SOC 383 - EXAM 1 Study
SOC 383 - EXAM 1 Study
Readings - Ch. 1.2.4.5
Types of Research:
Descriptive - statistics of general population
Exploratory - qualitative
Explanatory - theory
Evaluatory - research, impact of law, policy, etc. change
Scientific Method
Observations
Generalizations
Reasoning
Potentially re-evaluate
Errors/Bias
Selective observation - choose to observe select few
Inaccurate observations - general errors/misjudgements
Over generalization - assuming true for all/larger group
Illogical Reasoning - jumping to conclusions
Resistance to change
Variables
Independent Var: ~cause - factors that INFLUENCE
Dependent Var: ~effect - ARE influenced
IV → DV ++++++ (positive correlation)
IV → DV --------- (negative correlation)
IV → ^ Mediating concept^→ DV (all or part of relationship explained by outside mediator influencing both pieces)
Ex: social class → gpa (mediate) → school funding/resources
IV → \/ Moderating concept \/ → DV (the moderator changes the outcome based on its involvement from IV to DV)
Ex: education → gender → wages
IV X→X DV Confounder impacts both separately and they are not actually related
Ex: Ice Cream X→X Drowning - confounder is season
Types of Validity
Measurement validity: are we actually measuring the concepts we think we are?
Face validity: appears appropriate and valid on its face (layperson can understand)
Content Validity: fully capture concepts
Criterion Validity: values/measures can be compared with a more direct/already valid source of measure
Ex: direct/objective (statistics) vs indirect/subjective (self-report)
Generalizability: are results applicable to broader pop?
Construct Validity: no criterion is available, showing relationship between concepts based on prior research
“Generalizability of a study is the extent to which it can inform us about persons, places, or events that were not directly studied.” Ch. 1
Causal/Internal Validity: how confident that X→Y etc ?
Deduct from theory to data - DTD
Induct from data to theory - IDT
Conceptualization
Concept - Mental image that summarizes a set of similar observations, feelings, or ideas
Conceptualization - The process of specifying what we mean by a term; Used to make sense of related observations
Operationalism - The process of specifying the measures that will indicate the value of cases on a variable; Connects concepts to measurement
Concepts → Variables → Indicators
Levels of Measurement
Nominal: Categorical, no order or ranking
Ordinal: Categorical with ranking
Interval: Fixed value units with no meaningful zero point
Ratio: Fixed value units with a meaningful zero point
Measurement Reliability
Reliability is easier to test than validity
Multiple Times
Test-retest Reliability
Intra-rater Reliability
Alternate forms Reliability
Multiple Items Reliability
Inter-term Reliability
Split-half Reliability
Multiple Observers Reliability
Interobserver: Consistency in measurement by 2 different observers
Intercoder: Consistency in coding by 2 different coders
Sampling and Generalizability
Population - The entire set of individuals or other entities to which study findings are to be generalized
Sample - A subset of a population used to study the population as a whole
Elements - The individual members of the population whose characteristics are to be measured
Sampling Frame - list of all elements in the population to draw samples from (we often rely on to pull from)
Sample Unit: units lifted from each stage of multi-stage design
Sample Error: your statistics of sample are different from the population
Samples - A subset of a population used to study the population as a whole
Probability sampling
Systematic Random Samples - A method of sampling in which sample elements are selected from a list or from sequential files, with every nth element being selected after the first element is selected randomly
Simple Random Samples - A method of sampling in which every sample element is selected purely on the basis of chance through a random process
Stratified Random Samples
Proportionate - Sampling method in which elements are selected from strata in exact proportion to their representation in the population
Disproportionate - Sampling in which elements are selected from strata in proportions different from those that appear in the population
Multistage Cluster Sampling - randomly selecting at each level/stage of the clusters
Ex: country, state, county, town, neighborhood
Non-probability sampling
Convenience Sampling - Sampling in which elements are selected on the basis of convenience
Quota Sampling - A nonprobability sampling method in which elements are selected to ensure that the sample represents certain characteristics in proportion to their prevalence in the population.
Purposive Sampling - A nonprobability sampling method in which elements are selected for a purpose, usually because of their unique position
Snowball Sampling - A method of sampling in which sample elements are selected as successive informants or interviewees identify them
Research Design
Cross-sectional: A study in which data are collected at only one point in time.
Longitudinal: Multiple points in time
Panel: Same people over time
Trend: Repeated cross-section of the same population over time
Cohort: Trend or panel
Units of Analysis
Individuals vs Groups of Individuals
Don’t want to generalize from higher units to lower ones and vice versa