chapter 2_sampling_ilearn.pdf

8/14/2019 CHAPTER 2_Sampling_ilearn.pdf

1/31

CHAPTER 2

SAMPLING AND DATACOLLECTION METHODS

PREPARED BY SANIZAH AHMAD

[email protected]


2/31

LEARNING OUTCOMES

Explain the different types of samplingmethods

Apply the different sampling methods Explain different methods of collecting data

and the suitability to their tasks

Design questionnaires

[email protected] 2


3/31

Important statistical terms

Population:a set which includes all

measurements of interestto the researcher

(The collection of allresponses, measurements, or

counts that are of interest)CENSUS

Sample:A subset of the populationSAMPLE SURVEY

[email protected]


4/31

Process of selecting sample from population The sample must be selected in such a way so that it will

accurately represent its population

Sampling technique scientific method of selecting sample

from population (must be random and represent population)

Sampling Unit individuals or items to be sampled

Ex. Student, person who uses credit card

Sampling frame - LIST of individuals or items from whichthe samples can be obtained (list of sampling units).

Ex. Telephone directory, student list, customer list of creditcard users

WHAT IS SAMPLING?

[email protected] 4


5/31

Why sampling?

Get information about large populations when itsimpossible to study the whole population

Less costs

Less field time

Eliminate any BIAS

More accuracy i.e. Can Do a Better Job of Data

Collection

Once a sampling frame has been established, you can

choose a SAMPLING TECHNIQUE

[email protected]


6/31

Types of Sampling TechniquesNON-

PROBABILITY

SAMPLING

Convenience

sampling

Judgementalsampling

Snowballsampling

Quotasampling

PROBABILITYSAMPLING

Simplerandom

sampling

Systematicsampling

Clustersampling

Stratifiedrandom

sampling6


7/31

Non probability sampling

The selection of the items/individuals withouttheir probabilities of selection

Used when generalization concerning the

population is not required or when samplingframes are difficult to obtain

Advantage Quick, inexpensive and convenient

DisadvantageSample selected not representative of the

population

[email protected]


8/31

Non probability sampling

Convenience samplingpre-testing of questionnaires, gathering ideas andinsights, or forming hypothesis

Judgemental Sampling

selected based on the judgement of researcher

Snowball Sampling

select respondent at random. After interviewed, ask

respondent to identify others who are in the targetpopulation of interest

Quota Sampling

observes the specific characteristics of potential

respondent. [email protected]


9/31

Technique Strength Weakness

Conveniencesampling

Less expensive,

less time, convenient

No need list of pop

Selection bias,

Not representative of the

pop

Judgemental

sampling

Less expensive, less time,

convenient

Bias due to experts

belief may make sampleunrepresentative

Quota sampling Sample can be controlled for

certain characteristics

High bias because

sample units not

independent Time consuming

Snowballsampling

Useful in reaching/locating

rare populations/characteristic

Selection bias maybe in

researchers clasification

of subjects

Time consuming

Strengths and Weaknesses of

Non-Probability sampling

[email protected] 9


10/31

Probability sampling

The items/individuals are selected randomly,based on known probabilities

Random means the item has an equal chance of beingselected (unbias)

Used when a researcher plans to makeinferences about the population

Advantage The sample represent the population

DisadvantageSample selected not representative of the population

[email protected]


11/31

Simple random sampling (SRS)Item/subject is selected from the population in such a way that eachitem have the same chance of being selected as a sample.

How to use simple random sampling: STEP 1: Prepare sampling frame

i.e.: Write everyone's name on a slip of paper or assigned number to

each of the people.

STEP 2: Select sample by using: Lucky draw method Table of random numbers

Calculator random number generator

Notation: N = population sizen = sample size

[email protected] 11


12/31

Simple Random Sampling

List of clients = N

Random sample = n

[email protected]


13/31

Let say we get 2, 5, 8, and

10. Our sample would then

look this:

Suppose you want

to select a sample of 4

peoplefrom a group of 12.

