Fichas
SOC 318
Fichas del conjunto (149)
- 1GeneralizabilityMaking predictions or assertions about a population based on a sample
- 2OvergeneralizeImproperly generalizing to a population from a small and/or non-representative sample
- 3Non-representativeWhen a sample fails to represent the population on some key characteristic or variable
- 4Sample/population generalizabilityWhen a conclusion based on a sample of a population holds true for that population
- 5Cross-population/external generalizabilityWhen findings about one population hold true for other populations
- 6Unit of analysisThe level/unit of social life on which a research question is focused
Population elements - 7PopulationThe entire set of individuals or entities to which the findings will be generalized
- 8SampleA subset of the population that is used to study the population as a whole
- 9CensusResearch in which the entire population is included
Population = sample - 10Sampling frameThe set of all cases from which the sample is actually selected
- 11Probability samplingSampling method that relies on random, or chance, selection methods so that the probability of selection methods is known
- 12Why use probability sampling?Reduce systematic bias
Allows us to estimate sample distributions and inferential statistics - 13Simple random samplingEvery combination of elements has the same probability of being selected
Will always have some level of bias - 14Replacement samplingSample elements are returned to the sampling frame after being selected
Can be sampled more than once
Rarely used - 15Systematic samplingEvery "Nth" case is selected
Every case has an equal probability of being selected
Every combination does not have an equal probability of being selected - 16Sampling intervalThe number of cases from one sampled case to another
- 17Stratified samplingPopulation is divided into strata, and then sample elements are selected from within those strata
Must be weighted - 18Proportionate samplingElements are selected in exact proportionate to their representation in the population
- 19Disproportionate representationElements are selected in different proportions from those in the population
- 20Cluster samplingElements are selected in two or more stages: first stage is random selection of clusters, second stage is selection of elements within the clusters
Must be weighted - 21Non-probability samplingSampling method in which the probability of selection of population elements is unknown
- 22Why use non-probability sampling?Practical reasons: population is unclear and/or hard to reach, limited resources
Don't need to generalize to a population - 23Availability/convenience samplingElements are selected because they are easy to find
- 24Quota samplingElements are selected to represent certain characteristics in proportion to their prevalence in the population
Ex. African Americans are 13% of US population, so should be 13/100 interviews - 25Purposive samplingElements are selected for a specific purpose
- 26Snowball samplingElements are selected by referral from informants or interviewees
Network sampling - 27ParameterThe value of a characteristic in a population
- 28StatisticThe value of a characteristic in a sample
Used to estimate parameters - 29Random sampling errorDifferences between the population and sample that are due only to chance
Larger sample = lower standard error
Will always exist, why we do statistical tests of significance - 30Coverage errorSampling frame is not representative in some way
Ex. telephone polls - 31Non-response biasDifferences between those who responded and those who did not
Ex. political polls - 32Sampling distributionThe probability distribution of a sample statistic
How close is the statistic (sample average) to the parameter (population average)?
Bigger samples = lower error, smaller samples = larger error
Bigger samples -> narrower distribution -> more confidence in statistic - 33Confidence intervalMeasure of the reliability of an estimate
The "plus or minus" range of an estimate
Typically want less than or equal to 5% - 34Confidence levelRepresents how often the true percentage of the population who would pick an answer lies within the confidence interval
Typically want greater than or equal to 95% - 35Criteria for causality1. Association
2. Direction of influence
3. Non-spuriousness - 36AssociationAssociation = correlation
One variable is related to the other - 37Direction of influencePlausible cause and effect
- 38How do we establish direction of influence?Longitudinal: observations over time
Experiment: manipulate the independent variable - 39Non-spuriousnessNot due to a third variable
- 40Spurious relationshipThere is a third variable that is the cause of both the independent and dependent variable
- 41Cross-sectional research designData are collected at one point in time
- 42Longitudinal research designData are collected at two or more points in time
- 43True experimentExperiment in which subjects are assigned randomly to an experimental group that receives a treatment or other manipulation of the independent variable, and a comparison group that does not receive the treatment or receives some other manipulation. Outcomes are measured in a posttest.
