# 1 Sampling Techniques handout by linzhengnd

VIEWS: 19 PAGES: 5

• pg 1
```									                        Sampling Techniques
THE GOALS OF SAMPLING:

1. To select a small group of participants to _____________________ a larger population;
the small group (the sample) should be selected in a way that reflects the
____________________________________ of people in the population.
2. To do the selection __________________, ____________________, and _________________.

THE BASIC STRATEGY:

1. Obtain a list of every member of the population. Assign a number to each person
that is eligible for your study.

2. Good sampling techniques are objective; that is, the surveyor is not allowed to
choose which particular individuals they want to be part of the sample. Instead, a
machine, table, dice or other method is used to pick the participants.

3. Good sampling techniques are random; that is, no particular person from the
population is more likely or less likely than others to be selected for the study. Every
person has an equal chance of being chosen.

THE KINDS OF SAMPLING TECHNIQUES:

SIMPLE RANDOM SAMPLING

Randomly select people from the population one at a time until you have enough
participants. Make sure that you do not select the same person more than once!

Method 1: ____________________________________                                  Red die
1   2     3    4    5    6
Method 2: ____________________________________                        1 1   2     3    4    5    6
Green die

2 7   8     9   10    11 12
Method 3: ____________________________________
3 13 14 15      16   17   18
Method 4: ____________________________________                        4 19 20 21      22   23   24
5 25 26 27      28   29   30
6 31 32 33      34   35   36
Observation: ____________________________________
SYSTEMATIC RANDOM SAMPLING: Sampling every nth person

(1) Calculate a sampling interval by dividing the total number of people in the
population by the number of people you want to select for your study. Round
your answer to a whole number.
(2) Use a simple random sampling method to select one participant from the first
sampling interval.
(3) Starting from the chosen participant from step (2) above, count off another
sampling interval from there and select that participant. Repeat this step until you
reach the end of the population.

Easy example:                               Realistic example:

28 people in MDM class                      22369 Bolton residents
10 wanted for sample                        250 wanted for sample

sampling interval = ----- =                sampling interval = ------------ =    

1     2 3 4 5 6 7                           Rnd # generator = .022287
8     9 10 11 12 13 14                      First participant =       X       =     
15   16 17 18 19 20 21                      Other participants: _______________________
22   23 24 25 26 27 28                      __________________________________________

CLUSTER RANDOM SAMPLING: Sampling everyone within a random group

THE CLASSIC METHOD:
(1) Calculate the number of clusters by dividing the total number of people in the
population by the number of people you want to select for your study. Round
your answer to a whole number.
(2) Use a simple random sampling method to select one of the clusters.
(3) To find your first participant, multiply the cluster # (from step 2) by the number of
people you wish to sample. Select every participant after that until you have your
sample size.

NOTE: The classic method described above divides the population into perfectly
equal groups (clusters). In real life, the participants are usually already in groups and
these groups can be of different sizes (although not dramatically different). For
example, Peel students are already divided into schools, and some schools have more
students than other schools do.

Example:                                                               SCHOOLS
Roughly 19800 students in 18 schools                        1    2 3 4 5 6
This means an average of 1100 students/school               7    8 9 10 11 12
To get a sample of 2200 students, survey 2 schools          13   14 15 16 17 18
Rnd # generator = .388 Rnd # generator = .605
STRATIFIED RANDOM SAMPLING: Sampling fairly by classifying people by variables

(1) Obtain relevant background data about all members of the population.
(e.g., gender, age, grade level, etc.)
(2) Create separate lists of people (called strata) that have similar backgrounds.
(e.g., females age 15, males age 15, females age 16, males age 16, etc.)
(3) Find the sample size as a percentage of the population.
(e.g., 150 in sample / 800 in population = 18.75% of the pop’n is being sampled)
(4) To determine how many people from each strata to sample, multiply the percent
of the pop’n being surveyed by the number of people in that strata.
(e.g., 240 females age 15 in pop’n X .1875 = 45 females age 15 in sample)
(5) Randomly select participants within each strata to create your sample.
(e.g., use simple random sampling to select 45 females age 15 out of the 240 in the
pop’n)

Example:

SAC wants to hire a band for a coming dance. They want to know what kinds of
music the students prefer. The SAC has the time and effort to sample 15% of the 1340
students in the school.

The enrolment is as follows:

# OF
GRADE                           CALCULATIONS
STUDENTS
9            403

10            326

11            276

12            188

TOTAL

MULTI-STAGE SAMPLING

This method is just a unique combination of sampling techniques from those already
discussed.

Example 1: Randomly select 5 schools from within Peel using a systematic technique,
then you do a cluster sampling of homeroom classes within those schools.

Example 2: Divide a city into neighbourhoods, then do a simple random sample of
residents within each neighbourhood.
BAD SAMPLING TECHNIQUES (BUT EASIER TO DO)
VOLUNTARY RESPONSE SAMPLING

The researcher invites people in the population to participate in the study. Those
people who volunteer become part of the study and the ones who don’t volunteer
are not part of the sample.

Example 1: _________________________________________

Example 2: _________________________________________

Example 3: _________________________________________

Example 4: _________________________________________

CONVENIENCE SAMPLING

The researcher asks/begs people who are readily available to complete their survey.

Example 1: _________________________________________

Example 2: _________________________________________

Example 3: _________________________________________
[represent, various kinds, quickly, cheaply, simply]
[mail-in surveys, phone-in shows, newspaper subscription surveys, website polls]

[find people in the caf, ask friends or family, ask people waiting at the checkout at a
grocery store]

```
To top