Statistics Final Exam Review

Front 1

There are four main components of statistics; what does the first component, or the Design component, refer to?

Back 1

A plan on how to OBTAIN DATA needed to answer the question.

Front 2

What is the Description Component refer to?

Back 2

Using tables, graphs, etc to summarize and analyze the data.

Front 3

What is the third component (probability) refer to?

Back 3

Determine how sample differs from the population.

Front 4

What is the fourth and final component refer to (also called the inference)?

Back 4

Make decisions and predictions.

Front 5

What is the difference between Descriptive Statistics and Inferential Statistics?

Back 5

Descriptive Statistics uses graphs and numerical summaries; used with samples or populations. Inferential we use data from SAMPLES to make predictions about populations

Front 6

When is Inferential Statistics not necessary?

Back 6

When we already have information about a population and there is no need to make a prediction (i.e. mean age of a population).

Front 7

A survey of all graduating students was taken and numerical summaries are taken of average starting salary and the percentage of students earning more than $30k/year.

1.) Are these Descriptive or Inferential?

2.) And, are the numerical summaries statistics or parameters?

Back 7

1.) Descriptive - they summarize data from a population

2.) Parameter - they refer to a population

Front 8

What is a Response Variable?

Back 8

The variable we are interested in measuring (dependent variable; y-axis).

Front 9

What is a Component?

Back 9

What you are actually simulating through the use of a random device.

Front 10

True or False?

In a particular study, you could use descriptive statistics or inferential statistics, but you would rarely use both.

Back 10

False. We often want to describe the sample and make inferences (predictions) about the population.

Front 11

Inferential Statistics are often used to draw a conclusion. What are some methods for drawing measuring the reliability and conclusions about a population?

Back 11

Things like confidence intervals and hypotheses.

Front 12

What is a categorical variable?

Back 12

A qualitative, non-numerical variable with different categories. Ex: types of pets.

Front 13

What is a Quantitative variable?

Back 13

numerical, measurable variable. Ex: GPA

Front 14

Which of Qualitative and Quantitative variables are further categorized into discrete and continuous?

Back 14

Quantitative.

Front 15

What is a discrete, quantitative variable?

Back 15

values that for a set of separate numbers. Ex: shoe size, rolling a die, or something that can be counted. Often whole numbers. Something we count.

Front 16

What is a continuous, quantitative variable?

Back 16

values form a continuum of values over a real number line. Ex: height, weight, time, age. Something we can measure.

Front 17

What kind of charts are used for Categorical data?

Back 17

Pie, Bar, Pareto (specialized bar; ordered in relative frequency)

Front 18

When would a data set have the same median and mode?

Back 18

A symmetric distribution.

Front 19

If the mean is greater than the median the data are likely skewed which way?

Back 19

Right. Mean>Median = right-skewed

Front 20

If the median is larger than the mean the data are likely skewed...

Back 20

Left. Median>Mean = left-skewed

Front 21

What percent of a data set is said to be one standard deviation from the mean (in a normal distribution)? Two standard deviations? Three?

Back 21

68%

95%

Nearly all for three standard deviations.

Front 22

A graph of the 5 number summary is just a...?

Back 22

Boxplot - min, Q1, median, Q3, max

Front 23

Why is the standard deviation (s) preferred over the range?

Back 23

the range is more affected by an outlier and the standard deviation uses all of the data.

Front 24

Why is the IQR sometimes preferred to (s)?

Back 24

IQR is not affected by the outlier

Front 25

What is the advantage of (s) over the IQR?

Back 25

the standard deviation takes into account all observations.

Front 26

How do you describe data distributions?

Back 26

SOCS - shape, outliers, central tendency, and spread.

Front 27

What does shape refer to in SOCS?

Back 27

Modality (# of peak) - uni, bi, multimodal.

Skewness & Symmetry - left (negative), right (positive), symmetric (bell-shaped or normal)

Front 28

What does Outliers refer to in SOCS?

Back 28

Unusual values. A data value with a z-score greater than zero.

Front 29

What does the Central Tendency of SOCS refer to?

Back 29

Mean for symmetric distributions.

Median for skewed distributions

Front 30

The spread of SOCS refers to?

Back 30

Standard deviation (mean)

IQR and Range (median)

Front 31

What is an explanatory variable?

Back 31

The varibale being manipulated (x-axis; independent variable). Explanatory explains the response variable.

Front 32

What is a lurking variable?

Back 32

unobserved variable that influences the association between explanatory and response variables.

Front 33

True or False? An observational study can establish causation.

Back 33

False. An observational study can reveal association or correlation.

Front 34

How is Margin of Error calculated?

Back 34

= 1/[sqrt(n)] x 100%

Front 35

What is the significance of the Margin of Error? Why is it used?

Back 35

Used to give a range of plausible values for the population parameter.

Front 36

How do you find the range of plausible values for a given parameter?

Back 36

= sample statistic (+/-) margin of error

ex: 78%(+/-)2.9% = (75.1%, 80.9%)

Front 37

What is sampling bias?

Back 37

Sampling method that tends to obtain non-representative samples.

Front 38

What is under-coverage?

Back 38

Sampling frame does not represent all parts of the population; portion is not sampled or has smaller representation in sample than it does the population.

Front 39

What is over-coverage?

Back 39

Members that are not in the population of interest are included in the sample.

Front 40

What is non-response bias?

Back 40

Sampled subjects cannot be reached or refuse to participate. Or some may not respond to some questions further resulting in missing data.

Front 41

Name three poor ways to sample.

Back 41

Convenience sample (individuals who are easy to sample, which may not rep pop)

Volunteer sample (common type of convenient sample)

Large, non-representative sample

Front 42

What is Cluster Random Sampling?

Back 42

Heterogenous "clusters" that resemble the population; a census is taken of each cluster.

Front 43

What is Stratified Random Sampling?

Back 43

Homogenous groups (to reduce variability); then use simple random sampling to choose members from each strata. Each stratum will be different from the others.