Set of procedures to organize, summarize, and communicate information

GOAL: learn something about empirial reality

Information represented by numbers;

can be subject of statistical analysis

1. Ask a question

2. Formulate hypothesis

3. Collect data

4. Analyze data

5. Evaluate hypothesis

The variable to be explained;

outcome variable;

the effect

Explanatory variable;

Predictor variable;

the cause

Numbers assigned to non-ordered categories with no quantitative values

Ex:  Male=1  Female=2

Numbers assigned to rank ordered categories ranging from low to high

Ex: Social class

Working=1  Middle=2  Upper=3

Responses that are expressed all in the same unit and have quantitative value

Ex: Age, Income, SAT scores

Pros: ask a variety of questions, can change questions

Con: different people each time

Pros: same people over time (gives a better idea of cause & effect)

Cons: same questions and attrition

Phone

Internet=most honest answers

Mail

Face-to-face=highest response rate

Combination

Organize and describe data from a sample and/or population

-Measures of central tendency (mean, median, mode)

-Measures of variation (IQV, variance, standard deviation)

Inferential statisics;

Surveying a small portion that is supposed to represent the population

formula: % = f/N*100

f=frequency

N=total # of cases

Nominal-categories don't need to be in a specific order

Ordinal-categories must be in a logical order (low to high)

Interval-ratio-few categories, show frequencies and %s; many categories, combine categories

Divide the # of actual occurences in a given time period by the # of possible occurences

Ex: Divorce, Murder, Unemployment

-Proportion/Percentage

-Per 100, p 1,000, p 100,000 people

#s that describe what is average or typical of the distribution

Mean, Median, Mode

Category with highest frequency(%) in distribution

-Reflects most important element of distribution

-Nominal and Ordinal variables

Score that divides distribution into 2 equal parts

1. Rank from low to high

2. Odd #s=(N+1)/2

3. Even #s=(N+1)/2 then take 2 scores & divide by 2

-Ordinal variables (Interval-ratio when distribution skewed)

The average

-Add up each case & divide by the total # of cases

-Interval-ratio varibles only

#s that describe diversity or variability in the distribution

-IQV, Range, Interquartile range

measure of variability for nominal variables

-always between 0 & 1

0=no variation; all cases in 1 category

1=max variation; all cases evenly distributed

-For interval-ratio

-The average of the squared deviations from the mean 

[image]

The square root of the average of the squared deviation from the mean (the square root of the variance)

-0=No deviation

-Greater the #, the more variation or distribution