Statistics - Full Review

raw data

unprocessed info collected for a study

variable

quantity being measured

two types of variable

1. continuous

2. discrete

continuous variable

any val within a range

discrete

limited number of values (countable) → 1, 2 ,3, 4

qualitative research

analysis of data such as words, pictures, objects

quantitative research

analysis of numerical data (measurable)

bar graphs

category based, order doesn’t matter

histograms

looking at numerical values

order matters

no gaps, unlike bar graphs although looks similar to bar graphs

frequency polygon

• looks like a line graph

• cojmpare diff data between diff groups

cumulative polygon

looks like frequency polygons except in an upward trajectory

circle graph

literally a circle

grouped data

it’s grouped bro based on category

sampling frame

teh group of individuals who can actually be selected

stat techniques should be:

RANDOM

simple rrandom sample

• items selected randomly

• everyone has an equal chance of being selected

systematic sample

work through the popu sequentially selecting items at regular intervals

• randomly select a staring item in the first interval

• interval length = popu size/sample size

stratified sample

• popu is split into groups of members who share similar traits

• a strat sample has the same PROPORTION of members from each stratum as the popu does

cluster sample

• popu is divided into groups (clusters) and a random sample of these clusters are selected

sampling bias

occurs when the sample frame does not reflect the characteristics of the popu

non-response bias

occurs when particular groups are underrepresented bc they had the choice not to participate

measurement bias

when data collection method consistently under/over estimates a characteristic of the popu

response bias

when participates in a survey deliberately give false/misleading answers

outliers

extreme values that can greatly skew the mean

deviation

difference between an individual value and the mean of the data set