Statistics for the Social Sciences

variable

a property of people or objects that takes on two or more values

dichotomous variable

can only take on two values

discrete variable

cannot be subdivided

continuous variable

can be subdivided

three conditions for determining causality

correlation, time order, nonspuriousness

conceptualization

the process of defining what we are interested in

dimensions

different aspects of a single concept that help us measure the concept in its totality

indicators

variables that indicate the presence of the concept that we have chosen to study

population

a group that includes all the cases in which the researcher is interested

sample

a subset of cases selected from a population

sampling unit

the thing you are actually selecting

probability sampling

specifying the probability of selection for each case

descriptive statistics

procedures that help us organize and describe data collected from either a sample or a population

inferential statistics

the logic and procedures concerned with making predictions or inferences about a population from observations and analyses of a sample

systematic random sample

a method of sampling in which every kth member in the population is chosen for inclusion in the sample after the first member of the sample is selected at random from among the first k members of the population

stratified random sample

a method of sampling obtained by dividing the population into subgroups based on one or more variables central to our analysis and drawing a simple random sample from each of the subgroups

weighting

fixes the problem of bias by trying to make the sample match the demographics of the population

multistage cluster sample

samples members/elements grouped geographically

panel data

multiple time periods with same people

simple random sample

a sample where every sampling unit and combination of sampling units within the population has an equal chance of being selected

frequency table

shows how respondents answer a question

positive skew

when the mean is higher than the median

negative skew

when the mean is lower than the median

dispersion

how widespread a variable’s distribution is; measured using standard deviation

variance

average of the squared deviations from the mean

squared deviation

a measurement of dispersion that increases the influence of outliers

percentage within first, second, and third standard deviations

68, 95, 99

descriptions of interval-ratio variables

should describe the variable, include the mean and median, state if there is a skew, and say the standard deviation and range

descriptions of nominal variables

should describe the variable, include the mode, state if there is a skew, and say the standard deviation and range

descriptions of ordinal variables

should describe the variable, include the mean and mode, state if there is a skew, and say the standard deviation and range

descriptions of crosstabs

should disclose the relationships, percentages and percentage point differences in different categories, interpretation of chi-square tests, strength of the relationship, patterns within the percentages, and if there is evidence of a strong relationship between variables

bivariate analysis

used for interval-ratio dependent variables, dummy/dichotomous, and nominal or ordinal (categorical) independent variables

t-tests

used for nominal and ordinal independent variable and interval-ratio dependent variable

dummy variable

a dichotomous variable that is coded 0 and 1

Pearson’s R

measurement of correlation

bivariate analysis

used with only one independent variable

multivariable regression

discusses multiple independent variables and correlation coefficients

matrix

shows the correlation between multiple independent variables and a dependent variable

description of multivariable regression

should contain the regression coefficient, the average increase in the dependent variable by increase in an independent variable, and the independent variable

standardized coefficient

used when comparing the effects of different independent variables

R²

the percentage change in an independent variable in multivariable regression explained by the model