[email protected]


14/31

Table of random numbers

2 0 4 0 2 9 2 7 3 2 1 5 6 3 2 1 4 0

5 8 2 0 3 2 1 5 4 7 8 5 9 6 2 0 2 4

3 6 2 3 3 3 2 5 4 7 8 9 1 2 0 3 2 5

9 8 5 2 6 3 0 1 7 4 2 4 5 0 3 6 8 6

[email protected]


15/31

Simple random sampling

[email protected]


16/31

Systematic Random Sampling

Let N= pop size and n = sample size.

Number units in population from 1 to N.

Decide on the n that you want or need. Let the interval size be k= N/n.

Randomly select a number from 1 to k. Let

the number ber.

Take every kth unit until the sample size isobtained.

(r

+k)

th

, (r

+ 2k)

th,

Procedure:

[email protected]


17/31

SYSTEMATIC SAMPLING

Suppose you want to select a sample

of 4 people from a group of 12

STEPS in using systematic sampling:

1. Find the range k= 12/4 = 3

2. Select first sample, rusing SRS of

every 3rd people. Let say you getnumber 2.

3. Find:

i. 2nd element = 2 + 3 = 5

ii. 3rd element = 2 + (2x3)= 8

iii. 4th element = 8 + 3 = [email protected]


18/31

Systematic Random

Sampling

1 26 51 76

2 27 52 77

3 28 53 78

4 29 54 79

5 30 55 80

6 31 56 817 32 57 82

8 33 58 83

9 34 59 84

10 35 60 85

11 36 61 86

12 37 62 87

13 38 63 8814 39 64 89

15 40 65 90

16 41 66 91

17 42 67 92

18 43 68 93

19 44 69 9420 45 70 95

21 46 71 96

22 47 72 97

23 48 73 98

24 49 74 99

25 50 75 100

N= 100

Want n= 20

N/n = 5

Select a random number from 1-5:

chose 4

Start with #4 and take every 5th unit

The samples are 4, 9, 14, 19, 24,

until the 20th sample

[email protected]


19/31

Stratified Sampling

Example:A company has a total of 360employees in four different

categories:

Managers 36

Drivers 54Administrative Staff 90

Production Staff 180

How many from each

category should be includedin a stratified random sample

of size 20 ?

Solution:

[email protected]


20/31

Stratified Sampling

Divide the population into several mutually

exclusive groups (strata) and randomly sample

from each of these strata

Involves a 2 step process

STEP 1: Divide population into groups called strata

Note: Elements within each stratum should be

homogeneous, whereas the differences between

strata should be heterogeneousSTEP 2:

Select elements from each stratum by a random

procedure, usually SRS

[email protected]


21/31

Stratified Sampling

List of clients

Strata

Malays OthersChinese

N

N1 N2 N3

n1 n2 n3

[email protected]


22/31

Cluster Sampling

The target population is first divided intosubpopulations or clusters.

Then a random sample of clusters is selectedbased on a probability sampling techniquesuch as SRS.

For each selected cluster, all elements areincluded in the sample.

[email protected]


23/31

Cluster Sampling1

2

3

Population

[email protected]


24/31

Strengths and Weaknesses of

Probability sampling

Techniques Strength Weakness

Simple randomsampling

Easy to apply and analyze

Results can be projected on

population

Difficult to obtainsampling frame,expensive, notrecommended fordescriptive research

Systematicsampling

Easier to apply than SRS Decrease the no of

respondents if a certain

pattern is exist (periodic)

Stratified

sampling

Includes all important

subpopulations,precision is improved

Require accurate

information in each stratum

Cluster

sampling

Easy to implement, costeffective and work isreduced

Difficult to assign the

element in the cluster

Not easy to interpret results

[email protected] 24


25/31

Multi-Stage Sampling

Designed to reduce time and cost whenworking with samples from very largepopulations.

Example:

Suppose we need a random sample of 2000residents from the Malaysian population.

How to choose the sample using multi-stage

sampling?

