9 - Sampling Issues

Question 1

Population

Answer 1

Entire collection of all observations of interest to the researcher

Question 2

Parameter

Answer 2

Descriptive measure of characteristic of the entire population of all observations of that characteristic

Question 3

Statistic

Answer 3

Any descriptive measure of a sample and serves as an estimate of the parameter

Question 4

Random sampling

Answer 4

All the members of the population have an equal change of being selected

Question 5

Census

Answer 5

Complète énumération of the population

Cons

• practicality
• cost in money and manpower
• inaccessibility
• varying population

Pros

• small and identifiable population
• sampling might eliminate important cases from the study
• credibility requires consideration of all members of the pop

Question 6

Sampling error

Answer 6

Difference between the unknown population parameter and the sample statistic.

• random error: choose atypical elements unknowingly
• sampling bias: tendent to favor certain characteristics

Question 7

Standard error form the mean

Answer 7

SE= SD/sqrd(n)

Question 8

Central limit theorem

Answer 8

As n increases n>30, the sampling distribution of same-sized means approaches a normal distribution with the sampling distribution mean = to the population mean.

Question 9

Standard error of the proportion

Answer 9

SE=sqrd(pq/n)

Question 10

Sampling process

Answer 10

Balance the needs of gathering information efficiently interns of cost time and numbers with accurate generalisation

1- defining population: unambiguous + differentiating
2- identifying sampling frame: full list of target pop
3- chose sample size
4- selecting sampling method

Question 11

Probability sampling methods

Answer 11

• random sampling
• systematic sampling (nth pb if periodic cycle exists)
• stratified sampling (take proportions of each strata, reduce error and increase precision without changing n + provide info on each strata)
• cluster sampling (one -> multistage cluster sampling)

Question 12

Non probability sampling methods

Answer 12

Sample that is not selected by chance. Quick and inexpensive but do not allow generalisations

• opportunity sampling: convenient
• judgement sampling: identified by expert
• quota sampling: freely selected but predefined proportion (interviewers tend to approach people similar to themselves -> under representation)
• purposive sampling: specific group with specific characteristic
• snowball/referral sampling: initial contract provide futher contacts