Pol 280 Deck 2

Quantitative Research

Uses numbers and statistical methods

• Associated with large sample size

Examples of quantitative research

Survey research, experiments, inferential statistics, content analysis

Qualitative Research

Uses non-numerical data to understand individuals' social reality (attitudes, beliefs, motivations)

• In-depth, case-oriented study of a relatively small # of cases

Examples of qualitative research

Interviews, focus groups, comparative case studies, Scientific Research

Correlation

A statistical measure of covariation which summarizes the direction (positive or negative) and strength of the linear relationship between two variables Covariation

Causation

A cause and effect relationship in which one variable controls the changes in another variable. Hypothesis

Independent variable

A variable that is theorized to cause variation in the dependent variable

Dependent variable

A variable for which at least some of the variation is theorized to be caused by one or more independent variables

Empirical research

Based on real-world observation -Positivists prefer this

Example of empirical question

Does listening to music once a day improve happiness?

Normative research

Statements about how the world ought to be -Critical theorists

Example of a normative question

Should we listen to music to feel happier?

Variable

A definable quantity that take on two or more values

Positive relationship

Higher values of the independent variable tend to coincide with higher values of the dependent variable (EX: Increased temperatures lead to increased ice cream sales)

Negative relationship

Higher values of the independent variable tend to coincide with lower values of the dependent variable (EX: excessive sugar intake leads to decreased oral hygiene)

Cross-sectional measure

Time dimension is the same for all cases and the cases represent multiple spatial units (EX: GDP in 2005 for 22 random countries - what percentage is military spending?)

Time-series measure

Spatial dimension is the same for all cases and the dependent variable is measured at multiple points in time (EX: monthly level of U.S. presidential approval displayed from 1995 to 2005)

Confounding Variable

A variable that is correlated with both the independent and dependent variables and that somehow alters the relationship between those two variables (EX: Ice cream consumption (X) and homicide rates (Y)... one potential variable Z is average monthly temperature)

Probabilistic relationship

Increases in X are associated with increases (or decreases) in the probability of Y occurring, but those probabilities are not certainties (EX: democracies are highly unlikely to go to war with one another)

Deterministic relationship

If some cause occurs, then the effect will occur with certainty (EX: democracies don't go to war with one another)

Spurious Variables

Not what it appears to be, false, when two variables appear causally related, but are not (EX: Master's degrees and box office revenue: it's more likely than an increasing global population is causing both to increase)

Theory

A tentative conjecture about the causes of some phenomenon of interest

Null hypothesis

A theory-based statement about what we would observe if there were no relationship between an independent variable and dependent variable (EX: animal-assisted therapy has no impact on one's recidivism)

Alternative research hypothesis

A statement that rejects the null hypothesis

Non-directional (two-tailed) hypothesis

When a hypothesis does not state a direction but simply states that one variable affects the other, or that it's correlational.

Selection bias/sample of convenience

A sample of cases from the underlying population in which the mechanism for selection cases is not random

Control group

In an experiment, the subset of cases that is not exposed to the main causal stimulus under investigation

Experimental/ treatment group

In an experiment, the subset of cases that is exposed to the main causal stimulus under investigation

External validity

The degree to which we can be confident that the results of our analysis apply not only to the participants and circumstances in the study, but also to the population more broadly construed

• Replicability, generalizability, representativeness

• Large random sample increases external validity

• Quantitative analyses

Field experiment

An experimental study that occurs in the natural setting where the subjects normally lead their lives

Internal validity

The degree to which a study produces high levels of confidence about whether the independent variable causes the dependent variable

• "Is the relationship truly causal"?

Observational studies

Research designs in which the researcher does not have control over values of the independent variable, which occur naturally, it is necessary that there be some degree of variability in the independent variable across cases, as well as variation in the dependent variable

Random assignment

When the participants for an experiment are assigned randomly to one of several possible values of X, the independent variable

Random sampling

A method for selection individual cases for a study in which every member of the underlying population has an equal probability of being selected

Research designs

The strategies that a researcher employs to make comparisons with the goal of evaluating causal claims

Replication

A scientific process in which researchers implement the same procedures repeatedly in identical form to see if the relationships hold in a consistent fashion

Survey experiment

A survey research technique in which the interviewing process includes experimental randomization in the survey stimulus

Categorical variable

A variable for which cases have values that are either different from or the same as the values for other cases, but about which we cannot make any universally holding ranking distinctions (EX: Religious identification or eye color) - can be nominal or ordinal

Central tendency

Typical values for a particular variable at the center of its distribution

• EX: mean, median, mode

Continuous variable

A variable whose metric has equal unit differences such that one-unit increase in the value of the variable indicates the same amount of change across all values of that variable (EX: household income, the weight of a truck in a truck-weighing station)

Dispersion

The spread or range of values of a variable

Histogram

A visual depiction of the distribution of a single variable that produces a two-dimensional figure in which the horizontal dimension (x axis) displays the values of the variable and the vertical dimension (y axis) displays the relative frequency of cases

Mean value

The arithmetical average of a variable equal to the sum of all values divided by the total number of cases

Measurement metric

The type of values that a variable takes on

Median value

The value of the case that sits at the exact center of our cases when we rank the values of a single variable from the smallest to the largest observed values

Mode

The most frequently occurring value of a variable

Interval variable

Associated with a specific value

• continuous -EX: age (exact difference between 5 and 12)

Nominal variable

-Categorical -No set difference between one and the next -EX: eye color, religion

Ordinal variable

-Categorical -Have a specific order -Not equal unit differences -Lichert scales (strongly disagree -> disagree) -EX: education level

Outlier

A case for which the value of the variable is extremely high or low relative to the rest of the values for that variable

Skewness

A statistical measure indicating the symmetry of the distribution around the mean

Standard deviation

A statistical measure of the dispersion of a variable around its mean, square root

Variance

A statistical measure of the dispersion of the variable around its mean

68-95-99 rule

A useful characteristic of the normal distribution which states that moving +/- 1, 2, and 3 standard deviations from the mean will leave 68, 95, and 99 percent of the distribution's area under the curve

Central limit theorem

A fundamental result from statistics indicating that if one were to collect an infinite number of random samples and plot the resulting sample means, those sample means would be distributed normally around the true population mean

Confidence interval

A probabilistic statement about the likely value of a population characteristic based on the observations in a sample

Frequency distribution

A distribution of actual scores in a sample

Normal distribution

A bell-shaped statistical distribution that can be entirely characterized by its mean and standard deviation

Population

Data for every possible relevant case

Random sample

A sample such that each member of the underlying population has an equal probability of being selected

Sample

A subset of cases that is drawn from an underlying population

Sampling distribution

A hypothetical distribution of sample means

Standard error of the mean

The standard deviation of the sampling distribution of sample means

Statistical inference

The process of using what we know about a sample to make probabilistic statements about the broader population

R-squared statistic

A goodness-of-fit measure that varies between 0 and 1 representing the proportion of variation in the dependent variable that is accounted for by the model

Omitted variable bias

Bias that results from the failure to include a variable that belongs in the regression model

Autocorrelation

Occurs when the randomly determined terms for any two or more cases are systematically related to each other.

• can tell you the wrong dependent variable -two variables that are highly correlated with each other