Week 8 --- Univariate Analysis

Univariate Analysis

descriptive statistics

data analysis used to describe a sample

inferential statistics

data analysis used to predict and infer about the variables

univariate statistics analysis

only one variable examined at a given time

bivariate statistics

two variables, examined in relation to each other

multivariate statistics

3 or more variables, in relation to each other

value distribution

frequency, central tendency, and variance

frequency

how often values occur

nominal and ordinal variables

frequency count

the total number of times a specific value, event, item, or behavior occurs within a given dataset or observation period

central tendency

• how values of a variable cluster around a center point

• mean, median, and mode

• usually interval and ratio

proportion

(frequency count)/total sample

sample

the entire group

subsample

smaller group within the sample

variance

• how values are spread in the data

• summary of overall spread of all values from the mean

• average distance of each value from the mean

standard deviation

square root of variance, a specific distance away from the mean, more commonly identifiable and translatable

frequency distribution

visual representation of all values in a data set

unimodal curve

bimodal curve

multimodal curve

symmetrical curve

skewed curve

kurtosis

• a statistical measure that describes the "tailedness" of a data distribution

• tells you whether a dataset has heavy tails (more extreme outliers) or light tails (fewer extreme outliers) compared to a normal bell curve

parts of a curve

normal distribution

• standardized pattern of distributing values, the same ratio every time

• same value for mean, median, and mode

• unimodal and symmettrical

• mesokurtic

z-score formula

z-score

measures how many standard deviations a specific data point is away from the average (mean) of a data set

empirical rule

• states that for any data set that follows a normal distribution (bell curve), almost all data will fall within three standard deviations of the average

• 68 - 95 - 99.7

percentile rank

• shows your relative position compared to a group of peers

• percentage rank in comparison to other values