Estimating Sample Size And Power

Question 1

What are the steps for estimating sample size?

Answer 1

1. State the null hypothesis and a one- or two-sided alternative hypothesis
2. Select the appropriate statistical test based on the type of predictor variable and outcome variable in those hypotheses
3. Choose a reasonable effect size (and variability, if necessary).
4. Set alpha and beta (specify a two-sided alpha unless the alternative hypothesis is clearly on-sided).
5. Use the appropriate table or formula to estimate the sample size

Question 2

Can the t test be used for a study with participants fewer than 30?

Answer 2

No

Question 3

How can the effect size and variability be estimated?

Answer 3

They can be estimated by using previous studies in the literature and consultation with experts.

Question 4

What is a standardized effect size?

Answer 4

It is the unitless quantity that makes it possible to estimate a sample size when an investigator cannot obtain information about the variability of the outcome variable. (effect size divided by the standard deviation of the outcome variable.

Question 5

What is the shortcut formula to estimating sample size using a t test?

Answer 5

When more than 30 subjects will be needed:

Sample size=16/(standardized effect size)^2

Question 6

What is relative risk?

Answer 6

It’s the risk ratio, is the ratio of the event of something occurring in an exposed group to the event of something occurring in a comparison, non-exposed.

Question 7

Is correlation coefficient calculation important in clinical studies outside of behavioral medicine?

Answer 7

No

Question 8

What does one get when we square the correlation coefficient?

Answer 8

Represents the proportion of the spread (variance) in an outcome variable that results from its linear association with a predictor variable, vice versa

Question 9

True or false: Dropouts count in the sample size.

Answer 9

False

Question 10

What is the equation to use when when accounting for lost follow-up participants?

Answer 10

1/[1-C]

C= estimated percentage of participants that will be lost

Question 11

What is one thing that must be done to adjust for confounding variables?

Answer 11

Increase sample size

Question 12

The magnitude of the increase in sample size when handling confounding variables depends on what factors?

Answer 12

• Prevalence of confounder
• the strength of the association between the predictor and the confounder
• the strength of the association between confounder and outcome

Question 13

What is one statistical technique we can use to adjust for confounder and for differences in length of follow-up?

Answer 13

Cox proportional hazards

Question 14

How many ordinal variables must there be to be treated as continuous variables?

Answer 14

six or more and if averaging the values of the variable makes sense

Question 15

True or false: Some tests are more interested in finding out that the null hypothesis is true.

Answer 15

True

Question 16

Equivalence studies have 1._______ power and 2.________ effect size which force it to have a 3._________ sample size.

Answer 16

1. High
2. Small
3. Very large

Question 17

What is an equivalence study?

Answer 17

A study in which the study should show no substantial association between the predictor and the outcome variable; the null hypothesis is correct

Question 18

What is the main issue with equivalence studies?

Answer 18

The loss of usual safeguards. There is no Type I error.

Question 19

What do descriptive studies lack?

Answer 19

They do not have predictor or outcome variables, nor do they compare different groups.

Question 20

What do descriptive studies commonly report?

Answer 20

Confidence intervals. Investigator must specify desired level and width of the confidence interval.

Question 21

If the sample size is fixed or limited, what must the investigator do?

Answer 21

Researcher must work backwards from the fixed sample size. Use sample size equations or work backwards using table.

Question 22

What are the five strategies for minimizing sample size and maximizing power?

Answer 22

1. Use continuous variables
2. Use paired measurements
3. Use more precise variables
4. Use unequal group sizes
5. Use a more common outcome

Question 23

What is an unpaired two-sample t test?

Answer 23

A test that looks at the mean of two groups and assesses the difference between the groups

Question 24

What is a paired two-sample t test?

Answer 24

It is a test that looks at the change in a pair of measurements in two groups separately

Question 25

What is the difference between one- and two-sample t test?

Answer 25

One-sample t test only assesses one group. Two-sample assesses two groups.

Question 26

When the outcome is dichotomous, using a ________ outcome, up to a frequency of .5, is usually one of the best ways to increase power.

Answer 26

More frequent

Question 27

What are ways to make an outcome more common?

Answer 27

• Enroll subjects that are at greater risk of developing outcome
• Extend follow-up period
• Loosen definition of what constitutes an outcome

Question 28

True or false: Studies with rare outcomes require large sample sizes.

Answer 28

True

Question 29

When are pilot studies advisable?

Answer 29

Studies that involve new instruments, measurement methods, or recruitment strategies.

Question 30

What trick can be applied to find out the standard deviation of a group of continuous variables?

Answer 30

S.D. Can be estimated as one-quarter of the difference between high and low ends of the range of values that occur commonly, avoiding outliers.

Question 31

When the mean and standard deviation for continuous or categorical variable are in doubt one can ___________ that variable.

Answer 31

Dichotomize.

By doing so, this strategy avoids having to calculate the S.D. but still we must report proportions of the variable

Question 32

Should the sample size be estimated early or late in a project?

Answer 32

Early when fundamental changes can still be made

Question 33

Does the sample size account for those that need to be enrolled?

Answer 33

No. It accounts for number of subjects with outcome data.