Statistics, Census, GIS-Maps Flashcards
definition / population
the group you want to generalize to
entire set of persons with at least one common characteristic
sample selected from population
definition / sample
subset of population
too expensive and time consuming to collect data on entire population66
UD Decennial Census
attempts to count every person in short for m
what makes a survey valid statistically
must employ a method everyone has a measureable chance of being selected
definition / sampling error
error expected in probability sampling
composed of population parameter
sample size
standard error
difference than bias
what is typical margin of error
in simple cases, margin of error in a sample with 95% confidence equals
1/ sqrt(x) with x being number of respondents
i.e 1,000 samples - 0.0316
definition / sampling frame
listing of the accessible population from which you’ll draw your sample
example - phone book
non-probability sampling (two types)
non-probability sampling…convenience sampling - only those who are accessible
volunteer sampling - people volunteer to be sampled
ok for descriptive results, introduces bias
probability sampling (4 types)
simple random - each individual has equal chance of being selected
systematic - every xth individual is selected from list, starting at randomly chosen point
stratified - population may have two or more groups in study. provides best results because it ensures even coverage of population but maintains random selection probabilities
cluster - used when stratified or simple random sampling would be difficult and/or expensive
definition / mode
most frequent score in a distribution, can have more than 1, can be bi-modal
definition / mean
sum of scores divided by number of scores
definition / median
number occurring in middle
definition / range
difference between highest and lowest score
variance
measure of how spread out a distribution is
average squared deviation of each number
standard deviation
square root of variance
68% within 1 standard deviation
95% should be within 2 standard deviations
99% within 3 standard deviations
definition/dependent variable
the fact the researcher is interested in explaining (number of trips)
definition/independent variable
variable used to explain variation (GFA, number of units)
definition/ bias
difference between expected value of an estimator and real value of the parameter
sampling error vs bias
sampling error is natural consequence of sample size being smaller than population
bias is due to faulty design of survey. not a result of sample size so increasing sample size will not reduce bias in estimates
two categories of bias
measurement bias - due to errors in sampling
non representative sampling bias - also known as selection bias due to not random methods during selection. convenience sampling often happens here. can happen when non responses from certain groups is significant.
measurement bias (3 causes)
error occurring when recording data
intentionally asking leading questions, framing question in way to lead responses
inadvertent false responses by respondents (d0n’t understand q)