Chapter 6: Generalizations Flashcards
Generalizations
based on fleeting experiences, impressions or anecdotes.
Often have weak evidence, but can be difficult to eject from our assortment of beliefs.
Statistical Inference
inferences about a population based on a sample of data
Statistical generalization
an inference made about a population based on features of a sample
moves outward from facts about a group to draw conclusion about the group at large
Statistical Instantiation
an inference made about a sample based on features of a population
moved inward from a fact about the larger group to draw a conclusion about a sample
Convenience Sample
a set of observations that is small and carelessly selected
Sampling Bias
a selection effect in a sample created by the way in which we are sampling the population
Sample Size Matters
Small samples provide insufficient evidence
Appropriate sample size leads to higher strength factor
i.e. probabilities of (E given H) and (E given not-H) are very different
The Law of Large Numbers
The larger a sample, the more likely its proportions reflect the population as a whole
How Big is Big Enough For a Sample?
Whether a sample is big enough depends on how narrow test’s margin of error needs to be
Margin of Error
the size of the confidence on either side of the given value
Confidence Interval
the margin in which true population percentage falls, with that degree of confidence
Survey Pitfalls
a) participation biases
b) response bias
Language of survey can create these biases
Measures of Centrality:
Arithmetic Mean: the average of values
Median: the midpoint on spectrum of collected values
Mode: the most common value
Geometric Mean
Helps track changes in proportions of each value
Gives best “central” value
Difference Between Arithmetic Mean and Geometric Mean
Arithmetic mean: adding up values and dividing by number of cases
Geometric mean: multiplying values and using number of cases as root