Round 2 Lecture 1

Question 1

How to Determine Sample Size

Answer 1

1) Heuristics ( 100/group, Minimum of 20 to 50 per subgroup, Disproportionate sampling)
2) Budget constraints( Sample size = total budget/cost per subject)
3) Previous Research
4) Formal Statistics

Question 2

precision

Answer 2

the size of the estimating interval

Question 3

confidence

Answer 3

the degree of certainty that the true value falls within the estimating interval

Question 4

formal statistics

Answer 4

trade off btw precision and confidence

Question 5

Confidence interval for a population mean is:

Answer 5

X + Z* (s / sq. rt n)

Question 6

T =

Answer 6

margin of error you are willing to tolerate, i.e.,

zone around sample mean where population mean lies, or half CI width

Question 7

T = (equation)

Answer 7

z * (s / sq. rt. n)

Question 8

Three pieces of information are needed for estimating for a mean sample size

Answer 8

1. Margin of error T (zone around sample mean where population mean lies)
2. Your confidence level
   For 90% confidence, z = 1.645 For 95% confidence, z = 1.96 For 99% confidence, z= 2.575
3. Estimate of the population standard deviation (SD) s

Question 9

Estimating the population SD

Answer 9

Review prior research Conduct a pilot study

Use conservation guesses

Question 10

conservation guess for 5-item rating/Likert scale

Answer 10

use s=1.4

Question 11

conservation guess for 7-item rating/Likert scale

Answer 11

use s=2

Question 12

conservation guess for 9, 10, 11-item rating scale

Answer 12

use s=3

Question 13

conservation guess for responses measured by ratio scales

Answer 13

use s=range/4

Question 14

Three pieces of information are needed for estimating a proportion sample size

Answer 14

1. Margin of error T
2. Confidence level
   For 90% confidence, z = 1.645 For 95% confidence, z = 1.96 For 99% confidence, z= 2.575
3. Estimate of the population proportion p

Question 15

Estimating the Population Proportion p

Answer 15

 Review prior research
 Find secondary data (e.g., US census)  Conduct a pilot study
 Make an informed guess
 Use the most conservative guess: p=0.5

Question 16

Sample Size Required for a Survey with Multiple Questions

Answer 16

• calculate ideal sample size for each question

- go with highest among all questions and the most critical questions

Question 17

total sampling elements (TSE) =

Answer 17

n / { (1- BCI) (1-U) (1-R) (1-I) }

Question 18

BCI =

Answer 18

proportion that are expected to have bad contact

Question 19

U =

Answer 19

proportion that are expected to be unreachable

Question 20

R =

Answer 20

proportion that are expected to refuse to participate

Question 21

I =

Answer 21

proportion that are expected to be ineligible

Question 22

assume that the population is

Answer 22

very large