[email protected]


26/31

[email protected]


27/31

DATA COLLECTION METHOD

Data collection

Interview

Face-to-face

Telephone

DirectObservation

Questionnaire

Direct

Indirect

Others (e-mail;video record)

27


28/31

Advantage Disadvantage

Face-to face

interview

A.k.a personal

interview

Interviewer initiate to

get information from

respondent

Allow interviewer to clarify

term to respondent

have high response rate

expensive (cost of

travelling

error inrecording

interviewer bias

use a lot of time

Telephone

interview

the questioned

asked based on

prepared

questionnaire

less expensive than personal

interview

speed of data collection

only short question can

be asked

restricted to respondent

who have telephone

limited duration

Mail questionnaire cheapest

easiest

no interviewer influence

cover wide area

respondent has more time to

answer

low response rate

simple question can be

asked

Direct observation not influenced by others

perception

not effected by the

respondent itself

need a high skilled and

unbiased

Others internet Same as mail 28


29/31

Designing a questionnaire Before you begin drafting your questionnaire, it is important to consider:

Who is the questionnaire for?

What is it intending to find out or measure?

Guidelines in Designing a questionnaire

Design questions to meet the objective of the research.

Questionnaires should be short , simple and easy to understand. Begin with simple and less controversial questions.

Avoid: doubt, confusion, and vagueness.

bias questions. sensitive questions. double barrel question. asking questions that are beyond the respondents' capabilities. questions that involve calculation.


30/31

Questionnaire checklist

i. Objectives of the study

ii. Answers sought from the study

iii. Variables used in the studyiv. Methods of data analysis

Once the above procedures are understoodby the researchers, a proper questionnairecan be designed.


31/31

REFERENCES

1. Laura Lake. Types of Data. resources.jorum.ac.uk

2. http://rchsbowman.wordpress.com/2009/08/16/stati

stics-notes-sampling-techniques-2/

3. http://faculty.elgin.edu/dkernler/statistics/ch01/4-1.html

4. http://www.encyclopedia.com/video/sYRUYJYOpG0-

stratified-sampling.aspx

5. http://www.cimt.plymouth.ac.uk/projects/mepres/book9/bk9i18/bk9_18i3.html

31
http://rchsbowman.wordpress.com/2009/08/16/statistics-notes-sampling-techniques-2/http://rchsbowman.wordpress.com/2009/08/16/statistics-notes-sampling-techniques-2/http://faculty.elgin.edu/dkernler/statistics/ch01/4-1.htmlhttp://faculty.elgin.edu/dkernler/statistics/ch01/4-1.htmlhttp://www.encyclopedia.com/video/sYRUYJYOpG0-stratified-sampling.aspxhttp://www.encyclopedia.com/video/sYRUYJYOpG0-stratified-sampling.aspxhttp://www.encyclopedia.com/video/sYRUYJYOpG0-stratified-sampling.aspxhttp://www.encyclopedia.com/video/sYRUYJYOpG0-stratified-sampling.aspxhttp://www.encyclopedia.com/video/sYRUYJYOpG0-stratified-sampling.aspxhttp://www.encyclopedia.com/video/sYRUYJYOpG0-stratified-sampling.aspxhttp://www.encyclopedia.com/video/sYRUYJYOpG0-stratified-sampling.aspxhttp://faculty.elgin.edu/dkernler/statistics/ch01/4-1.htmlhttp://faculty.elgin.edu/dkernler/statistics/ch01/4-1.htmlhttp://faculty.elgin.edu/dkernler/statistics/ch01/4-1.htmlhttp://rchsbowman.wordpress.com/2009/08/16/statistics-notes-sampling-techniques-2/http://rchsbowman.wordpress.com/2009/08/16/statistics-notes-sampling-techniques-2/http://rchsbowman.wordpress.com/2009/08/16/statistics-notes-sampling-techniques-2/http://rchsbowman.wordpress.com/2009/08/16/statistics-notes-sampling-techniques-2/http://rchsbowman.wordpress.com/2009/08/16/statistics-notes-sampling-techniques-2/http://rchsbowman.wordpress.com/2009/08/16/statistics-notes-sampling-techniques-2/http://rchsbowman.wordpress.com/2009/08/16/statistics-notes-sampling-techniques-2/http://rchsbowman.wordpress.com/2009/08/16/statistics-notes-sampling-techniques-2/http://rchsbowman.wordpress.com/2009/08/16/statistics-notes-sampling-techniques-2/http://rchsbowman.wordpress.com/2009/08/16/statistics-notes-sampling-techniques-2/

chapter 2_sampling_ilearn.pdf

Documents