Stats- levels of measurement

what are levels of measurement

rules for assigning numbers to objects

what type of number has the lowest number of rules

nominal

which type of number has the highest number of rukes

ratio

order from lowest to highest number of rules

nominal, ordinal, interval, ratio

rules for nominal categories (lowest form)

• must be exclusive (fits into only one category)

• must be exhaustive (fits into at least one category)

example of nominal

gender, race/ ethnicity, picking fav food

ordinal

categories that can be rank ordered

ordinal rules

• Categories must be exhaustive and mutually exclusive

• each category or box must be recognized as higher or lower or better or worse than another category

• there is not a set number between one subjects score and another score

• do not have a continuum of values with equal distances between them

• pain scale ( no set distance between numbers in pain scales)

what are Nonparametric or distribution-free analysis techniques used for

to analyze nominal and ordinal level variables to examine relationships

what are assumptions for Nonparametric or distribution-free analysis techniques

• Values dont need to be normally distributed

• The level of measurement is usually ordinal or nominal

nominal level data measurement of central tendency

mode

mode

most frequently occuring in a data set

Ordinal Level Data: Measurement of Central Tendency

median

median

middle value in a data set

what are descriptive statistical analyses

frequencies and percentages

what are frequencies and percentages used to describe

demographic variables

what is range used to determine

the dispersion or spread of values of a variable measured at the ordinal level

interval

The distances between intervals of the scale are numerically equal

interval rule

• There is no absolute zero

• The score of zero does not indicate that the property being measured is absent (can go negative)

• continuous variable

• ex: temp

ratio is the

highest form of measurement

ratio rules

• numerically equal intervals of a scale

• absolute zero 0 and up

• continuous variables ( same distance between each number)

• age, blood pressure, pulse

what are parametric statistics conducted on

interval and ratio levels of data

what do parametric statistics describe

variables, examine relationships among variables and determine difference between groups

parametric statistics assumptions

• Distribution of scores is expected to be a normal distribution or approximately normal

• Variables are continuous, measured at the interval or ratio level

• Data can be treated as though they were obtained from a random sample

significant results for stats

p values less than or equal to 0.05

levels of measurement review

nominal variable analysis

chi-square

ordinal variable analysis

spearman ranked order coreelation coefficient and the mean whitney U and wilcoxen signed ranks

what are parametric statistics

powerful analysis

means and standard deviations

use interval or ratio levels

pearsons correlation coefficient

determines relationships between variables

T-test and ANOVA

calculated to determine significant differences between groups

what happens when mean, medium mode is positively or negatively skewed

Positvely skewed: mode, medium, mean (high to low)

negative skewed: mean medium mode (low to high)