Midterm 1 Flashcards
(59 cards)
Statistics
science that involves the
-collection of data
- organization
- analyzing
- interpret
collection of data examples
- interviews
- questionaiires
- calls
- conducting experiments
organization of data examples
- graphs
- bar graph
- histograms
- scatter plots
- dot plot
- stem leaf plot
- lowest to highest
- five number summary
analyzing data means
- to make conclusions
- normal distribution
- testing hypothesis
discrete data
- finite numbers
- whole numbers
continuous data
- infinite
- mixture of whole and decimals
qualitative data broken down into:
ordinal and nominal
nominal data
- no organization to the order
- religion, race, flavours
ordinal data
- order, mathematical sense
- order of grades: A, B, C, D
- ranks (junior officer, senior officer)
- flavours (how good they are)
interval data
- quantitative broken down further
- no natural zero, zero has no meaning
- temperature
- years
ratio data
- quantitative broke down more
- has natural zero
- bank account
- rental cars
- time of zero for delivery
sampling
process of getting samples for analysis
statistic
characteristic of a sample
parameter
characteristic of a population
sampling methods
random
systematic
stratified
cluster
convinence
random sampling and advantage
put everything into basket and picking, no bias.
n must be equal or greater than 30.
- without replacement
- advantage- allows equal chances for all samples, not biased
systematic
- order, every third for fourth person, pick randomly
- selecting the kth item
stratified sampling and advantage
- put the communities into similar characteristics (S, N, E, W)
- those are statas
- then you go to each strat and do random sampling
- n will be properly represented
- youre sure you will get north side representation
convinence sampling and disadvantage
pick samples ased on info that is already out there. take your sampling form the good side of town, etc
- biased
cluster sampling and disadvantage
similar to stratified
- put into clusters (just like stratas)
- but instead of random sampling, you pick everyone out of the clusters.
- there are so many people, ususally not the best method
descriptive stats
- organizing and analyzing
inferential stats
- conclusion
- interpreting the data
- testing hypothesis
Range
- every set of data has a range
= H-L
outliars
extreme values