descriptive statistics, parametric and non parametric

frequency distribution

number of times a score or value occurs within a set of data

absolute frequency

how many times each score occurs

relative frequency

absolute frequency/total

mean

what to have to use

sum of measurements/number of measurements

have interval or ratio to use

what is median used with

ordinal data

what is mode used with

nominal data

variability of the data set

how far away from the central point the rest of the data lie

what does variability of data include

• range

• variance

• standard deviation

how to calculate variance

• find mean

• mean-each measurement and square result

• find average of squared difference

standard deviation and how to calculate

absolute distance all data points are from mean

square root of variance

central tendency includes

• mean

• median

• mode

frequency includes

• absolute frequency

• relative frequency

what does distribution include

• range

• variance

• standard deviation

z-score

describe an individual compared to a know distribution

how to calculate z-score

z=X/S

x=individual score-mean

s=standard deviation

negatively skewed curve

tail pulled to left

pulls mean and median to left

p-value

how much risk youre willing to take in committing type 1 error

null hypothesis

there is NO relationship/difference

alternative hypothesis

there IS a difference/relationship

type 1 error

when there is no difference/relationship and you say there is

type 2 error

when there is a difference/relationship and you say there is not

types of inferential stats

• analysis of relationship

• analysis of difference

steps of choosing an analysis of difference stats

• how many groups

• are groups related

• determine readiness

groups that are independent of eachother

not related- different groups

groups that are dependent of eachother

related-same group

same group measured at two points of time

parametric tests

uses SD, mean, variance

normally distributed

interval or ratio data

non parametric tests

uses rank or frequency, median

not normal distribution

nominal or ranked data

parametric test used for not related , one independent variable with two levels

independent t test

parametric test used for related, one independent variable with two levels

paired t test

parametric test used for related one independent variable with three or more levels

repeated measures ANOVA

parametric test used for non-related one independent variable with three or more levels

one way ANOVA

non parametric alternative tests to independent t test

• mann-whitney (ranks)

• wilcoxon rank sums (rank)

non parametric alternative test for one way ANOVA

kruskal wallis (ranks)

non parametric alternative test for paired t test

• wilcoxon signed rank (ranks)

• mcnemar (frequencies/nominal)

non parametric alternative test for repeated measures ANOVA

friedmans ANOVA (ranks)

what does an independent t test allow you to test

null hypothesis

p-value of >0.05

accept null hypothesis (no difference)

p-value of <0.05

reject null hypothesis (there is a difference)

what study is a paired t test used in

pre-test/post-test intervention

one group is measured at two different times

post-hoc test

use when ANOVA says there is a difference

tell where the difference is

what is difference between one way and two way ANOVA

two way is the difference between two independent variables _

more likely for the data to be normal and representative of the population when ___ of participants in a study

large number

what does a chi-square test determine

if a distribution of observed frequencies differs from expected frequencies