CH 1 : Introduction To Social Statistics Flashcards
Variable
Is not constant and can vary, something that varies for everybody. E.g. Age (like a scale, a number line)
Imagination
Is connected to variables, try imagining a group with a mean age of 5. What would the group look like?
Inferential Statistics
Inferences are usually made while describing something for example the political views of all Canadians uses a sample to infer the views of the whole population.
Descriptive Statistics
Summarizes one variable, summarizes the relationship between two variables
Summarizes the relationship between 3 or more variables
One variable
Univariate
Two variables
Bivariate
3 or more variables
Multivariate
Score
A value of the variable
Inferential statistics
Generalizes and infers from a sample to a population
Bivariate
Describes the strength and direction of a relationship between 2 variables
Univariate
Tools to summarize or describe the distribution of a single variable
Multivariate
Describe the relationships between three or more variables
Population
All cases in which the research is interested
Sample
Carefully chosen subset of the population
Ways to classify variables
A. Independent or dependent
B. Discrete or continuous.
C. Levels of measurement: Nominal, Ordinal or interval-ratio variables
Independent Variable (X)
Cause = independent variable.
Cause produces something else ‘
‘If I do this, that happens”
The independent variable is what’s making something happen.
Dependent variable (Y)
This is the outcome you are trying to improve. It depends on the other variable.
When the cause changes it also changes the effects.
Discrete
Measures in definite units that cannot be subdivided e.g. the number of people in a room or home
Continuous
Variable measured in units that can be subdivided infinitely
e.g. Age, weight
Level of measurement
Looking at one variable at a time. Mathematical quality of the value of the variable (score)
Nominal level of measurement
Classifying observations into categories (Names)
Scores are different but cannot be treated as numbers. Has no intrinsic value (so basically names)
Criteria for Nominal Level Variables
Mutually exclusive: Only one category for each case
Exhaustive: A category must exist for every possible value of a variable that can be found
Homogenous = Categories should include cases that are comparable ( a general sense of being the same, would you say these values all belong in the same category?)
Ordinal level variables
Values of the variables that can be ranked from high to low or more to less
Scores represent only one position with respect to other scores ( one score is above the other, the other is below another)
Usually survey items measuring opinions and attitudes.
Interval-Ratio Level Variables
Scores are actual numbers that have equal determined intervals between them
The distance from score to score is exactly defined
E.g. Age (In years)
Income (In dollars)
Number of children