nominal & ordinal levels
|
|
Qualitative validity
|
based on participants views & accounts; authenticity: assuming participants are giving the honest truth from their perspective (asking in multiple ways to know for sure if someone knows the material)
|
|
nominal (level of measurement)
|
categories by names (social class, race)
|
|
oridinal (level of measurement)
|
one is greater than another (letter grades; completely agree, agree, disagree);nominal too
|
|
Interval (level of measurement)
|
(oridinal & nominal too!) no absolute zero-> it's arbitrary; where you cannot say something is 3 or 4x bigger; ie. temperature (bc you have neg. degrees)
|
|
ratio (level of measurement)
|
(nominal, ordinal, interval too!)real zero; you can say something is 2x something else (ie. age, weight, $)
|
|
scales
|
measure of intensity, direction, level of a construct; oridinal: attitude (measured in continuum)
strongly agree <----> strongly disagree; likert is used most frequently |
|
indexes
|
measure that adds or combines several indicators; I or R; look @ weighted measure; cumulative score that is supposed to measure faculty member qualitiy by student perceptions
|
|
good measurement
|
mutually exclusive (only one possibility for an answer); exhaustive (all possible choices must be listed); unidimensional (within an index all scales should measure one construct)
|
|
Literature Review
|
scholarly journals, books, dissertations, gov't documents
|
|
ethnography
|
study on a community; studies on neighborhoods, street corners
|
|
Quantitative(issue in research)
|
linear path: logical sequence of path, rigid research question is much more narrowed & focused
|
|
Qualitative (issue in research)
|
circular path; flexible, you might get new ideas while researching and refine your approach, "How much" "How many"
|
|
Qualitative research
|
1. interpretations & meanings (many perspectives), inductive reasoning, empirical>observation, record, document, examine real events; 2. grounded in theory: "grounded in the data"-drives the theory & research question; 3. context-"the meaning", applied to a certain time, place, culture, setting you cannot move the situation out of the context; 4. cases of the unit of analysis(casual relationship), each individual becomes your unit of analysis; 5. interpretation (steps) a.participant's voice (record them verbatim) b. researcher as an "outsider" c. connect study to sociology
|
|
Quantitative Research Methods
|
variables, attributes, independent variable, hypothesis, level of analysis & unit of analysis, Types of error
|
|
variables
|
different concepts, looking @ relationships among different variables (race, ethnicity)
|
|
attributes
|
characteristics or categories of variables (ie. adolscence/young adulthood, male/female)
|
|
independent variable
|
cause leading to the dependent variable
|
|
intervening variable
|
ex. good parent support---> high self esteem (when actually, it's bc of good school performance)
|
|
hypothesis
|
able to predict a cause & effect relationship (edu. guess); linked to research question & theory; can be tested as false, but you cannot neccessarily test if a hypothesis is true
|
|
null hypothesis
|
test that there is no relationship bn the variables (no association)
|
|
alternative hypothesis
|
test that a relationship exists bn 2 variables
|
|
level of analysis
|
macro(social category or institution) & micro (unit individual)
|
|
unit of analysis
|
what level? individual, family, institutional, social categories
|
|
ecological fallacy
|
making an error in infering relationships bn 2 variables will also hold true at the individual level; assume some group behavior can be understood @ the individual level; ie. unemployment increases so does mental illness, so they must be related
|
|
reductionism
|
fallacy of non equivalence; using a psychological explanation; use 1 factor to explain a range of behaviors
|
|
spuriousness
|
no relationship bn the variables; each of those variables are correlated to a 3rd variable(has more impact, NOT casual); ie. consuming an herb & longevity--> 3rd variable = wealth
|
|
to avoid error, your level of analysis =
|
your unit of analysis!
|
|
nonprobability sampling (qualitative) non random (4 types)
|
Purposive sampling, snowball sampling, deviant case sampling, sequential sampling
|
|
purposive sampling
|
field research, exploratory, do not know if cases are selecting are representative of population (if possible you want to get every person in the population)
|
|
snowball sampling
|
researcher selects the cases from a network of people; start w/ 1 case that might lead to another case (indirect or direct links w/ individuals); used for further topics/referrals; very effective with "sensitive" topics
|
|
Deviant Case sampling
|
studying cases of different social patterns, that do not fit the general pattern; ie. highschool dropout whose parents went to grad. school
|
|
sequential sampling
|
researcher has acquired enough cases that he is satisfied (acquired the minimal #); * there is no way you can say that your sample is representative of the WHOLE subculture
|
|
probability sampling
|
(quantitative); always taking sample of larger population
|
|
sampling frame
|
specific list constructed to approximate the population
|
|
parameter
|
a characteristic of the larger population; summary measure of the entire population
|
|
random
|
everyone has an equal chance, good representative of entire population
|
|
sampling error
|
how much a sample deviates from being representitive of the population
|
|
simple random probability sampling
|
a random sample in which a reasearcher creates a sampling frame and uses a pure random process to select cases.
|
|
simple random WITH replacement
|
draw sample out, put it back; draw from SAME smple out
|
|
simple random WITHOUT replacement
|
draw sample out; draw another sample
|
|
sampling distribution
|
a distribution of different samples that shows the frequency of different sample outcomes from many separate random samples
|
|
systematic sampling
|
a type of random sampling in which a researcher selects every xth (ie. 9th) case in the sampling frame using sampling interval
|
|
stratified sampling
|
researcher divides population into strata (subpopulation), then draws a random sample from each subpopulation; used to control the size of population; every single category will be represented
|
|
cluster sampling
|
take a sample of an already existing cluster, then randomly select from that cluster
|