Soc methods midterm

Sociological Imagination

Ability to see the connection between the larger world and our personal lives

Generalizable

can explain a broad class of events

Agency

our capacity to make our own choices and act autonomously

Basic Research

seeks to answer theoretically informed questions or resolve fundamental intellectual puzzles about social behavior

Applied research

seeks to answer a question or concrete problem in the real world or to evaluate a policy or program

Mixed-Methods Approach

a general research approach that uses qual and quant in a single study

Triangulation

the use of multiple methods to study one research question (more general than mixed methods)

Cross-Sectional Study Design

a study in which data are collected at only one point in time

Repeated cross-sectional study design/trend design

a type of longitudinal study in which data are collected at multiple time points, but from different subjects at each time point

Panel Design

a type of longitudinal study in which data are collected on the same subjects at multiple time points

Causality

a relationship where one factor or variable is dependent on another factor or variable
Studies that track individuals at multiple points in time are better suited for ascertaining cause and effect than a single cross-sectional design

Units of Analysis

refers to the level of social life about which we want to generalize

Ecological fallacy

a mistake that researchers make by drawing conclusions about the micro level based on some macro level analysis

The Scientific Method

the systematic process of asking and answering questions in a rigorous and unbiased way

The Scientific Method steps (5)

1. Identify an important question that needs an answer

2. Construct a hypothesis about the answer to this question

3. Gather data that allow the researcher to assess the accuracy of this prediction

4. Analyze the data to determine whether the prediction is accurate

5. Draw and report conclusions

Value-Free

the goal of being objective and not biased by personal ideologies

Subjectivity

the way research is influenced by the perspectives, values, social experiences, and viewpoint of the researcher

Reflexivity

a process of attending systematically to the context of knowledge construction, especially to the effect of the researcher, at every step of the process

Descriptive Research

the what. documents or describes trends, variations, and patterns of social phenomena

Exploratory Research

tends to answer questions of how, with the goal of documenting precisely how particular processes and dynamics unfold

Explanatory Research

documents the causes and effects of social phenomena, thus addressing questions of why

Purpose of theory

describe, explain, predict, control

Empiricism

the idea that the world can be subjected to observation, which is the use of the senses to gather data about social phenomena

Inductive Approach

the process by which scientists draw up a general understanding of social phenomenon through empirical observations

Inductive Approach goal

observe patterns and the build up to explanation. Concrete to abstract; specific to general; bottom up

Deductive Approach

the translation of general theory into specific empirical analysis

Deductive Approach goal

create an argument to organize and guide empirical activities. Abstract to concrete; general to specific; top down

Mediation

the expected relation between two concepts is channeled through a third concept that links them to each other

Moderation

the strength of the association between two variables is made weaker or stronger by a third variable

Null Hypothesis

the hypothesis that no relationship between concepts exists or no difference in the dependent variable between groups exists

Spuriousness

when an apparent relation between two concepts is actually the result of some third concept (confound(er)) influencing both of them. unlikely to be a part of a theory and more likely to come up when testing a theory

Confound

a third variable that is linked to two concepts in a way that makes them appear to be related even when they are not

Conceptualization

the process of precisely defining ideas and turning them into variables

Operationalization

the process of linking the conceptualized variables to a set of procedures for measuring them. The process of identifying a plan for measurement. It involves making trade-offs between the potential benefits of using proven measures and developing novel ways to measure variables

categorical variables

Have a finite set of possible values
No known distances between values
Includes nominal and ordinal variables

Nominal variables

categorical variables that have catalog states or statuses that are parallel and cannot be ranked or ordered
no mathematical interpretation (Ex: race, state)

Ordinal variables

categorical variables for which the categories have a natural ordering

Continuous Variables

Have an infinite set of possible values
Values have fixed distances between them
Includes interval and ratio variables

Interval variables

Continuous variables that have a continuum of values with meaningful distances between them, but not true zero. The values can be compared directly, but they cannot be used in proportions or mathematical operations (Ex: SAT score, temperature)

Ratio variables

Continuous variables that do have a true zero, and the distance between values can be measured, and values can be expressed as proportions
(Ex: school size, income in US dollars)

Indicator

the values assigned to a variable to provide the blueprint for measurement
(How many weeks/months, wages over the weeks, who you vote for)

Reductionism

a mistake that researchers make by drawing conclusions about the macro-level unit based on analysis of micro-level data

Reliability

a quality of a measure concerning how dependable it is

Validity

a quality of a measure concerning how accurate it is

Composite Variables

variables that average a set of items to measure the same concept

Cronbach's Alpha

a calculation that measures a specific kind of reliability for a composite variable
Score of 0-1, conventional standard is 0.7+ for a highly reliable composite measure

Internal Reliability

the degree to which the various items in a composite variable lead to a consistent response and, therefore, are tapping into the same concept

Intercoder Reliability

reveals how much different coders or observers agree with one another when looking at the same data

Precision

a quality of measurement referring to how detailed and specific it is
more categories \= more precision

Robustness

a quality of an operational protocol referring to how well it works

Split-Half Method

assesses robustness by testing the similarity of results after administering one subset of an item to a sample and then another subset

Test-Retest Method

assesses robustness by administering a measure to the same sample at two different times

Pilot Testing

a method of administering some measurement protocol to a small preliminary sample of subjects as a means of assessing how well the measure works

Internal Validity of a Study

the degree to which the study establishes a causal effect of the independent variable on the dependent variable

Internal Validity of a Measure

the degree to which the measures truly and accurately capture concepts

Face Validity

a dimension concerning whether a measure looks valid

Criterion-Related Validity

concerns how closely a measure is associated with some other factor (Concurrent validity & Predictive validity)

Content Validity

concerns how well a measure encompasses the many different meanings of a single concept.

