random

Question 1

If BMI = weight/(height)2. Name the dependent and the independent variable in this equation.

Answer 1

BMI is the dependent variable as it depends on the independent variables weight and height

Question 2

What does it mean if correlation is -1, 0 or 1.

Answer 2

-1 means a perfect negative linear relationship 0 means no linear relationship 0.5 means a moderate positive linear relationship

Question 3

If the output for a hypothesis test for correlation is the one below. What would be your
conclusion?
Pearson’s product-moment correlation
data: Var_1 and Var_2
t = 7.6064, df = 2, p-value = 0.01685
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.4004041 0.9996629
sample estimates:
cor
0.9831516

Answer 3

The result is statistically significant. Therefore, there is evidence that the true correlation is not equal to
zero and so we reject the null hypothesis.

Question 4

What is a QQ-plot used for?

Answer 4

To check normality

Question 5

Can we have x2 in a linear model?

Answer 5

yes

Question 6

Can we have a categorical variable as a predictor in a linear model?

Answer 6

yes

Question 7

What is the null hypothesis for the regression parameter when we fit a simple linear
regression?

Answer 7

That the coefficient is equal to zero

Question 8

What is a multivariable linear regression?

Answer 8

Regression with multiple independent (predictor) variables in one model.

Question 9

Experimentation begins by

Answer 9

formulating a number of research hypotheses.

Question 10

A significance level of 0.05 means that if you ran many statistical tests you would expect the tests to show there was a significant difference even when there was no real difference about 1 time out of every

Answer 10

20

Question 11

The significance level

Answer 11

is a measure of how strong the sample evidence must be before determining the results are statistically significant.

Question 12

if alpha is 0.05, your analysis has a what chance of producing a significant result when the null hypothesis is correct.

Answer 12

5% chance

Question 13

Why would an experimenter choose to use a randomized block design?

(A) To test the effect of a blocking variable on a dependent variable.
(B) To assess the interaction between a blocking variable and an independent variable.
(C) To control unwanted effects of a suspected nuisance variable.
(D) None of the above.
(E) All of the above.

Answer 13

c

Question 14

Which, if any, of the following attributes does not describe a good blocking variable?

(A) It is included as a factor in the experiment.
(B) It is not of primary interest to the experimenter.
(C) It affects the dependent variable.
(D) It affects the independent variable.
(E) All of the attributes describe a good blocking variable.

Answer 14

d
A good blocking variable is not related to an independent variable. When the blocking variable and treatment variable are related, tests of treatment effects may be biased.

Question 15

You are comparing the seed weights of four varieties of winter wheat, and you have weighed 50 seeds of each variety.
What null hypothesis should you test, and which test should you use?

Answer 15

the null hypothesis is that there is no significant difference between the mean weight of each of the varieties.

one-way ANOVA.

Question 16

You are investigating how the swimming speed of fish depends on their length, and you have measured both for 30 fish.
What is your null hypothesis, and which statistical test should you use to test it?

Answer 16

no significant association between
them
regression analysis.

Question 17

You are investigating the relationship between the weight and social rank of domestic hens, and you have observed 34 birds.

Which test should you use?

Answer 17

null hypothesis is that there is no relationship between them

measurements are affected by each other – weight can affect rank, and rank can also affect weight

Since social rank is by definition rank data, you should choose the nonparametric option

= rank correlation

Question 18

clinician wants to find out if there is any link between energy intake (in calories) and heart rate in old people.
She collects data on both of them from 150 volunteers.
What is her null hypothesis? Which statistical test should she choose

Answer 18

Her null hypothesis is that there is no significant relationship between energy intake
and heart rate.
The statistical test to use is correlation. She is looking at measurements, looking for an association between two sets of measurements, and neither variable is clearly independent of the other.

Question 19

An ecologist collects data about the numbers of individuals that belong to five species of
crow feeding in three different habitats: farmland, woodland and mountain.
He wants to determine whether different crows are distributed non-randomly in different habitats.
What is his null hypothesis, and how will he analyse his data

Answer 19

His null hypothesis is that there is no significant association between particular habitats
and species.

The statistical test to use is the x2 test for association.

He is looking at frequencies in different categories, and looking for an association between two types
of category (species and habitat).

Question 20

A doctor wants to find out if there is any difference in insulin levels between three races
of people: Afro-Caribbean, Asian and Caucasian. He collects data on insulin levels from 30 people of each race.
What is the null hypothesis, and which statistical test should he use to answer his question?

Answer 20

His null hypothesis is that there is no significant difference between the insulin levels
of the three races.
The statistical test to use is one-way ANOVA. He has taken measurements and is looking for differences between groups; there are more than two groups;
the measurements are not matched; and he is investigating just one factor (race).

Question 21

An ecologist wants to find out whether the levels of pesticide residue found in kestrels differ at different times of the year.
She measures pesticide levels in ten birds, repeating
the measurements on each bird every 2 months.
What is the null hypothesis, and which statistical test should she perform

Answer 21

His null hypothesis is that there is no significant difference in pesticide levels between the birds at different times of year.
The statistical test to use is repeated measures ANOVA.
He has taken measurements and is looking for differences between different sets of measurements; there are more than two sets of measurements and these are related, since they are taken on the same birds. Finally, pesticide levels are continuous
variables which are likely to be normally distributed

Question 22

A new medication to lower blood pressure is being tested in field trials.
Forty patients were tested before and after taking the drug.
Which test should the clinicians use to best determine whether it is having an effect?

Answer 22

Their null hypothesis is that there is no significant difference between blood pressure before and after taking the drug.
The statistical test to use is the paired t test.

Question 23

what is bootstrapping

Answer 23

statistical procedure that resamples a single dataset to create many simulated samples

Question 24

central assumption for bootstrapping

Answer 24

the orginal data accurately represents the actual population

Question 25

advantages of bootstrapping

Answer 25

doesnt make assumptions about the data
use for a wide range of distrubtions

Question 26

when is bootstraping not appropriate

Answer 26

when population variance is infinite
values are discontinuous at the median

Question 26

pseudoreplication

Answer 26

observations are not statisically independent but treated as if they are

Question 27

how can use of ANOVA lead to pseudoreplication

Answer 27

use a on-way ANOVA for paired comparisons
this can only be used in a completely random design