Stats: Identifying Levels of Measurements + Sampling Methods

nominal

lowest form of measurement

Data catergories must be exhaustive (each datum will fit into at least one category) 

Example: Gender, Race/ethnicity, eye color 

Nominal level data - mode 

ordinal

Includes categories that can be rank-ordered 

Categories must be exhaustive and mutually exclusive

Each category must be recognized as higher or lower of better or worse than another category (ex: pain scale, grades) 

Do not know exactly how much higher or lower one subjects score is in relation to another subjects score

Could be either words or numbers 

Ordinal level data - median 

interval

distance between intervals of the scale is numerically equal 

No absolute zero 

Continuous variable 

Ex: temperature 

ratio

Highest form of measurement 

Numerically equal intervals of a scale 

There is an absolute zero 

Continuous variables 

Ex: pulse, bp, age

parametric statistics

Distribution of scores is expected to be a normal distribution or approx normal 

means and standard deviations

use interval or ratio levels

Pearson’s correlational coefficient

determines relationships between variables

Significant results

usually identified by * or p values less than or equal to 0.05

population

particular group of individuals being studied

target

determined by sampling criteria

sample

focus of particular study

power

probability that statistics can detect relationships or differences in population studied

alpha/level of significance

usually 0.05

standard power

usually 0.80 or 80%

effect size

strength of relationships

refusal rate

(# refusing to participate/# approached) 100

attrition rate

% of students dropping out of a study or passively when lost to follow up

(#dropping out/total sample size) 100

null hypothesis

hypothesis that suggests there will be no statistically significant effect on variables being studied

alternate hypothesis

hypothesis that observations are influenced by non-random elements, hypothesis that researcher is interested in

sampling method

process of selecting people, elements, behavior, and being representative of population being studied

simple random sampling

random selection of numbers from sampling frame

stratified random sampling

used when researcher knows some variables to affect representativeness of sample (ex: age, gender, ethnicity, medical diagnosis

cluster sampling

used when time/travel are necessary

systematic sampling

selection through process that accepts every nth member on the list using randomly selected starting pt

convenience sampling

most frequently used

enroll subjects who are accessible and available to participate in study

quota sampling

used to ensure adequate representation of all types of subjects likely to be underrepresented

purposive sampling

occurs when researcher consciously selects subjects, elements, events, or incidents to include in study

network/snowball

makes use of social networks and fact that friends have common characteristics

theoretical sampling

data gathered from individuals or groups who can provide relevant info for theory generation