Statistics Exam 2 Flashcards
population
the entire group of objects to be studied
sample
a subgroup of a population on which data is actually collected
mean
average
x̅ is mean of sample(statistic)
μ is mean of population (parameter)
median
a number where as much of a list is below it as above it
1) put list in numerical order
2) determine number of observations (n)
3) determine the index(location) of the median
mode
response that occurred most (highest frequency)
*cant be mode if 1 or 0
*can be tied for highest 2
2 modes-bimodel, 2+, multimodel
numerical summary is said to be resistant if…
it is not affected by extreme values
skewed left shape
mean
skewed right shape
median
range
difference between highest and lowest numbers
standard deviation
how far each number in the list is from the average
z scores
x-μ/σ or x-x̅/s
percentiles
the kth percentile is the number where k percent of the data falls below that number (median is 50th percentile)
Quartiles
pieces of data that split the data into 4 equal parts
Q1 and Q3 are medians of
the lower and the upper lists receptively
*if mean is a number in the list include it in upper and lower
Interquartile range
Q3-Q1
Upper fence
Q3+1.5(IQR)
Lower Fence
Q1-1.5(IQR)
A piece of data is an outsider if…
it is outside the upper or lower fences
Five number summary
list of minimum, Q1, median, Q3, and maximum (no commas)
Box plot
Use numbers from 5 number summary, box is Q1, median and Q3, whiskers are up to numbers that don’t go past fence
Shape determined by box plot
If line is in middle of the box its symmetric
to the left distribution is skewed right
to the right distribution is skewed left
Scatter plot
a graph in which the explanatory variable is on the horizontal axis and the response variable is on the vertical axis and we plot a point for each indv. in the study
Positive Correlation
↑↑ or ↓ ↓
Negative Correlation
↑↓ or ↓↑
Explanatory Variable
The predictor variable
Response Variable
The affected variable
r being close to 0
no correlation
Residual Value
the difference between expected and observed values of y
Least-Square Regression Line
y^b1=bo
where b1= r x sy/sx
and bo= ȳ-(b1)(x̄)
Squared residuals
measure how well a line fits
linear correlation coefficient (r)
r= Σ(zxi)(zyi)/n-1
