Please enable JavaScript.
Coggle requires JavaScript to display documents.
Definitions - start of descriptive stats - Coggle Diagram
Definitions - start of descriptive stats
Functions
Groups of pre-written code that are pre-defined
they all have a name followed by a pair of brackets , filled with arguments
return output to console(if defined) e.g cbind
Meaning can be checked by using a question mark after the function name in the console
Function_name(argument_1 = input 1, argument_2 = input 2)
Arguments
Input points for required info
Input names are variables - chosen by author
Include argument names even when not needed
Tells the function or variable all of the info e.g time = 60
Accept variables. numeric values, text strings or logical values
Data types
Numeric e.g 1,2,3,4,1.4
Text strings e.g. university of Bristol
Logical values = TRUE, FALSE
Variable names e.g Betsy = ....
comments - done with hashtag and don't mean anything -just there to describe code e.g. ## this line shows
Quantification
data + analysis
provides empirical evaluation - does it support or refute hypothesis
First collect primary data
Secondary data - generally less reliable - usually for different purpose
gov data is quite reliable
can get data from primary or secondary sources
analysis allows us to see patterns/relationships
Geographical data frame
= case(number or name) +geographical reference(place)+ variables
Variable = attributes/measurements on the cases
continuous measurement = decimal point data
Discrete measurement = qualitative text label e.g. poor or very poor
from data frames, different data can be drawn to create relationship or change
Data
Signal + noise
Signal = main trend
noise = numbers below and above trend
Analysis = techniques to uncover the signal
To find the signal, application of technique e.g., rounding is often used
Some times changing axis on a graph can show the signal
Reasons for data analysis
Inference
Description/ exploration
uncovers patterns in data to generate research questions
Summarise 1 or 2 representative values
expose anomalies
provide faithful picture of data
to convey pattern to non-specialist audience
modelling a relationship
aims to uncover underlying relationship between variables in presence of uncertainty, noise etc....
how close are the relationships
what would happen if
useful to see how relationships ay go both ways
Aim of data analysis
Distinguishes a pattern from chance results
Whether it's significant and so likely to be genuine
How significant? did it occur by chance
Different techniques are used for different purposed