Designing a study

Question 1

Given a bivariate data set, what would indicate no linear relationship between X and Y?

Answer 1

When r approaches 0

Question 2

Define an experiment

Answer 2

The systematic procedure carried out under controlled conditions in order to discover an unknown effect.

Question 3

What defines a binomial distribution data set?

Answer 3

• a fixed number of trials
• each trial has only two outcomes (P/F)
• probability of falling in range is always the same
• outcomes are independent of each other

Question 4

Define a random variable

Answer 4

A variable whose possible values are numerical outcomes of a random phenomenon. There are two types of random variables- discrete and continuous.

Question 5

Describe the difference between discrete random variable and continuous random variable

Answer 5

D: variable which can only take a countable number of variables.
C: variable takes on values within an interval or it has so many possible values that they might as well be considered continuous.

Question 6

Define confidence interval

Answer 6

a range of values so defined that there is a specified probability that the value lies within it.

Question 7

Define confidence level

Answer 7

Corresponds to the percentage of the time the result would be correct if numerous random samples were taken.

Question 8

Define population variability

Answer 8

Variability (also called spread or dispersion) refers to how spread out a set of data is. Variability is measured in terms of standard errors/ deviations.

Question 9

How do you set up H0 and Ha in a hypothesis test?

Answer 9

The null hypothesis is set up so that H0 is true unless some data statistics demonstrate otherwise. If we have sufficient evidence against H0 (p-value<0.05), it can be rejected in favour of Ha.

Question 10

How would you define a “strong linear relationship” between X and Y?

Answer 10

-0.6 > r
0.6 < r

Question 11

What is blocking?

Answer 11

(stratification) separates nuisance variables in our experiment which contribute to undesired variation.

Question 12

What is confounding?

Answer 12

Confounding factors are variables that influence both the inputs and the outputs causing false association.

Question 13

What are the positives of factorial experiments?

Answer 13

(+): Evaluates interactive effects
(+): All of the results are used to evaluate the effects
(+): The results are applicable over a wide range of experimental conditions.

Question 14

What are the negative of factorial experiments?

Answer 14

(-): The required sample size may be large
(-): Explanation/ interpretation of some interactions may be complex

Question 15

What are interaction graphs?

Answer 15

Interaction graphs occur when the effect of one process parameter depends on the level of the other process parameters.

Question 16

Define accuracy

Answer 16

the maximum amount by which the result differs from the true value. The nearness of the measured values to its true value often expressed as a percentage.

Question 17

Define precision

Answer 17

consistent reproducibility of a measurement

Question 18

Define drift

Answer 18

variation caused in the output of an instrument, not caused by any change in the input.

Question 19

Define zero stability

Answer 19

the ability of an instrument to return to the zero reading after the input signal comes back to the zero value and other variations have been eliminated.

Question 20

In a designed experiment, what is the definition of “responses”?

Answer 20

The measured outcomes of an experiment.

Question 21

In an experiment, what are the “uncontrollable factors”?

Answer 21

Factors that affect the results and are difficult to control

Question 22

Why should good experiments be always comparative?

Answer 22

Without monitoring the “no treatment” case for comparison, you have no basis for quantifying the effect of the variables you are investigating.

Question 23

Why should you always replicate your experiments?

Answer 23

To characterise the variability
within measurements

Question 24

In the design of experiments ambit, what is “randomization”?

Answer 24

Deliberate process designed to eliminate potential biases from the conclusions.

Question 25

The basic principles of the design of experiments include:

Answer 25

Replication, randomization, and stratification (or blocking).

Question 26

In metrology, what is the definition of static calibration?

Answer 26

The variable involved remain constant while calibrating a given instrument

Question 27

What is the definition of “random errors”?

Answer 27

A random error provides a measure of random deviations when measurements of a physical quantity are carried out repeatedly.