15. Test Yourself

Question 1.

An investigator plans an experiment with a control and treated group, but reduces the numbers in the treated group and increases those in the control group because she fears that the treated animals may experience pain. Is this an example of:

Replacement
No, nothing is replaced
Refinement
Yes because the aim
is to reduce pain
Reduction
No because that increases
the numbers

Question 2
Which is the best way to randomise 12 animals all in the same cage (individually identified)  to three treatment groups A, B and C?

1. Roll a die and assign animal one  to group A if the die shows 1 or 2, to group B if it shows 3 or 4 or to group C if it shows 5 or 6.

2. Assign animal one to group A, the next to Group B and the third to group C, and repeat this four times

3. Assign the first four animals to group A, the next four to group B and the last four to group C

4. Use EXCEL to randomise a column with 4 As, 4 Bs and 4 Cs and assign the animals according to the random sequence

Click

No, it would be difficult to get
exactly 4 animals per treatment
this way
One
No. This might create a bias
because we don't know how numbers
were assigned to mice originally
Two
Three
No, this might create bias. We don't
know how the animals were 
originally numbered
Yes. This could be done in the office
and the result printed out and taken
to the animal house
Four

Question 3

Inbred
strains

Cheaper to buy

Phenotypically more uniform

Genetically more stable

Easier genetic quality control

Most commonly used by toxicologists

Most commonly used by geneticists

Like an immortal clone of identical individuals

Large strain differences

May change following selective breeding

Established and well used strain nomenclature

Click
Click
Click
Click
Click
Click
Click
Click
Click
Click
Outbred
stocks Both
Yes, correct
Click
Click
Click Click
Yes, correct
Yes, correct
Click Click
Click Click
Yes, correct
Click
Click
Yes, correct
Click Click
Yes, correct
Click Click
Yes, correct
Click
Click
Yes, correct
Click
Click
Yes, correct
Click Click
Yes, correct

Question 4.
In the Power Analysis method of determining sample size, if you make the changes noted below, how would it alter the group (sample) size needed?

Power.   Increase it from 80% to 90%

Significance level.  Increase it from 0.05 to 0.10

Standard deviation. Decreased by choosing uniform animals

Alternative hypothesis.  Make it one instead of two sided

Effect size. Increase it by increasing the dose

Required sample size would be:

Increased
Decreased
Yes
Increased
Decreased
Yes
Increased
Decreased
Yes
Decreased
Increased
Yes
Decreased
Increased
Yes

Question 5
An investigator wants to find out whether a drug affects activity of mice when given access to a running wheel over a period of a week. The drug will be given by injection.

She plans to use 2 strains of mice, 3 dose levels and both sexes in a factorial design with 6 mice per group.

Using the Resource Equation method Calculate E. Is this number of animals:

 

Too few
About right
Too many
No
No
Yes
Feedback
2 (strains) x 3 (dose levels) x 2 (sexes)
=12 groups x 6 = 72 mice
Therefore E=72-12 =60
E should be between 10 and 20 
So she is using too many animals
Home
Top