Class Notes (1,100,000)

CA (630,000)

U of G (30,000)

SOAN (900)

SOAN 2120 (400)

David Walters (100)

Lecture

# SOAN 2120 Lecture Notes - Central Tendency, Squared Deviations From The Mean, Decimal Mark

by OC97365

School

University of GuelphDepartment

Sociology and AnthropologyCourse Code

SOAN 2120Professor

David WaltersThis

**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

Video

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