Engineering Data Analysis Reviewer  Ratio: scales that have measurable

intervals
Chapter 1
 Data: facts and figures from which Chapter 2
conclusions can be drawn
 Data set: the data that are collected for a Graphically Summarizing Qualitative Data
particular study
Elements: may be people, objects, events, or  Frequency distribution: A table that
other entries summarizes the number (or frequency) of
 Variable: any characteristic of an items in each of several non-overlapping
element and may change from one object classes
to another in the population.  Relative frequency: summarizes the
 Univariate: data set consists of proportion of items in each class
observations on a single variable.
 Bivariate: data when observations are Formula:
made on each of two variables.
 Multivariate: data arises when
observations are made on more than one
variable. Steps in Constructing a Frequency Distribution
 Measurement: A way to assign a value of
a variable to the element 1. Find the number of classes
 Quantitative: the possible measurements 2. Find the class length
of the values of a variable are numbers 3. Form non-overlapping classes of equal
that represent quantities width
 Qualitative: the possible measurements 4. Tally and count the number of
fall into several categories measurements in each class
 Cross-sectional data: Data collected at 5. Graph the histogram
the same or approximately the same point
in time Contingency Tables:
 Time series data: data collected over
different time periods Classifies data on two dimensions
 Rows classify according to one dimension
Data Sources  Columns classify according to a second
dimension
 Existing sources: data already gathered Requires three variables
by public or private sources like  The row variable
o Internet  The column variable
o Library  The variable counted in the cells
o US Government
o Data collection agency Scatter Plots:
 Used to study relationships between two
 Experimental and observational variables
studies: data we collect ourselves for a  Place one variable on the x-axis
specific purpose  Place a second variable on the y-axis
 Place dot on pair coordinates
Response variable: variable of interest
Independent Variable: related to the variable Types of Relationships
of interest and will be measured.  Linear: A straight line relationship
Experimental study: able to manipulate the between the two variables
independent variables  Positive: When one variable goes up, the
Observational: unable to control IV. other variable goes up
 Negative: When one variable goes up,
Population and Samples the other variable goes down
 No Linear Relationship: There is no
 Population: A set of all elements about coordinated linear movement between the
which we wish to draw conclusions two variables
 Census: An examination all of the
population of measurements Chapter 3
 Sample: A subset of the elements of a
population Measures of Central Tendency
 Mean,  : The average or expected value
 Descriptive Statistics: The science of  Median, Md : The value of the middle
describing the important aspects of a set point of the ordered measurements
of measurements  Mode, Mo: The most frequent value
 Statistical Inference: The science of
describing the important aspects a set of Measures of Variation
measurements  Range: Largest minus the smallest
measurement
Scales of Measurement  Variance: The average of the squared
 Nominative: tags or named variables deviations of all the population
 Ordinal: ordering and ranking data measurements from the population mean
 Interval: known equal intervals of the  Standard Deviation: The square root of
same distance the variance.
Empirical Rule  The mean of all these returns is the
68.26% : within one standard deviation means calculated as the geometric mean:
95.44%: within two standard deviation means
99.73%: within three standard deviation means Formula:

Z-scores: the number of standard deviations

that x is from the mean
o A positive z score is for x above the mean Chapter 4
o A negative z score is for x below the mean  Experiment: is any process of
o The mean has a z score of zero observation with an uncertain outcome
 Sample space: The possible outcomes for
Formula: an experiment also known as
experimental outcomes and sample space
outcomes
 Probability is a measure of the chance
Percentiles, Quartiles that an experimental outcome will occur
 The first quartile Q1 is the 25th percentile when an experiment is carried out
 The second quartile (or median) is the 50th o If E is a sample space outcome,
percentile then P(E) denotes the probability
 The third quartile Q3 is the 75th percentile that E will occur and:
 The interquartile range IQR is Q3 - Q1 Conditions:
o 0  P(E)  1 such that:
Formula: o If E can never occur, then P(E) = 0
o If E is certain to occur, then P(E) =
1
o The probabilities of all the sample
space outcomes must sum to 1
Covariance
 A positive covariance indicates a positive Sample Space
linear relationship between x and y  Sample Space: The set of all possible
o As x increases, y increases experimental outcomes
 A negative covariance indicates a negative  Sample Space Outcomes: The
linear relationship between x and y experimental outcomes in the sample
o As x increases, y decreases space
 Event: A set of sample space outcomes
Formula:  Probability: The probability of an event is
the sum of the probabilities of the sample
space outcomes that correspond to the
event

