Class Notes (1,100,000)
CA (630,000)
U of G (30,000)
SOAN (900)
SOAN 2120 (400)

SOAN 2120 Lecture Notes - Central Tendency, Decimal Mark, Square Root

Sociology and Anthropology
Course Code
SOAN 2120
David Walters

This preview shows half of the first page. to view the full 3 pages of the document.
September 25, 2012
Standards Deviants
- measure of central tendency
- measures of dispersion
- population versus sample
- z-scores and probability
Statistics language test simple math relationship b/w terms
- Taking a portion out of that data and studying it
- Small groups tell us about big groups
- How to make sense of data
Statistical Problems
- exists when something is unknown about a population
- a population is the group you are interested in learning about
o population can be animate or inanimate it’s just a group
o large group you gather the data from
- inferences drawing conclusions from a piece of data
- sample randomly selected portion of the population
- step 1 Identify question being asked
- step 2 Determine design of experiment (use whole population or sample?)
- step 3 establish a method to collect and Analyze data
o data set
- step 4 choose a procedure for making Inferences
- step 5 see how Reliable your inferences are (test accuracy)
- “IDAIR” – identify, determine, analyze, inference, reliable
Graphical Display
- relative frequency histogram
o divides the data into intervals/classes
o put data into those classes
o box-like display of data
o classes, class width, class boundaries, class frequency, class relative
frequency etc.
o knowing the range = how many classes to use
too few classes = not good. Values very different in size are
limped together and then the histogram is too big
too many classes = most empty, histogram becomes misleading
o class width how wide classes are --- divide range by the number of
classes we want if it’s not a whole number, round up
all classes need to be of equal width
class boundaries where class starts and ends
- distribution curve
o essentially the same as histogram
o bigger data sets, looks like a bell
- class relative frequency = number of measurements in a class divided by total
number of measurements
You're Reading a Preview

Unlock to view full version