BILD 5: MIDTERM 1

categorical

identify: in the context of R, identify fxn of $ (operator)

serves to extract specific columns/components of a large, complex data set.

identify: in the context of R, names of columns & rows

columns; variables. row; observation.

define: in the context of R, what’s an object?

a stored value/data structure that can be manipulated/analyzed/reused.

identify: in the context of R, what are the allowed characters to name an object? (letters, digits, punctuation [2]?)

letters A-Z, digits 0-9, the period, and the underscore

identify: in the context of R, what are the three conventions of naming an object?

name must begin with a letter, no spaces, case-sensitive

define: in the context of R, define the fxn of print()

outputs text of an object

define: in the context of R, define the fxn of c()

creates list (vector)

define: in the context of R, define the fxn of userinfo()

runs a background check on the person using the computer

define: in the context of R, define the fxn of mean()

returns sample mean

define: in the context of R, define the fxn of runif()

returns a random number

define: in the context of R, define the fxn of str()

summarizes a data set

define: in the context of R, define the fxn of boxplot()

draws a box-and-whisker plot

define: numerical data

expressed in numbers

define: categorical data

observations sorted into groups.

define: core bias

treating one study as whole, accurate representation for one population

identify: amplifiers of core bias (4)

authority bias, illusion of explanatory depth, black-and-white thinking, confirmation bias

define: anchoring bias

over reliance on the first piece of information recieved “anchor”, which serves as a baseline of future comparison & judgment

define: dunnning-kruger effect

dumb people overestimate their intelligence and skill

define: PPDAC (investigative cycle)

problem, plan, data, analysis, conclusion

define: bar plot

one continuous variable +

define: scatterplot

relationship between two continuous variables

define: line graph

relationship between two continuous variables; usually one variable is ordered

define: histogram

distribution of one continuous variable

define: box plots

Shows the distribution of numerical data using quartiles, median, and potential outliers.

define: violin plots

Shows the distribution of numerical data across categorical groups.

define: ordinal data

ex: order of higher education certifications

categorical data with meaningful order

identify: types of numerical data

continuous, count

identify: types of categorical data

ordinal, binomial

define: continuous data (NUMERICAL)

a measurement that can take any value within a range

define: count data (NUMERICAL)

a number representing how many of something there is

define: binomial data (non numerical)

a category with two different outcomes

define: median

average

define: median

middle of all ranked values

define: mode

the most common value

define: robust

an overall measure being resilient— less susceptible to change because of outliers

define: central limit theorem

the distribution of a sample mean approaches a normal distribution as sample size increases, regardless of the original shape of the original population distribution.

define: standard error of mean

measures how much a sample mean is expected to vary from the true population mean

define: standard deviation

measure of how much the values in the data set deviate from the mean

define: negative control

sample treated with factor known to produce no effect

define: positive control

sample treated with a factor known to produce a specific, expected outcomed

define: p-value

probability of observing an effect– or something more extreme– under the null hypothesis

define: signal

observed component which carries significance as to understanding the true underlying pattern

define: noise

amount of variation between groups

define: z-score

measures how many standard deviations a data point is from the mean

define: t-score

measures how many estimate errors a sample ean is away from a hypothesized population mean

define: chi-square statistic

does the observed data match an expected distribution?

define: R command for the mean of a colum

mean(data$column)

define: IQR

Q3-Q1, spread of the middle 50% of observations

define: 95% confidence interval

range of numbers that is likely to contain the true population value

• so if you ran this experiment 100 times, 95 of them would successfully capture the true population

define: power

the probability of correctly detecting a real effect

define: power equation

1 - beta, beta = the probability of a type II error (false negative)

define: effect size

a measure of how prominent a difference is between groups and/or variables

define: alpha

the significance threshold chosen before a hypothesis test (p = 0.05)

define: type I error

false positive; a test detects an effect that doesn’t actually exist— rejecting the null.

define: type II error

false negative; test fails to detect effect that actually exists— failing to reject a false null hypothesis

what constitutes tidy data?

each variable is a column, each observation is a row, and each value is a cell