Correlation Coefficient: a measure of the Probability Rules

strength of the relationship that does not depend  Addition Rule
on the magnitude of the data o If A and B are mutually exclusive,
then the probability that A or B will
Formula: occur is P(AB) = P(A) + P(B)

o If A and B are not mutually

exclusive:
Weighted Means: calculated by multiplying the P(AB) = P(A) + P(B) – P(A∩B),
weight or probability associated with a particular where P(A∩B) is the joint
event or outcome. probability of A and B both
occurring together
Formula:
 Conditional Probability
o The probability of an event A, given
that the event B has occurred, is
Example: called the conditional probability of
A given B. Denoted as P(A|B)
Weight (%) Grade (%) o Further, P(A|B) = P(A∩B) / P(B);
Q1 10 70 P(B) ≠ 0
Q2 10 65 o Likewise, P(B|A) = P(A∩B) / P(A)
Q3 30 70
Q4 50 85  Multiplication Rule
Weighted Mean 77% o Given any two events, A and B
P(A∩B) = P(A)P(B|A)
= P(B)P(A|B)

Geometric Mean:
 For rates of return of an investment, use Chapter 5
the geometric mean  Random variable: a variable that
 Suppose the rates of return are R1, R2, …, assumes numerical values determined by
Rn for periods 1, 2, …, n the outcome of an experiment
 o Discrete random variable:  Properties of f(x): f(x) is a continuous
Possible values can be counted or function such that
listed o f(x) ≥ 0 for all x
Examples o The total area under the curve of
 The number of defective units in a batch f(x) is equal to 1
of 20  Essential point: An area under a
 A rating on a scale of 1 to 5 continuous probability distribution is a
o Continuous random variable: probability
May assume any numerical value in
one or more intervals Uniform Distribution: all values between a
Examples minimum and maximum value have the same
 The waiting time for a credit card probability.
authorization Formula:
 The interest rate charged on a business
loan

Discrete Probability Distribution

 Probability distribution: is a table,
graph or formula that gives the probability Normal Probability Distribution
associated with each possible value that  A straight line indicates a normal
the variable can assume. distribution
o Denote the values of the random
variable by x and the value’s Formula:
associated probability by p(x)

 Binomial Experiments
1. Experiment consists of n identical trials
2. Each trial results in either “success” or
“failure” Finding Normal Probabilities
3. Probability of success, p, is constant from 1. Formulate the problem in terms of x
trial to trial values
4. Trials are independent 2. Calculate the corresponding z values, and
o If x is the total number of restate the problem in terms of these z
successes in n trials of a binomial values
experiment, then x is a binomial 3. Find the required areas under the
random variable standard normal curve by using the table

 Poisson Distribution: finding the Exponential Distribution

probability of an independent event that is
occurring in a fixed interval of time and Chapter 7
has a constant mean rate.  Random Sample: every set of n
Formula: elements in the population has the same
chance of being selected
 Probability sampling: is a sampling
Example: where we know the chance that each
In a café, the costumers arrives at a mean rate of element in the population will be included
2 per min. Find the probability of arrival of 5 in 1 in the sample
min. o Allows making statistical inferences
 Convenience sampling: is where we
select elements because they are easy or
convenient to sample
 Hypergeometric Distribution:  Voluntary response sampling: is where
determine the probability of certain participants self-select
number of success in a series of draws  Judgment sampling: is where a
made without replacements from a fixed knowledgeable person selects population
population. elements
Formula:  Sampling distribution of the sample
mean: is the probability distribution of
the population of the sample means
obtainable from all possible samples of
size n from a population of size N
 Sampling distribution of the sample
proportion: distribution of all possible
sample proportions
Chapter 6 o Formula:

Continuous Probability Distribution

 A continuous random variable may
assume any numerical value in one or
more intervals Central Limit Theorem: non-normal
o Car mileage population
o Temperature o The larger n, the better the
Properties of Continuous Probability Distribution approximation