- 44Experimental/treatment groupReceives treatment or experimental manipulation
- 45Comparison groupExposed to a different treatment (or value of the independent variable) than the experimental group
- 46Control groupComparison group that receives no treatment
- 47RandomizationUnits are randomly assigned to the experimental, comparison, or control groups
- 48MatchingPairing similar units, assigning one to experimental and one to control
- 49PosttestMeasurement of the outcome in both groups after the experimental group has received the treatment
ALL true experiments have a posttest - 50PretestMeasurement of the dependent variable prior to experimental intervention
Many true experiments have a pretest - 51Internal validityConfidence in the causal relationship between the independent variable and the dependent variable
Improved by: random assignment, consistency of conditions/manipulation/dependent variable measurement - 52External validityConfidence that the findings apply outside of the experimental setting
Improved by: replication (especially across diverse settings and samples), ensuring experimental setting resembles the natural world - 53Field experimentsStudies that meet all the requirements of a true experiment but are conducted in a natural setting
- 54Natural experimentsStudies in which the experimental conditions happen naturally
Ex. two states have the same obesity rates (dependent variable). One state adopts a law (experimental), the other does not (control) - 55Survey researchCollection of information from a sample of individuals through their responses to questions
Surveys are the most popular form of social research - 56Attractions to survey research1. Versatility
2. Efficiency
3. Generalizability - 57SatisficingRespondents reduce effort required to answer a question by interpreting questions superficially and giving what they think will be an acceptable answer
- 58How can measurement error in surveys be reduced?Clear and interesting questions
Well-organized questionnaire
Questions appropriate for specific population - 59Social exchange theoryBehavior is motivated by cost-benefit analysis, so people won't respond if they think the costs outweigh the benefits
- 60Types of surveys1. Face-to-face
2. Mail survey
3. Phone survey
4. Internet survey
5. Group survey - 61Omnibus surveysSurveys that cover a wide range of topics of interest to social researchers
Ex. general social survey, Wisconsin longitudinal survey - 62Split-ballot designQuestion subsets are administered to randomly selected subsets of the sample
Allows for more questions to be included - 63Double-barreled questionsSingle survey question that actually asks two questions
- 64Double-negative questionsQuestion or statement that includes two negatives
Unclear and confusing - 65Mutually exclusive questionsEvery case can only have one attribute
"Check only one box" - 66Exhaustive questionsEvery case can be classified into one of the categories
- 67Fence-sittersPeople who see themselves as neutral
- 68FloatersRespondents who will chose a substantive answer when really they are neutral or don't know
- 69Forced choice questionsDo not include "don't know" or "neutral" response
Forcing someone to state an opinion
Ex. political polls - 70Leading questionsQuestions themselves suggest that there is a "correct" answer
- 71Filter questionDesigned to identify a subset of respondents
- 72Contingent questionAsked of only a subset of survey respondents
- 73Skip patternCombination of questions created by filter and contingent questions
- 74Interpretive questionsQuestions included in a questionnaire to help explain answers to other important questions
- 75Secondary data analysisUsing pre-existing data in a different way or to answer a different research question than intended by those who collected the data
- 76Content analysisSystematically analyzing and making inferences from recorded human communications
- 77ICPSRInter-university consortium for political and social research
- 78Unobtrusive methodDoes not involve interacting with people
- 79Manifest contentConcrete terms contained in a communication
- 80Latent contentUnderlying meaning of communication
- 81Key steps of content analysis1. Operationalize concepts
2. Identify units of analysis (and units of observation)
3. Sample
4. Code - 82Descriptive statisticsDescribe the distribution of, and relationship among, variables
- 83Inferential statisticsEstimate the degree of confidence that can be placed in generalization from a sample to the population from which the sample was selected
- 84Data cleaningProcess of checking data for errors after the data have been entered in a computer file
- 85ImputationReplacing missing data with estimates
- 86Mean imputationReplace missing data with the average
- 87Multiple imputationReplace missing data with estimates based on other values
- 88Cross-tabulationThe distribution of one variable for each category of another variable
- 89Null hypothesisThe hypothesis that there is no relationship between two or more variables
- 90Hypothesis testing"Rejecting the null"
"There's probably not no relationship" - 91Evaluation researchResearch conducted for a distinctive purpose, to investigate social programs and/or policies
- 92StakeholdersIndividuals and groups who have some basis of concern with the research
- 93InputsThe resources, raw materials, clients, and staff that go into a program
- 94Program processComplete treatment or service delivered by the program
Designed to have some impact on the cases - 95OutputsThe services delivered or new products produced by the program process
Ex. clients served, managers trained - 96OutcomesImpact of the program process on the cases processed
Ex. improved test scores, higher rates of job retention - 97Needs assessmentType of evaluation research that attempts to determine the needs of some population that might be met with a social program
- 98Evaluability assessmentType of evaluation research conducted to determine whether it is feasible to evaluate a program's effects within the available time and resources
