Qmb 3200 Test 1

Front 1

The science of collecting, analyzing, & interpreting data

Back 1

Statistics

Front 2

A data "detective"

Back 2

Statistician

Front 3

Can be quantitative (numerical) or qualitative (categorical) information.

Back 3

Data

Front 4

Measured outcome whose data values vary

Back 4

Variable

Front 5

The object upon which measurements are taken

Back 5

Experimental unit

Front 6

Five elements common to inferential statistical problem:

Back 6

1. Population _ a set of exp. units that you are interested in.

2. Variable - quantitative or qualitative

3. Sample _ a subset of the population.

4. Inference _ statement about pop based on info in sample.

5. Reliability _ measure of how good inference is; level of confidence.

Front 7

Rare data points; have a small probability of occurrence.

Back 7

Outliers

Front 8

95% Confidence Interval Practical Interpretation

Back 8

“We are 95% confident that the true mean (variable) for all (experimental units) fallsbetween (lower bound) and (upper bound) (units of measurement).”

Front 9

Explain (theoretically) the meaning of the phrase "95% Confident".

Back 9

In repeated sampling, 95% of all similarly constructed confidence intervals will capture the true population mean.

Front 10

What is a type I error?

Back 10

Reject Ho when Ho is true. Alpha level is the probability of a Type I error.

Front 11

What is a type II error?

Back 11

Fail to reject Ho when Ho is false

Front 12

Measure of central tendency

Back 12

Mean.

The mean is an average.

Front 13

Measure of variation or spread

Back 13

Standard deviation

Front 14

When is a data point considered unusual?

Back 14

When its z-score is great than 3 in absolute value.

Front 15

(T/F)A population is a subset of data selected from a sample.

Back 15

False. A sample is a subset of data selected from a population.

Front 16

In a sample of n=100 college business students, the mean number of course credits taken this semester is x-bar = 17.5. Give a practical interpretation of the mean.

Back 16

The average number of credits taken this semester by the 100 students is 17.5 credits.

Front 17

In a sample of n=100 female CEOs, the mean age = 35 years and the standard deviation = 5 years. Give an interval that is likely to contain about 95% of the sampled female CEOs' ages.

Back 17

(25, 45)

Front 18

What are the 3 properties of the sampling distribution of x-bar?

Back 18

1. For large sample size, n, the theoretical distribution of sample means is approximately normal.

2. The mean (average) of the theoretical distribution of sample means is equal to the true mean of the population, μ.

3. The standard deviation of the theoretical distribution of sample means is smaller than the true standard deviation of the population, σ.

Front 19

Interpret a 95% confidence interval for the true mean first-year salary of all law students who pass the Florida Bar exam is ($75,000, $95,000).

Back 19

We are 95% "confident" that the true mean first-year salary of all law students who pass the Bar is between $75,000 and $95,000.

Front 20

Know how to identify a suspect outlier.

Back 20

(2 < z < 3)

Front 21

Know how to identify an outlier.

Back 21

(z >3)

Front 22

Interpret the Sample mean (x-bar):

Back 22

“The average (QN variable) for the (#) (experimental units) sampled is (x-bar) (Units Of Measurement).”

Front 23

Sample Std. Deviation calculations and interpretation.

Back 23

Step 1: Do Calculations: x-bar + and – 2s

Step 2: Interpret: “About 95% of the (variable) for the (#)(experimental units) sampled falls between (x-bar - 2s) and(x-bar + 2s) (units of measurement).”

Front 24

The ________ promises a normal distribution for sample mean for large samples (n ≥ 30)

Back 24

Central Limit Theorem

Front 25

plan for selecting the data

Back 25

Design

Front 26

object upon which variables are measured

Back 26

Experimental unit

Front 27

Response (dependent) variable:

Back 27

y (We want to compare means of the y's)—QN, e.g., GPA

Front 28

Factors (independent variables):

Back 28

x’s (variables related to y) – usually QL, e.g., Gender

Front 29

Factor levels :

Back 29

Pre-selected values of a factor (e.g., Male & Female)

Front 30

Treatments :

Back 30

Combinations of factor levels (form the groups to be compared)

Front 31

ANOVA F test assumptions

Back 31

(1) Distribution of y’s for each treatment is normally distributed

(2) Var(y) for each treatment is the same (equal variances)

Front 32

Use the Tukey technique to Construct each confidence interval @ adjusted α level to achieve desired overall Type I error rate Overall Type I error called the ___________.

Back 32

experiment-wise error rate

Front 33

Chi square test assumption

Back 33

χ 2 approx is good as long as n is large (i.e., the expected # for each cell is > or = 5).

Front 34

Null and alternative hypotheses for a chi square test for independence.

Back 34

Ho: QL variable 1 and QL variable 2 are Independent

Ha: QL variable 1 and QL variable 2 are Dependent.

Front 35

Confidence interval for a population mean assumptions:

Back 35

n large- no assumptions (CLT)

n small - assume population is normally distributed.

Front 36

Hypothesis test for a mean assumptions

Back 36

n large- no assumptions (CLT)

n small - assume population is normally distributed.

Front 37

ANOVA null and alternative hypotheses

Back 37

Ho: mean1=mean2=mean3.....

Ha: At lease 2 means differ

Front 38

Tukey method assumptions

Back 38

1. Dependents variable is normal for all treatments.

2. The treatment Variances are equal.

Front 39

What test is used for 1QL-2 levels?

Back 39

CI/test: p=P(s)

Front 40

What test is used for 1QL-3+ levels?

Back 40

multinomial chi squared test

Ho: Proportions are equal

Front 41

What test is used for 2 QL's?

Back 41

Chi squared test of independence.

Ho: QL1 and QL2 are independent.