Construct Validity

how well multiple indicators are connected to some underlying factor

External Validity

concerns the degree to which the results of a study can generalize beyond the study

Probability Sample

A sample chosen via random selection
Random chance is used to select participants for the sample, where each individual has a probability of being selected that can be calculated

Target Population

a group about which social scientists attempt to make generalizations
They do not necessarily refer to groups or individuals. It can also include nations, etc.

Census

a study that includes data on every member of a population

Population Parameter

a number that characterizes some quantitative aspect of a population

Nonprobability Sample

a sample that is not drawn using a method of random selection

Convenience Sample

the simplest type of nonprobability sample, for which the cheapest and easiest observations are selected

Unbiased

the sample estimate is the same as the population parameter except for the difference caused by random chance

Systematic Error

a flaw built into the design of the study that causes a sample estimate to diverge from the population parameter

Margin of Error

the amount of uncertainty in an estimate; equal to the distance between the estimate and the boundary of the confidence interval

Sampling Distribution

a set of estimates that would be observed from a large number of independent samples that are all the same size and drawn using the same method

Confidence Level

the probability that an estimate includes the population parameter (typically 95%)

Confidence Interval

the range of values implied by the margin of error

Sampling Frame

a list of population members from which a probability sample is drawn

Simple Random Sample

a type of probability sample in which each individual has the same probability of being selected, and each pair of individuals has the same probability of being selected

Systematic Sample

a probability sampling strategy in which sample members are selected by using a fixed interval, such as taking every 5th person on a list of everyone in the population
Each pair in this sample does not have an equal chance of selection

Cluster Sample

a probability sampling strategy in which researchers divide up the target population into groups, or 'clusters,' first selecting clusters randomly and then selecting individuals within those clusters

Advantages to cluster sampling (2)

Researchers can take probability samples when a sampling frame-a list of all population members-doesn't exist
Researchers can conduct many more surveys at a much lower cost than they would incur with simple random sampling

Stratified Sample

A type of probability sampling in which the population is divided into groups with a common attribute and a random sample is chosen within each group

Oversampling

a group that is deliberately sampled at a rate higher than its frequency in the population

Weighting

determines how much sample members "count" when producing estimates

Members of stratified samples may be weighted differently, to account for the fact that the sample is no longer representative
A group that is oversampled should receive less weight than other members of the sample

Variable-Oriented Research

scientists study a large number of cases, but gather only a limited amount of data (or variables) about each

Case-Oriented Research

scientists gather large amounts of data about a single case or small number of cases
Choosing a single case to study is still considered sampling!

Purposive Sampling

a sampling strategy in which cases are selected on the basis of features that distinguish them from other cases

Sequential Sampling

enables researchers to make decisions about what additional data to collect based on their findings from data they've already collected

Key Informants

individuals who have intimate knowledge of a subject and are willing to share it with the researcher

Saturation

when new materials fail to yield new insights and simply reinforce what the researcher already knows

Snowball Sampling

a strategy in which the researcher starts with one respondent who meets the requirements for inclusion and asks them to recommend other people to contact

Snowball sampling pros

Useful for studying populations we would otherwise know very little about

Random Assignment

distributes individual differences equally across conditions
Ensures that the only difference between the experimental group and the control group is the independent variable manipulated in the experimental group

3 advantages of experiments

1. The best research method to establish causality

2. Can uncover mechanisms that produce an outcome, explaining both if and why

3. Can be used to evaluate abstract theories about the social world

3 conditions for establishing causality

1. 2 variables must be correlated

2. The cause (independent variable) must precede the effect (dependent variable)

3. The relationship between the independent variable and dependent variable must not be spurious

Concurrent validity

how closely the measure is associated with a preexisting measure

Predictive validity

how closely the measure is correlated with something it should be correlated with

Experimenter Effects

unintended changes in subjects' behavior due to cues inadvertently given by the experimenter

Selection Bias

in an experiment, unintended differences between the participants in different groups

Laboratory Experiments

take place in laboratories, giving researchers the maximum amount of control over the environment in which the experiment is conducted

strengths of Laboratory Experiments (2)

1. high degree of internal validity

2. Highly artificial setting allows researchers to assess causality and test abstract theories

Field Experiments

takes place in a natural or 'real-world' setting
Often used to evaluate the success of interventions to improve educational and health outcomes

Audit Studies

a type of field experiment that assess whether the characteristics such as gender, race, and sexual orientation lead to discrimination in real labor and housing markets

Factorial Design

have two or more independent variables. This allows researchers to measure various characteristics at once