Test 3 review

Question 1

Graphs for categorical data

Answer 1

Bar Chart (best),
Pie Chart (okay, but no scale),
Pictogram (often
misleading)

Question 2

p vs. p-hat

Answer 2

p is the population parameter, p-hat is the sample statistic

Question 3

How to compute p-hat

Answer 3

x / n

Question 4

standard deviation of p-hat

Answer 4

σ = sqrt(p(1-p) / n)

Question 5

standard error of p-hat

Answer 5

sqrt(phat(1-phat) / n)

Question 6

margin of error for p-hat

Answer 6

z*sqrt(phat-hat(1-phat) / n)

Question 7

One-Sample z Confidence interval for p conditions for use?

Answer 7

• Large sample size or normality
  FOR LARGE SAMPLE SIZE:
  np > 10, (1-p) > 10
• Randomization

Question 8

Sampling distribution of p-hat?

Answer 8

All sample proportions
From all possible random samples
Of the same size
Taken from a population

Question 9

Definition of confidence interval

Answer 9

Range of reasonable values which we predict with certain accuracy the parameter of interest falls under

Question 10

Computing confidence interval for one sample z confidence interval for proportions

Answer 10

p̂ ± z*√(𝑝̂(1−𝑝̂) / 𝑛)

Question 11

Interpretation of confidence interval

Answer 11

We are ___% confident that (parameter of interest) is contained in the interval (lower bound, upper bound)

Question 12

Minimum sample size for categorical data

Answer 12

np > 10, n(1-p) > 10

Question 13

How conditions are different for confidence intervals and hypothesis tests for p

Answer 13

For confidence intervals you always use p-hat because there is no null hypothesis
For hypothesis tests, you always assume p-null is true, and thus you use it for np rules and test statistics

Question 14

One-sample z Test for proportions conditions for use?

Answer 14

• Large sample size or normality
  FOR LARGE SAMPLE SIZE:
  np > 10, (1-p) > 10
• Randomization

Question 15

One-sample z Test for proportions hypotheses?

Answer 15

H0 p = #, Ha p >, <, =/= #

Question 16

One-sample z Test for proportions compute test statistic?

Answer 16

z = (p-hat - p) / sqrt((p(1-p)) / n)

Question 17

One-sample z Test for proportions write conclusion?

Answer 17

We reject (fail to reject) the null hypothesis and conclude we do (not) have sufficient evidence to conclude that the null hypothesis is (not) true.

Question 18

Role/Type classifications for relationships

Answer 18

C -> C (proportions)
Q -> C (We do not discuss)
C -> Q (means) (boxplots)
Q -> Q (means)

Question 19

How to tell if something is Matched Pairs?

Answer 19

Either both treatments are applied to each individual or two very similar individuals are given the same treatment

Question 20

Matched Pairs t Confidence Interval?

Answer 20

Used to estimate the population parameter for means

Question 21

Matched pairs conditions?

Answer 21

• Randomization
• Roughly normal distributions or n > 30

Question 22

Matched pairs hypothesis wording?

Answer 22

The mean difference between all participants

Question 23

Matched Pairs t Confidence Interval computation?

Answer 23

average of differences +- t * (s / sqrt(n))
degrees of freedom: number of PAIRS

Question 24

Matched Pairs t test computation?

Answer 24

dbar / (s / sqrt(n))

Question 25

Two sample t confidence interval / test conditions?

Answer 25

• Randomization
• Roughly normal distributions or n > 30
• big sd / small sd < 2 (equal standard deviations)

Question 26

When to use Anova?

Answer 26

Several groups of means

Question 27

When to use Chi-squared?

Answer 27

Several groups of proportions

Question 28

Two sample t tests for means degrees of freedom?

Answer 28

n1 + n2 - 2

Question 29

Two sample t tests for means write hypotheses?

Answer 29

μ1 = μ2, μ1 <, >, =/= μ2
OR
μ1 - μ2 = 0, μ1 <, >, =/= μ2

Question 30

Two sample t tests formulas?

Answer 30

sd * sqrt(1 / n1 + 1 / n2)

(x-bar1 - x-bar2) / sd * sqrt(1 / n1 + 1 / n2)

(x-bar1 - x-bar2) +- t * s * sqrt(1 / n1 + 1 / n2)

Question 31

Chi-Square when to use?

Answer 31

Multiple proportions

Question 32

Anova Hypotheses?

Answer 32

μ1 = μ2 = μ3 = μ4
vs
At least one differs from the rest