QUIZ 2-Ch. 2 (1/30) Flashcards
3 measures of central location:
mean, median, mode
calculating mean/average:
sum/count
X̄
“x-bar” represents the mean of a variable
(usually a mean of a sample)
weighted mean definition:
not all values contribute equally
useful when data is displayed in a frequency table
think: hw, tests, quizzes all worth different amounts of your overall grade in class
μ
“mu” represents the population mean
mean definition:
the balance point for a set of data
most commonly used measure of central location
all values used in calculation
mean is affected by:
extreme values (causes skewed distribution)
it gets pulled in the direction of the skew
median definition:
midpoint of a set of data
not all values used in its calculation
how to find a median:
- order the data
- position= (n+1)/2
median is NOT affected by:
extreme values
calculation property of a mean:
Σ(Xi-x̄)=0
will always equal 0!
think: when calculating variance third column (x-x̄) all of those values have to sum up to 0!
mode definition:
the value (s) that occurs most frequently
it’s the only numerical summary that is useful for qualitative data
when skewed right:
mode<median<mean
when skewed left:
mean<median<mode
3 measures of dispersion (variation):
range, variance, standard deviation
range definition:
max-min
weak measure because only looks at the two values on the end
variance definition:
reflects the dispersion of all the values around the mean of the data
it is in squared units of measurement
sample variance calculation:
s^2=Σ(Xi-x̄)^2/(n-1)
when calculating variance table: last column (take sum & divide by n-1)
Empirical Rule: x̄ ±1s
1s= 1 standard deviation
approx. 68% of the observations will lie with x̄ ±1s
Empirical Rule: x̄ ± 2s
2s= 2 standard deviations
approx. 95% of the observation will lie with x̄ ± 2s
Empirical Rule: x̄ ± 3s
3s= 3 standard deviations
approx. 99.7% of the observations will lie with x̄ ± 3s
Chebyshev’s Theorem:
the proportion of observations in any sample that lie within “k” standard deviation of the mean is at least 1-1/(k^2) for k>1
percentiles definition:
% of the data less than or equal to a value
think: SAT if you were in the 80th percentile
Q1
first quartile
25% of the data ≤ Q1
Q2
second quartile
75% of the data ≤ Q3
IQR definition:
Interquartile Range
spread of the “middle 50%” of the data
*IQR=Q3-Q1