- 99Process evaluationType of evaluation research that investigates the process of service delivery
- 100Impact analysisType of evaluation research that analyzes the extent to which a treatment or other service has an effect
- 101Efficiency analysisType of evaluation research that compares program costs with program effects. Can be cost-benefit analysis or cost-effectiveness analysis
- 102Selective observationChoosing to look only at things that are in line with our preferences or beliefs
- 103Inaccurate observationAn observation based on faulty perceptions of empirical reality
- 104Ecological fallacyConclusions are drawn about individuals from group-level data
(inverse of overgeneralization) - 105Errors in survey research1. Measurement error
2. Non-response error
3. Coverage error
4. Sampling error - 106Illogical reasoningPrematurely jumping to conclusions or arguing on the basis of invalid assumptions
- 107Resistance to changeReluctance to change our ideas in light of new information
Causes: ego-based commitments, excessive devotion to tradition, uncritical agreement with authority - 108Exploratory researchSeeks to find out how people get along in the setting under question, what meanings they give to actions, and what issues concern them
Investigation without expectation - 109Explanatory researchSeeks to identify causes and effects of social phenomena and to predict how changes in phenomenon will cause change or variation in another
- 110Evaluation researchResearch that describes or identifies the impacts of social policies and programs
- 111Deductive reasoningTheory --> data
General --> specific
The conclusion is absolutely certain if the evidence is true - 112Inductive reasoningData --> theory
Specific --> general
The conclusion is uncertain even if the evidence is true - 113Parts of a research questionTopic, question, and motivation/significance
- 114TriangulationThe use of two or more different methods to study one research question
Strengthens measurement, causal explanations
Reduces error - 115Dependent variableProposed outcome or effect
- 116Independent variableProposed influence or cause
- 117Control variableVariable that affects the dependent variable but is not a variable "of interest"
- 118ConstantConcept that does not vary
- 119Continuous variableTakes on any value along an interval
Ex. height, weight, time, distance - 120Discrete variablesAssume isolated points along an interval
Ex. number of cars, speed limits, students' grade levels - 121Antecedent variableVariable that occurs prior in time to both the independent and dependent variable
- 122Intervening variableOccurs between the independent and the dependent variable
It is an effect of the independent variable and a cause of the dependent variable - 123Dichotomous variable"Binary variables"
Any variable with only two response categories
Type of nominal (categorical) variable - 124HypothesisA tentative statement about empirical reality involving a relationship between two or more variables
- 125Parts of a hypothesisIndependent variable, dependent variable, direction of effect/relationship
- 126Positive associationAs one variable increases, the other increases
As one variable decreases, the other decreases - 127Negative associationAs one variable increases, the other decreases
- 128ConceptualizationThe process of specifying what we mean by a term
- 129OperationalizationProcess of connection concepts to observations
Specifies the values or categories of a variable - 130Nominal (categorical) level of measurementVariables whose values have no mathematical interpretation
Ex. race, nationality, gender, occupation - 131Ordinal level of measurementA measurement of a variable in which the numbers indicating a variable's values specify only the order of the cases, permitting greater than and less than distinctions
Ex. liekert scale - 132Interval level of measurementThe numbers indicating a variable's values represent fixed measurement units, but have no absolute or fixed zero point
Ex. fahrenheit temperatures, age - 133Ratio level of measurementNumbers indicating a variable's value represent fixed measuring units and have an absolute or fixed zero point
Ex. years of education - 134ReliabilityThe extent to which an assessment tool produces consistent results
- 135ValidityState that exists when statements or conclusions about empirical reality are correct
- 136Measurement validityThe extent to which measures indicate what they intend to measure
- 137Face validityThe appropriateness of a concept "on its face"
- 138Content validityThe measure covers the full range of the concept's meaning
- 139Criterion validityThe measures can be accurately compared with a more direct or already validated measure of the same concept
- 140Construct validityWhen a measure accurately reflects the theory
- 141IndexComposite measure composed of accumulated scores to several indicators
All items are equally important - 142ScaleComposite measure composed of several items that have a logical or empirical structure among them
Some items are more important than others
Weights - 143Frequency distributionA list or image of the values of observations
- 144HistogramGraphical representation of the distribution of data
Graph of frequencies (or probabilities) of values - 145Normal distributionMean = median = mode
Bell curve
Statistical tests assume normality of variables - 146MeanAverage computed by summing the values of several observations and dividing by the number of observations
Can be distorted by extremes or outliers - 147MedianThe value representing the middle case of a number of observations
- 148ModeMost frequently occurring value of a set of observations
- 149Standard deviationIndex of the amount of variation in the observations
Low standard deviation: a "thin" curve
High standard deviation: a "fat" curve
Lower standard deviation means higher consistency and confidence in mean
Información
- Descripción
- Fichas sobre SOC 318, creado por Amanda Jayne el 30/04/2015.
- Creado por
- Amanda Jayne
- Visitas
- 0
- Valoración